THE YEAR IN ANAESTHESIA AND CRITICAL CARE VOLUME 1
THE YEAR IN ANAESTHESIA AND CRITICAL CARE VOLUME 1 EDITED BY
JENNIFER HUNTER, TIM COOK HANSJOACHIM PRIEBE, MICHEL STRUYS
CLINICAL PUBLISHING OXFORD DISTRIBUTED WORLDWIDE BY TAYLOR & FRANCIS LTD BOCA RATON • LONDON • NEW YORK • WASHINGTON, DC Clinical Publishing
An imprint of Atlas Medical Publishing Ltd This edition published in the Taylor & Francis e-Library, 2005. “ To purchase your own copy of this or any of Taylor & Francis or Routledge’s collection of thousands of eBooks please go to http://www.ebookstore.tandf.co.uk/.” Oxford Centre for Innovation Mill Street, Oxford OX2 OJX, UK Tel: +44 1865 811116 Fax: +44 1865 251550 Web: http://www.clinicalpublishing.co.uk/ Distributed by: Taylor & Francis Ltd 6000 Broken Sound Parkway NW, Suite 300 Boca Raton, FL 33487, USA E-mail:
[email protected] Taylor & Francis Ltd 23–25 Blades Court Deodar Road London SW15 2NU, UK E-mail:
[email protected] © Atlas Medical Publishing Ltd 2005 First published 2005
All rights reserved. No part of this publication may be reproduced, stored in a retrieval system, or transmitted, in any form or by any means, without the prior permission in writing of Clinical Publishing or Atlas Medical Publishing Ltd Although every effort has been made to ensure that all owners of copyright material have been acknowledged in this publication, we would be glad to acknowledge in subsequent reprints or editions any omissions brought to our attention A catalogue record for this book is available from the British Library ISBN 0-203-02492-3 Master e-book ISBN
ISBN 1-904392-47-4 (Print Edition) ISSN 1745-9508 The publisher makes no representation, express or implied, that the dosages in this book are correct. Readers must therefore always check the product information and clinical procedures with the most up-to-date published product information and data sheets provided by the manufacturers and the most recent codes of conduct and safety regulations. The authors and the publisher do not accept any liability for any errors in the text or for the misuse or misapplication of material in this work Project Manager: Rosemary Osmond
Contents Contributors
vi
Foreword Peter Simpson
ix
Part I Perioperative care Editor: Hans-Joachim Priebe Editorial: Perioperative care, Hans-Joachim Priebe 1. Perioperative intravenous fluid therapy Joachim Boldt 2. Perioperative blood transfusion therapy Tim Walsh 3. Off-pump coronary artery bypass surgery Charlotte Dempsey, Joseph Arrowsmith 4. Protecting the heart in non-cardiac surgery Simon Howell, Peter Kimpson Part II Clinical pharmacology Editor: Michel Struys Editorial: What is new in anaesthetic pharmacology? Michel Struys 5. New hypnotics Harald Ihmsen 6. ‘New’ analgesics Martin Luginbühl 7. Pharmacogenetics and anaesthesia Richard Smiley 8. Clinical applications of pharmacokinetics and pharmacodynamics Frédérique Servin Part III Monitoring and equipment Editor: Tim Cook Editorial: New technology in anaesthesia and critical care, Tim Cook
1
16 35 58 82 113
120 130 153 169 191
9. Less invasive cardiac output monitoring Gernot Marx 10. Depth of anaesthesia monitoring Hugo Vereecke, Michel Struys 11. New airway equipment Takashi Asai 12. Ultrasound techniques for central venous access and regional anaesthesia Gordon Chapman, Andrew Bodenham Part IV Critical care Editor: Jennifer Hunter Editorial: The Critical Care Outreach conundrum: does the end justify the means? Jane Harper, Jennifer Hunter 13. Choice of resuscitation fluid in the critically ill Simon Finfer, James Cooper, Deborah Cook 14. Glucose control in the critically ill Johan Groeneveld, Emmy Rijnsburger, Marina Honing 15. Non-invasive ventilation Jonathon Truwit 16. Brain death Gilbert Park, Balachandra Maiya
205 226 246 269 289
301 320 344 368
List of abbreviations
394
Index of papers reviewed
400
General index
411
Editors and contributors Editors
Tim M Cook, BA, MBBS, FRCA Consultant Anaesthetist, Department of Anaesthesia, Royal United Hospital, Bath, UK Jennifer M Hunter, MB chB, phD, FRCA Professor of Anaesthesia, University of Liverpool, Royal Liverpool University Hospital, Liverpool, UK Hans-Joachim Priebe, MD, FRCA, FFARCSI Professor of Anaesthesia, Department of Anaesthesia, University Hospital, Freiburg, Germany Michel MRF Struys, MD, PhD Professor in Anaesthesia and Research Coordinator, Department of Anaesthesia, Ghent University Hospital, Ghent, Belgium
Contributors
Takashi Asai, MD, PhD Assistant Professor, Department of Anaesthesiology, Kansai Medical University, Moriguchi City, Osaka, Japan Joseph E Arrowsmith, MBBS, MD, FRCP, FRCA Consultant Anaesthetist, Papworth Hospital, Cambridge, UK Andrew R Bodenham, FRCA Consultant in Intensive Care and Anaesthesia, The General Infirmary at Leeds, Leeds, UK Joachim Boldt, MD, PhD Professor, Department of Anesthesiology, Department of Anesthesiology and Intensive Care Medicine, Klinikum der Stadt Ludwigshafen, Ludwigshafen, Germany Gordon A Chapman, MBChB, FRCA Research Fellow in Anaesthesia and Intensive Care, Academic Unit of Anaesthesia, The General Infirmary at Leeds, UK Deborah J Cook, MD, FRCPC Professor of Medicine and Clinical Epidemiology and Biostatistics, Faculty of Health Sciences, McMaster University Medical Center, Hamilton, Ontario, Canada
Tim M Cook, BA, MBBS, FRCA Consultant Anaesthetist, Department of Anaesthesia, Royal United Hospital, Bath, UK D James Cooper, BMBS, MD, FRACP, FANZCA, FJFCM Clinical Associate Professor, Medicine, Surgery and Epidemiology and Preventive Medicine, Faculty of Medicine, Monash University, Intensive Care Department, Alfred Hospital, Commercial Road, Melbourne, Victoria, Australia Charlotte M Dempsey, MBBS, FRCA Fellow in Cardiothoracic Anaesthesia, Papworth Hospital, Cambridge, UK Simon R Finfer, MBBS, FRCP, FRCA, FJFICM Clinical Associate Professor (Intensive Care), Northern Clinical School, Faculty of Medicine, University of Sydney; Senior Staff Specialist in Intensive Care, Royal North Shore Hospital, St Leonards, New South Wales, Australia. AB Johan Groeneveld, MD, PhD, FCCP, FCCM Professor, Department of Intensive Care, Intensive Care, Vrije Universiteit Medical Centre, Institute for Cardiovascular Research, Amsterdam, The Netherlands S Jane Harper, MD CM, FRCA Consultant in Intensive Care and Anaesthesia, Department of Anaesthesia and Critical Care, Royal Liverpool University Hospital, Liverpool, UK Marina LH Honing, MD, PhD Researcher of Internal Medicine, Intensive Care, Vrije Universiteit Medical Centre, Institute for Cardiovascular Research, Amsterdam, The Netherlands Simon J Howell, MA(cantab), MRCP, FRCA, MSc, MD Senior Lecturer in Anaesthesia, Academic Unit of Anaesthesia, Leeds General Infirmary, Leeds, UK Jennifer M Hunter, MB chB, PhD, FRCA Professor of Anaesthesia, University of Liverpool, Royal Liverpool University Hospital, Liverpool, UK Harald Ihmsen, MSc, PhD Research Scientist, Department of Anaesthesiology, University Hospital, University of Erlangen-Nuremberg, Erlangen, Germany Peter Kimpson MB chB, BSC, FRCA Lecturer, Academic Unit of Anaesthesia, St James University Hospital, Leeds, UK Martin Luginbühl, MD Head, Division of Anaesthesia for Orthopaedic and Hand Surgery, Department of Anesthesiology, University Hospital of Bern, Inselspital, Bern, Switzerland Balachandra Maiya, FRCA John Farman Intensive Care Unit, Addenbrooke’s Hospital, Cambridge, UK Gernot Marx, MD Professor in Anaesthesia and Intensive Care Medicine, Department of Anaesthesia and Intensive Care Medicine, Friedrich-Schiller University, Jena, Germany Gilbert Park, MD, FRCA Consultant in Anaesthesia and Intensive Care, John Farman Intensive Care Unit, Addenbrooke’s Hospital, Cambridge, UK Hans-Joachim Priebe, MD, FRCA, FFARCSI Professor of Anaesthesia, Department of Anaesthesia, University Hospital, Freiburg, Germany
Emmy R Rijnsburger, MD Consultant in Intensive Care, Intensive Care, Institute for Cardiovascular Research, Vrije Universiteit Medical Centre, Amsterdam, The Netherlands Frédérique Servin, MD Consultant, Service d’Anesthésie-Réanimation Chirurgicale, Hôpital Bichat, Paris, France Richard Smiley, MD, phD Professor of Clinical Anesthesiology, Department of Anesthesiology, College of Physicians and Surgeons of Columbia University; Chief, Division of Obstetric Anesthesia, Columbia University Medical Center, New York, New York, USA Michel MRF Struys, MD, PhD Professor in Anaesthesia and Research Coordinator, Department of Anaesthesia, Ghent University Hospital, Ghent, Belgium Jonathon D Truwit, MD E Cato Drash Professor of Medicine, Senior Associate Dean for Clinical Affairs, Chief of Pulmonary and Critical Care Medicine, University of Virginia Health Systems, Charlottesville, Virginia, USA Hugo EM Vereecke, MD Staff Anaesthesiologist, Department of Anaesthesia, Ghent University Hospital, Ghent, Belgium Tim Walsh, BSC(Hons), MBCHB(Hons), MRCP, FRCA, MD Consultant in Anaesthetics and Intensive Care, Edinburgh Royal Infirmary; Part-time Senior Lecturer, University of Edinburgh, Edinburgh, UK
Foreword PETER J SIMPSON MD, FRCA Consultant Anaesthetist, Frenchay Hospital Frenchay, Bristol, UK
It is difficult to overestimate the importance of Continuing Medical Education and Professional Development (CEPD) for practising clinicians, and the major advances which continue to be made in anaesthesia and critical care must occupy a key role in this. It is all too easy for us, with the ever-increasing workload and demands of complex patient care, to feel we have no time to update our knowledge and skills, preferring to employ techniques with which we are familiar, supported by the sound clinical knowledge and judgment which we have acquired over many years. However, the development of anaesthetic techniques, equipment and pharmacology does not stand still. Equally, the sophistication of diagnostic techniques for our patients means that we are much better informed than in the past. In some ways, the more we know, the more difficult it is to make a balanced judgment on the optimal care of a patient. Yet how much more informative it is to know the actual gradient across a stenosed aortic valve than simply a judgment made on clinical signs. The demands made of anaesthetists both in theatre, in pain management and in the intensive care unit mean that being able to find sufficient time to attend CEPD meetings, particularly outside one’s own local environment let alone overseas, is increasingly difficult and therefore, for many of us, there is limited opportunity to concentrate and learn in an undisturbed way. Importantly, we need to hear the views of experts in their chosen field and, inevitably, we rely to a great extent on information supplied in academic journals and textbooks together, increasingly, from the Internet. Unless one has an enormous amount of time and a range of journals available, it is often difficult to obtain a balanced view about new areas of anaesthesia and critical care. What we really need is an expert to assess the current topic in question and to produce an objective commentary and judgment on the various papers that have been written during recent months. The Year in Anaesthesia and Critical Care is, for me, a new concept of book-based CEPD, which concentrates on extracting information from a number of recent papers and assessing it in a meaningful and readable way. New books appear on a regular basis but most are either orientated towards examinations and assessment processes during training, or are expert textbooks in a particular field of either anaesthesia or critical care. Few are aimed at CEPD for established career grade clinical anaesthetists, but this, with its new innovative format and concept, is such a book. No book of this nature should attempt to be comprehensive, and what the contributors have done is to select four key areas of recent development, namely perioperative care, anaesthetic pharmacology, new technologies in anaesthesia, and critical care. Individual key subjects are dealt with in more detail within these four sections, and organised in
each case by an editor who provides an excellent and objective editorial section at the beginning, including a brief overview of the subject, a summary of the papers, individual comment and references. The subject areas are reviewed by experts in the field who concentrate on a number of key publications that have occurred during the past year, looking at the key findings and recommendations of each and then coordinating these to provide more detailed response and comment. The reader is thus able to have expert opinion and comment at two levels both for the individual detailed subject area and more generally in the field which is the subject of the section of the book. Laid out as it is, in sections and subsections each with comprehensive headings, it is both easy to read and concentrate upon. Importantly, one only needs to read a small section at a time to gain relevant information, and interruptions do not disturb one’s flow of thought and learning unduly. There is no doubt that the opportunity to concentrate on specific topical areas rather than the need to be comprehensive in terms of subject content makes for a much more readable and interesting publication. Even at the level of studying for examinations, key comment and essays from experts do much to help one’s understanding of a subject and the research and scientific basis behind new developments and techniques. This selective approach is exemplified in all parts of the book. The section on perioperative care, for example, concentrates on perioperative blood and intravenous fluid therapy and protecting the heart in coincidental surgical operations, together with one review specifically related to cardiac surgery. For clinical anaesthetists wishing to read material to increase their knowledge, this combination has all the right ingredients. It develops themes about which they already know a considerable amount but wish to be updated, leading them through new ideas and techniques. Finally, it provides a detailed look at a more specialist area of work which would be of interest even if they were not acutely involved with it. The second section on anaesthetic pharmacology follows a similar theme identifying clinical aspects of basic pharmacology, leading into discussions on some of the newer hypnotics and analgesics which will undoubtedly alter clinical practice in the future. Finally, it provides an update on the influence of pharmacogenetics on anaesthesia, a topic which is becoming increasingly important for us. The section on new technology in anaesthesia and critical care is designed to take clinicians away from their current techniques, to show them what new developments have occurred and why these might be beneficial both for the quality of patient care and importantly for patient safety. This section then leads on to new equipment which few of us have had the opportunity of using. Although many of us are probably content to use our tried and tested techniques for central venous access and regional anaesthesia, if one talks to trainees, the vast majority find the use of ultrasound and Seldinger techniques second nature, and cannot imagine why we do not routinely use them in the same way. A book like this has the opportunity to discuss the advantages and disadvantages of new developments at length, and to try and reason through recent research and publications which inform the choice of technique. At present, there is considerable emphasis on the need to monitor depth of anaesthesia. Yet informed judgment is undoubtedly necessary if we are to avoid simply purchasing vast numbers of monitors in response to external demands and pressures. It is for this reason as much as anything else that, as anaesthetists, we need to develop our thoughts and reasoning logically. We are very fortunate to have this undertaken for us in such a readable and understandable way.
Although the last section contains a very valuable editorial on the issues surrounding critical care outreach and allows us to take a balanced view on its value in different hospitals, other articles in this section deal with a variety of other problems in intensive care medicine. Critical care outreach is again something about which one needs to make an informed judgment, before being pressurised to institute it, particularly if one feels that the decision to set up such a scheme is based upon little hard evidence and rather more on a perceived need because of a shortfall in ward-based care. Review articles are an excellent way of keeping up to date and enhancing one’s knowledge. Inevitably, they concentrate on specific topics and contain much of the authors’ own opinions rather than looking at the variety of views from others and allowing the reader to exercise their own judgment. This book allows objective discussion and comment to be made about all the articles written around a certain subject by others. It then allows the reader to make a balanced judgment based upon the recommendations made. For continuous education to have an impact on one’s opinion and clinical practice, one must be allowed to judge for oneself and not simply feel spoonfed by others’ opinions. The innovative format of this book, and the ease with which one can read and concentrate on it, makes it an ideal opportunity to enhance one’s education in anaesthesia and critical care. The breadth of subjects covered will, I am sure, have widespread appeal.
Part I Perioperative care
Perioperative care HANS-JOACHIM PRIEBE Introduction Numerous aspects of management comprise perioperative care. This brief overview will concentrate on four broad topic areas: perioperative intravenous fluid therapy, perioperative blood transfusion therapy, off-pump coronary artery bypass surgery and perioperative cardiac protection. Perioperative intravenous fluid therapy Type of fluid Even after decades of ongoing controversy, the debate on crystalloids versus colloids is still not settled. The debate is enlarged by the debate over the optimal type of crystalloid. Crystalloid administration may be associated with considerable adverse effects and these effects may differ considerably between different crystalloids. Metabolic effects Resuscitation with large volumes of high-concentration chloride solutions causes hyperchloraemic, non-anion gap metabolic acidosis. This type of acidosis may adversely affect urination, renal and gastric perfusion and function, pulmonary artery pressure and post-operative wellbeing |1|. Lack of awareness of this condition may result in the erroneous diagnosis and treatment of presumed anion gap or respiratory acidosis. Immunological effects Different intravenous fluids have variable effects on immune function. This may be of considerable clinical relevance because trauma and surgery activate the immune system, which, in turn, may aggravate post-traumatic cell injury and organ dysfunction. It must be of concern that isotonic crystalloids (in particular lactated Ringer’s solution) can cause immune activation and cellular injury |2|. The effect of hypertonic saline is less pronounced and plasma and albumin seem to be void of immuneactivating activity. Lactated Ringer’s and hydroxyethyl starch solutions may even cause acute lung apoptosis |3|. © Atlas Medical Publishing Ltd
The clinical relevance of these findings needs to be determined. Crystalloid solutions are being developed that not only possess volume-expanding but also antiinflammatory characteristics |4|. Haematological effects In general, non-protein colloids are associated with impaired haemostasis, platelet dysfunction and increased bleeding, while albumin and gelatins have the least effect |5|. The effect of hydroxyethyl starch on coagulation is clearly dependent on the type of preparation, as defined by concentration (low 3%, medium 6% and high 10%), the degree of substitution of hydroxyethyl starch residues per mole of glucose (low 0.4, medium 0.5 and high 0.62–0.7), mean molecular weight (low 70 kDa, medium 130–260 kDa and high >450 kDa) and the ratio of C2:C6 hydroxyethylation |6, 7|. Conclusion It is highly unlikely that there will ever be an intravenous fluid that ‘fits all’. Only knowledge of the kinetics of fluid therapy, of the physico-chemical and pharmacokinetic characteristics of the fluids and of the pathophysiology of the underlying disease entity requiring fluid resuscitation will enable an appropriate choice and monitoring of volume replacement. Volume of fluid administration The past It is still common practice to administer large volumes of fluid during elective surgical procedures |8|. This practice is based on the paradigm upheld for many decades that surgical patients require high intra-operative fluid volumes, irrespective of specific haemodynamic measurements |9|. This paradigm seemed to be supported by the principle of ‘goal-directed’ therapy, which improved outcome in high-risk surgical patients |10| and in patients with severe sepsis and septic shock |11| by ‘optimizing’ cardiac output. However, data from studies involving patients with sepsis, shock and trauma cannot automatically be extrapolated to patients undergoing elective surgical procedures (and vice versa). Furthermore, most studies of goal-directed therapy have examined the effect of fluid therapy strategy in combination with inotropic support and the results of these studies have not been unanimous. Studies on goal-directed therapy looking at the influence of fluid administration alone have shown differences in the amount of perioperatively administered fluid between groups that were of questionable clinical relevance |12–15|, which makes interpretation of the findings difficult. The most extensive recent trial on goal-directed therapy in major surgery not only failed to demonstrate benefit of such an approach, but pointed out some serious disadvantages |15|. The strategy of replacing blood loss by crystalloids at a 3:1 ratio is still common practice. Interestingly, the Saline versus Albumin Fluid Evaluation (SAFE) Study reported a mere 1.4:1 ratio of administered crystalloid:colloid |16|. This finding is consistent with the possibility that the accepted practice of replacing blood loss with
crystalloids at a 3:1 ratio leads to clinically relevant acute over-transfusion with crystalloids. The present In contrast to what is commonly practised, there is now mounting evidence that the extracellular fluid shift (‘third spacing’) is less than previously assumed, that excess fluid can be detrimental and that excess perioperative fluid administration may worsen and fluid restriction may improve the perioperative outcome |17, 18|. Thus, evidence suggests that careful perioperative fluid restriction may be indicated in selected patient populations undergoing selected elective surgical procedures. A decrease in perioperative urine output may often be a reflection of the perioperative stress response rather than of inadequate fluid therapy. Timing of volume administration The timing of volume of fluid resuscitation has recently received increasing attention. Inadequate fluid resuscitation (identified as being the most common management deficiency in trauma deaths) must be carefully weighed against increasing evidence that overly aggressive fluid replacement adversely affects the outcome in trauma patients |19, 20|. Perioperative blood transfusion therapy Although red blood cell transfusion increases oxygen-carrying capacity, a liberal transfusion strategy is usually not associated with improved outcome. This would suggest that either anaemia is beneficial or transfusion is detrimental. The beneficial effects of anaemia per se are difficult to postulate. On the other hand, the detrimental effects of transfusion leading to increased morbidity and mortality are well defined: infection, haemolytic reaction, contamination, allergic-anaphylactic reactions, transfusion-related lung injury |21| intravascular volume-related pulmonary oedema and immunomodulation. Transfusion trigger Uncertainty remains as to the appropriate haemoglobin (Hb) concentration that should trigger a blood transfusion. Despite evidence that a restrictive transfusion strategy (Hb 7.0–9.0 g/dl) lowers in-hospital mortality and reduces blood transfusions without adversely affecting the outcome in patients without cardiac disease |22|, transfusion practice in the care of critically ill patients seems to have changed little during the past 10 years. Worldwide, the mean pre-transfusion Hb concentrations have consistently remained around 8.5 g/dl |23, 24|. Only patients with acute coronary syndromes may benefit from a more liberal transfusion strategy |25–27|.
Immunomodulation Transfusion-related immune modulation is receiving continued and increasing attention. It is postulated that transfused allogeneic leucocytes trigger the immune response, possibly resulting in more frequent infections, earlier development of malignancies and even increased mortality |28, 29|. Evidence does in fact suggest that universal leukoreduction may improve outcome |30| and, in particular, decrease post-transfusion non-haemolytic fever |31–34|. However, a recent meta-analysis of ten randomized trials did not demonstrate statistically significant benefits from leukoreduction |35|. Leukoreduction was only associated with a 40% reduction in post-operative infections if the analysis was restricted to those patients who actually received blood transfusions. Although the scientific proof for the benefit of leukoreduction is still pending, based on existing evidence universal adoption of this practice has been recommended |28|. Storage lesion The clinical relevance of the blood ‘storage lesion’ possibly associated with the transfusion of ‘old’ blood continues to be debated. A blood storage lesion describes the reduced erythrocyte deformability caused by changes in the erythrocyte membrane due to storage-induced depletion of 2,3 diphosphoglycerol and adenosine triphosphate (ATP). Such erythrocyte defect may impair tissue oxygenation via impaired capillary blood flow and oxygen unloading from Hb. The findings of two recent trials are controversial. One showed a trend towards a worse outcome in patients who received relatively old blood (mean age of blood 21 days) |24|. The other did not show adverse effects of transfusion of stored, leukodepleted red cells to anaemic, critically ill patients on gastric tonometry and global indices of oxygen delivery |36|. The issue remains far from being settled |37|. Off-pump coronary artery bypass surgery Who benefits? Considerable progress has been made during the past years in the technique of coronary artery bypass graft (CABG) surgery without cardiopulmonary bypass (off-pump CABG or off-pump coronary artery bypass [OPCAB] surgery). The advance in the technique of OPCAB surgery is driven by the considerable risk of post-operative cognitive dysfunction and stroke associated with on-pump CABG in high-risk patients on the one hand and concern with poor long-term graft patency possibly associated with OPCAB on the other. Thus, a reduced incidence of post-operative cognitive dysfunction and stroke, combined with a comparable quality of revascularization surgery will be the ultimate criteria by which to judge the benefit of OBCAB over on-pump surgery |38|. Despite considerable effort and progress in this area, it may be somewhat disappointing that several recent randomized clinical trials comparing OPCAB to onpump CABG surgery failed to document clear advantages of one technique over the other |39–42|. Although OPCAB caused less myocardial damage than on-pump CABG and was
as safe, the graft patency rates were lower at 3 months post-operatively, which may adversely influence the long-term outcome |39|. There were no advantages of OPCAB over on-pump CABG in terms of morbidity (transfusion requirements, perioperative myocardial infarction, stroke, new atrial fibrillation and sternal wound infection), length of hospitalization and mortality |40, 41|. A very recently published meta-analysis of 37 randomized trials involving 3369 patients confirmed the lack of difference between on-pump and off-pump CABG surgery in primary outcomes (mortality, myocardial infarction, stroke, renal dysfunction, requirement for an intra-aortic balloon pump, wound infection, rethoracotomy and reintervention) at 30 days and mortality at 2 years post-operatively |43|. Data on graft patency and post-operative cognitive function were inconclusive. Only selected shortterm and mid-term clinical (atrial fibrillation, requirement for transfusion and inotropic support, respiratory infections, ventilatory support and length of stay in the intensive care unit and hospital) and resource outcomes favoured OPCAB surgery. In contrast, most non-randomized trials (including large database observational studies) found significant differences in mortality, neurological, pulmonary, renal and bleeding complications usually in favour of the off-pump approach |44–47|. Particularly in subsets of patients considered to be at high risk for perfusion, off-pump surgery showed clear benefits |48, 49|. Whereas in most randomized studies high-risk patients were excluded and the procedure-related risk was low for both procedures |43|, the nonrandomized trials contained relatively more high-risk patients in the on-pump groups than either the on- or off-pump groups in the randomized trials. Obviously, excluding highrisk patients from the randomized studies will make it more difficult for off-pump surgery to demonstrate superiority. Furthermore, in the randomized trials the mean age was only 63 years, the aggregate risk of stroke was only 1% (which is much lower than the usually reported rate of 2–3%), graft patency was not analysed quantitatively and the follow-up time was limited to 2 years. Thus, the final verdict is not out yet. However, it is likely that different surgical approaches are indicated in different patient populations in order to optimize the shortand long-term outcomes |50|. Current evidence would suggest that, in goodrisk patients, it is not worth compromising optimal revascularization for the purpose of performing surgery without cardiopulmonary bypass |51|. In contrast, patients with a risk of aortic cannulation or previous stroke and elderly patients may frequently benefit from the offpump approach. Intra-operative management The success of OPCAB surgery will largely depend on optimal positioning of the heart in order to expose the target coronary vessel, on maximal reduction in local cardiac wall motion to perform the distal anastomosis and on techniques directed at minimizing myocardial injury during temporal occlusion of the target coronary artery necessary for visualizing the site of the distal anastomosis. Lifting and rotating the heart and use of cardiac wall stabilization devices during OPCAB may impair cardiovascular performance, frequently requiring intravenous administration of fluid and inotropic support and the Trendelenburg position. Displacement of the heart, fixation of the cardiac wall and temporary occlusion of the
target coronary vessel contribute to impaired coronary blood flow and subsequent regional myocardial ischaemia, which, in turn, worsens haemodynamic instability |52|. With this much interference of routine surgery with cardiovascular performance, it is to be expected that, like the surgical, the anaesthetic management for OPCAB surgery is more demanding than that for on-pump CABG. Anticipation and intimate knowledge of the aetiology of the side effects associated with OPCAB-typical surgical interventions, close cooperation with the surgeon, preventive rather than therapeutic measures for counteracting the surgery-induced impairment in cardiovascular performance and, foremost, a highly skilled surgeon, are likely to affect outcomes more than a particular anaesthetic technique or drug. Perioperative cardiac protection The perioperative period induces large, unpredictable and unphysiological alterations in coronary plaque morphology, function and progression and may trigger a mismatch of myocardial oxygen supply and demand |53|. With many and diverse factors involved, it is highly unlikely that one single intervention will successfully improve cardiac outcomes following non-cardiac surgery. A multifactorial, step wise approach is indicated. Two principal strategies have been employed in an attempt to reduce the incidence of perioperative ischaemic cardiac events and complications: pre-operative coronary revascularization, and pharmacological treatment. Pre-operative coronary revascularization Controversy remains as to the appropriate management of patients identified preoperatively as having relevant but correctable coronary artery disease (CAD). The effectiveness of pre-operative coronary revascularization in this population continues to be debated. Proponents of ‘prophylactic’ coronary revascularization in selected patients argue that it improves both perioperative as well as long-term outcomes |54|. Opponents of this approach point out that the morbidity and mortality of percutaneous coronary intervention and CABG surgery in high-risk elderly vascular patients are substantial and outweigh any benefit, that recovery from such major morbidity substantially delays and even prevents the surgery for which the intervention was undertaken, that it does not differentiate between young and old age and between patients with symptomatic CAD and those with CAD discovered by cardiac stress testing only, that only survivors of coronary revascularization are included in the various reports and, most importantly, that no prospective randomized trial exists that demonstrates the effectiveness of preoperative coronary revascularization in improving the short- and long-term cardiac outcomes and mortality in high-risk patients undergoing high-risk surgery. Pre-operative percutaneous intervention Patients who have recently been subjected to coronary stenting run a high risk of suffering a perioperative myocardial infarction and serious bleeding (summarized in |53|). If a (bare metal) coronary stent is placed, elective non-cardiac surgery should, therefore,
be delayed for an absolute minimum of 2 weeks, but ideally for 4–6 weeks. Today, stents eluting anti-proliferative drugs that delay endothelialization are increasingly being placed. As this may well increase the risk of early and late stent thrombosis, a 6–12month period of anti-platelet treatment has been recommended |55, 56|. Pre-operative surgical coronary revascularization Survivors of coronary revascularization tend to have better perioperative and long-term cardiac outcomes than patients with comparable CAD without pre-operative coronary revascularization. However, this analysis does not take into account the high cardiac morbidity and mortality associated with pre-operative coronary angiography and coronary revascularization (pre-operative percutaneous intervention or CABG) in highrisk patients. In addition, survivors of pre-operative percutaneous intervention face the perioperative risk of coronary (stent) thrombosis or haemorrhage associated with discontinuation or continuation of dual anti-platelet therapy, respectively. The overall outcome may thus be comparable between preoperatively revascularized and nonrevascularized patients—it may be even worse in individual revascularized patients. The first prospective, randomized trial demonstrated that prophylactic coronary revascularization before elective surgery does not alter the long-term outcome |57|. However, the patients were not selected according to the guidelines for pre-operative cardiac assessment issued by the American College of Cardiology and the American Heart Association |58| and less than half had significant CAD. Therefore, the applicability of the findings to high-risk patients remains unanswered. The decision for or against pre-operative coronary revascularization and for or against pre-operative percutaneous intervention or CABG should therefore be based entirely on universally accepted medical indications for coronary revascularization and the appropriate technique. The philosophy of performing pre-operative coronary revascularization merely ‘to get the patient through surgery’ is contrary to all available evidence. If the decision for pre-operative coronary revascularization is made, timing with respect to the subsequent surgery appears crucial. If these caveats are being observed, it is conceivable that carefully selected patients might benefit from preoperative coronary revascularization |54|. Pharmacological protection Perioperative β-blocker therapy Numerous cardiovascular and other effects (anti-arrhythmic, anti-inflammatory, altered gene expression and receptor activity and protection against apoptosis) of β-blockers may account for their cardioprotective effect in the operative and non-operative setting |59|. Although perioperative β-blocker therapy has been designated as one of eleven specific practices with sufficient clinical-based evidence for patient safely to justify immediate and widespread implementation |60| and although its use continues to be highly recommended |61|, the implementation of aggressive perioperative β-blocker therapy is slow |61–63|.
However, before a final recommendation for liberal use of perioperative β-blockade can be made safely, several caveats have to be kept in mind (summarized in |53|). All studies that support the use of perioperative β-blocker therapy have included rather small numbers of patients (as few as 26). Often the recruitment of patients was highly selective and consecutive (recruitment rate as low as 8%), excluding application of the results to an unselected surgical population. Furthermore, the beneficial effects were probably not only due to a rather aggressive therapy (targeted heart rates maximally 80 beats/min), but also (and perhaps even more importantly) due to continuous close monitoring of the patient. This will ensure both optimal cardioprotection and patient safety. A more uncontrolled but equally aggressive post-operative administration of β-blockers on ordinary surgical floors might well result in more adverse side effects, possibly negating any beneficial effects. Although current evidence suggests that selected patients are likely to benefit from perioperative β-blocker therapy, we have to acknowledge that the data on the risks and benefits of such therapy are still few and inconclusive |64–66|. α2-adrenoceptor agonists α2-adrenoceptor agonists may improve cardiovascular morbidity and mortality following non-cardiac and cardiac surgery |67|. The mechanism of the protective effect is likely to be manifold. α2-adrenoceptor agonists attenuate perioperative haemodynamic instability, inhibit central sympathetic discharge, reduce peripheral norepinephrine release and dilate post-stenotic coronary vessels. Statins Perioperative use of statins may be associated with reduced perioperative mortality in patients undergoing major vascular and non-cardiac surgery |68–70|. ‘Pleiotropic’ effects of statins independent of their lipid-lowering action have been proposed as the mechanisms of their beneficial effects. These pleiotropic effects include reversal of endothelial dysfunction |71|, modulation of macrophage activation |73|, immunological effects |72| and anti-inflammatory |72, 73|, anti-thrombotic |71| and anti-proliferative action (possibly mediated by the induction of haeme oxygenase-l) |74|. The direct effect of statins on vascular function may result in coronary plaque stabilization. Anaesthetic preconditioning An increasing number of experimental investigations demonstrate that volatile anaesthetics can acutely induce a cardioprotective memory effect that lasts beyond their elimination |75|. This mechanism attenuates the deleterious effects of myocardial ischaemia and reperfusion injury and is of considerable potential clinical relevance. However, advanced age |76|, diabetes mellitus |77|, the use of oral hypoglycaemic drugs, the cardiodepressant effect of higher concentrations of volatile anaesthetics and the constraints of a complicated preconditioning protocol limits the applicability of this mechanism in routine clinical practice (for a review see |75|). We will have to await the results of large, well-designed clinical trials to see whether the results of basic research can be transferred into clinical practice.
Conclusion The aetiology of perioperative ischaemic cardiac events is multifactorial |53, 78|. The perioperative period induces large, unpredictable and unphysiological changes in sympathetic tone, cardiovascular performance, coagulation and inflammatory response (to name just a few). These changes in turn induce unpredictable alterations in plaque morphology, function and progression. Simultaneous perioperative alterations in homeostasis and coronary plaque characteristics may trigger a mismatch of myocardial oxygen supply and demand by numerous mechanisms. If not alleviated in time, it will ultimately result in myocardial infarction, irrespective of its aetiology (morphologically, haemodynamically, inflammatory or coagulation induced). With these many and diverse factors involved, it is highly unlikely that one single intervention will successfully improve cardiac outcome following non-cardiac surgery. A multifactorial, stepwise approach is indicated |79–81|. Based on increasing knowledge of the nature of atherosclerotic CAD and in view of the poor positive predictive value of non-invasive cardiac stress tests and the considerable risk of coronary angiography and coronary revascularization in highrisk patients, the paradigm is shifting from an emphasis on extensive non-invasive pre-operative risk stratification to an emphasis on a combination of selective noninvasive testing (to identify reliably those patients who truly benefit from pre-operative intervention, such as cancellation of surgery, pre-operative coronary revascularization and initiation or optimization of cardioprotective medication) and aggressive pharmacological perioperative therapy |81, 82|). Perioperative plaque stabilization by pharmacological means (statins, aspirin and β-blockers) may be as important in the prevention of perioperative myocardial infarction as is an increase in myocardial oxygen supply (by coronary revascularization) or a reduction in myocardial oxygen demand (by β-blockers or α2-agonists). References 1. Wilkes NJ, Woolf R, Mutch M, Mallett SV, Peachey T, Stephens R, Mythen MG. The effects of balanced versus saline-based hetastarch and crystalloid solutions on acid—base and electrolyte status and gastric mucosal perfusion in elderly surgical patients. Anesth Analg 2001; 93: 811– 16. 2. Rhee P, Koustova E, Alam HB. Searching for the optimal resuscitation method: recommendations for the initial fluid resuscitation of combat casualties. J Trauma 2003; 54: S52–62. 3. Deb S, Sun L, Martin B, Talens E, Burris D, Kaufmann C, Rich N, Rhee P. Lactated Ringer’s solution and hetastarch but not plasma resuscitation after rat hemorrhagic shock is associated with immediate lung apoptosis by the up-regulation of the bax protein. J Trauma 2000; 49: 47– 55. 4. Sims CA, Wattanasirichaigoon S, Menconi MH, Ajami AM, Fink MP. Ringer’s ethyl pyruvate solution ameliorates ischemia/reperfusion-induced intestinal mucosal injury in rats. Crit Care Med 2001; 29: 1513–18. 5. Boldt J. Fluid choice for resuscitating trauma: review of physiological/pharmacological/ clinical evidence. Can J Anesth 2004; 51: 500–13.
6. De Jonge E, Levi M. Effects of different plasma substitutes on blood coagulation: a comparative review. Crit Care Med 2001; 29: 1261–7. 7. Wilkes MW, Navickis RJ, Sibbald WJ. Albumin versus hydroxyethyl starch in cardiopulmonary bypass surgery: a meta-analysis of post-operative bleeding. Ann Thorac Surg 2001; 72: 527–34. 8. Holte K, Sharrock NE, Kehlet H. Pathophysiology and clinical implications of perioperative fluid excess. Br J Anaesth 2002; 89: 622–32. 9. Shires T, Williams J, Brown F. Acute change in extracellular fluids associated with major surgical procedures. Ann Surg 1961; 154: 803–10. 10. Shoemaker WC, Appel PL, Kram HB, Waxman K, Lee TS. Prospective trial of supra-normal values of survivors as therapeutic goals in high-risk surgical patients. Chest 1988; 94: 1176–86. 11. Rivers E, Nguyen B, Havstad S, Ressler J, Muzzin A, Knoblich B, Peterson E, Tomlanovich M for the Early Goal-directed Therapy Collaborative Group. Early goal-directed therapy in the treatment of severe sepsis and septic shock. N Engl J Med 2001; 345: 1368–77. 12. Gan TJ, Soppitt A, Maroof M, El-Moalem H, Robertson KM, Moretti E, Dwane P, Glass PSA. Goal-directed intraoperative fluid administration reduces length of hospital stay after major surgery. Anesthesiology 2002; 97: 820–6. 13. Conway DH, Mayall R, Abdul-Latif MS, Gilligan S, Tackaberry C. Randomised controlled trial investigating the influence of intravenous fluid titration using oesophageal Doppler monitoring during bowel surgery. Anaesthesia 2002; 57: 845–9. 14. Price J, Sear J, Venn R. Perioperative fluid volume optimization following proximal femoral fracture. Cochrane Library 2002 (http://www.cochrane.org/). 15. Sandham JD, Hull RD, Brant RF, Knox L, Pineo GF, Doig CJ, Laporta AP, Viner S, Passerini L, Devitt H, Kirby A, Jacka M. A randomized, controlled trial of the use of pulmonary artery catheters in high-risk surgical patients. N Engl J Med 2003; 348: 5–14. 16. Finfer S, Bellomo R, Boyce N, French J, Myburgh J, Norton R; SAFE Study Investigators. A comparison of albumin and saline for fluid resuscitation in the intensive care unit. N Engl J Med 2004; 350: 2247–56. 17. Brandstrup B, Tønnesen H, Beier-Holgersen R, Hjortsø E, Ørding H, Lindorff-Larsen K, Rasmussen MS, Lanng C, Wallin L and the Danish Study Group on Perioperative Fluid Therapy. Effects of intravenous fluid restriction on post-operative complications: comparison of two perioperative fluid regimens. A randomized assessor-blinded multicenter trial. Ann Surg 2003; 238: 641–8. 18. Lobo DN, Bostock KA, Neal KR, Perkins AC, Rowlands BJ, Allison SP. Effect of salt and water balance on recovery of gastrointestinal function after elective colonic resection: a randomised controlled trial. Lancet 2002; 359: 1812–18. 19. Kwan I, Bunn F, Roberts I on behalf of the WHO Pre-hospital Trauma Care Steering Committee. Timing and volume of fluid administration for patients with bleeding. Cochrane Database Syst Rev 2003: Issue 3. 20. Moore FA, McKinley BA, Moore E. The next generation in shock resuscitation. Lancet 2004; 363: 1988–96. 21. Looney MR, Gropper MA, Matthay MA. Transfusion-related acute lung injury: a review. Chest 2004: 126: 249–58. 22. Hill SR, Carless PA, Henry DA, Carson JL, Hebert PC, McClelland DBL, Henderson KM. Transfusion thresholds and other strategies for guiding allogeneic red blood cell transfusion. Cochrane Database Syst Rev 2002: Issue 2. 23. French CJ, Bellomo R, Finfer SR, Lipman J, Chapman M, Boyce NW. Appropriateness of RBC transfusion in Australasian ICU practice. Med J Aust 2002; 177: 548–51. 24. Corwin HL, Gettinger A, Pearl RG, Fink MP, Levy MM, Abraham E, MacIntyre NR, Shabot MM, Duh MS, Shapiro MJ. CRIT Study: anemia/blood transfusion in the critically ill—clinical practice in the USA. Crit Care Med 2004; 32: 39–52.
25. Hébert PC, Yetisir E, Martin C, Blajchman MA, Wells G, Marshall J, Tweeddale M, Pagliarello G, Schweitzer I. Is a low transfusion threshold safe in critically ill patients with cardiovascular diseases? Crit Care Med 2001; 29: 227–34. 26. Wu W-C, Rathore SS, Wang Y, Radford MJ, Krumholz HM. Blood transfusion in elderly patients with acute myocardial infarction. N Engl J Med 2001; 345: 1230–6. 27. Hébert PC , Fergusson DA. Do transfusions get to the heart of the matter? JAMA 2004; 292: 1610–12. 28. Carson JL, Berlin JA. Will we ever know if leukoreduction of red blood cells should be performed? Can J Anesth 2004; 51: 407–11. 29. Raghavan M, Marik PE. Anemia, allogenic blood transfusion, and immunomodulation in the critically ill. Chest 2005; 127: 295–307. 30. Hébert PC, Fergusson D, Blajchman MA, Wells GA, Kmetic A, Coyle D, Heddle N, Germain M, Goldman M, Toye B, Schweitzer I, Van Walraven C, Devine D, Sher GD for the Leukoreduction Study Investigators. Clinical outcomes following institution of the Canadian universal leukoreduction program for red blood cell transfusions. JAMA 2003; 289: 1941–9. 31. Paglino JC, Pomper GJ, Fisch GS, Champion MH, Snyder EL. Reduction of febrile but not allergic reactions to RBCs and platelets after conversion to universal pre-storage leukoreduction. Transfusion 2004; 44: 16–24. 32. King KE, Shirey RS, Thoman SK, Bensen-Kennedy D, Tanz WS, Ness PM. Universal leukoreduction decreases the incidence of febrile non-hemolytic transfusion reactions to RBCs. Transfusion 2004; 44: 25–9. 33. Van Hilten JA, Van de Watering LMG, Van Bockel JH, Van de Velde CJH, Kievit J, Brand R, Van den Hout WB, Geelkerken RH, Roumen RMH, Wesselink RMJ, Koopman-Van Gemert AWMM, Koning J, Brand A. Effects of transfusion with red cells filtered to remove leucocytes: randomised controlled trial in patients undergoing major surgery. BMJ 2004; 328: 1281–8. 34. Bilgin YM, Van de Watering LMG, Eijsman L, Versteegh MIM, Brand R, Van Oers MHJ, Brand A. Double-blind, randomized controlled trial on the effect of leukocyte-depleted erythrocyte transfusions in cardiac valve surgery. Circulation 2004; 109: 2755–60. 35. Fergusson D, Khanna MP, Tinmouth A, Hebert PC. Transfusion of leukoreduced red blood cells may decrease post-operative infections: two meta-analyses of randomized clinical trials. Can J Anesth 2004; 51: 417–24. 36. Walsh TS, McArdle F, McLellan SA, Maciver C, Maginnis M, Prescott RJ, McClelland DB. Does the storage time of transfused red blood cells influence regional or global indexes of tissue oxygenation in anemic critically ill patients? Crit Care Med 2004; 32: 364–71. 37. Goodnough LT, Shander A, Brecher ME. Transfusion medicine: looking to the future. Lancet 2003; 361: 161–9. 38. Floyd T, Fleisher LA. Off-pump coronary artery bypass surgery and the hypothesis from which it grew: is it yet to be tested? What are the downsides of the lingering questions? Anesthesiology 2005; 102: 3–5. 39. Khan NE, De Souza A, Mister R, Flather M, Clague J, Davies S, Collins P, Wang D, Sigwart U, Pepper J. A randomized comparison of off-pump and on-pump multivessel coronaryartery bypass surgery. N Engl J Med 2004; 350: 21–8. 40. Légaré J-F, Buth KJ, King S, Wood J, Sullivan JA, Friesen CH, Lee J, Stewart K, Hirsch GM. Coronary bypass surgery performed off pump does not result in lower in-hospital morbidity than coronary artery bypass grafting performed on pump. Circulation 2004; 109: 887–92. 41. Gerola LR, Buffolo E, Jasbik W, Botelho B, Bosco J, Brasil LA, Branco JN. Off-pump versus on-pump myocardial revascularization in low-risk patients with one or two vessel disease: perioperative results in a multicenter randomized controlled trial. Ann Thorac Surg 2004; 77: 569–73. 42. Puskas JD, Williams WH, Mahoney EM, Huber PR, Block PC, Duke PG, Staples JR, Glas KE, Marshall JJ, Leimbach ME, McCall SA, Petersen RJ, Bailey DE, Weintraub WS, Guyton RA.
Off-pump versus conventional coronary artery bypass grafting: early and 1-year graft patency, cost, and quality-of-life outcomes: a randomized trial. JAMA 2004; 291: 1841–9. 43. Cheng DC, Bainbridge D, Martin JE, Novick RJ, the Evidence-based Perioperative Clinical Outcomes Research Group. Does off-pump coronary artery bypass reduce mortality, morbidity, and resource utilization when compared with conventional coronary artery bypass? A metaanalysis of randomized trials. Anesthesiology 2005; 102: 188–203. 44. Mack MJ, Duhaylongsod FG. Through the open door! Where has the ride taken us? J Thorac Cardiovasc Surg 2002; 124: 655–9. 45. Sabik JF, Gillinov AM, Blackstone EH, Vacha C, Houghtaling P, Navia J, Smedira NG, McCarthy PM, Cosgrove DM, Lytle BW. Does off-pump coronary surgery reduce morbidity and mortality? J Thorac Cardiovasc Surg 2002; 124: 698–707. 46. Al-Ruzzeh S, Ambler G, Asimakopoulos G, Omar RZ, Hasan R, Fabri B, El-Gamel A, DeSouza A, Zamvar V, Griffin S, Keenan D, Trivedi U, Pullan M, Cale A, Cowen M, Taylor K, Amrani M. Off-pump coronary artery bypass (OPCAB) surgery reduces riskstratified morbidity and mortality: a United Kingdom multicentre comparative analysis of early clinical outcome. Circulation 2003; 108(Suppl I): II1–8. 47. Racz MJ, Hannan EL, Isom OW, Subramanian VA, Jones RH, Gold JP, Ryan TJ, Hartman A, Culliford AT, Bennett E, Lancey RA, Rose EA. A comparison of short- and long-term outcomes after off-pump and on-pump coronary artery bypass graft surgery with sternotomy. J Am Coll Cardiol 2004; 43: 557–64. 48. Hirose H, Amano A, Takahashi A. Off-pump coronary artery bypass grafting for elderly patients. Ann Thorac Surg 2001; 72: 2013–19. 49. Sharony R, Bizekis CS, Kanchuger M, Galloway AC, Saunders PC, Applebaum R, Schwartz CF, Ribakove GH, Culliford AT, Baumann FG, Kronzon I, Colvin SB, Grossi EA. Off-pump coronary artery bypass grafting reduces mortality and stroke in patients with atheromatous aortas: a case control study. Circulation 2003; 108(Suppl II): II15–20. 50. Magee MJ, Coombs LP, Peterson ED, Mack MJ. Patient selection and current practice strategy for off-pump coronary bypass surgery. Circulation 2003; 108(Suppl 1): II9–14. 51. Lytle BW, Sabik JF. On-pump and off-pump bypass surgery: tools for revascularization. Circulation 2004; 109: 810–12. 52. Verma S, Fedak PWM, Weisel RD, Szmitko PE, Badiwala MV, Bonneau D, Latter D, Errett L, LeClerc Y. Off-pump coronary artery bypass surgery. Fundamentals for the clinical cardiologist. Circulation 2004; 109: 1206–11. 53. Priebe H-J. Perioperative myocardial infarction—aetiology and prevention. Br J Anaesth 2005; 95 (in press). 54. Landesberg G, Mosseri M, Shatz V, Akopnik I, Bocher M, Mayer M, Anner H, Berlatzky Y, Weissman C. Cardiac troponin after major vascular surgery; the role of perioperative ischemia, preoperative thallium scanning, and coronary revascularization. J Am Coll Cardiol 2004; 44: 569–75. 55. Grube E, Silber S, Hauptmann KE, Mueller R, Buellesfeld L, Gerckens U, Russell ME. TAXUS I: six- and twelve-month results from a randomized, double-blind trial on a slowrelease paclitaxel-eluting stent for de novo coronary lesions. Circulation 2003; 107: 38–42. 56. Auer J, Berent R, Weber T, Eber B. Risk of non-cardiac surgery in the months following placement of a drug-eluting coronary stent (letter to the editor). J Am Coll Cardiol 2004; 43:713. 57. McFalls EO, Ward HB, Moritz TE, Goldman S, Krupski WC, Littooy F, Pierpont G, Santilli S, Rapp J, Hattler B, Shunk K, Jaenicke C, Thottapurathu L, Ellis N, Reda DJ, Henderson WG. Coronary-artery revascularization before elective major vascular surgery. N Engl J Med 2004; 351: 2795–804. 58. Eagle KA, Berger PB, Calkins H, Chaitman BR, Ewy GA, Fleischmann KE, Fleisher LA, Froehlich JB, Gusberg RJ, Leppo JA, Ryan T, Schlant RC, Winters Jr WL, Gibbons RJ, Antman EM, Alpert JS, Faxon DP, Fuster V, Gregoratos G, Jacobs AK, Hiratzka LF, Russell RO, Smith
Jr SC; American College of Cardiology; American Heart Association. ACC/AHA guideline update for perioperative cardiovascular evaluation for noncardiac surgery—executive summary: a report of the American College of Cardiology/American Heart Association Task Force on Practice Guidelines (Committee to Update the 1996 Guidelines on Perioperative Cardiovascular Evaluation for Noncardiac Surgery). J Am Coll Cardiol 2002; 39: 542–53. 59. London MJ, Zaugg M, Schaub MC, Spahn DR. Perioperative β-adrenergic receptor blockade: physiologic foundations and clinical controversies. Anesthesiology 2004; 100: 170–5. 60. Shojania KG, Duncan BW, McDonald KM, Wachter RM. Safe but sound: patient safety meets evidence-based medicine. JAMA 2002; 288: 508–13. 61. Lindenauer PK, Fitzgerald J, Hoople N, Benjamin EM. The potential preventability of postoperative myocardial infarction; underuse of perioperative β-adrenergic blockade. Arch Intern Med 2004; 164: 762–6. 62. Rapchuk I, Rabuka S, Tonelli M. Perioperative use of beta-blockers remains low: experience of a single Canadian tertiary institution. Can J Anesth 2004; 51: 761–7. 63. Siddiqui AK, Ahmed S, Delbeau H, Coner D, Mattana J. Lack of physician concordance with guidelines on the perioperative use of β-blockers. Arch Intern Med 2004; 164: 664–7. 64. Devereaux PJ, Leslie K, Yang H. The effect of perioperative beta-blockers on patients undergoing non-cardiac surgery—is the answer in? Can J Anesth 2004; 51: 749–55. 65. Devereaux PJ, Yusuf S, Yang H, Choi PTL, Guyatt GH. Are the recommendations to use perioperative β-blocker therapy in patients undergoing non-cardiac surgery based on reliable evidence? CMAJ 2004; 171: 245–7. 66. Giles JW, Sear JW, Foex P. Effect of chronic β-blockade on perioperative outcome in patients undergoing non-cardiac surgery: an analysis of observational and case control studies. Anaesthesia 2004; 59: 574–83. 67. Wallace AW, Galindez D, Salahieh A, Layug EL, Lazo EA, Haratonik KA, Boisvert DM, Kardatzke D. Effect of clonidine on cardiovascular morbidity and mortality after noncardiac surgery. Anesthesiology 2004; 101: 284–93. 68. Poldermans D, Bax JJ, Kertai MD, Krenning B, Westerhout CM, Schinkel AF, Thomson IR, Lansberg PJ, Fleisher LA, Klein J, Van Urk H, Roelandt JR, Boersma E. Statins are associated with a reduced incidence of perioperative mortality in patients undergoing major non-cardiac vascular surgery. Circulation 2003; 107: 1848–51. 69. Kertai MD, Boersma E, Westerhout CM, Van Domburg R, Klein J, Bax JJ, Van Urk H, Poldermans D. Association between long-term statin use and mortality after successful abdominal aortic aneurysm surgery. Am J Med 2004; 116: 96–103. 70. Lindenauer PK, Pekow P, Wang K, Gutierrez B, Benjamin EM. Lipid-lowering therapy and inhospital mortality following major non-cardiac surgery. JAMA 2004; 291: 2092–9. 71. Vaughan CJ, Gotto Jr AM. Update on statins: 2003. Circulation 2004; 110: 886–92. 72. Almog Y, Shefer A, Novack V, Maimon N, Barski L, Eizinger M, Friger M, Zeller L, Danon A. Prior statin therapy is associated with a decreased rate of severe sepsis. Circulation 2004; 110: 880–5. 73. Muhlestein JB, Anderson JL, Home BD, Carlquist JF, Blair TL, Bunch TJ, Pearson RR for the Intermountain Heart Collaborative Study Group. Early effects of statins in patients with coronary artery disease and high C-reactive protein. Am J Cardiol 2004; 94: 1107–12. 74. Lee T-S, Chang C-C, Zhu Y, Shyy JYJ. Simvastatin induces heme oxygenase-1; a novel mechanism of vessel protection. Circulation 2004; 110: 1296–301. 75. Riess ML, Stowe DF, Warltier DC. Cardiac pharmacological preconditioning with volatile anesthetics: from bench to bedside? Am J Physiol Heart Circ Physiol 2004; 286: H1603–7. 76. Sniecinski R, Liu H. Reduced efficacy of volatile anesthetic preconditioning with advanced age in isolated rat myocardium. Anesthesiology 2004; 100: 589–97. 77. Ebel D, Müllenheim J, Frassdorf J, Heinen A, Huhn R, Bohlen T, Ferrari J, Sudkamp H, Preckel B, Schlack W, Thamer V. Effect of acute hyperglycaemia and diabetes mellitus with
and without short-term insulin treatment on myocardial ischaemic late preconditioning in the rabbit heart in vivo. Pflügers Arch 2003; 446: 175–82. 78. Priebe H-J. Triggers of perioperative myocardial ischaemia and infarction. Br J Anaesth 2004; 93: 9–20. 79. Cohn SL, Goldman L. Preoperative risk evaluation and perioperative management of patients with coronary artery disease. Med Clin N Am 2003; 87: 111–36. 80. Grayburn PA, Hillis LD. Cardiac events in patients undergoing non-cardiac surgery: shifting the paradigm from non-invasive risk stratification to therapy. Ann Intern Med 2003; 138: 506– 11. 81. Mukherjee D, Eagle KA. Perioperative cardiac assessment for non-cardiac surgery: eight steps to the best possible outcome. Circulation 2003; 107: 2771–4. 82. Henke PK, Blackburn S, Proctor MC, Stevens J, Mukherjee D, Rajagopalin S, Upchurch Jr GR, Stanley JC, Eagle KA. Patients undergoing infrainguinal bypass to treat atherosclerotic vascular disease are under-prescribed cardioprotective medications: effect on graft patency, limb salvage, and mortality. J Vasc Surg 2004; 39: 357–65.
1 Perioperative intravenous fluid therapy JOACHIM BOLDT Introduction Perioperative volume deficits are common. Hypovolaemia may lead to maldistribution of the blood flow, which prevents the circulation fulfilling its nutritive role. Many manifestations of organ failure after seemingly successful primary resuscitation may result from peripheral (micro-) circulatory derangements (Fig. 1.1). Restoration of ‘normal’ systemic haemodynamics is not necessarily accompanied by restoration of organ and tissue perfusion. During low output states the organism compensates for perfusion deficits by redistribution of the blood flow to vital organs (e.g. the heart and brain) at the expense of under-perfusion of non-vital organs (the splanchnic bed and kidney). Inflammatory mediators and vasoactive substances are released in this situation and contribute to the development of impaired perfusion. Activation of the sympathetic nervous system and the renin—angiotensin system are compensatory mechanisms for maintaining peripheral perfusion. Although such compensatory neurohumoral activation is beneficial at first, it may however contribute to an adverse outcome in the hypovolaemic surgical patient, even after initially successful volume resuscitation. Controversy continues about what constitutes the ideal volume replacement regimen. Crystalloids (e.g. normal saline and Ringer’s lactate), the natural colloid human albumin, synthetic colloids (dextrans, gelatins and hydroxyethyl starch preparations) (Fig. 1.2) and hypertonic solutions are all available for treating perioperative volume deficits. All plasma substitutes have their merits and demerits. Some well-known side effects (e.g. bleeding following the administration of dextrans) considerably limit the routine use of some of them. A new controversy surrounds the adequate amount of perioperative fluid administration. For decades the philosophy has been to keep the patient normovolaemic (‘well hydrated’) in the perioperative period. Newer findings now suggest that keeping the patient ‘dry’ may improve outcomes. Overall evidence indicates that an adequate volume replacement therapy is a cornerstone in the management of the surgical patient. It is a prerequisite in the © Atlas Medical Publishing Ltd
Perioperative intravenous fluid therapy
17
Fig. 1.1 Hypovolaemia-related changes in the macrocirculation result in a deteriorated microcirculation with deleterious consequences for tissue perfusion and oxygenation. This may result in organ dysfunction or even multiple-organ failure. efforts at improving organ function and reducing patient morbidity and possibly even mortality. However, in a prospective review of 111 consecutive in-hospital deaths in trauma patients, inadequate fluid resuscitation was the most common management failure |1|.
The year in anaesthesia and critical care
18
What is the ideal substance for perioperative fluid therapy? Haemodynamic efficacy Due to their different physico-chemical characteristics the various solutions differ in their haemodynamic efficacy. After infusion, crystalloids rapidly shift from the
Fig. 1.2 The physico-chemical characteristics of hydoxyethyl starch solutions have changed over the years resulting in different generations of hydroxyethyl starch solutions with different concentrations (3, 6 and 10%), mean molecular weights (70, 200 or 130 kDa) and molar substitutions (e.g. 0.5, 0.62 or 0.7). intravascular to the interstitial compartment and subsequently possess only a limited volume-replacing capability. Consequently, if haemodynamic stability is to be guaranteed, crystalloids have to be administered at three to five times the volume lost. Due to the subsequent interstitial dilution the interstitial colloid oncotic pressure decreases, resulting in interstitial oedema formation. According to their widely differing colloid oncotic pressures, colloids are separated into hypo-oncotic (e.g. 3.5% gelatine and 4% albumin), iso-oncotic (e.g. all 6% hydroxyethyl starch preparations) and hyperoncotic (e.g. 10% hydroxyethyl starch preparation, 10% dextran and 20% human albumin). Colloids also differ in their water-binding capacity.
Perioperative intravenous fluid therapy
19
3.5% urea-linked gelatin is as effective as 6% HES 200/0.5 for volume management in cardiac surgery patients Van der Linden PJ, De Hert SG, Daper A, et al. Can J Anaesth 2004; 51: 236–41 BACKGROUND. The efficacy of volume expansion with either 3.5% gelatine (25.8±4.8 ml/kg) or 6% hydroxyethyl starch 200/0.5 (24.5±6.0 ml/kg) was assessed in patients undergoing cardiac surgery. The second objective was to compare blood loss and the allogeneic blood transfusion exposure rate. INTERPRETATION. Hydroxyethyl starch was not associated with a better plasma expansion than gelatine, but resulted in a higher need for allogeneic blood transfusion.
Fig. 1.3Volume effects of gelatine and a second-generation hydroxyethyl starch solution. Source: modified from |2| and |3|). Comment The finding of comparable haemodynamic effects is surprising because several previous studies have shown a lower volume-replacing efficacy of gelatine than of hydroxyethyl starch (Fig. 1.3). The molecular weights of gelatins range from 5000 to 50000 Da, with a weight-average molecular weight of 30000 to 35000 Da. As the molecular weight is lower than the renal threshold, gelatins are rapidly cleared from the bloodstream by glomerular filtration, resulting in the shortest intravascular halflife (less than 2 h) of all colloids. The finding of a significantly higher total blood loss in the hydroxyethyl starch group (11.0±7.8 ml/kg) than in the gelatine group (8.7±4.0 ml/kg) is equally surprising. A previous meta-analysis showed that blood loss in cardiac surgery is comparable during
The year in anaesthesia and critical care
20
volume therapy with either 6% hydroxyethyl starch 200/0.5 or human albumin, the ‘gold standard’ of volume therapy that did not negatively affect coagulation |4|. Side effects of plasma substitutes used for treating perioperative hypovolaemia Bleeding For numerous reasons the surgical patient is at increased risk of developing coagulation abnormalities perioperatively, resulting in an increased bleeding tendency. All fluids used for volume replacement lower the plasma concentration of clotting proteins by haemodilution. Some fluids possess additional specific (negative) effects on haemostasis (e.g. dextrans). Albumin and gelatins appear to be almost void of relevant negative effects on haemostasis. The continued reluctance to use hydroxyethyl starch for volume replacement is based on reports of abnormal coagulation and the increased bleeding tendency associated with its use |4, 5|. However, such findings were probably related to the use of first-generation, high molecular weight (mean molecular weight 450 kDa), high degree of molar substitution (>0.62) preparations. Recent studies using newer hydroxyethyl starch preparations have shed new light on this issue. A modified first-generation high molecular weight hydroxyethyl starch (Hextend™) (molar substitution 0.7, weight-aver age molecular weight approximately 670 kDa and mean molecular weight 550 kDa) impaired coagulation less than conventional first-generation high molecular weight hydroxyethyl starch (Hetastarch) |5|. Similarly, a third-generation, low molecular weight (molecular weight 130 kDa), low substitution (molar substitution 0.4) hydroxyethyl starch preparation |6| appeared to be associated with less adverse effects on coagulation and post-operative bleeding |7–12|.
Inhibition of platelet function by hydroxyethyl starch solutions in chronic pain patients undergoing peridural anesthesia Scharbert G, Deusch E, Kress HG, Greher M, Gustorff B, Kozek-Langenecker SA. Anesth Analg 2004; 99: 823–7 BACKGROUND. The effects of 6% hydroxyethyl starch (molecular weight 130 kDa and molar substitution 0.4), 6% hydroxyethyl starch (molecular weight 200 kDa and molar substitution 0.62) and lactated Ringer’s solution (10 ml/kg each administered intravenously for 30 min) on platelet function (using the platelet function analyser PFA-100′) and haemodynamics were studied in patients with chronic low back pain scheduled for peridural analgesia. INTERPRETATION. Low and medium molecular weight hydroxyethyl starch solutions inhibited platelet function. This effect was more pronounced during administration of hydroxyethyl starch 200/0.62 than administration of hydroxyethyl starch 130/0.4.
Perioperative intravenous fluid therapy
21
Comment Although the changes in the platelet function analyser closure times (a marker for platelet function) after infusion of 6% hydroxyethyl starch 130/0.4 were significant, they were small and the mean data remained within the normal range. In contrast, the changes induced by hydroxyethyl starch 200/0.62 were more pronounced and sometimes exceeded normal values. Interestingly, previous studies from the same group of investigators |7| and from other investigators |8–12| have not reported adverse effects of the newest generation of hydroxyethyl starch (hydroxyethyl starch 130/0.4) on platelet function |7|, haemostasis |7–12| and bleeding |7–12|. Unfortunately, the number of patients in this study by Scharbert et al. was too small to allow valid con-clusions regarding the effects of hydroxyethyl starch 130/0.4. Based on available data, it seems unnecessary to avoid hydroxyethyl starch 130/0.4 in patients scheduled for epidural anaesthesia.
The effects of high molecular weight hydroxyethyl starch solutions on platelets Deusch E, Thaler U, Kozek-Langenecker SA. Anesth Analg 2004; 99: 665–8 BACKGROUND. The anti-platelet effect of a novel high molecular weight, balanced hydroxyethyl starch preparation (Hextend™) was studied in vitro. For this purpose the availability of the glycoprotein llb-llla receptor was assessed on nonstimulated and on agonist-induced platelets using flow cytometry. INTERPRETATION. Hextend™ showed an unexpected platelet-stimulating effect that is unique among the currently available hydroxyethyl starch preparations. This effect may be induced partly by its solvent containing calcium chloride dihydrate. Comment The adverse effects of hydroxyethyl starch on haemostasis appear to depend on its mean molecular weight and molar substitution. Higher values of these factors lead to greater expected adverse effects |13–15|. Most reports of impaired haemostasis are based on the study of first-generation high molecular weight, highly substituted hydroxyethyl starch (Hetastarch). Infusion of such a hydroxyethyl starch preparation may result in a type I Von Willebrand-like syndrome with reduced factor VIII coagulant activity and decreased Von Willebrand’s factor VIII-related antigen and factor VIII-related ristocitin cofactor |13, 14|. In addition, large amounts of high molecular weight hydroxyethyl starch caused platelet swelling and reduced platelet adhesion |13, 14|. In contrast, low molecular weight hydroxyethyl starch influences the concentrations of VIII-related antigen and VIII-related ristocitin cofactor significantly less than high molecular weight hydroxyethyl starch |13, 14|. The hydroxyethyl starch in Hextend™ is a high molecular weight hydroxyethyl starch with a very high molar substitution (0.7). It is surprising that small amounts of a solvent would be able to blunt the negative effects of this hydroxyethyl starch preparation on
The year in anaesthesia and critical care
22
haemostasis. Other investigators were unable to confirm that modification of a high molecular weight hydroxyethyl starch (Hextend™) eliminates the negative effects on coagulation |16|. Furthermore, the clinical relevance of such in vitro findings has to be questioned: in vivo endothelial cell and neurohumoral function markedly modify the coagulation process and, subsequently, interact with the effects of plasma substitutes on haemostasis.
Differential platelet receptor expression following hydroxyethyl starch infusion in thrombocytopaenic orthotopic liver transplantation recipients Jüttner B, Kuse ER, Elsner HA, et al. Eur J Anaesthesiol 2004; 21: 309–13 BACKGROUND. After orthotopic liver transplantation, patients present with thrombocytopenia and associated bleeding problems which may be aggravated by the interaction of hydroxyethyl starches with platelets. Surface expression of glycoprotein llb/llla and P-selectin were quantified by flow cytometry and the percentage of platelet—leukocyte complexes. INTERPRETATION. Infusion of 6% hydroxyethyl starch 200/0.5 in clinically relevant doses (10 ml/kg) did not alter glycoprotein llb/llla expression in thrombocytopenic patients with pre-existing platelet dysfunction after orthotopic liver transplantation. Comment In contrast to the in vitro study discussed above |17|, this in vivo study did not demonstrate any adverse effect of 6% hydroxyethyl starch 200/0.5 on platelet function. Infusion of this preparation may possibly exert a beneficial effect on microvascular graft perfusion, resulting from the haemodilution and reduced P-selectin expression. The latter may, in turn, reduce leukocyte—platelet complexes and endothelial adhesion.
Volume efficacy and reduced influence on measures of coagulation using hydroxyethyl starch 130/0.4 (6%) with an optimised in vivo molecular weight in orthopaedic surgery: a randomised, double-blind study Jungheinrich C, Sauermann W, Bepperling F, Vogt NH. Drugs R D 2004; 5: 1–9 BACKGROUND. The differences between different hydroxyethyl starch preparations on blood coagulation may partly be related to their different in vivo molecular weights. INTERPRETATION. Hydroxyethyl starch 130/0.4 and hydroxyethyl starch 200/0.5 showed comparable volume efficacy. However, sensitive coagulation parameters returned more rapidly to normal in the hydroxyethyl starch 130/0.4-treated patients than in the hydroxyethyl starch 200/0.5-treated patients.
Perioperative intravenous fluid therapy
23
Comment This double-blind study demonstrated that hydroxyethyl starch 130/0.4 is superior to the second-generation hydroxyethyl starch 200/0.5 with regard to certain coagula-tion data: in contrast to the hydroxyethyl starch 200/0.5 group, in the hydroxyethyl starch 130/0.4 group factor VIII and Von Willebrand factor had returned to near baseline values by 5 h post-operatively. Lower in vivo molecular weight and more rapid excretion of hydroxyethyl starch 130/0.4 are the likely explanations for the smaller influence on coagulation. Consistent with these findings, there was a (statistically non-significant) trend towards lower blood loss in the hydroxyethyl starch 130/0.4 group. Other investigators have confirmed these beneficial effects |18, 19|.
Hydroxyethyl starch as a priming solution for cardiopulmonary bypass impairs hemostasis after cardiac surgery Kuitunen AH, Hynynen MJ, Vahtera E, Salmenpera MT. Anesth Analg 2004; 98: 291–7 BACKGROUND. Hydroxyethyl starch is often reported to impair coagulation and increase the bleeding tendency in cardiac surgery patients. INTERPRETATION. A dose of 20 ml/kg of low molecular weight hydroxyethyl starch (molecular weight 120 kDa and molar substitution 0.7) and high molecular weight hydroxyethyl starch (molecular weight 400 kDa and molar substitution 0.7) added to the cardiopulmonary bypass prime compromised haemostasis after cardiac surgery more than equal amounts of 4% human albumin. This effect appeared to be related to the formation of a less stable thrombus. Comment Thrombelastogram was used for assessing the influence of the three plasma substitutes on coagulation. The reaction time (r) and coagulation time (r+k) did not differ between the groups. Only the maximal amplitude and speed of solid clot formation (alpha angle) were different between the hydroxyethyl starch- and albumintreated patients. However, the normal range of thrombelastogram values was not provided and the differences were rather small (e.g. maximal amplitude values of 46 and 40 for albumin and hydroxyethyl starch 120/0.7, respectively). Bleeding did not differ significantly between the albuminand hydroxyethyl starch 120/0.7-treated patients, but was higher in the hydroxyethyl starch 400/0.7 group, confirming data from a meta-analysis |4|. The results of this study suggest that high molecular weight, high substitution hydroxyethyl starch preparations should be avoided in cardiac surgery. Additional effects of plasma substitutes Besides its haemodynamic effects, those on organ perfusion and microcirculation may be as important when looking for the optimal volume replacement strategy |20|. However,
The year in anaesthesia and critical care
24
microcirculation, organ perfusion and tissue oxygenation are difficult to assess in humans. Unspecific surrogates are mostly used (e.g. gastromucosal tonometry [pHi]). Whether such surrogates reflect tissue perfusion and correlate with outcomes remains to be determined. In addition to improving organ perfusion and the microcirculation, optimization of the intravascular volume may have an important impact on immune responses |21|. In this regard, the amount rather than the composition of the administered fluid is assumed to be the main determinant. However, only a few studies have focused on this aspect of volume replacement therapy.
The effects of hydroxyethyl starch on lung capillary permeability in endotoxic rats and possible mechanisms Tian J, Lin X, Guan R, Xu JG. Anesth Analg 2004; 98: 768–74 BACKGROUND. The effects of hydroxyethyl starch 200/0.5 on lung capillary permeability in endotoxic rats (induced by lipopolysaccharide) were studied. INTERPRETATION. Hydroxyethyl starch 200/0.5 reduced increased lung capillary permeability in endotoxaemia, most likely due to an anti-inflammatory effect. Comment In this investigation in rats, hydroxyethyl starch 200/0.5 at doses of 3.75 and 7.5 ml/kg ameliorated lipopolysaccharide-induced increases in lung capillary permeability. Hydroxyethyl starch inhibited lung neutrophil accumulation, cytokine-induced neutrophil chemoattractant protein, NF-κB activation and CD 11b expression in a dose-dependent manner. Lack of immunosuppression by synthetic colloidal fluids has previously been suggested by findings of unimpaired T-cell activation and mitogenic response during the administration of dextran, gelatins or hydroxyethyl starch 450/0.7 |22, 23|. Furthermore, hydroxyethyl starch did not affect the time- and dose-dependent generation of a chemotactic cytokine and did not by itself induce cytokine generation or change human monocyte chemotaxis and spontaneous migration |24|. In patients undergoing minor urological surgery, different hydroxyethyl starch preparations (hydroxyethyl starch 70/0.5, 200/0.5 and 450/0.7) did not affect the phagocytic activity of polymorphonuclear neutrophils (PMNs). In contrast, gelatine decreased the phagocytic capacity of both neutrophils and monocytes |25|. Evidence would thus suggest that hydroxyethyl starch might be used in patients at increased risk of inflammatory response.
Influence of different volume replacement strategies on inflammation and endothelial activation in the elderly undergoing major abdominal surgery
Perioperative intravenous fluid therapy
25
Boldt J, Ducke M, Kumle B, Papsdorf M, Zurmeyer EL. Intensive Care Med 2004; 30: 416–22 BACKGROUND. Major surgery activates an inflammatory cascade, which in turn triggers counter-regulatory mechanisms that are aimed at controlling the intensity of the inflammatory response. Different intravenous fluids may have different effects on the inflammatory response. INTERPRETATION. In elderly surgical patients, volume replacement with a thirdgeneration hydroxyethyl starch preparation (6% hydroxyethyl starch 130/0.4) was associated with a lesser inflammatory response (lower concentrations of interleukins) and endothelial injury/activation (lower concentrations of circulating adhesion molecules) compared to volume replacement with either Ringer’s lactate or normal saline. Comment The reasons for the different inflammatory response and endothelial injury/activation between the hydroxyethyl starch- and crystalloid-treated patients remain unclear. The microvasculature is a key ‘battleground’ for the inflammatory response, with evidence of a central role for the endothelium in modulating inflammation. The hydroxyethyl starch molecule may exert a direct, substance-specific effect on endothelial cells resulting in a decreased release of adhesion molecules |26|. The beneficial effect may also be due to improved organ function (e.g. via improved splanchnic perfusion) and microcirculation.
Intra-operative colloid administration reduces post-operative nausea and vomiting and improves post-operative outcomes compared with crystalloid administration Moretti EW, Robertson KM, El-Moalem H, Gan TJ. Anesth Analg 2003; 96: 611–17 BACKGROUND. The choice of intravenous fluid may affect the post-operative patient recovery profile. INTERPRETATION. In patients undergoing major elective non-cardiac surgery, administration of a non-balanced, standard hydroxyethyl starch solution (1301±1079 ml) or a balanced hydroxyethyl starch solution (1448±759 ml) was associated with a lower incidence of nausea and vomiting, use of rescue anti-emetics, severe pain, periorbital oedema and double vision compared to crystalloid administration (5946±1909 ml Ringer’s lactate). Comment This is the first study to suggest that the choice of intravenous fluid may affect postoperative recovery and that colloids may, in this regard, be superior to crystalloids. However, additional studies are required before a final conclusion is possible.
The year in anaesthesia and critical care
26
Does albumin have a place for treating the hypovolaemic patient? Albumin is still widely considered the ‘best’ solution, particularly in the critically ill with compromised organ function. However, cost containment is becoming increasingly important in medical decision making. Although comparison of costs is very difficult because of local differences in medico-economic systems, albumin is considerably more expensive than modern synthetic colloids |27|. Consequently, albumin use has declined over the past few years. Not surprisingly, the albumin industry launched an international promotion programme worth US$ 2.2 million in order to promote albumin |28|, triggering the publication of several studies that addressed the value of albumin, several of them, however, being reviews, retrospective analyses or meta-analyses. The Cochrane Injuries Group published a meta-analysis (including 30 studies with a total of 1419 patients) that compared albumin with other intravenous volume replacement fluids (crystalloids and synthetic colloids) |29|. The use of albumin was associated with an overall excess mortality of 6.8% (or approximately six additional deaths for every 100 patients treated with albumin). It is beyond the scope of this review to address the topic of correction of hypoproteinaemia in the critically ill intensive care patient.
Safety of human albumin—serious adverse events reported worldwide in 1998–2000 Vincent JL, Wilkes MM, Navickis RJ. Br J Anaesth 2003; 91: 625–30 BACKGROUND. All serious adverse event reports and total doses of albumin distributed worldwide from the beginning of 1998 to the end of 2000 were reviewed. INTERPRETATION. Within a total distribution group of 1.62×107 units of albumin, 198 non-fatal and 13 fatal serious adverse events were reported, amounting to an incidence of 5.28 per 106 doses and 4.65 per 106 doses for serious non-fatal and fatal adverse events, respectively. No patient death was attributed directly to albumin administration. Comment This study was supported by a grant from the Plasma Protein Therapeutics Association. The findings are in complete contradiction to those of the meta-analysis by the Cochrane Injuries Group, which suggested that albumin might be increasing mortality |29|. There was no differentiation between treatments for hypovolaemia or hypoproteinaemia.
Volume expansion with albumin decreases mortality after coronary artery bypass graft surgery
Perioperative intravenous fluid therapy
27
Sedrakyan A, Gondek K, Paltiel D, Elefteriades JA. Chest 2003; 123: 1853–7 BACKGROUND. As albumin and non-protein colloids (starches, dextran and others) are used frequently as blood volume expanders in coronary artery bypass graft (CABG) surgery, this study aimed at determining differences between colloids with regard to patient characteristics and mortality rates. INTERPRETATION. In this retrospective analysis of 19 578 patients undergoing CABG surgery, mortality was lower in the albumin (n=8084) than in the non-protein colloid group (2.47 versus 3.03%) (P=0.02), which included first-generation hydroxyethyl starch 450/0.7 and dextrans. Albumin was associated with 25% lower odds of mortality compared to non-protein colloid use (odds ratio 0.80 and 95% confidence interval [Cl] 0.67–0.96). Comment This retrospective analysis had several limitations. No information regarding preoperative myocardial infarction, congestive heart failure and renal failure was provided. As the authors did not differentiate between the non-protein colloids (hydroxyethyl starch versus dextrans), it is impossible to determine which group had the highest mortality. Both nonprotein colloids used in this study are known to interfere with the coagulation cascade resulting in an increased bleeding tendency |30|. Accordingly, the blood transfusion requirements were three times lower in the albumin than in the non-protein colloid groups. Use of non-leukocyte-depleted packed red blood cells is associated with several adverse effects |31, 32|, including increased inflammation and impaired coagulation. Thus, the higher number of transfusions in the non-protein colloid groups might have been responsible for the differences in mortality. It is conceivable that, had albumin been compared with a modern colloid (e.g. hydroxyethyl starch 130/0.4), the results might have been different.
A comparison of albumin and saline for fluid resuscitation in the intensive care unit Finfer S, Bellomo R, Boyce N, French J, Myburgh J, Norton R; SAFE Study Investigators. N Engl J Med 2004; 350: 2247–56 BACKGROUND. Volume therapy using 4% albumin was compared with normal saline solution in approximately 7000 critically ill intensive care unit (ICU) patients in a multicentre, randomized double-blind trial. INTERPRETATION. In a heterogenous population of ICU patients (43% surgical and 57% medical patients), fluid resuscitation with either 4% albumin or normal saline resulted in comparable 28-day morbidity and mortality. Length of stay in the ICU and in the hospital as well as days on the ventilator and days on renal replacement therapy were also comparable.
The year in anaesthesia and critical care
28
Comment Although this study in hypovolaemic ICU patients was not directly related to the topic of perioperative fluid replacement therapy, it deserves mentioning because it is the largest available study to date on volume replacement strategies. Although the results seriously question any advantage of albumin over crystalloids in the critically ill, the results are not necessarily applicable to all colloids. It is questionable whether 4% albumin is the ideal colloid for treating the hypovolaemic critically ill. Moreover, the criteria for volume replacement were not strictly defined (no ‘goal-directed’ volume therapy) and the ratio of administered albumin to crystalloids was 1:1.4. Considering usually accepted ratios of between 1:3 and 1:5, it is possible that the patients receiving crystalloids were insufficiently volume resuscitated. Thus, even the SAFE (Saline versus Albumin Fluid Evaluation) Study is unable to settle the ‘age-old’ crystalloid versus colloid debate. Hypertonic volume replacement strategy Enthusiasm has surrounded the treatment of hypovolaemic shock with hypertonic or hypertonic-hyperoncotic solutions |33|. The sodium concentrations of these solutions range from 3 to 7.5%. Hypertonic solutions appear to improve cardiovascular function by multiple mechanisms. 1. The displacement of tissue fluid into the blood compartment with subsequent plasma volume expansion (main mechanism). 2. Direct vasodilatory effects in the systemic and pulmonary circulation. 3. A reduction in venous capacitance. 4. Positive inotropic effects through direct actions on myocardial cells. Due to the hypertonicity of the solutions, only a small volume of fluid (approximately 4 ml/kg) is required for restoring cardiovascular function effectively (the principle of ‘small volume resuscitation’). The initial improvement in cardiovascular function (e.g. increase in cardiac output) seems to be mediated by the hypertonicity rather than by the solute composition of the solution |33|. As the beneficial effects of hypertonic solutions seem to be rather transient, hypertonic solutions were mixed with colloids (dextran or hydroxyethyl starch), which significantly prolonged their action |33|.
Hypertonic versus near isotonic crystalloid for fluid resuscitation in critically ill patients Bunn F, Roberts I, Tasker R, Akpa E. Cochrane Database Syst Rev 2004; 3: CD 002045 BACKGROUND. Due to their physico-chemical properties, rapid administration of small volumes of hypertonic solutions may be superior to the administration of crystalloids in expanding blood volume and elevating blood pressure. INTERPRETATION. This meta-analysis reviewed 14 randomized trials (including a total of 956 patients) that compared hypertonic with isotonic and near isotonic
Perioperative intravenous fluid therapy
29
crystalloids in patients with trauma, burns or undergoing surgery. The pooled relative risk for death was 0.84 in trauma patients (95% Cl 0.69–1.04), 1.49 in burn patients (95% Cl 0.56–3.95) and 0.51 in patients undergoing surgery (95% Cl 0.09–2.73). Comment This review did not provide sufficient data for allowing final judgement as to whether a hypertonic crystalloid is better than isotonic and near-isotonic crystalloids for the resuscitation of patients with trauma, burns or those undergoing surgery. As mortality was not the primary outcome variable in most studies on volume replacement regimens, the validity of meta-analyses that focus on mortality has been questioned |34|. Outcome may also be defined as the time spent on ventilator support, incidence of organ dysfunction, overall costs or length of stay in hospital/ICU. We do not need more metaanalyses that pool old-to-very-old data, but well-controlled studies in well-defined groups of patients (separated by disease and type of surgery) that compare clearly defined volume replacement strategies (crystalloids, albumin, gelatins, dextrans and different hydroxyethyl starch preparations) and outcome variables. Is less more or is more better? The ‘dry versus wet’ philosophy in managing the surgical patient has spiked enormous interest in recent years |35|. Intravenous fluid overload in the perioperative period is associated with decreased tissue oxygen tension, delayed recovery from gastrointestinal function and even adverse outcomes |36, 37|. On the other hand, inadequate volume resuscitation of the critically ill patient undergoing major surgery may equally result in organ dysfunction and an adverse outcome. The microcirculation represents the final common pathway of the respiratory and circulatory system. Under certain conditions (e.g. shock, ischaemia and cardiopulmonary bypass), interactions between the endothelium and cellular elements of the blood and endothelial swelling prevent effective microcirculatory flow |38|. Poor capillary perfusion impairs oxidative killing in the wound and triggers additional release and activation of mediators that subsequently promote cell adhesion and vaso-constriction. Increased permeability (‘capillary leakage’) causes endothelial swelling in the capillaries. The deteriorated microcirculation initiates a vicious cycle of progressive tissue damage that may ultimately lead to the development of multiple-organ failure.
Effects of intravenous fluid restriction on post-operative complications: comparison of two perioperative fluid regimens: a randomized assessor-blinded multicenter trial Brandstrup B, Tønnesen H, Beier-Holgersen R, and the Danish Study Group on Perioperative Fluid Therapy. Ann Surg 2003; 238: 641–8
The year in anaesthesia and critical care
30
BACKGROUND. As the amount of perioperatively administered fluid may affect outcomes, the effects of a restricted versus a standard intravenous fluid regimen were compared in patients undergoing colorectal resection. INTERPRETATION. In this randomized observer-blinded multicentre trial in 141 patients restricted perioperative fluid administration (median 2740 ml and range 1100– 8050 ml versus median 5388 ml and range 2700–11 083 ml in the control group) was associated with fewer post-operative complications. The mortality was comparable between the groups. Comment The total number of major complications (see Table 1.1) was significantly lower in the fluid-restricted than in the standard group. The main limitation of the study was the administration of fixed amounts of fluid rather than ‘goal-directed’ fluid administration. Due to the fixed-volume replacement protocol, some patients were likely to have been fluid overloaded. The range of values of administered volume indicates significant overlap between groups. In fact, 15% of patients in the Restricted group’ received more volume, while 24% of the ‘standard group’ received less volume than dictated by the protocol. No haemo-
Table 1.1 Major complications between the two groups: only differences with more than three complications are listed Complication
‘Restricted group’
‘Standard group’
Sepsis
0
4
Bleeding
1
5
Pulmonary oedema
0
4
Bradycardia
0
4
Source: Brandstrup et al. (2003).
dynamic data (e.g. central venous pressure and cardiac output) were provided. Goaldirected volume therapy in the critically ill |39| and pre-operative increases in cardiac output to ‘supra-normal’ values (particularly in critically ill patients undergoing complex surgical procedures) |40, 41| may improve outcome.
Supplemental perioperative fluid administration increases tissue oxygen pressure Arkilic CF, Taguchi A, Sharma N, et al. Surgery 2003; 133: 49–55 BACKGROUND Hypoperfused tissues experience low oxygen tension that is
Perioperative intravenous fluid therapy
31
insufficient for allowing adequate oxidative killing or wound healing. The hypothesis was tested that supplemental perioperative fluid administration increases tissue perfusion and tissue oxygen pressure. INTERPRETATION. ‘Aggressive’ fluid management (10 ml/kg of crystalloids given prior to surgery and 6–18 ml/kg/h given intra-operatively and during the first postoperative hour) in patients undergoing colon resection resulted in higher muscle tissue oxygenation (measured at the patient’s upper arm) than conservative fluid management (8–10 ml/kg/h of crystalloids). The findings indicate that supplemental perioperative fluid administration may significantly increase tissue perfusion and tissue oxygen partial pressure. Optimizing tissue perfusion will require providing more fluid than may be indicated by the usual clinical criteria or routine invasive monitoring. Comment Improved tissue perfusion and oxygenation can be expected to be of particular benefit in patients who are at increased risk of reduced tissue perfusion (e.g. patients with diabetes mellitus). Even moderate hypovolaemia (that is sometimes difficult to recognize clinically) may be associated with a seriously impaired microcirculation. It is important to note that no colloids were administered throughout the study. However, colloids may improve tissue oxygenation more than crystalloids |42, 43|. It remains to be determined whether muscle tissue oxygenation reflects tissue oxygenation in other, clinically more relevant organs. Conclusion The beneficial and adverse effects of the various types of fluid continue to be debated |44|. Besides the natural colloid albumin, several synthetic colloids are increasingly being used as plasma substitutes. A colloid versus colloid debate can now be added to the ageold crystalloid versus colloid debate. Several messages can be extracted from the recent publications on perioperative volume replacement therapy. 1. Crystalloids may adversely affect coagulation (hypercoagulability) and the metabolic state (acidosis). They seem to lack beneficial effect on the microcirculation and organ perfusion. Even ‘aggressive’ administration of large amounts of crystalloids may not adequately restore microcirculatory blood flow. 2. Hydroxyethyl starch is the most studied plasma substitute. Conflicting results regarding its effects on various organ perfusions and functions are mostly due to the use of different hydroxyethyl starch preparations and varying study designs. 3. Available evidence does not support the use of albumin over less expensive alternatives in treating perioperative hypovolaemia. 4. It remains to be determined whether the quantity or composition of the intravenous fluid is the main determinant of the treatment effectiveness in the hypovolaemic critically ill patient.
The year in anaesthesia and critical care
32
We are living in an era of meta-analyses and evidence-based medicine. At times this has produced more confusion than it has help in finding the optimal volume replacement regimen. Although the quality of study design has improved, standardized protocols for volume replacement are frequently still missing. The choice for or against a particular intravenous fluid should take into account its effects on tissue perfusion, organ perfusion and function, endothelial function, inflammatory response and various other physiological variables. The challenge will be to determine which patients benefit most from which type of volume replacement strategy. References 1. Deane SA, Gaudry PL, Woods P, Cass D, Hollands MJ, Cook RJ, Read C. The management of injuries—a review of deaths in hospital. Aust NZ J Surg 1988; 58: 463–9. 2. Kroll W, Polz W, Colombo T, Steindorfer P. Degree of substitution and volume expanding effect of various medium molecular weight hydroxyethyl starch solutions. Wien Klin Wochenschr 1994; 106: 416–21. 3. Kroll W, Gerner P, Colombo T, Ramschak H, Hinghofer-Szalkay H, List WF. The effect of 6% HES 200/0.6–0.66 on plasma volume and blood coagulation. Infusionsther Transfusionsmed 1992; 19: 171–80. 4. Wilkes MM, Navickis RJ, Sibbald WJ. Albumin versus hydroxyethyl starch in cardiopulmonary bypass surgery: a meta-analysis of post-operative bleeding. Ann Thorac Surg 2001; 72: 527–33. 5. Martin G, Bennett-Guerrero E, Wakeling H, Mythen MG, El-Moalem H, Robertson K, Kucmeroski D, Gan TJ. A prospective, randomized comparison of thrombelastographic coagulation profile in patients receiving lactated Ringer’s solution, 6% hetastarch in a balancedsaline vehicle, or 6% hydroxyethyl starch in saline during major surgery. J Cardiothorac Vasc Anesth 2002; 16: 441–6. 6. Waitzinger J, Bepperling F, Pabst G, Opitz J, Müller M, Baron JF. Pharmacokinetics and tolerability of a new hydroxyethylstarch (HES) specification (HES 130/0.4) after single-dose infusion of 6% or 10% solution in healthy volunteers. Clin Drug Invest 1998; 16: 151–60. 7. Franz A, Bräunlich P, Gamsjäger T, Felfernig M, Gustorff B, Kozek-Langenecker SA. The effects of hydroxyethyl starches of varying molecular weights on platelet function. Anesth Analg 2001; 92: 1402–7. 8. Haisch G, Boldt J, Krebs C, Suttner S, Lehmann A, Isgro F. Influence of a new hydroxyethylstarch preparation (HES 130/0.4) on coagulation in cardiac surgical patients. J Cardiothorac Vasc Anesth 2001; 15: 316–21. 9. Haisch G, Boldt J, Krebs C, Kumle B, Suttner S, Schulz A. The influence of intravascular volume therapy with a new hydroxyethyl starch preparation (6% HES 130/0.4) on coagulation in patients undergoing major abdominal surgery. Anesth Analg 2001; 92: 565–71. 10. Boldt J, Haisch G, Suttner S, Kumle B, Schellhaas F. In vivo changes of hemostasis with a new modified, balanced hydroxyethyl starch preparation (Hextend). Br J Anaesth 2002; 89: 722–8. 11. Kasper SM, Meinert P, Kampe S, Gorg C, Geisen C, Mehlhorn U, Diefenbach C. Large-dose hydroxyethyl starch 130/0.4 does not increase blood loss and transfusion requirements in coronary artery bypass surgery compared with hydroxyethyl starch 200/0.5 at recommended doses. Anesthesiology 2003; 99: 42–7. 12. Jungheinrich C, Sauermann W, Bepperling F, Vogt NH. Volume efficacy and reduced influence on measures of coagulation using hydroxyethyl starch 130/0.4 (6%) with an optimised in vivo molecular weight in orthopaedic surgery: a randomised, double-blind study. Drugs R D 2004; 5: 1–9.
Perioperative intravenous fluid therapy
33
13. Treib J, Haass A, Pindur G, Seyfert UT, Treib W, Grauer MT, Jung F, Wenzel E, Schimrigk K. HES 200/0.5 is not HES 200/0.5—influence of the C2/C6 hydroxyethylation ratio of hydroxyethyl starch (HES) on hemorheology, coagulation and elimination kinetics. Thromb Haemost 1995; 74(6): 1452–6. 14. Treib J, Baron JF, Grauer MT, Strauss RG. An international view of hydroxyethyl starches. Intensive Care Med 1999; 25: 258–68. 15. Treib J, Haass A, Pindur G. Coagulation disorders caused by hydroxyethyl starch. Thromb Haemost 1997; 78: 974–83. 16. Boldt J, Haisch G, Suttner S, Kumle B, Schellhaas F. In vivo changes of hemostasis with a new modified, balanced hydroxyetyhl starch preparation (Hextend). Br J Anaesth 2002; 89: 722–8. 17. Schabert G, Deusch E, Kress HG, Greher M, Gustorff B, Kozek-Langenecker SA. Inhibition of platelet function by hydroxyethyl starch solutions in chronic pain patients under-going peridural anesthesia. Anesth Analg 2004; 99: 823–7. 18. Gallandat Huet RCG, Siemons AW, Baus D, Van Rooyen-Butijn WT, Haagenaars JA, Van Oeveren W, Bepperling F. A novel hydroxyethyl starch (Voluven™) for effective perioperative plasma volume substitution in cardiac surgery. Can J Anaesth 2000; 47: 1207–15. 19. Langeron O, Doelberg M, Ang ET, Bonnet F, Capdevila X, Coriat P. Voluven, a lower substituted novel hydroxyethyl starch (HES 130/0.4), causes fewer effects on coagulation in major orthopedic surgery than HES 200/0.5. Anesth Analg 2001; 92: 855–62. 20. Funk W, Baldinger V. Microcirculatory perfusion during volume therapy. Anesthesiology 1995; 82: 975–82. 21. Wilson MA, Chou MC, Spain DA, Downard PJ, Qian Q, Cheadle WG, Garrison RN. Fluid resuscitation attenuates early cytokine mRNA expression after peritonitis. J Trauma 1996; 41: 622–7. 22. Sillett HK, Whicher JT, Trejdosiewicz LK. Effects of resuscitation fluids on T cell immune response. Br J Anaesth 1998; 81: 242–3. 23. Sillett HK, Whicher JT, Trejdosiewicz LK. Effects of resuscitation fluids on non-adaptive immune responses. Transfusion 1997; 37: 953–9. 24. Eastlund DT, Douglas MS, Choper JZ. Monocyte chemotaxis and chemotactic cytokine release after exposure of hydroxyethyl. Transfusion 1992; 32: 855–60. 25. Jaeger K, Jüttner B, Heine J, Ruschulte H, Scheinichen D, Piepenbrock S. Effects of hydroxyethyl starch and modified fluid gelatine on phagocytic activity of human neutrophils and monocytes—results of a randomized, prospective clinical study. Infus Ther Transfus Med 2000; 27: 256–60. 26. Collis RE, Collins PW, Gutteridge CN, Kaul A, Newland AC, Williams DM, Webb AR. The effect of hydroxyethyl starch and other plasma volume substitutes on endothelial cell activation; an in vitro study. Intensive Care Med 1994; 20: 37–41. 27. Grootendorst AF, Van Wilgenburg MG, De Laat PH, Van der Hoven B. Albumin abuse in intensive care medicine. Intensive Care Med 1988; 14: 554–7. 28. Roberts I, Bunn F. Egg on their faces. The story of human albumin solution. Eval Health Prof 2002; 25: 130–8. 29. Cochrane Injuries Group Albumin Reviewers. Human albumin administration in critically ill patients: systematic review of randomised controlled trials. BMJ 1998; 317: 235–40. 30. De Jonge E, Levi M. Effects of different plasma substitutes on blood coagulation: a comparative review. Crit Care Med 2001; 29: 1261–7. 31. Landers DF, Hill GE, Wong KG, Fox IJ. Blood transfusion-induced immunomodulation. Anesth Analg 1996; 82: 187–204. 32. Agarwal N, Murphy JG, Cayten CG, Stahl WM. Blood transfusion increases the risk of infection after trauma. Arch Surg 1993; 128: 171–7.
The year in anaesthesia and critical care
34
33. Kreimeier U, Messmer K. New perspectives in resuscitation and prevention of multiple organ system failure. In: Baethmann A, Messmer K (eds). Surgical Research: Recent Concepts and Results. Berlin, Heidelberg: Springer, 1987: pp 39–50. 34. Astiz ME, Rackow EC. Crystalloid-Colloid controversy revisited. Crit Care Med 1999; 27: 34– 5. 35. Kudsk KA. Evidence for conservative fluid administration following elective surgery. Ann Surg 2003; 238: 649–50. 36. Lobo DN, Bostock KA, Neal KR, Perkins AC, Rowlands BJ, Allison SP. Effect of salt and water balance on recovery of gastrointestinal function after elective colonic resection: a randomised controlled trial. Lancet 2002; 359: 1812–18. 37. Holte K, Kehlet H. Compensatory fluid administration for preoperative dehydration—does it improve outcome? Acta Anaesthesiol Scand 2002; 46: 1089–93. 38. Intaglietta M. Objectives for the treatment of the microcirculation in ischemia, shock, and reperfusion. In: Vincent JL (ed). Update in Intensive Care and Emergency Medicine. Berlin, Heidelberg, New York: Springer, 1990: pp 293–8. 39. Rivers E, Nguyen B, Havstad S and the Early Goal-Directed Therapy Collaborative Group. Early goal-directed therapy in the treatment of severe sepsis and septic shock. N Engl J Med 2001; 345: 1368–77. 40. Shoemaker WC, Appel PL, Kram HB. Role of oxygen debt in the development of organ failure sepsis, and death in high-risk surgical patients. Chest 1992; 102: 208–15. 41. Lobo SM, Salgado PF, Castillo VG, Borim AA, Polachini CA, Palchetti JC, Brienzi SL, De Oliveira GG. Effects of maximizing oxygen delivery on morbidity and mortality in high-risk surgical patients. Crit Care Med 2000; 28: 3396–404. 42. Lang K, Boldt J, Suttner S, Haisch G. Colloids versus crystalloids and tissue oxygen tension in patients undergoing major abdominal surgery. Anesth Analg 2001; 93: 405–9. 43. Standl T, Burmeister MA, Schroeder F, Currlin E, Schulte am Esch J, Freitag M, Schulte am Esch J. Hydroxyethyl starch (HES) 130/0.4 provides larger and faster increases in tissue oxygen tension in comparison with prehemodilution values than HES 70/0.5 or HES 200/0.5 in volunteers undergoing acute normovolemic hemodilution. Anesth Analg 2003; 96: 936–43. 44. Henry S, Scalea TM. Resuscitation in the new millennium. Surg Clin North Am 1999; 79: 1259–67.
2 Perioperative blood transfusion therapy TIM WALSH Introduction Approximately 40–50% of all blood transfusions are administered to surgical patients in the perioperative period |1|. Allogeneic blood transfusion was developed as a life-saving therapy for haemorrhagic shock, but many red cell transfusions are now administered either to anaemic normovolaemic patients or to patients who have lost blood but are not severely shocked. Anaesthetists, intensivists and surgeons must balance the benefits of transfusion against the risks of receiving allogeneic blood. The risks of transfusions have been increasingly well documented. Some of these risks have high public and political profiles even though they are rare or even unproven. The first two cases of probable transmission of variant Creutzfeld—Jakob disease (vCJD) through blood transfusion were reported in 2004 |2|. In the UK there are currently known to be 17 blood donors who later developed vCJD. Fifty recipients of blood from those donors have been identified, 18 of whom are still alive. Two of these 50 recipients are known as having become infected with vCJD: one died of the disease and the other died from unrelated causes but had histological signs of infection. In an attempt to minimize this risk all donated blood in the UK is leucodepleted prior to storage, at a cost of £70 million annually. New and emerging technologies could more than double this cost in the near future |3|. Two important editorials relating to blood transfusion were recently published in the British Medical Journal |3, 4|. McClelland and Contreras |3| argued that additional processing and testing of donated blood were resulting in exponential increases in cost, yet there is little high-quality evidence that blood transfusions are effective for many conditions. The authors noted that those studies that had compared restrictive with liberal transfusion strategies in surgical or critically ill patients had mostly found no benefit from liberal use of blood and, in some cases, harm. This finding was confirmed by the most recent Cochrane systematic review |5|. The authors argued that the huge costs of improving blood safety needed to be balanced by similar investment in clinical trials in order to determine how to use blood effectively. In another editorial Copplestone pointed out that the donor pool had been cut as a result of a precautionary approach driven by the vCJD disaster and that, unless clinicians learn to use © Atlas Medical Publishing Ltd
The year in anaesthesia and critical care
36
Table 2.1 Developments in the microbiological safety of blood transfusion in the UK since 1970 and projected further tests that are likely to be introduced during the next 5–10 years Decade
Test or restriction
Date of introduction
1970s
Hepatitis B tests
1970
1980s
HIV donor exclusions
1984
HIV antibody test
1985
Hepatitis C antibody test
1992
Hepatitis C PCR test
1997
AH UK plasma imported for fractionation
1999
HTLV1 antibody test
2001
HIV PCR test
2001
Universal leucodepletion of blood products
2001
Clinical plasma imported
2004
Bacterial diversion system during donation
2004
Pathogen-reduced clinical plasma
2002–2004
SARS exclusions
2003
West Nile virus exclusions
2003
Ban on previously transfused donors
2004
Selected West Nile virus PCR testing
2004
Selected HbcAb testing
2004
1990s
2000–2005
Likely future testing
Prion removal filtration Pathogen-reduced red cells and platelets Test for vCJD agent in blood donors
HbcAb, hepatitis B core antibody; HIV, human immunodeficiency vims; HTLV1, human T cell lymphoma virus 1; PCR, polymerase chain reaction; SARS, severe acute respiratory syndrome. Source: Copplestone (2005) |4|.
blood effectively, there maybe inadequate supplies for meeting current demands |4|. The editorial indicated that there was a real risk of cancellation of elective surgery unless action to implement blood-sparing measures is taken now. The current estimated cost of 1 unit of red cells is £120: this cost has doubled over the past 5 years suggesting that bloodsparing measures are likely to be increasingly cost-effective. The dramatic cost increase has resulted from donor exclusions and increased processing costs (Table 2.1).
Perioperative blood transfusion therapy
37
This chapter will examine some of the key papers concerning clinical transfusions published recently. The first section will examine recent publications concerning the effectiveness of blood transfusions, the second will address the complications of transfusions, including current evidence concerning the importance of leucodepletion and the third will consider three recent meta-analyses of transfusion-sparing techniques. Effectiveness of blood transfusion Broadly speaking, there are three approaches to investigating the effectiveness of blood transfusions. The first is to study population cohorts using epidemiological analyses for examining the association between blood transfusions and clinical outcomes in predefined patient groups. These have been widely used, but are often associated with major methodological problems. The following study by Rao et al. is an excellent example of this approach. The second approach and the gold standard is to carry out randomized controlled trials, but there are few such studies in the transfusion literature. The study by McIntyre et al. is a subgroup analysis of the Transfusion Requirements in Critical Care (TRICC) Trial, which is the highest quality evidence concerning transfusion triggers currently available |6|. The third approach is to use physiological oxygenation end-points for investigating the effectiveness of blood transfusions under certain conditions. The studies by Walsh et al. and Suttner et al. are recent well-performed studies that illustrate nicely the challenges and limitations of this approach.
Relationship of blood transfusion and clinical outcomes in patients with acute coronary syndromes Rao SV, Jollis JG, Harrington RA, et al. JAMA 2004; 292: 1555–62 BACKGROUND. It is unclear when blood transfusions should be given to patients with ischaemic heart disease. A previous retrospective cohort study of patients aged >65 years with acute myocardial infarction found an association between anaemia at hospital admission and higher mortality and a lower mortality among anaemic patients who received transfusions |7|. In this cohort study the association between blood transfusion and mortality was investigated among patients with acute coronary syndromes who developed bleeding or anaemia or both during their hospital course. INTERPRETATION. The authors analysed the 24112 patients enrolled in three international trials of patients with acute coronary syndromes, 10% of whom received at least one blood transfusion. The authors used complex statistical analyses with multiple models for adjusting for potential confounders, the effect of in-hospital events such as bleeding and procedures and treatment by time effects. They found that blood transfusion was associated with an increased risk of short-term mortality and myocardial infarction, which persisted after adjustment for patient characteristics, nadir haemoglobin (Hb) concentrations, bleeding and in-hospital procedures in a Cox proportional hazard model (Table 2 2) Transfusions were not associated with better outcomes when the nadir
The year in anaesthesia and critical care
38
haematocrit levels were 20–25% and were associated with worse outcomes when the values were >30%. Comment Given the prevalence of ischaemic heart disease, we need to know what level of anaemia is safe and when these patients should be transfused. The earlier retrospective cohort study by Wu et al. |7| suggested benefit from blood transfusion: however, their analysis was based only on admission haematocrit levels and made limited adjustments for confounders and time-dependent effects and the patients had low
Table 2.2 Unadjusted rates of outcomes and adjusted results of a Cox regression predicting 30day death and death or recurrent myocardial infarction using transfusion as a time-dependent covariate Hazard ratio (95% cl)* Underwent transfusion, no. (%) Outcomes Yes No P (n=2401) (n=21 value 711)
Adjusted Adjusted for for baseline transfusion characteristics† propensity
Adjusted tor baseline characteristics, bleeding and transfusion propensity and nadir haematocrit†
Death
192 (8.00) 669 (3.08)
<0.001 3.77 (3.14– 4.52)
3.54 (2.96–4.23)
3.94 (3.26–4.75)
MI
604 (25.16)
1771 (8.16)
<0.001 …
…
…
Composite (death/MI)
702 (29.24)
2176 <0.001 2.79 (2.45– (10.02) 3.18)
2.80 (2.45–3.19)
2.92 (2.55–3.35)
CI, confidence interval; MI, myocardial infarction. Ellipses indicate data not computed. * P<0.001 for all comparisons. † Baseline characteristics adjusted to include US vs non-US site, age, race, weight in Kilograms, diabetes mellitus, systolic blood pressure, diastolic blood pressure, heart rate at baseline, time from symptom onset to hospitalization, prior stroke, prior MI, sex, history of angina prior to qualifying episode, hypertension, hyperiipidaemia, family history of coronary artery disease, history of congestive heart failure, peripheral vascular disease, prior percutaneous coronary intervention, prior coronary artery bypass graft surgery, Killip class, baseline haematocrit, maximum creatine kinase ratio at baseline, chronic renal insufficiency, ST-segment elevation or depression on initial electrocardiogram, β-blocker use at baseline, calcium channel blocker use at baseline, nitrate use at baseline, and current smoking. Source: Rao et al. (2004).
Perioperative blood transfusion therapy
39
transfusion rates. The Rao et al. study used prospectively collected data from high-quality trials that were probably very accurate and made adjustments for many confounders and the time and transfusions in hospital. They also examined a possibly higher risk group of younger patients undergoing invasive procedures and at relatively high risk for bleeding. An excellent accompanying editorial discussed the possible meaning of these studies and offered sound advice |8|. Based on available evidence, patients with ischaemic heart disease should not be transfused if their Hb is >100 g/1 and should be transfused when their Hb is <70 g/1. In the presence of an acute coronary syndrome single units of red cells should be given to maintain a Hb in the range 80–90 g/1, bearing in mind the risk of fluid overload. What is really needed is a well-designed randomized controlled trial.
Is a restrictive transfusion strategy safe for resuscitated and critically ill trauma patients? McIntyre L, Hebert PC, Wells G, et al. J Trauma 2004; 57: 563–8 BACKGROUND. The TRICC Trial is the best available evidence that haemodynamically stable critically ill patients should be managed with a restrictive transfusion strategy |6|. In the trial a transfusion trigger of 70 g/1 was used aiming to maintain the Hb range at 70–90 g/1 during intensive care unit (ICU) stay and was compared with a trigger of 100 g/1 aiming for a Hb concentration of 100–120 g/1. There were no differences in the 30-day mortality. The mortality was lower in the restrictive group in patients aged <55 years and with Acute Physiology and Chronic Health Evaluation (APACHE) II scores of <20. The authors had previously published subgroup analyses for patients undergoing prolonged mechanical ventilation and for those with cardiovascular disease, describing no statistically significant differences among these groups. This paper examined the subgroup of patients admitted to the ICU with trauma. INTERPRETATION. Overall, 203 of the 838 patients enrolled in the TRICC Trial had trauma diagnoses (100 in the restrictive group and 103 in the liberal group). The patients’ demographics, injury severity, APACHE II scores and organ failures were similar at study entry. Approximately 90% of the patients suffered blunt trauma and received a median of 3–3.5 red cell units prior to randomization. The average daily Hb in the restrictive group was 82.7 g/1 in the ICU and these patients received 57% less red cell units than the liberal group (mean 2.3 units per patients). The patients’ 30-day mortality, organ failures, infections and ICU length of stay were all similar between the groups with no clinically or statistically significant differences between the restrictive and liberally transfused patients (Table 2.3). Interestingly, the patients in the restrictive group were more likely to have a pulmonary artery catheter (17 versus 8%). Comment Although this was another subgroup analysis of the TRICC Trial, it is the only large randomized trial of transfusion triggers in resuscitated trauma patients. It is import-
The year in anaesthesia and critical care
40
Table 2.3Outcome of all trauma patients* Patient characteristics
Liberal group (n=103)
Restrictive group (n=100)
P-values
9(9)
10(10)
0.81
10(10)
10(10)
1.00
6(6)
8(8)
0.59
10(10)
10(10)
1.00
MODS (n=202)
7.7±3.9
7.9±4.4
0.69
∆MODS (n=202)
0.6±3.8
0.0±4.4
0.29
MODS† adjusted for death
9.0±6.0
9.2±6.3
0.81
∆MODS†—change in MODS adjusted for death
1.9±5.7
1.2±6.1
0.44
ICU (days)
10.2±8.7
9.8+8.1
0.73
Hospital (days)
33.7±17.7
31.4±17.1
0.34
5.4±4.3
2.3±4.4
<0.0001
Proportion transfused, n (%)
103 (100)
65 (65)
<0.0001
Average daily haemoglobin concentrations
104.3±12.2
82.7±6.2
<0.0001
13 (13)
8 (8)
0.28
7 (7)
3 (3)
0.33
Mortality rates, n (%) 30-day 60-day (n=202) ICU Hospital Organ failure and dysfunction
Length of stay
Transfusions during ICU admission Transfusions (units per patient)
Proportion who developed infection, n (%) Physician non-adherence, n (%)
* Plus—minus values are means ± standard deviations. ICU, intensive care unit. Because of rounding, percentages may not equal 100. Cl, confidence interval; MODS, multiple organ dysfunction score; ∆MODS, change in multiple organ dysfunction score from baseline values; † non-survivors are considered to have all organs failing on date of death. Source: McIntyre et al. (2004).
ant to remember that patients were only entered after they were stabilized and no longer bleeding, so there may have been some selection bias in excluding patients with continuing haemorrhage over the 72 h during which enrolment was possible. It is also not clear how many patients were not entered because of physician refusal, which was a much-discussed aspect of the original trial. Despite this, the data are highly suggestive
Perioperative blood transfusion therapy
41
that restrictive transfusion strategies are safe and effective among resuscitated trauma patients. Given the association between lower illness severity and worse outcome with liberal blood use in the main trial, it is also highly suggestive that trauma patients not requiring intensive care should only be transfused when their Hb is approximately 70 g/l and even then they should receive only 1 red cell unit at a time. Another caveat was the lack of association between liberal blood use and organ failures. This contradicts the series of papers by Moore’s group who used regression analysis of cohort data sets for arguing that blood transfusion, particularly with older red cells, may actually cause organ failures and infections |9, 10|.
Does the storage time of transfused red blood cells influence regional or global indices of tissue oxygenation in anemic critically ill patients? Walsh TS, McArdle F, McLellan SA, et al. Crit Care Med 2004; 32: 364–71 BACKGROUND. It is often suggested that stored red cells may be less effective than fresh or autologous red cells because of storage changes such as decreased deformability, loss of red cell 2,3 diphosphoglycerate or decreased red cell survival in vivo. Some authors have suggested that stored red cells might even have adverse effects. Despite this widely held view, no previous randomized, double-blind studies have compared fresh with stored red cells in humans. One of the studies most widely quoted as evidence for a detrimental effect of stored red cells was a post hoc analysis of a non-randomized observational study examining the effect of transfusing 3 units of blood to patients with severe sepsis |11|. The authors found that transfusing older red cells, particularly those stored >15 days, seemed to result in worsening of gastric intramucosal pH (pHi), an index of gastric oxygenation. In the study by Walsh et al. anaemic non-bleeding haemodynamically stable ICU patients were randomized double blind to receive 2 units of red cells either aged <5 days or aged >20 days. The primary end-point was the change in the gastric intramucosal PCO2 to arterial PCO2 gap (PCO2 gap) and pHi, with the arterial acid—base status and lactate as secondary end-points.
The year in anaesthesia and critical care
42
Fig. 2.1 Changes in the patients’ Hb concentrations during the study period. All values are medians (first and third quartiles). No significant differences exist between the groups. Source: Walsh et al. (2004).
Perioperative blood transfusion therapy
43
Fig. 2.2 Changes in the arterial CO2 pressure (PaCO2), gastric to arterial PCO2 gap (PgCO2-PaCO2 gap) and gastric pHi in relation to the mean of the values for baseline measurement period. All values are medians (first and third quartiles). No significant
The year in anaesthesia and critical care
44
differences exist between the groups. Source: Walsh et al. (2004). INTERPRETATION. The authors randomized 22 patients (ten to fresh red cells and twelve to stored red cells) to receive transfusions according to a controlled protocol and then measured gastric tonometry and systemic acid-base variables during and after the transfusion (a total of 8 h of data collection). Most patients had values for the PCO2 gap and pHi at baseline considered outside the healthy range, although none had severe systemic acidosis. Despite using various statistical approaches, the authors could find no evidence that any oxygenation end-point improved with either fresh or stored red cells. There was no detectable difference between the two groups for any oxygenation variable and no evidence that stored red cells made any oxygenation variable worse over the 8-hour study period (Figs. 2.1 and 2.2). Comment This exploratory study illustrated several important issues with studies of red cell efficacy. First, although the choice of end-points was appropriate, currently available oxygenation indices, including gastric tonometry values, are probably too insensitive and non-specific. Second, although the authors selected patients typical of the majority transfused in intensive care, they may not have had an oxygen supply dependency because they were relatively stable and only moderately anaemic (mean baseline Hb 85.8 g/l with standard deviation of 8.4 g/l). Although the study was small and underpowered, the authors’ main aim was to confirm or refute the findings of the earlier non-randomized study. In the patients they studied they pointed out that, for the patients who received stored red cells, the observed confidence intervals (CIs) for the average changes in the pHi and PCO2 gap made clinically important worsening of gastric tonometry variables very unlikely after transfusion with red cells stored on average for 28 days. This cannot exclude a difference for sicker patients, particularly those with a major oxygen debt and this needs to be further investigated. Perhaps the most important message from the study is that it is very difficult to demonstrate a physiological effect on tissue oxygenation from transfusion even under rigorous, carefully controlled conditions.
The influence of allogeneic red blood cell transfusion compared with 100% oxygen ventilation on systemic muscle oxygen tension after cardiac surgery Suttner S, Piper SN, Kumle B, et al. Anesth Analg 2004; 99: 2–11 BACKGROUND. In this study two different oxygenation end-points from those used in the previously described paper were assessed following three strategies for improving oxygen delivery after cardiac surgery. These were global oxygen delivery and oxygen consumption calculated using a pulmonary artery catheter and the
Perioperative blood transfusion therapy
45
tissue oxygen tension (PO2) measured in the deltoid muscle using a tissue electrode. INTERPRETATION. Fifty-one volume-resuscitated, mechanically ventilated, elective coronary artery bypass graft patients received one of three strategies for improving oxygen delivery at a Hb value of 75–85 g/l. The first strategy (n=17) was a single red cell unit during 40% oxygen ventilation, the second (n=17) was a 2-unit red cell transfusion during 40% oxygen ventilation and the third (n=17) was no transfusion but 100% oxygen ventilation. The authors found that all three strategies increased the calculated global oxygen delivery, as would be expected, but none increased the global oxygen consumption (Fig. 2.3).
The year in anaesthesia and critical care
46
Fig. 2.3 Changes in the systemic oxygen delivery index (DO2I), systemic oxygen consumption index (VO2I) and skeletal muscle PO2 (PtiO2). Values are reported at a nadir Hb concentration in the range of 7.5–
Perioperative blood transfusion therapy
47
8.5 g/dl (baseline) at 30, 60, 120 and 180 min after increasing the inspired fraction of oxygen to 1.0. The timepoints of 30 and 60 min denote the completion of the infusion of 1 and 2 units of packed erythrocytes in the transfusion groups, respectively. * P<0.05/6 (Bonferroni corrected) and significantly different from baseline; † P<0.05/6 (Bonferroni corrected) compared with other groups. Source: Suttner et al. (2004). The two transfusion strategies had no effect on the tissue oxygen tension, but in the 100% oxygen group this variable increased significantly. Of note in relation to the previous paper, the policy in the authors’ centre was to administer only red cells stored ≤14 days to cardiac surgery patients. Comment This was a well-designed and well-executed study based in the ‘real’ world of cardiac surgery, which remains a major user of allogeneic blood. However, the results are not really surprising. The patients did not have an oxygen debt, so transfusion was unlikely to increase their global oxygen consumption. Although the use of the tissue PO2 was interesting this is not really a measure of the volume of oxygen made available to tissues, but it is an indicator of the likely diffusion gradients present. However, skeletal muscle is not an organ considered at high risk for tissue hypoxia and measurement of the gut mucosal PO2 is probably more valid but impractical in most clinical settings. In addition, the skeletal muscle PO2 has been shown to be normal or high even during severe sepsis, so the clinical meaning of this measurement is uncertain. This study therefore reiterated the fact that currently available clinical indices of tissue oxygenation are inadequate and that demonstrating effectiveness from transfusions is difficult. Complications of blood transfusion Effectiveness of blood transfusions should be balanced against the potential adverse effects. The incidence of viral infection from blood transfusion is now well documented and extremely small. The clinical importance of other potential complications, such as acute lung injury and immunomodulation, are much less certain. The following papers consider these issues.
The year in anaesthesia and critical care
48
Acute lung injury after blood transfusion in mechanically ventilated patients Gajic O, Rana R, Mendez JL, et al. Transfusion 2004; 44: 1468–74 BACKGROUND. Transfusion-related acute lung injury is a recognized and probably under-reported complication of the transfusion of blood products |12|. It is thought to result from transfusion of anti-neutrophil antibodies in donor plasma reacting with recipient neutrophils, although an alternative aetiology involving the transfusion of bioactive substances has also been proposed. Blood product transfusion has the potential to act as a second ‘hit’ in mechanically ventilated patients, resulting in a higher incidence or severity of acute lung injury. INTERPRETATION. The authors retrospectively analysed a database of 181 ventilated, critically ill patients who received a blood transfusion within the first 48 h of mechanical ventilation. The potential predictive factors for acute lung injury during the first 48 h were recorded, such as sepsis, shock, aspiration and trauma. In addition, the number and age of the red cell units given were extracted together with the units of plasma-rich products (fresh frozen plasma [FFP] and platelets). A logistic regression model was developed for exploring the association between the potential causative factors during the first 48 h and the occurrence of acute lung injury. Thirty-three per cent of the patients developed acute lung injury, usually on day 2 of ventilation. There was no association between the storage age of the transfused red cells and the development of acute lung injury. The main risk factors for acute lung injury in the multivariate analysis were FFP transfusion and the presence of thrombocytopenia. Comment This study is typical of many retrospective cohort studies exploring adverse effects from blood transfusions. The main problem is defining outcomes, in this case acute lung injury, rigorously and identifying sufficient relevant variables in order to avoid the effects of confounding in the analyses. The authors went to some lengths to identify acute lung injury accurately using established definitions. They also tried hard to include as many variables as possible that could be confounders in the analysis. Unfortunately, they did not define the transfusion thresholds used or whether all the products were nonleucodepleted (although, as the study was from a US centre, this was likely). The patients did appear to receive a large number of red cell units (median 13 units for the acute lung injury group). It is reassuring that there was no strong signal relating to the age of the red cells but, because it was an observational study, individual patients received a wide range of red cell ages. The association with FFP transfusion was consistent with a possible effect from anti-leucocyte antibodies, but this study did not prove that this relationship was causative. It may have been that sicker patients received more FFP and were more prone to acute lung injury anyway, but that this was not adequately adjusted for in the
Perioperative blood transfusion therapy
49
analysis. The relation with thrombocytopenia, which is known to correlate closely with outcome in ICUs, could have a similar explanation. In conclusion, this study provided no evidence that red cell transfusions or their storage age are a contributing factor to acute lung injury in mechanically ventilated patients, but it does not exclude this possibility. With plasmareduced red cell products transfusion-related acute lung injury is most closely linked to FFP or platelet transfusions, which the presented data support. The importance of leucodepletion of red blood cells The importance of donor leucocytes in allogeneic blood transfusion has been debated for many years. There seems little doubt that febrile transfusion reactions are associated with the leucocytes, but studies with contradictory findings exist concerning whether transfused leucocytes cause a clinically important transfusion-associated immunomodulation. It has been hypothesized that transfusion-associated immunomodulation could increase post-operative infection rates and in cancer patients increase recurrence rates. The following two papers are a randomized trial from a Dutch group that has long-standing clinical trials experience in this area and a metaanalysis of published trials by a leading transfusion epidemiologist.
Effects of transfusion with red cells filtered to remove leucocytes: randomised controlled trial in patients undergoing major surgery van Hilten JA, van der Watering LMG, van Bockel JH, et al. BMJ 2004; 328: 1281–9 BACKGROUND. In an earlier trial in cardiac surgery, this group found that leucodepletion of red blood cells resulted in lower post-operative infections and mortality from multiple-organ failure, but only for patients receiving >4 red cell units |13|. This study was carefully designed for determining whether leucodepletion decreased mortality, ICU length of stay, multiple-organ failure, infections and hospital length of stay in patients undergoing aortic aneurysm surgery (ruptured or non-ruptured) and gastrointestinal cancer surgery. INTERPRETATION. The authors performed a clear and careful power analysis for the trial from pilot work. They needed 400 transfused patients in each group in order to detect an 8% decrease in mortality and a decrease of 2 days in intensive care with leucodepletion. Although their total randomization numbers were close to those intended (1200), the transfusion rates were lower than expected (approximately 50%) and the amount of blood they received was smaller than predicted. The numbers of evaluable transfused patients were only 278 and 267 in the two groups. No significant differences were found in mortality and infections (Fig. 2.4) on an intention-to-treat analysis or in the mean ICU stay (mean length of ICU stay 0.4 days shorter for the leucodepleted group with 95% Cl −1.6 to +0.6 days). Patients in the leucodepleted group had a shorter mean hospital stay (average 2.4 days), which just reached statistical significance (P=0.05) and a lower chance of multiple organ failure (odds ratio [OR] 0.70 with 95% Cl 0.49 −1.00) (P=0 05) Analysis of the groups of patients separately suggested that patients undergoing
The year in anaesthesia and critical care
50
gastrointestinal oncology surgery might gain some benefit in relation to multi-organ failure and mortality (Fig. 2.4). Comment This must have been a very difficult study to perform and it is disappointing that it was ultimately under-powered. The authors could not extend their recruitment because the Dutch Ministry of Health introduced universal leucodepletion in an effort to reduce the risk of vCJD transmission. This irony illustrates the relative influence of the ‘precautionary principle’ and evidence-based medicine in the blood industry (see below). It is difficult to assess the significance of the difference in organ failures between the groups. The authors used serial APACHE scores rather than the more widely used descriptive systems for assessing organ failures. In addition, it is difficult to believe that the difference was clinically important because there were no differences in ICU mortality or length of stay. Despite its limitations, this study
Fig. 2.4 An intention-to-treat analysis of the primary and secondary endpoints. The effects on mortality, multiorgan failure and infection are depicted as odds ratios (ORs)
Perioperative blood transfusion therapy
51
(filtered/nonfiltered) with 95% CIs. Source: Van Hilten et al. (2004). provided no convincing evidence that leucodepletion has a significant impact on mortality or infections. One of the subgroup analyses did suggest a greater benefit in the gastrointestinal cancer patients. Although the numbers of transfusions given were less than expected, the authors found no effect of blood dose in contrast to their earlier work in cardiac patients.
White blood cell containing allogeneic blood transfusion, post-operative infection and mortality: a meta-analysis of observational before and after studies Vamvakis EC. Vox Sanguinis 2004: 86: 111–19 BACKGROUND. In addition to the Van Hilten et al. study, there have been eight randomized controlled trials of leucodepleted versus non-leucodepleted allogeneic blood transfusions for various types of surgery. The data from these trials are contradictory and, in particular, the most recent trials failed to demonstrate an association between non-leucodepleted blood transfusion and infection. The most recently published meta-analysis of these trials |14| suggested an effect on post-operative infection only in cardiac surgery, but the author pointed out that, if the results of a subsequently published randomized controlled trial were added, this effect disappeared. The other approach to addressing this question has been ‘before and after’ observational studies done around the time of introduction of universal leucodepletion. This study was a meta-analysis of available studies. INTERPRETATION. This systematic review identified six studies that had reported on post-operative infection and/or mortality before and after introducing leucodepletion. The author performed a number of different meta-analytical approaches in an attempt to adjust for the potential confounders that are present in observational studies and for the different ways that post-operative infection was measured. In unadjusted analysis the six studies suggested a small but statistically significant decrease in the risk of post-operative infection, but this was small (OR 0.93; 95% Cl 0.88–0.99). For the studies in which adjustment was possible the OR was similar but no longer statistically significant (OR 0.94; 95% Cl 0.85–1.04). There was no association between leucodepletion and decreased mortality regardless of how the meta-analysis was performed, although it was noteworthy that the largest individual study in adults had found a small but statistically significant decrease in mortality. Comment Before and after studies of transfusion are notoriously difficult to interpret because, even if potential confounders are included, there is a high risk of residual confounding by factors that were unmeasured. These can include the complexity of surgery, changes in
The year in anaesthesia and critical care
52
techniques or the way health systems are organized. The results of this analysis, taken with the meta-analysis of randomized trials and the Van Hilten et al. study, suggest that the presence of leucocytes in allogeneic blood is unlikely to have clinically important effects on post-operative infection rates or mortality. Whether other more subtle effects exist will probably never be known because in many countries a political decision has resulted in the introduction of leucodepletion. This decision and other expensive blood safety measures were driven largely by the ‘precautionary principle’ |15|. This means that the potential threat of adverse effects over-rides scientific certainty. A consequence is that different health systems now have different blood products without really knowing how clinically important this is. For example, in Canada, Australia and most of Europe red cells are universally leucodepleted and stored for 35 days, whereas in much of the USA red cells are stored longer (42 days), but are not leucodepleted. Perioperative blood conservation Perioperative blood conservation techniques can be considered to fall into two categories. First, methods designed for increasing autologous transfusion (the patient’s own blood) and decreasing allogeneic transfusion (stored donated blood from another individual). These methods include pre-operative acute normovolaemic haemodilution (ANH), preoperative autologous blood donation (PAD) and peri-operative cell salvage (CS). The alternative methods include techniques for minimizing blood loss, including surgical techniques and the use of drugs for modifying coagulation such as aprotonin and tranexamic acid. The following two reviews addressing methods of increasing autologous infusion were recently published.
Pre-operative acute normovolemic hemodilution: a meta-analysis Segal JB, Blasco-Colmenares E, Norris EJ, Guallar E. Transfusion 2004; 44: 632–44 BACKGROUND. ANH is the technique whereby autologous blood is drawn immediately prior to surgery and replaced with non-blood products in order to achieve intravascular normovolaemia. The blood can be re-transfused to the patients during or after surgery. The technique has the theoretical advantage of avoiding allogeneic blood transfusion with its associated risks and re-infusion of fresh whole blood that includes coagulation factors. Animal experiments and the volunteer studies of Weiskopf et al. |16| suggest that most normovolaemic patients tolerate acute decreases in their Hb level to 70–80 g/l without demonstrable adverse effects. However, the techniques are time-consuming and labour-intensive and, therefore, are relatively expensive. This meta-analysis examined published studies of the technique. INTERPRETATION. The literature search only examined full publications in English that were controlled and included a non-ANH arm. The primary outcomes chosen were the proportion of patients receiving allogeneic transfusion and the average amount of allogeneic blood received by patients The secondary outcomes measured were
Perioperative blood transfusion therapy
53
perioperative blood loss and reported adverse events. The authors also tried to distinguish large- from small-volume ANH. Forty-two trials were examined. Twenty-three compared ANH to no conservation, eleven compared ANH to no ANH, but with other blood conservation techniques used in both arms and eleven compared ANH to an alternative blood conservation method. The risk of allogeneic transfusion was similar among patients receiving ANH and those receiving usual care (relative risk [RR] 0.96; 95% Cl 0.90– 1.01) or another blood conservation method (RR 1.11; 95% Cl 0.96–1.28). The ANH volume did not influence this finding. ANH did result in a lower mean volume of allogeneic blood transfusion per patient, but there was considerable variation between trials. The variation correlated with the volume of ANH and the transfusion trigger: the lowest benefit was seen in the most recent trials. ANH was associated with less total perioperative blood loss (mean 91 ml and 95% Cl 25–195 ml) compared to usual care, but not compared to alternative blood conservation methods. The adverse event reporting was inconsistent. Comment There are so many factors that could produce confounding when comparing blood conservation trials that meta-analysis is difficult. One of the most important of these is the transfusion trigger used, which has decreased in recent years. Despite these limitations, this analysis does not support the clinical effectiveness of ANH, particularly as the most recent trials showed the least benefit. The lower limit of the CI for the pooled estimate suggests a best possible scenario of a 10% reduction in the transfusion rate and a saving of approximately 2 units of red cell per patient. Although this would be clinically relevant, these benefits were much less likely with more recent trials or those comparing ANH with other blood conservation methods. It is worth noting that a previous metaanalysis, including only trials published up to 1998, was far more positive |17|. The current analysis did not systematically review all the literature and, of note, excluded abstracted and non-English language studies. It would not fulfil the criteria for a Cochrane systematic review and could be subject to trial selection or publication bias. However, as it is generally negative trials that fail to be published, this paper suggests that, on current evidence, ANH should not be the first-choice perioperative blood conservation method.
Autologous transfusion techniques: a systematic review of their efficacy Carless P, Moxey A, O’Connell D, Henry D. Transfus Med 2004; 14: 123–44 BACKGROUND. This group of authors was the core that conducted the Cochrane systematic reviews of the blood transfusion literature. Their approach was very rigorous and is the gold standard against which other reviews should be measured. In this paper the authors summarized their analysis of the literature concerning PAD, ANH and CS. Individual detailed descriptions of the reviews can be seen on the Cochrane database of systematic reviews The conclusions for each of
The year in anaesthesia and critical care
54
the methods assessed are considered separately. Pre-operative autologous blood donation INTERPRETATION. Eight randomized controlled trials that included 1119 subjects were analysed. There was a lot of heterogeneity between the studies, which made generalization difficult and the trial quality was poor, particularly for concealment from the intervention. PAD decreased the chance of receiving allogeneic blood by 63% (RR 0.37; 95% CI 0.26–0.54). PAD appeared more effective in orthopaedic surgery trials (average decrease 84%) than the cancer trials (average decrease 51%). However, the effectiveness was less obvious when clear transfusion protocols were used compared with trials that did not have protocols. In addition, PAD was associated with an 80% blood transfusion rate (autologous or allogeneic) compared with only 60% in the control groups (RR 0.20; 95% CI 0.10–0.29). The observational studies also showed higher overall transfusion rates associated with PAD. Comment The message from this systematic review is clear. PAD does decrease the chance of exposure to allogeneic blood transfusion, but patients tend to receive more blood. The assumption is that autologous blood is administered even if not clinically indi-cated. There is some experimental evidence that even autologous stored blood could have some adverse effects, although whether these are clinically important and/or offset the benefit of transfusion is unknown. The main limitation of this analysis is the poor quality of the studies included, in particular the lack of transfusion protocols, which tend to decrease blood transfusions anyway. The transfusion rates in both the PAD and control groups for these studies were very high, making it uncertain whether this approach is effective with protocols that include lower transfusion thresholds. Acute normovolaemic haemodilution INTERPRETATION. The review included 30 randomized controlled trials (twelve with cardiac surgery, seven with orthopaedic surgery and eleven with various other procedures) most of which were small. The methodological quality of the trials was poor, none were blinded and there was considerable heterogeneity. ANH decreased the risk of exposure to allogeneic transfusion by 23% for cardiac surgery (RR 0.77; 95% Cl 0.57– 1.04) and that in orthopaedics by 21% (RR 0.79; 95% Cl 0.60–1.06). The risk reduction was less for trials that included transfusion protocols than those that did not. Similarly, the average volume of allogeneic blood saved per patient was less for trials that included transfusion protocols than for those that did not (1 versus 3 units). The data for adverse event reporting were poor. Comment The trials examined in this analysis were a different mix from those included in the metaanalysis by Segal et al. This analysis examined trials comparing ANH with standard techniques only, which may explain why it appeared more positive. Again, a major
Perioperative blood transfusion therapy
55
problem was the poor trial methodology and in particular the confounding effect of the transfusion protocols. Perioperative cell salvage INTERPRETATION. The review examined 30 randomized controlled trials (14 with cardiac surgery, twelve with orthopaedic surgery and four with vascular surgery) and the trial methodology was again poor with considerable heterogeneity. A greater decrease in the risk of allogeneic blood transfusion was observed in orthopaedic surgery than in cardiac surgery (RR 0.35; 0.24–0.52 versus 0.82 and 95% Cl 0.70–0.95). As with the other techniques, there was a dramatic difference between trials that used transfusion protocols and those that did not (average decrease in exposure 38 versus 86%). There was a modest average saving (0.91 units/patient) (95% Cl 0.51–1.31) for the 17 trials that reported the volumes of allogeneic blood transfused. Comment As for the other techniques assessed, the generalizability of the findings was limited by the methodological quality of the trials. In particular, there was a general lack of blinding and concealment of treatment allocation. This may have tended to bias the use of transfusions, particularly when the study conditions were not controlled by protocols. There was also wide variation in the type of CS (washed versus unwashed) and the timing (intra-operative versus post-operative). Despite these limitations, if these net savings were generalizable, CS is likely to be economically viable as the current costs of consumables per case are less than the estimated current cost of 1 unit of allogeneic red cells. These authors have published an updated version of this meta-analysis in the Cochrane library (including 49 randomized controlled trials), which has similar conclusions but with slightly more modest benefits (average saving in allogeneic red cell units per patient 0.64) (95% CI 0.46–0.86). Section summary Overall, these systematic reviews illustrated important factors limiting our understanding of the effectiveness of these technologies. The trial qualities were poor both in terms of methodology and power. In addition, unless the baseline and/or control practice is clearly described and represents the current standard of care it is very difficult to know who should use these techniques and for which type of surgery. It is well known that transfusion rates vary widely between institutions and individual surgeons for the same procedure. The reasons for this are complex and can include surgical and anaesthetic techniques, the transfusion protocols used and the use of other blood-sparing methods such as anti-fibrinolytics. Future studies that take these factors into account are needed. It is also uncertain what end-points are the most relevant for these studies. The traditional use of avoidance of allogeneic transfusion and volume of blood saved are important, but economic end-points also need to be considered. Given current concerns about the infectious risks of allogeneic blood, particularly vCJD, it could be argued that greater
The year in anaesthesia and critical care
56
weight should be placed on avoidance for younger patients than older ones. At present, most ‘expert interpretation’ of perioperative blood conservation trials favours the use of CS over other techniques, but there is an urgent need for more well-designed and wellconducted studies. Conclusion Blood transfusion developed as a life-saving therapy for major haemorrhage. It is remarkable how little high-quality data are available concerning why patients receive blood transfusions, but those available suggest that only approximately 20% of transfusions are to treat immediately life-threatening bleeding |1|. Given the number of surgical and critically ill patients who receive allogeneic transfusions it is further remarkable how little we know about when we should give transfusions and how effective they are. One reason for this is that non-randomized trials are very difficult to interpret because of the large number of potential confounders involved, as illustrated by the Rao et al. study reviewed here. The only way that key questions of effectiveness are likely to be answered is through well-designed randomized controlled trials with appropriate clinical end-points. The Van Hilten et al. study, together with a meta-analysis of earlier randomized controlled trials, is highly suggestive that leucodepletion does not have a major impact on mortality or post-operative infection. It is ironic that leucodepletion has been implemented anyway because of concerns regarding vCJD. This emphasizes the future importance of answering other questions through trials with clinically relevant end-points. The studies by Walsh et al. and Suttner et al. illustrated how physiological end-points are unlikely to provide definitive answers as to when transfusions should be given or whether they are effective. These trials reassure us that restrictive transfusion triggers are justified and unlikely to cause tissue hypoxia in most patients. The McIntyre et al. subgroup analysis of the TRICC Trial was highly suggestive that trauma patients, once stabilized, do not benefit from transfusions for maintaining Hb concentrations greater than 70–90 g/l. What is now needed are further randomized controlled trials designed specifically for adding further information about appropriate transfusion triggers for specific subgroups of patients, such as the elderly or patients with severe coronary disease. There are randomized controlled trials comparing transfusion triggers currently under way for patients undergoing orthopaedic surgery and in critically ill children. These data and further trials should be eagerly awaited and the results translated into clinical practice. Ideally, allogeneic blood transfusion should be avoided altogether. Autologous transfusion techniques are a potential method of reducing or avoiding the exposure of individual patients and conserving blood supplies. The two meta-analyses described illustrate the lack of high-quality evidence concerning which types of patients are most likely to benefit and the need for more trials. These trials need to incorporate evidencebased transfusion protocols and adjust for confounders such as variation in surgical and anaesthetic techniques and the effect of other blood-sparing therapies. Recent advances in CS technology make this technique the most likely to become widely used. Future evaluation needs to include health economic assessment and direct comparison with the
Perioperative blood transfusion therapy
57
increasing cost of allogeneic transfusion. This will be particularly important in countries such as the UK, in which the health service budgets for blood provision are separate from the hospital budgets for surgery. In these situations, the service development of technologies such as CS may be difficult to justify unless it can be directly offset against the cost saving in allogeneic blood. References 1. Wells AW, Mounter PJ, Chaoman CE, Stainsby D, Wallis JP. Where does all the blood go? Prospective observational study of red cell transfusion in north England. BMJ 2002; 325: 803– 6. 2. Peden AH, Head MW, Ritchie DL, Bell JE, Ironside JW. Preclinical vCJD after blood transfusion in a PRNP codon 129 heterozygous patient. Lancet 2004; 364: 527–9. 3. McClelland B, Contreras M. Appropriateness and safety of blood transfusion. BMJ 2005; 330: 104–5. 4. Copplestone JA. Changes in blood supplies, regulations, and transfusion practice. BMJ 2005; 330: 268–9. 5. Hill SR, Carless PA, Henry DA, Carson JL, Hebert PC, McClelland DBL, Henderson KM. Transfusion thresholds and other strategies for guiding allogeneic red blood cell transfusion (Cochrane review). Oxford: The Cochrane Library 2002; Issue 2. 6. Hebert PC, Wells G, Blajchman MA, Marshall J, Martin C, Pagliarello G, Tweeddale M, Schweitzer I, Yetisir E. A multicenter, randomized, controlled clinical trial of transfusion requirements in critical care. Transfusion Requirements in Critical Care Investigators, Canadian Critical Care Trials Group. N Engl J Med 1999; 340: 409–17. 7. Wu WC, Rathore SS, Wang Y, Radford MJ, Krumholz HM. Blood transfusion in elderly patients with acute myocardial infarction. N Engl J Med 2001; 345: 1230–6. 8. Hebert PC, Fergusson DA. Do transfusions get to the heart of the matter? JAMA 2004; 292: 1610–12. 9. Zallen G, Offner PJ, Moore EE, Blackwell J, Ciesla DJ, Gabriel J, Denny C, Silliman CC. Age of transfused blood is an independent risk factor for postinjury multiple organ failure. Am J Surg 1999; 178: 570–2. 10. Offner PJ, Moore EE, Biffl WL, Johnson JL, Silliman CC. Increased rate of infection associated with transfusion of old blood after severe injury. Arch Surg 2002; 137: 711–16. 11. Marik PE, Sibbald WJ. Effect of stored-blood transfusion on oxygen delivery in patients with sepsis. JAMA 1993; 269: 3024–9. 12. Wallis JP. Transfusion-related acute lung injury: under-diagnosed and under-reported. Br J Anaesth 2003; 90: 573–6. 13. Van de Watering LM, Hermans J, Houbiers JG, Van den Broek PJ, Bouter H, Boer F, Harvey MS, Huysmans HA, Brand A. Beneficial effects of leukocyte depletion of transfused blood on postoperative complications in patients undergoing cardiac surgery: a randomized clinical trial. Circulation 1998; 97: 562–8. 14. Vamvakas EC. WBC-containing allogeneic blood transfusion and mortality: a metaanalysis of randomized controlled trials. Transfusion 2003; 43: 963–73. 15. Moreno JD. ‘Creeping precautionism’ and the blood supply. Transfusion 2003; 43: 840–2. 16. Weiskopf RB, Viele MK, Feiner J, Kelley S, Lieberman J, Noorani M, Leung JM, Fisher DM, Murray WR, Toy P, Moore MA. Human cardiovascular and metabolic response to acute, severe isovolemic anemia. JAMA 1998; 279: 217–21. 17. Bryson GL, Laupacis A, Wells GA. Does acute normovolemic hemodilution reduce perioperative allogeneic transfusion? A meta-analysis. The International Study of Perioperative Transfusion. Anesth Analg 1998; 86: 9–15.
3 Off-pump coronary artery bypass surgery CHARLOTTE DEMPSEY, JOSEPH ARROWSMITH Introduction The term ‘off-pump coronary artery bypass’ (OPCAB) encompasses a number of techniques during which myocardial revascularization is performed without employing extra-corporeal circulation. Although the term ‘beating heart surgery’ has been used interchangeably with OPCAB, it should be borne in mind that coronary artery bypass grafting (CABG) can be performed on a beating heart ‘on-pump’. For the purposes of this review, OPCAB is considered to be multi-vessel CABG through a median sternotomy without the use of cardiopulmonary bypass. Historical note It was during the first decade of the last century that the Nobel laureate Alexis Carrel conceived the idea of coronary artery bypass and performed such procedures on dogs |1|. From his own accounts it is clear that Carrel was all too well aware of the difficulties of operating on a beating heart and the potential for serious complications secondary to haemodynamic instability. However, until the advent of clinically useful cardiopulmonary bypass the early pioneers of modern cardiac surgery had no choice but to operate on the beating heart. Carrel’s ingenious ideas were seemingly ‘lost’ for over 40 years until, in 1952, Demikhov described internal mammary—coronary anastomosis in dogs. The following year Mustard attempted carotid—coronary anastamosis under mild hypothermia and Gibbon successfully used cardiopulmonary bypass for facilitating the closure of an atrial septal defect |1|. The chance discovery of coronary angiography in the early 1960s and the development of cardioplegia provided the additional catalysts necessary for the evolution of conventional CABG. Since popularization of the procedure in the late 1960s, CABG with cardiopulmonary bypass has grown to become one of the most commonly practised surgical procedures, with over 1000 000 being performed annually worldwide. © Atlas Medical Publishing Ltd
Off-pump coronary artery bypass surgery
59
Fig. 3.1 The changing face of coronary revascularization. Source: Medtronic, Inc |3|. Despite the obvious advantages of cardiopulmonary bypass—a ‘quiet’ and bloodless operative field and vital organ perfusion—extra-corporeal circulation has long been known to be associated with major organ system dysfunction. Nevertheless, the place of cardiopulmonary bypass in CABG surgery went largely unchallenged for over two decades, until the introduction of cardiac stabilization devices and new surgical techniques prompted a reawakening of off-pump techniques. The growth of percutaneous intervention (angioplasty and stenting) has resulted in a fall in the number of patients referred for surgical revascularization, while the proportion of patients undergoing OPCAB has increased (Fig. 3.1) |2|. By 2008, it is estimated that over half of all CABG procedures will be performed off-pump |3|. Why off-pump coronary artery bypass? Eliminating cardiopulmonary bypass from CABG should, at least in theory, reduce or completely prevent many of the complications ascribed to it (Table 3.1). One could be forgiven for thinking that a series of large, multicentre, prospective, randomized studies would have provided most of the proof by now. However, the truth is that the vast majority of published clinical investigations have failed to inform the evaluation process unambiguously and, instead, stimulated much debate |4–10|. The confounding factors include small numbers of patients, highly selective recruitment, non-randomization, institutional and operator bias, the use of historical control groups and the use mortality, morbidity and composite clinical measures as endpoints. However, it should be recalled that it took a decade to produce the evidence that, for certain groups of patients, conventional CABG was superior to medical therapy |11|.
The year in anaesthesia and critical care
60
Table 3.1 Putative benefits of surgery without cardiopulmonary bypass Efficacy equivalent to CABG with cardiopulmonary bypass
Reduced stroke rate
Reduced overall mortality
Reduced cognitive dysfunction
Reduced overall morbidity
Reduced renal dysfunction
Reduced duration of ventilation
Reduced gastrointestinal complications
Reduced length of ICU stay
Reduced infective complications
Reduced length of hospital stay
Reduced pulmonary dysfunction
Reduced cost
Reduced myocardial dysfunction Reduced long-term MACE Reduced post-operative dysrhythmia Reduced blood loss/transfusion
In a recent review, Chassot et al. |12| reported that less than 20 published studies provided level I or level II evidence. The weight of currently available evidence suggests OPCAB is associated with reduced overall mortality and morbidity |13–15|, reduced mortality and morbidity in high-risk patients |13, 16|, shorter duration of intensive care unit (ICU) and hospital stay |17–19|, reduced cost |17, 19–21|, reduced transfusion requirements |18, 21–23| and reduced perioperative myocardial infarction |15, 18|. Less certain is the impact of OPCAB on long-term graft patency and freedom from major adverse cardiac events. Although early angiographic graft patency rates following OPCAB have been shown to be excellent |24|, mid-term patency rates for saphenous vein grafts to the posterior coronary circulation are lower than expected |25|.
Comparison of coronary artery bypass surgery with and without cardiopulmonary bypass in patients with multivessel disease Mack MJ, Pfister A, Bachand D, et al. J Thorac Cardiovasc Surg 2004; 127(1): 167–73 BACKGROUND. This study was a retrospective analysis of patients undergoing isolated CABG with cardiopulmonary bypass (n=10118) (58.1%) or OPCAB (n=7283) (41.9%) at four US cardiac centres between 1999 and 2000. The centres were selected on the basis of the total number of procedures performed annually and the experience of surgeons with OPCAB. Using a propensity score, derived from twelve independent perioperative variables, 5774 (79%) OPCAB cases were matched to a similar number of CABG cases for subgroup analysis. INTERPRETATION. The incidences of heart failure, chronic pulmonary disease, renal failure cerebrovascular disease peripheral vascular disease prior stroke and prior
Off-pump coronary artery bypass surgery
61
CABG were significantly greater in patients undergoing OPCAB. Although there was no significant
Table 3.2 Predictors of mortality in specific subgroups Variable
Odds ratio
Confidence interval
P-value
Female gender
1.99
1.15–3.46
0.02
Redo surgery
4.60
2.25–9.40
<0.001
Cardiopulmonary bypass used
2.13
1.20–3.76
0.01
Dialysis
17.87
9.95–32.07
0.05
Cerebrovascular accident
3.08
1.04–9.13
0.04
Recent myocardial infarction
3.12
1.39–7.00
0.006
Cardiopulmonary bypass used
3.37
1.59–7.17
0.002
Dialysis
4.57
1.79–11.7
0.002
Cerebrovascular accident
2.02
1.0–4.1
0.05
Recent myocardial infarction
2.7
1.65–4.73
<0.001
Cardiopulmonary bypass used
1.7
1.07–2.69
0.02
Redo surgery
3.88
1.94–7.76
<0.001
Age >75 (n=1479; 20%)
Redo CABG (n=460; 6.2%)
Women (n=1929; 26.1%)
Source: Mack et al. (2004).
difference in predicted mortality (CABG versus OPCAB, 6.6 versus 5.8%), the observed mortality was significantly lower following OPCAB (3.5 versus 1.9%) (P<0.001). In the propensity-matched sample, mortality was lower following OPCAB (3.7 versus 2.0%) (P<0.001) despite a similar predicted mortality risk (CABG versus OPCAB, 6.33±5.86 versus 6.46±5.91%). In order of importance, the independent predictors of mortality in the matched sample were found to be ‘redo’ surgery, renal failure, recent myocardial infarction, the use of Cardiopulmonary bypass, female gender and prior stroke. Cardiopulmonary bypass was predictive of mortality in female patients, patients aged ≥75 years and patients undergoing reoperation (Table 3.2). OPCAB was associated with a significant reduction in major organ system complications and bleeding complications. Comment As the authors themselves noted, the ever-increasing volume of evidence in support of OPCAB makes it more and more difficult to conduct large, randomized studies. Despite
The year in anaesthesia and critical care
62
the obvious shortcomings of the study design, this publication represents a significant contribution. It confirmed previous reports that OPCAB confers a significant short-term survival benefit in high-risk, female, elderly and ‘redo’ patients. The widely varying rate of OPCAB at each of the four centres (29.2–86.2%) suggests that other factors, such as surgeon predilection and patient preference, influenced the decision to perform OPCAB. As in all previous studies, the influence of surgical experience and predilection for OPCAB was not considered as an independent predictor of outcome. This ignores an important factor and makes it difficult to general-ize the findings to other surgeons and centres. The paucity of data for procedures converted from OPCAB to CABG with cardiopulmonary bypass prevents an analysis on an intention-to-treat basis and would tend to worsen apparent outcomes from on-pump surgery. As expected, the Parsonnet score overestimated mortality and was unable to predict that expected mortality would be lower following OPCAB. Although the European System for Cardiac Operative Risk Evaluation |26, 27| may have more accurately predicted overall mortality, it is unlikely that it would discriminate between the two groups of patients. Despite the lack of a definitive, multicentre, randomized trial of sufficient size, it is hard to disregard several large, observational studies |28–30|, which provide strong circumstantial evidence that OPCAB reduces the risk of stroke. Whether OPCAB reduces post-operative cognitive dysfunction remains unclear, as much of the published evidence is conflicting.
Predictors of delirium after cardiac surgery delirium: effect of beatingheart (off-pump) surgery Bucerius J, Gummert JF, Borger MA, et al. J Thorac Cardiovasc Surg 2004; 127(1): 57– 64 BACKGROUND. Neurological injury following cardiac surgery is associated with increased morbidity and mortality and prolonged hospital stay. Stroke following CABG is relatively uncommon (~3%), but increases hospital mortality tenfold. In contrast, cognitive dysfunction, which is considerably more common (30– 70%), has little immediate impact on outcome. Delirium—invariably considered a benign, self-limiting condition—occurs in 3–8% patients and is associated with a fivefold increase in mortality. The purpose of this study was to use prospectively collected clinical data for determining the predicators of delirium after cardiac surgery. INTERPRETATION. Of 16 184 patients undergoing cardiac surgery between April 1996 and August 2001, cardiopulmonary bypass was used in 14 342 (89%). The overall prevalence of delirium—defined as a transient mental syndrome of acute onset characterized by global cognitive impairment, reduced conscious level, abnormalities of attention, increased or decreased psychomotor activity and a disordered sleep-wake cycle—was 1354 out of 16 184 (8.4%). Of these patients, 219 (16%) were found to have had a stroke. Patients with delirium were more likely to have post-operative respiratory insufficiency, require tracheal reintubation, develop sternal instability, require sternal wound debridement and have prolonged ICU and hospital stays In contrast the
Off-pump coronary artery bypass surgery
63
prevalence of delirium in patients undergoing beating heart surgery (OPCAB) was 2.3%. The prevalence of delirium in patients undergoing minimally invasive direct CABG was 0.8% (Table 3.3). Comment As pointed out in the editorial that accompanied this article, the ten independent risk factors for delirium identified in this study are in effect generic for any adverse outcome after cardiac surgery |31|. It is not therefore surprising that the incidence of
Table 3.3 Prevalence of independent predictors of post-operative delirium as determined by muitivariate logistic regression analysis, in relation to the different surgical procedures Prevalence (%) Variable
Total
CABG + CPB
Beating heart
Valve+CPB ± CABG
Variables associated with higher risk of delirium History of cerebrovascular disease
3.6
3.4
3.1
4.1
Atrial fibrillation
9.1
4.2
3.3
19.1*
Diabetes mellitus
36.8
41.2†
29.4
27.8
Peripheral vascular disease
20.0
21.8†
14.7‡
18.8
Left ventricular ejection fraction ≤30%
9.2
8.3§
6.1
11.7*
Pre-operative cardiogenic shock
15.5
10.1
8.5
26.6*
Urgent operation
23.7
22.0¶
12.9
30.1*
Operating time ≥180 minutes
21.6
19.4
18.8
26.1*
Intraoperative haemofiltration
11.2
7.4
−
17.5#
Red cell transfusion ≥2000 ml
10.3
8.8¶
4.1
14.9*
11.4
0
100
0
8.8
6.3†
12.5
11.7
Age >50 and <60 years
16.6
17.1
20.6**
14.4*
Age >60 and <70 years
38.3
40.4‡
37.9
34.9
Variables associated with lower risk of delirium Beating-heart surgery Age <50 years
* P<0.0001 versus CABG plus CPB (cardropulmonary bypass) and beating heart;
The year in anaesthesia and critical care
64
† P<0.0001 versus beating heart and valve plus CPB; ‡ P<0.0001 versus valve plus CPB; § P=0.001 versus beating heart; H P<0.0001 versus beating heart; # P<0,0001 versus CABG plus CPB; ** P=0.002 versus CABG plus CPB. Source: Bucerius et al. (2004).
delirium was lower in beating heart surgery patients when the risk factors for delirium were significantly lower in this group of patients. Of particular interest, however, is the very low prevalence of delirium in minimally invasive direct CABG patients, suggesting that surgical manipulation and instrumentation of the ascending aorta contributes to the genesis of delirium. This finding lends weight to the belief that OPCAB combined with ‘no touch’ aortic techniques eliminates a significant number of the major risk factors for stroke during conventional CABG |32, 33|. Anaesthetic management From the anaesthetist’s point of view, conventional CABG is conveniently divided into three readily identifiable epochs. 1. The period between induction of anaesthesia and the onset of cardiopulmonary bypass, when the chest is opened and conduits harvested. 2. The period of cardiopulmonary bypass itself, incorporating one or more episodes of myocardial ischaemia. 3. The period between the termination of cardiopulmonary bypass and the end of surgery. During cardiopulmonary bypass the pump flow rate (i.e. the ‘cardiac output’), systemic vascular resistance and arterial pressure are all amenable to manipulation. In contrast, during OPCAB the period of cardiopulmonary bypass is replaced by intermittent episodes of cardiac displacement and coronary artery occlusion, during which the distal coronary anastomoses are fashioned. Ischaemic myocardial depression, dysrhythmias, external cardiac compression, atrioventricular valve distortion and incompetence, caval (inflow) occlusion and right ventricular outflow obstruction may all conspire to reduce cardiac output and arterial pressure |21, 34–37|. It is during these periods that the anaesthetist is perhaps most active: providing the surgeon adequate surgical access to the lateral and posterior surfaces of the heart while minimizing the haemodynamic consequences. General considerations The duration of OPCAB procedures and the requirement for adequate anaesthesia is frequently no different from those performed ‘on-pump’. Of the many anaesthetic and analgesic regimes described, techniques based on volatile anaesthetic agents appear to offer greater cardioprotection than intravenous agents such as propofol and midazolam. The use of short-acting drugs such as remifentanil and the widespread use of heat-
Off-pump coronary artery bypass surgery
65
conserving and active-warming techniques has resulted in a trend towards earlier extubation and earlier discharge from an ICU. While early or even immediate ‘on-table’ extubation (so-called ultra-fast track) may not be associated with higher morbidity or mortality |38, 39| and may reduce costs, there is little evidence to show that it actually benefits patients |39, 40|.
Sevoflurane provides greater protection of the myocardium than propofol in patients undergoing off-pump coronary artery bypass surgery Conzen PF, Fischer S, Detter C, Peter K. Anesthesiology 2003; 99(4): 826–33 BACKGROUND. By a mechanism of ischaemic preconditioning, the halogenated volatile anaesthetics improve myocardial recovery following ischaemia and reduce myocardial infarct size. Intravenous anaesthetics, such as propofol, do not possess this property. Twenty patients scheduled for primary, elective OPCAB were randomized to receive either etomidate followed by sevoflurane (2% end-tidal) or propofol (target-controlled infusion to achieve a plasma concentration of 2–3 mcg/ml) as the primary anaesthetic agent. Myocardial injury was assessed using troponin I, creatine kinase and myocardial creatine kinase. C-reactive protein (CRP), procalcitonin and interleukin-6 (IL-6) were used as measures of the systemic inflammatory response. INTERPRETATION. Sevoflurane was associated with an increase in the cardiac index during surgery (2.3±0.4 to 3.1±0.6 l/min/m2) (P=0.006) and a greater rise in IL-6 in the first 6 h after surgery (P=0.027). Propofol was associated with a greater rise in troponin I following myocardial reperfusion (P=0.009) and a greater increase in CRP (P=0.002) in the first 24 h after surgery (Fig. 3.2).
The year in anaesthesia and critical care
66
Fig. 3.2 Cardiac troponin I concentrations in Sevoflurane- and propofol-anaesthetized patients during and after anaesthesia. Samples were obtained before induction of anaesthesia (T1), before ischaemia (T2), 15 min after reperfusion (T3), at arrival in the post-anaesthetic care unit (T4) and 3 h (T5), 6 h (T6), 12 h (T7), 18 h (T8) and 24 h (T9) after arrival. Source: Conzen et al. (2003).
Off-pump coronary artery bypass surgery
67
Comment The finding that sevoflurane is cardioprotective during OPCAB adds to the existing evidence for volatile anaesthetic myocardial protection during CABG with cardiopulmonary bypass |41, 42|. The study was not sufficiently powered for detecting differences in major cardiac outcomes. Although the patients studied were young (mean age <65 years), had well preserved left ventricular function and required only one or two bypass grafts, it seems likely that similar results might be expected in older, high-risk patients |42|. It should be borne in mind that troponin I is released in response to both ischaemic and direct (e.g. thermal, incisional and compressive) myocardial injury. More recent work in patients undergoing CABG with cardiopulmonary bypass suggests that sevoflurane or desflurane anaesthesia is associated with shorter length of ICU and hospital stay, a lower incidence of post-operative atrial fibrillation and a reduced requirement for prolonged inotropic support |43|. Further-more, the preservation of myocardial function appears greatest when the volatile anaesthetic is administered throughout the entire procedure, rather than during induction or during the period of myocardial ischaemia |44|. The use of intrathecal morphine in cardiac surgery was first reported in 1980 in a series of 40 patients with favourable results |45|. The putative benefits of intrathecal morphine include optimal analgesia, improved pulmonary function, reduced time to extubation and reduced duration of ICU stay. Recent studies using low-dose intrathecal morphine have demonstrated effective post-operative analgesia and minimal respiratory depression |46|. Despite these potential advantages, intrathecal morphine has not been widely adopted in cardiac surgery, presumeably due to concerns about the risks of neuraxial haematoma formation following systemic anticoagulation. Similar concerns influence the use of thoracic epidural analgesia in cardiac anaesthesia despite the potential benefits of cardiac sympathetic blockade and myocardial protection.
Spinal (subarachnoid) morphine for off-pump coronary artery bypass surgery Mehta Y, Kulkarni V, Juneja R, et al. Heart Surg Forum 2004; 7(3): E205–10 BACKGROUND. This study was a prospective, randomized, double-blind, placebo-controlled, single-centre study of 100 patients undergoing primary elective OPCAB. The patients were randomized to pre-operative administration of intrathecal morphine (8 mcg/kg) or subcutaneous saline. A single anaesthesiologist who was not involved with perioperative patient care administered all injections. Following surgery, all patients were sedated and mechanically ventilated for a period of 2–3 h. Tracheal extubation was performed by a blinded observer according to predefined criteria. The authors wanted to study the effect of intrathecal morphine on the time to extubation, post-operative pulmonary function and post-operative analgesia in OPCAB patients. INTERPRETATION Patients with well preserved left ventricular and pulmonary
The year in anaesthesia and critical care
68
function were investigated. Prior to induction of anaesthesia, the patients were placed in the lateral decubitus position and the skin over L2–3 or L3–4 was infiltrated with 1% lidocaine. The patients in group 1 (n=53) received intrathecal preservative-free morphine (8 mcg/kg in 2 ml of sterile saline) via a 25G Quincke needle, while the patients in group 2 (n=47) received 2 ml of sterile saline by subcutaneous injection over the landmarks L2– L3 or L3–L4. General anaesthesia was induced with midazolam, fentanyl, vecuronium and propofol and maintained with propofol by infusion, fentanyl and vecuronium. Heparin (2 mg/kg) was administered at least 90 min after lumbar puncture in order to achieve an activated clotting time of 250–300 s. During surgery the patients in group 2 received higher doses of fentanyl than the patients in group 1 (625±235 versus 432±125 mcg) (P=ns). Post-operatively all patients received tramadol intravenously every 8 h, supplemented with parenteral morphine in ventilated patients or intramuscular diclofenac in spontaneously breathing patients. Sixteen patients in group 1 and 26 patients in group 2 required rescue analgesics (P=n.s.). The patients in group 1 awakened earlier (133±86 versus 170±82 min) (P=0.033) and were extubated more quickly (9.47±3.83 versus 11.25±3.94 h) (P=0.028) than the patients in group 2. At 12, 24 and 36 h after surgery the patients in group 1 had significantly better pain scores both at rest and during coughing. A single case of respiratory depression following extubation in group 1 was successfully treated with naloxone. Changes in pulmonary function, relative to pre-operative spirometry, revealed that the forced vital capacity (39.66±15.42 versus 31.85±11.65%) (P=0.016), forced expiratory volume in the first second (44.8±16.18 versus 35.97±13.32%) (P=0.013) and peak expiratory flow rate (47.76±24.61 versus 37.37±4.33%) (P=0.031) fell significantly less in group 1. There was no death or neurological complication in either group. The time to mobilization and length of ICU and hospital stay were equivalent in both groups. The authors concluded that intrathecal morphine provides superior analgesia, which translates into better maintenance of postoperative lung function and facilitates earlier tracheal extubation. Comment The only other recent study examining intrathecal morphine in cardiac surgery is retrospective |47|. A single intrathecal dose of opiate will often provide analgesia within 15–45 min that lasts for 30–36 h |48|. The optimal dose of intrathecal morphine for adequate analgesia without respiratory depression remains unknown. In contrast to thoracic epidural analgesia, intrathecal analgesia offers a number of advantages. 1. Subarachnoid puncture at the level of the cauda equina reduces the risk of spinal cord injury. 2. Lumbar puncture can usually be accomplished more quickly and more reliably than placement of an epidural catheter. 3. The patient and nursing staff are spared an additional continuous drug infusion. 4. Intrathecal injection is associated with a lower risk of haematoma formation than epidural instrumentation because of the smaller size of needles used and the prominence of the venous plexus in the epidural space.
Off-pump coronary artery bypass surgery
69
The estimated incidence of haematoma formation as a result of a spinal injection in an anticoagulated patient is approximately 1 in 220 000 |49|. In this study, lumbar puncture was performed ≥90 min before heparinization. Although lumbar puncture was to be abandoned after three unsuccessful attempts and surgery postponed for 24 h in the event of a bloody tap, neither event occurred. The adverse events attributable to intrathecal morphine included urinary retention and delayed respiratory depression. The incidence of pruritus, nausea and vomiting was similar in both groups. The ventilatory depression with intrathecal morphine is dose dependent and exacerbated by other sedatives or opiates as well as advancing age, coexisting disease, patient position and increased intrathoracic pressure |50|. The incidence of significant respiratory depression with high-dose (i.e. 0.03 mg/kg) intrathecal morphine in cardiac surgery is 1.9% |51|. This demonstrates that intrathecal morphine offers a clear benefit in terms of postoperative analgesia and pulmonary function. However, the low reported incidence of adverse events must be balanced against the relatively small sample size and the even smaller risk of permanent neurological injury.
High thoracic epidural anesthesia for off-pump coronary artery bypass surgery Salvi L, Sisillo E, Brambillasca C, Juliano G, Salis S, Marino MR. J Cardiothorac Vasc Anesth 2004; 18(3): 256–62 BACKGROUND. This was a retrospective review of 106 patients undergoing elective OPCAB in a single unit. Patients received high thoracic epidural anaesthesia as an adjunct to standardized general anaesthesia. The authors’ objectives were to evaluate the feasibility of high thoracic epidural anaesthesia combined with sevoflurane and to identify the side effects of the technique and its impact on post-operative pain control. The effects of high thoracic epidural anaesthesia on central haemodynamics were studied in a subgroup of 51 patients. INTERPRETATION. It is known that cephalad extension of sensory and motor blockade is minimal when an epidural catheter is sited at a high thoracic level |52|, as was done in this study. A sensory block at C7–T1 or T6–7 was achieved in all patients. An epidural catheter could not be inserted in two patients. One of these patients had surgery deferred for 24 h because of a bloody tap. There were no dural taps. Heparin (300 IU/kg) was used in all cases and the activated clotting time maintained at >450 s. The mean ± standard deviation (SD) high thoracic epidural anaesthesia to heparinization interval was 100±34 min. The incidences of perioperative myocardial infarction, myocardial ischaemia and atrial fibrillation were 2.8, 7.5 and 10.6%, respectively. The mean time to extubation was 4.6±2.9 h (range 0.5Z–12 h). The average ICU stay was 1.5±0.8 days (range 0.5–6 days). Two patients died, one from multiple-organ failure and the other from myocardial infarction. One patient presented with paraplegia on the first post-operative day. Magnetic resonance imaging excluded medullary compression secondary to epidural or spinal haematoma. The possible explanation was acute ischaemia of the lumbar medulla caused by multiple emboli from aortic atheroma High thoracic epidural
The year in anaesthesia and critical care
70
anaesthesia caused no significant changes in the mean arterial pressure, heart rate, cardiac index or systemic vascular resistance. However, following induction of anaesthesia both the mean arterial pressure and cardiac index decreased (P<0.05) and remained stable thereafter. The lack of any demonstrable effect on the heart rate was attributed to concurrent administration of β-blockers and calciumantagonists. In order to maintain a mean arterial pressure >70 mmHg a vasopressor was required in 44% patients. The authors concluded that the application of high thoracic epidural anaesthesia as an adjunct to sevoflurane anaesthesia is a feasible technique in patients undergoing OPCAB surgery. It guarantees excellent post-operative analgesia and does not carry the risk of hypotension if the block is limited to the upper thoracic dermatomes. Comment Of the small new studies examining high thoracic epidural anaesthesia and general anaesthesia in OPCAB, this is the largest to date. Despite the reported advantages |53|, enthusiasm for high thoracic epidural anaesthesia in cardiac surgery has been limited to a relatively small number of centres. Foremost in the minds of many anaesthetists is the small risk of epidural haematoma formation in a fully anticoagulated patient. Although the exact incidence of this complication is unknown, estimates range from 1 in 150 000 to 1 in 1500 for cardiac surgery with cardiopulmonary bypass |54|. In this study, epidural catheters were inserted on the day of surgery at least 1 h prior to anticoagulation. Insertion of the epidural catheter was abandoned after three unsuccessful attempts and surgery deferred for 24 h in the event of a bloody tap. The pain scores on the first postoperative day were comparable to those achieved with high thoracic epidural anaesthesia in the setting of CABG with cardiopulmonary bypass |55|. At 7.5%, the incidence of perioperative myocardial ischaemia lies within the range (5.1–14.8%) reported for CABG with cardiopulmonary bypass |55, 56|. The lack of a control group in this study makes it difficult to comment on the time to extubation and length of ICU stay. Similarly, it is not possible to assess haemodynamic stability and vasopressor requirements. The safety and efficacy of high thoracic epidural anaesthesia in CABG or OPCAB cannot be reliably evaluated until prospective, randomized studies are undertaken. Monitoring As with conventional CABG, surface electrocardiography, invasive arterial and central venous pressure monitoring and the measurement of urine output and core temperature are considered mandatory |12|. However, during cardiac manipulation a reduction in cardiac output measured by thermodilution by up to 28% may be accompanied by little change in either the heart rate or mean arterial pressure |57|. A reliable, continuous measure of cardiac output and vascular resistances during OPCAB would therefore be a useful addition to the anaesthetist’s armamentarium. For this reason continuous cardiac output monitors such as the oesophageal Doppler monitor have become commonplace,
Off-pump coronary artery bypass surgery
71
although the systematic evaluation of cardiac output monitors during OPCAB is itself challenging! The detection of new segmental wall motion abnormalities by multi-plane transoesophageal echocardiography is thought to be the most sensitive indicator of myocardial ischaemia |58–60|. Thus, the use of transoesophageal echocardiography during OPCAB might be useful for guiding changes for limiting the duration of ischaemia in patients who develop marked changes in segmental wall motion, for example during placement of the cardiac stabilization device.
Cardiac output monitoring during off-pump coronary artery bypass grafting Grow MP, Singh A, Fleming MD, Young N, Watnik M. J Cardiothorac Vasc Anesth 2004; 18(1): 43–6 BACKGROUND. A prospective study of 19 consecutive OPCAB patients was undertaken in order to evaluate the performance of several measures of cardiac output during repositioning and stabilization of the heart. Changes in cardiac output were monitored using (1) the velocity-time integral (VTI) of the descending thoracic aortic blood flow—an indirect measure of stroke volume—determined with pulsed wave transoesophageal Doppler with automated integration, (2) mixed venous oxygen saturation (SvO2) and (3) a continuous cardiac output pulmonary artery catheter. Thermodilution cardiac output (TDCO) was considered the ‘gold standard’. Measurements were taken at baseline and 5 min after positioning for each distal anastomosis. A regression model was constructed in order to allow prediction of the percentage change from baseline values of the TDCO, in order to determine which variables characterize changes in cardiac output. INTERPRETATION. Cardiac output measurements from 17 patients undergoing 38 coronary artery anastomoses were analysed. The data from two patients were excluded from the analysis due to conversion to CABG with cardiopulmonary bypass. After positioning and stabilization for OPCAB, the mean ± standard error of the mean TDCO decreased by 25±2.3%, whereas the average continuous cardiac output decreased by 4±4.1%. The aortic VTI decreased by 23±2.9% whereas the VTI×heart rate product decreased by 22±3.1%. The SvO2 decreased by 7±1.3%. Using a ‘jack-knife’ (so-called ‘leave-one-out cross-validation’) technique, repeated modelling was used for evaluating the relationship between the measured and predicted change in TDCO. The changes in VTI×heart rate and SvO2 emerged as significant predictors for the measured change in TDCO using the following formula: calculated ∆TDCO=−13.15+0.35(∆VTI×heart rate)+0.61(∆Sv02), where ∆TDCO, ∆VTI and ∆SvO2 are the percentage changes in the TDCO, VTI and SvO2, respectively, from baseline values. Cross-validation revealed a significant relationship between the measured and calculated cardiac output (r2=0.32; P<0.05) (Fig. 3.3). The changes in the mean arterial pressure, pulmonary arterial pressure and continuous cardiac output did not correlate with the changes in TDCO. The authors concluded that the changes in VTI×heart rate and SvO2 could be used as surrogate markers for changes in cardiac output during OPCAB surgery.
The year in anaesthesia and critical care
72
Fig. 3.3 Correlation between the measured change in the TDCO and change in the TDCO calculated from the model based on changes in the VTI×heart rate and SvO2 during cardiac repositioning for OPCAB. Source: Grow et al. (2004). Comment Repeated measurements using TDCO are both impractical and, by their very nature, intermittent. Furthermore, TDCO has a reported margin of error of ±13% under even optimal conditions |61, 62|. Tricuspid regurgitation secondary to right ventricular distortion during OPCAB would be expected to introduce further error |63|. The poor correlation between the changes in continuous cardiac output and changes in TDCO (r=0.03; P=0.87), SvO2 (r=0.24; P=0.15) and mean arterial pressure (r= −0.31; P=0.06) was unexpected and somewhat disappointing. The authors postulated that catheter motion may interfere with mathematical reconstruction of the thermodilution curve. Although the authors suggested that changes in the VTI×heart rate and SvO2 may strengthen a clinical decision during cardiac manipulation, it is hard to imagine that their complex mathematical equation would be of use in clinical practice.
A comparison of CardioQ and thermodilution cardiac output during offpump coronary artery surgery Hullett B, Gibbs N, Weightman W, Thackray M, Newman M. J Cardiothorac Vasc Anesth 2003; 17(6): 728–32 BACKGROUND The purpose of this study was to compare cardiac output
Off-pump coronary artery bypass surgery
73
measurement with TDCO in 22 patients undergoing OPCAB using an oesophageal Doppler monitor. Paired measurements were made at intervals of 15 min throughout the procedure during periods of relative haemodynamic stability in order to permit three TDCO measurements to be made for each comparison. A total of 331 direct comparisons were made. The position of the oesophageal Doppler monitor probe was adjusted before each measurement in order to ensure an optimal signal. Measurements were not performed at the time of rapid administration of fluid or immediately after drug administration. Individual patient data and pooled data were analysed. INTERPRETATION. Adequate oesophageal Doppler monitor signals could not be obtained in two patients. The Pearson correlation coefficient between the oesophageal Doppler monitor and TDCO was 0.62 for the pooled data. For individual patients the coefficient ranged from −0.27 to 0.93. The mean difference between the oesophageal Doppler monitor and TDCO (bias) for the pooled data was −0.56 l/min. For individual patients the bias ranged from −1.35 to 0.27 l/min. The SD of the mean difference (precision) was 0.64 l/min for the pooled data and ranged from 0.24 to 0.74 l/min for individual patients (Fig. 3.4). There was no improvement in bias or precision over the course of the study. The wide limits of agreement (bias±SD) and the large inter-patient differences in both bias and precision led the authors to conclude that the oesophageal Doppler monitor could not be recommended as an alternative to TDCO. Comment This was a small but neat study with unfortunately discouraging results. Validation studies have shown acceptable agreement between cardiac output measured by an oesophageal Doppler monitor and by TDCO |64, 65|. This study suggested that, in OPCAB at least, there is little correlation between the oesophageal Doppler monitor and TDCO. The authors emphasized the limitations of TDCO and stressed the fact that it is not possible to prove that the oesophageal Doppler monitor is more or less accurate. However, any discrepancies were attributed to oesophageal Doppler monitor measurement error. Lefrant et al. recommended that a reliable measurement of cardiac output with an oesophageal Doppler monitor can be achieved after training in a dozen patients |66|. The need for frequent repositioning makes the oesophageal Doppler monitor operator dependent. Nevertheless, the oesophageal Doppler monitor offers some advantages, which would make it suitable for use during OPCAB. 1. The descending aorta is a fixed anatomical structure that is relatively unaffected by cardiac displacement. 2. During periods of hypotension, measurement of systolic flow time allows hypovolaemia to be readily distinguished from vasodilation. 3. In the presence of a normal mean arterial pressure a downward trend in cardiac output may provide early indication of cardiac dysfunction |67|.
The year in anaesthesia and critical care
74
Oesophageal Doppler monitor enthusiasts will find the results of this small series disappointing.
Fig. 3.4 Bias and precision for an oesophageal Doppler monitor (CardioQ) cardiac output compared with TDCO. A Bland—Altman plot of pooled data is shown on the left. The bias±2SD was −0.56±1.28 l/min. The data from individual patients are
Off-pump coronary artery bypass surgery
75
shown on the right: bars indicate bias±2SD. Source: Hullett et al. (2003).
Transesophageal echocardiography for monitoring segmental wall motion during off-pump coronary artery bypass surgery Wang J, Filipovic M, Rudzitis A, et al. Anesth Analg 2004; 99(4): 965–73 BACKGROUND. The authors carried out a prospective, observational study of 60 consecutive OPCAB patients designed for evaluating the use of transoesophageal echocardiography for monitoring left ventricular segmental wall motion. In addition to routine monitoring, 43 (72%) patients had a pulmonary artery catheter inserted. Mid-oesophageal two-chamber, four-chamber and long-axis views were recorded following sternotomy (baseline), following application of the stabilization device, following completion of the distal anastamosis and at the end of surgery. Using the 16-segment model and five-point wall-motion scale |68|, two echocardiographers independently performed off-line segmental wall motion analysis. The availability of ≥14 segments was considered adequate for reliable segmental wall motion monitoring. In order to maintain a high degree of specificity, evidence of ischaemia was defined as worsening of the segmental wall motion by two or more grades in two or more segments in the territory vascularized by the target coronary artery. A consensus opinion was obtained when the opinion of the echocardiographers differed. Electrocardiogram (ECG) evidence of ischaemia in the seven-lead ECG was defined as ≥1 mm horizontal or down-sloping ST segment depression or ≥1 mm horizontal ST segment elevation 60 ms after the J point. If baseline ST segment changes were present, an additional ≥2 mm ST segment shift was required for the diagnosis of ischaemia. INTERPRETATION. OPCAB was completed as planned in 55 patients, with 144 out of 148 grafts being performed off-pump. At baseline, adequate transoesophageal echocardiography recordings were obtained in 59 (98%) patients. Following cardiac displacement adequate recordings were obtained prior to left anterior descending artery revascularization in 58 out of 76 (76%) cases (P<0.01), left circumflex coronary artery revascularizations in 33 out of 40 (83%) cases (P<0.01) and right coronary artery revascularizations in 29 out of 31 (94%) cases (P>0.05). New, persistent segmental wall motion abnormalities were found in 22 patients following completion of surgery. Of the four patients who fulfilled ECG criteria for ischaemia, two developed new pathological Q waves consistent with myocardial infarction. At 6-months follow-up, segmental wall motion abnormalities had resolved in eleven out of 20 (55%). In the nine patients with persisting segmental wall motion abnormalities, six had been detected immediately following cardiac stabilizer placement, prior to any surgical interference with the supplying coronary artery. The authors concluded that transoesophageal echocardiography reliably permits visualization of left ventricular segmental wall motion
The year in anaesthesia and critical care
76
during OPCAB in most patients, a prerequisite for ischaemia monitoring. Although cardiac displacement decreased the number of left ventricular segments visualized by transoesophageal echocardiography, reliable segmental wall motion monitoring could be achieved in four out of five patients. Interestingly, cardiac displacement and stabilization alone may be associated with new segmental wall motion abnormalities that may persist for at least 6 months after surgery. Comment This study demonstrated that, with three standard mid-oesophageal views, transoesophageal echocardiography can be used for imaging most of the left ventricle in the majority of OPCAB patients. In contrast to off-line transoesophageal echocardiography analysis, however, it is unlikely that new segmental wall motion abnormalities would be detected with the same sensitivity in clinical practice. A further limitation of the study is that there was no systematic assessment of mitral valvular function. New or worsened mitral regurgitation may itself indicate acute ischaemia. The inci-dence of new segmental wall motion abnormalities in this study is considerably lower than previously reported |69, 70|. The rigorous criteria for the diagnosis of ischaemia (i.e. worsening of segmental wall motion by two or more grades in two or more segments) and the use of intracoronary perfusion catheters may, in part, explain this finding. The lingering question is the impact of cardiac stabilization devices: do they merely reduce local segmental wall motion or can they actually induce myocardial ischaemia? Predefined criteria for myocardial ischaemia were met after cardiac stabilization in 23% of the patients. EGG evidence of ischaemia was present after 9 out of 147 (6%) stabilizer placements. OPCAB was abandoned in three cases and in six cases a new segmental wall motion abnormality persisted for more than 6 months. Further studies are required in order to determine whether actions based on transoesophageal echocardiography findings influence outcomes following OPCAB. Conclusion Despite the paucity of large, randomized studies comparing short- and long-term outcomes following OPCAB and conventional CABG, there is now a substantial and persuasive body of circumstantial evidence suggesting that OPCAB is associated with lower mortality and short-term morbidity. In certain groups of patients—the elderly, those requiring reoperation, those with coexisting disease and women—OPCAB may confer additional benefits over CABG with cardiopulmonary bypass. A number of important questions remain unanswered. 1. What impact does the experience and personal preference of the surgeon and the anaesthetist have on outcome? 2. What is the optimal anaesthetic technique and which intra-operative monitors should be used routinely? 3. Does OPCAB result in improved long-term cognitive function?
Off-pump coronary artery bypass surgery
77
4. Is long-term graft patency following OPCAB at least equivalent to that achieved following conventional CABG?
References 1. Shumacker HB. Approaches to the management of myocardial ischaemia. In: Schumacker HB (ed). The Evolution of Cardiac Surgery. Bloomington: Indiana University Press, 1992:129–42. 2. CTSNet Editors. Outcome Evidence After OPCAB Surgery. 2004. (http://www.ctsnet.org/%20doc/7280). Accessed: 30 November 2004. 3. Medtronic, Inc. OPCAB: The Case for a New Standard of Care. 2004. (http://www.medtronic.%20com/cardsurgery/opcabstandard.html). Accessed: 30 November 2004. 4. Feneck R. OPCAB surgery: time for a reappraisal? J Cardiothorac Vasc Anesth 2004; 18(3): 253–5. 5. Van Dijk D, Kalkman CJ. Pro: OPCAB surgery is the ideal treatment for coronary artery disease. J Cardiothorac Vasc Anesth 2003; 17(3): 396–9. 6. Raja SG, Dreyfus GD. Will off-pump coronary artery surgery replace conventional coronary artery surgery? J R Soc Med 2004; 97(6): 275–8. 7. Raja SG, Dreyfus GD. Off-pump coronary artery bypass surgery: to do or not to do? Current best available evidence. J Cardiothorac Vasc Anesth 2004; 18(4): 486–505. 8. Raja SG. Off-pump and on-pump coronary artery bypass surgery: time to move on. Chin Med J (Engl) 2004; 117(6): 959 (author reply 960). 9. Ascione R, Caputo M, Angelini GD. Off-pump coronary artery bypass grafting: not a flash in the pan. Ann Thorac Surg 2003; 75(1): 306–13. 10. Sanisoglu I, Guden M, Bayramoglu Z, Sagbas E, Dibekoglu C, Sanisoglu SY, Akpinar B. Does off-pump CABG reduce gastrointestinal complications? Ann Thorac Surg 2004; 77(2): 619–25. 11. Anonymous. Myocardial infarction and mortality in the coronary artery surgery study (CASS) randomized trial. N Engl J Med 1984; 310(12): 750–8. 12. Chassot PG, Van der Linden P, Zaugg M, Mueller XM, Spahn DR. Off-pump coronary artery bypass surgery: physiology and anaesthetic management. Br J Anaesth 2004; 92(3): 400–13. 13. Al-Ruzzeh S, Ambler G, Asimakopoulos G, Omar RZ, Hasan R, Fabri B, El-Gamel A, DeSouza A, Zamvar V, Griffin S, Keenan D, Trivedi U, Pullan M, Cale A, Cowen M, Taylor K, Amrani M. Off-pump coronary artery bypass (OPCAB) surgery reduces risk-stratified morbidity and mortality: a United Kingdom multicentre comparative analysis of early clinical outcome. Circulation 2003; 108(Suppl 1): II1–8. 14. Muneretto C, Bisleri G, Negri A, Manfredi J, Metra M, Nodari S, Dei Gas L. Off-pump coronary artery bypass surgery technique for total arterial myocardial revascularization: a prospective randomized study. Ann Thorac Surg 2003; 76(3): 778–82 (discussion 783). 15. Angelini GD, Taylor FC, Reeves BC, Ascione R. Early and midterm outcome after off-pump and on-pump surgery in Beating Heart Against Cardioplegic Arrest Studies (BHACAS 1 and 2): a pooled analysis of two randomised controlled trials. Lancet 2002; 359(9313): 1194–9. 16. Sharony R, Bizekis CS, Kanchuger M, Galloway AC, Saunders PC, Applebaum R, Schwartz CF, Ribakove GH, Culliford AT, Baumann FG, Kronzon I, Colvin SB, Grossi EA. Off-pump coronary artery bypass grafting reduces mortality and stroke in patients with atheromatous aortas: a case-control study. Circulation 2003; 108(Suppl 1): II15–20. 17. Ascione R, Lloyd CT, Underwood MJ, Lotto AA, Pitsis AA, Angelini GD. Inflammatory response after coronary revascularization with or without cardiopulmonary bypass. Ann Thorac Surg 2000; 69(4): 1198–204. 18. Puskas JD, Williams WH, Duke PG, Staples JR, Glas KE, Marshall JJ, Leimbach M, Huber P, Garas S, Sammons BH, McCall SA, Petersen RJ, Bailey DE, Chu H, Mahoney EM, Weintraub
The year in anaesthesia and critical care
78
WS, Guyton RA. Off-pump coronary artery bypass grafting provides complete revascularization with reduced myocardial injury, transfusion requirements, and length of stay: a prospective randomized comparison of two hundred unselected patients under-going off-pump versus conventional coronary artery bypass grafting. J Thorac Cardiovasc Surg 2003; 125(4): 797– 808. 19. Kilger E, Weis FC, Goetz AE, Frey L, Kesel K, Schutz A, Lamm P, Uberfuhr P, Knoll A, Felbinger TW, Peter K. Intensive care after minimally invasive and conventional coronary surgery: a prospective comparison. Intensive Care Med 2001; 27(3): 534–9. 20. Ascione R, Lloyd CT, Underwood MJ, Lotto AA, Pitsis AA, Angelini GD. Economic outcome of off-pump coronary artery bypass surgery: a prospective randomized study. Ann Thorac Surg 1999; 68(6): 2237–42. 21. Lancey RA, Seller BR, Vander Salm TJ. Off-pump versus on-pump coronary artery bypass surgery: a case-matched comparison of clinical outcomes and costs. Heart Surg Forum 2000; 3(4): 277–81. 22. Ascione R, Williams S, Lloyd CT, Sundaramoorthi T, Pitsis AA, Angelini GD. Reduced postoperative blood loss and transfusion requirement after beating-heart coronary operations: a prospective randomized study. J Thorac Cardiovasc Surg 2001; 121(4): 689–96. 23. Van Dijk D, Nierich AP, Jansen EW, Nathoe HM, Suyker WJ, Diephuis JC, Van Boven WJ, Borst C, Buskens E, Grobbee DE, Robles De Medina EO, De Jaegere PP. Early outcome after off-pump versus on-pump coronary bypass surgery: results from a randomized study. Circulation 2001; 104(15): 1761–6. 24. Puskas JD, Thourani VH, Marshall JJ, Dempsey SJ, Steiner MA, Sammons BH, Brown III WM, Gott JP, Weintraub WS, Guyton RA. Clinical outcomes, angiographic patency, and resource utilization in 200 consecutive off-pump coronary bypass patients. Ann Thorac Surg 2001; 71(5): 1477–83. 25. Omeroglu SN, Kirali K, Guler M, Toker ME, Ipek G, Isik O, Yakut C. Midterm angiographic assessment of coronary artery bypass grafting without cardiopulmonary bypass. Ann Thorac Surg 2000; 70(3): 844–9. 26. Nashef SA, Roques F, Michel P, Gauducheau E, Lemeshow S, Salamon R. European System for Cardiac Operative Risk Evaluation (EuroSCORE). Eur J Cardiothorac Surg 1999; 16(1): 9– 13. 27. Roques F, Michel P, Goldstone AR, Nashef SA. The logistic EuroSCORE. Eur Heart J 2003; 24(9): 881–2. 28. Patel NC, Grayson AD, Jackson M, Au J, Yonan N, Hasan R, Fabri BM. The effect of offpump coronary artery bypass surgery on in-hospital mortality and morbidity. Eur J Cardiothorac Surg 2002; 22(2): 255–60. 29. Cleveland Jr JC, Shroyer AL, Chen AY, Peterson E, Grover FL. Off-pump coronary artery bypass grafting decreases risk-adjusted mortality and morbidity. Ann Thorac Surg 2001; 72(4): 1282–8 (discussion 1288–9). 30. Stamou SC, Jablonski KA, Pfister AJ, Hill PC, Dullum MK, Bafi AS, Boyce SW, Petro KR, Corso PJ. Stroke after conventional versus minimally invasive coronary artery bypass. Ann Thorac Surg 2002; 74(2): 394–9. 31. Taggart D. Off-pump surgery and cerebral injury. J Thorac Cardiovasc Surg 2004; 127(1): 7–9. 32. Kirali K, Kocak T, Güzelmeric F, Göksedef D, Kayalar N, Yakut C. Off-pump awake coronary revascularization using bilateral internal thoracic arteries. Ann Thorac Surg 2004; 78(5): 1598– 602. 33. Kapetanakis El, Stamou SC, Dullum MKC, Hill PC, Haile E, Boyce SW, Bafi AS, Petro KR, Corso PJ. The impact of aortic manipulation on neurologic outcomes after coronary artery bypass surgery: a risk-adjusted study. Ann Thorac Surg 2004; 78(5): 1564–71. 34. George SJ, Al-Ruzzeh S, Amrani M. Mitral annulus distortion during beating heart surgery: a potential cause for hemodynamic disturbance—a three-dimensional echocardiography reconstruction study. Ann Thorac Surg 2002; 73(5): 1424–30.
Off-pump coronary artery bypass surgery
79
35. Mathison M, Edgerton JR, Horswell JL, Akin JJ, Mack MJ. Analysis of hemodynamic changes during beating heart surgical procedures. Ann Thorac Surg 2000; 70(4): 1355–60 (discussion 1360–1). 36. Nierich AP, Diephuis J, Jansen EW, Borst C, Knape JT. Heart displacement during off-pump CABG: how well is it tolerated? Ann Thorac Surg 2000; 70(2): 466–72. 37. Mueller XM, Chassot PG, Zhou J, Eisa KM, Chappuis C, Tevaearai HT, Von Segesser LK. Hemodynamics optimization during off-pump coronary artery bypass: the ‘no compression’ technique. Eur J Cardiothorac Surg 2002; 22(2): 249–54. 38. Djaiani GN, Ali M, Heinrich L, Bruce J, Carroll J, Karski J, Cusimano RJ, Cheng DC. Ultrafast-track anesthetic technique facilitates operating room extubation in patients undergoing offpump coronary revascularization surgery. J Cardiothorac Vasc Anesth 2001; 15(2): 152–7. 39. Myles PS, Daly DJ, Djaiani G, Lee A, Cheng DC. A systematic review of the safety and effectiveness of fast-track cardiac anesthesia. Anesthesiology 2003; 99(4): 982–7. 40. Hemmerling TM, Prieto I, Choiniere JL, Basile F, Fortier JD. Ultra-fast-track anesthesia in offpump coronary artery bypass grafting: a prospective audit comparing opioid-based anesthesia vs thoracic epidural-based anesthesia. Can J Anaesth 2004; 51(2): 163–8. 41. Van der Linden PJ, Daper A, Trenchant A, De Hert SG. Cardioprotective effects of volatile anesthetics in cardiac surgery. Anesthesiology 2003; 99(2): 516–17. 42. De Hert SG, Cromheecke S, ten Broecke PW, Mertens E, De Blier IG, Stockman BA, Rodrigus IE, Van der Linden PJ. Effects of propofol, desflurane, and sevoflurane on recovery of myocardial function after coronary surgery in elderly high-risk patients. Anesthesiology 2003; 99(2): 314–23. 43. De Hert SG, Van der Linden PJ, Cromheecke S, Meeus R, ten Broecke PW, De Blier IG, Stockman BA, Rodrigus IE. Choice of primary anesthetic regimen can influence intensive care unit length of stay after coronary surgery with cardiopulmonary bypass. Anesthesiology 2004; 101(1): 9–20. 44. De Hert SG, Van der Linden PJ, Cromheecke S, Meeus R, Nelis A, Van Reeth V, ten Broecke PW, De Blier IG, Stockman BA, Rodrigus IE. Cardioprotective properties of sevoflurane in patients undergoing coronary surgery with cardiopulmonary bypass are related to the modalities of its administration. Anesthesiology 2004; 101(2): 299–310. 45. Mathews ET, Abrams LD. Intrathecal morphine in open heart surgery. Lancet 1980; 2(8193): 543. 46. Bettex DA, Schmidlin D, Chassot PG, Schmid ER. Intrathecal sufentanil-morphine shortens the duration of intubation and improves analgesia in fast-track cardiac surgery. Can J Anaesth 2002; 49(7): 711–17. 47. Metz S, Schwann N, Hassanein W, Yuskevich B, Nixon T. Intrathecal morphine for off-pump coronary artery bypass grafting. J Cardiothorac Vasc Anesth 2004; 18(4): 451–3. 48. Wang JK, Nauss LA, Thomas JE. Pain relief by intrathecally applied morphine in man. Anesthesiology 1979; 50(2): 149–51. 49. Van der Meulen EP, Van Aken H, Vermylen J. Anticoagulants and spinal-epidural anesthesia. Anesth Analg 1994; 79(6): 1165–77. 50. Chaney MA. Side effects of intrathecal and epidural opioids. Can J Anaesth 1995; 42(10): 891– 903. 51. Taylor A, Healy M, McCarroll M, Moriarty DC. Intrathecal morphine: one year’s experience in cardiac surgical patients. J Cardiothorac Vasc Anesth 1996; 10(2): 225–8. 52. Visser WA, Liem TH, Van Egmond J, Gielen MJ. Extension of sensory blockade after thoracic epidural administration of a test dose of lidocaine at three different levels. Anesth Analg 1998; 86(2): 332–5. 53. Scott NB, Turfrey DJ, Ray DA, Nzewi O, Sutcliffe NP, Lai AB, Norrie J, Nagels WJ, Ramayya GP. A prospective randomized study of the potential benefits of thoracic epidural anesthesia and analgesia in patients undergoing coronary artery bypass grafting. Anesth Analg 2001; 93(3): 528–35.
The year in anaesthesia and critical care
80
54. Ho AM, Chung DC, Joynt GM. Neuraxial blockade and hematoma in cardiac surgery: estimating the risk of a rare adverse event that has not (yet) occurred. Chest 2000; 117(2): 551– 5. 55. Liem TH, Hasenbos MA, Booij LH, Gielen MJ. Coronary artery bypass grafting using two different anesthetic techniques: part 2: post-operative outcome. J Cardiothorac Vasc Anesth 1992; 6(2): 156–61. 56. Priestley MC, Cope L, Halliwell R, Gibson P, Chard RB, Skinner M, Klineberg PL. Thoracic epidural anesthesia for cardiac surgery: the effects on tracheal intubation time and length of hospital stay. Anesth Analg 2002; 94(2): 275–82 (table of contents). 57. Watters MP, Ascione R, Ryder IG, Ciulli F, Pitsis AA, Angelini GD. Haemodynamic changes during beating heart coronary surgery with the ‘Bristol Technique’. Eur J Cardiothorac Surg 2001; 19(1): 34–40. 58. Comunale ME, Body SC, Ley C, Koch C, Roach G, Mathew JP, Herskowitz A, Mangano DT. The concordance of intraoperative left ventricular wall-motion abnormalities and electrocardiographic S—T segment changes: association with outcome after coronary revascularization. Multicenter Study of Perioperative Ischemia (McSPI) Research Group. Anesthesiology 1998; 88(4): 945–54. 59. Seeberger MD, Skarvan K, Buser P, Brett W, Rohlfs R, Erne JJ, Rosenthaler C, Pfisterer M, Atar D. Dobutamine stress echocardiography to detect inducible demand ischemia in anesthetized patients with coronary artery disease. Anesthesiology 1998; 88(5): 1233–9. 60. Smith MD, MacPhail B, Harrison MR, Lenhoff SJ, DeMaria AN. Value and limitations of transesophageal echocardiography in determination of left ventricular volumes and ejection fraction. J Am Coll Cardiol 1992; 19(6): 1213–22. 61. Bilfinger TV, Lin CY, Anagnostopoulos CE. In vitro determination of accuracy of cardiac output measurements by thermal dilution. J Surg Res 1982; 33(5): 409–14. 62. Nishikawa T, Dohi S. Errors in the measurement of cardiac output by thermodilution. Can J Anaesth 1993; 40(2): 142–53. 63. Heerdt PM, Blessios GA, Beach ML, Hogue CW. Flow dependency of error in thermodilution measurement of cardiac output during acute tricuspid regurgitation. J Cardiothorac Vasc Anesth 2001; 15(2): 183–7. 64. Valtier B, Cholley BP, Belot JP, De la Coussaye JE, Mateo J, Payen DM. Noninvasive monitoring of cardiac output in critically ill patients using transesophageal Doppler. Am J Respirat Crit Care Med 1998; 158(1): 77–83. 65. Baillard C, Cohen Y, Fosse JP, Karoubi P, Hoang P, Cupa M. Haemodynamic measurements (continuous cardiac output and systemic vascular resistance) in critically ill patients: transoesophageal Doppler versus continuous thermodilution. Anaesth Intensive Care 1999; 27(1): 33–7. 66. Lefrant JY, Bruelle P, Aya AG, Saissi G, Dauzat M, De la Coussaye JE, Eledjam JJ. Training is required to improve the reliability of esophageal Doppler to measure cardiac output in critically ill patients. Intensive Care Med 1998; 24(4): 347–52. 67. Kong RS, Trivedi U. Esophageal Doppler monitoring in off-pump cardiac surgery. J Cardiothorac Vasc Anesth 2004; 18(4): 539–40. 68. Shanewise JS, Cheung AT, Aronson S, Stewart WJ, Weiss RL, Mark JB, Savage RM, SearsRogan P, Mathew JP, Quinones MA, Cahalan MK, Savino JS. ASE/SCA guidelines for performing a comprehensive intraoperative multiplane transesophageal echocardiography examination: recommendations of the American Society of Echocardiography Council for Intraoperative Echocardiography and the Society of Cardiovascular Anesthesiologists Task Force for Certification in Perioperative Transesophageal Echocardiography. Anesth Analg 1999; 89(4): 870–84.
Off-pump coronary artery bypass surgery
81
69. Brown Jr PM, Kim VB, Boyer BJ, Lust RM, Chitwood Jr WR, Elbeery JR. Regional left ventricular systolic function in humans during off-pump coronary bypass surgery. Circulation 1999; 100(19 Suppl): II125–7. 70. Moises VA, Mesquita CB, Campos O, Andrade JL, Bocanegra J, Andrade JC, Buffolo E, Carvalho AC. Importance of intraoperative transesophageal echocardiography during coronary artery surgery without cardiopulmonary bypass. J Am Soc Echocardiogr 1998; 11(12): 1139– 44.
4 Protecting the heart in non-cardiac surgery SIMON HOWELL, PETER KIMPSON Introduction The size of the problem Patients with pre-existing cardiac disease undergoing anaesthesia and surgery are at increased risk of major cardiac complications in the perioperative period |1|. Those undergoing major vascular surgery represent a group at particular risk because the presence of arterial disease identifies them as being at risk of coronary artery disease (CAD) and cerebrovascular disease. Sixty per cent of 1000 patients scheduled to undergo major vascular surgery who underwent preoperative coronary angiography had significant CAD |2|. The high incidence of cardiac disease in these patients is reflected in a high cardiovascular complication rate |3|. The combination of significant cardiac risk factors and a significant surgical insult is common in patients under-going major vascular surgery, but is not restricted to this group of patients |4|. Major vascular and cardiac surgery seems to be associated with most perioperative cardiac complications |5|. The advent of the cardiac troponin assays has demonstrated that myocardial infarction is not a dichotomous event, but is accompanied by a spectrum of cardiac troponin release which correlates with the degree of myocardial injury in patients presenting with chest pain |6–8|. Even a relatively small release of cardiac troponin has prognostic significance |9|. This is also the case in patients undergoing major vascular surgery |10|. Mechanisms of infarction A question arises as to whether perioperative myocardial infarction is a result of an acute coronary syndrome with plaque rupture and subsequent thrombosis resulting in the occlusion of a coronary artery or is due to a myocardial oxygen supply/demand imbalance caused by perioperative haemodynamic aberrations. In favour of the latter possibility, perioperative myocardial ischaemia is strongly associated with perioperative cardiac ischaemic events |11, 12|. © Atlas Medical Publishing Ltd
In favour of the first possibility, retrospective post mortem findings showed a comparable incidence of plaque rupture and thrombosis in patients who had died of a perioperative or a non-operative myocardial infarction |13|. Furthermore, the severity of the underlying
Protecting the heart in non-cardiac surgery
83
coronary stenoses did not predict the myocardial infarction territory. These findings suggest that a critical coronary stenosis-producing electrocardiogram (EGG) and clinical evidence of myocardial ischaemia may be a marker for an increased risk of perioperative myocardial infarction rather than being the direct cause of it. Protracted myocardial ischaemia as well as coronary plaque rupture with subsequent thrombosis may both be involved in the aetiology of perioperative myocardial infarction |14|. Most ischaemic episodes begin at emergence from anaesthesia, a period also characterized by increases in heart rate, blood pressure, sympathetic tone and procoagulant activity. An increased sympathetic tone may produce an increase in heart rate, blood pressure, coronary artery tone and shear stress at coronary plaques. This may trigger haemodynamically induced myocardial oxygen demand/ supply imbalances, plaque disruption, coronary vasospasm, plaque disruption and thrombosis |14|. Preventive strategies The prevention of perioperative myocardial infarction and death has taxed physicians for many decades. Intra-operative haemodynamic stability is important. Intra-operative tachycardia |15| and hypotension |16| have been associated with perioperative myocardial injury. However, although haemodynamic stability is important, the majority of perioperative myocardial ischaemia is not associated with changes in the heart rate and blood pressure and other strategies for the protection of the heart have been sought. β-blockade It has been known for several decades that perioperative β-blockade reduces the incidence of perioperative myocardial arrhythmia and ischaemia. Patients randomized to receive a single dose of labetalol, atenolol or oxprenolol with their pre-medication demonstrated a lower incidence of myocardial ischaemia at induction and emergence from anaesthesia than patients not β-blocked |17|. In a non-randomized study of patients undergoing aortic aneurysm repair, metoprolol reduced the incidence of perioperative myocardial infarction (3 versus 18%) and arrhythmias (12.5 versus 57%) compared to patients not receiving metoprolol |18|. While these and several other studies added to the evidence base for β-blockade, it was only in 1996 that a double-blind, randomized controlled trial showed a significant improvement in 2-year survival in patients receiving atenolol perioperatively |19|. However, this study has been criticized on a number of grounds |20|. The patients already receiving β-blockers had these discontinued and were randomized. Six in-hospital deaths were excluded from the analysis. The patients receiving placebo had a higher incidence of cardiac disease and were potentially at higher risk of cardiac complications. Finally, patients who were randomized to receive β-blockers were more likely to continue to receive β-blockers following discharge from hospital. As the majority of the benefit from β-blockade accrued in the 6 months following surgery and β-blockers are known to be effective cardiovascular secondary prevention, this may explain the benefit from perioperative β-blockade in this study. Thus, while this work is important as the first randomized controlled trial of perioperative β-blockade its results should be interpreted with caution.
The year in anaesthesia and critical care
84
More recently, a study of 112 patients with proven CAD undergoing major vascular surgery showed a startling difference in outcome between patients randomized to either bisoprolol or standard care |21|. The incidence of death or non-fatal myocardial infarction was 3.4% in the bisoprolol group as compared with 34% in the standard care group. However, once again the results of the study must be interpreted with caution. The study was not blinded, not placebo controlled and terminated prematurely. The study population was highly selected. Of the more than 1300 patients initially screened, less than 10% (n=112) were finally included. Patients with very severe CAD were excluded. Thus, the results provide evidence for β-blockade in patients at high risk, as identified by non-invasive cardiac testing, but do not necessarily support β-blocking every patient who has a history of cardiac disease or who has cardiac risk factors. While there is little doubt that β-blockers reduce the burden of perioperative myocardial ischaemia |22|, there are other potential mechanisms whereby β-blockers may yield benefit. These include a reduction in the incidence of perioperative arrhythmias, changes in platelet function, reductions in cardiomyocyte apoptosis in response to sympathetic stimulation, changes in myocardial gene expression, reduced downregulation of β-adrenergic receptors and alterations in myocardial bioenergetics (reviewed in |23|). Sympatholytic therapies α2-agonists α2-agonists (e.g. clonidine and mivazerol) offer an alternative pharmacological means of reducing the activity of the sympathetic nervous system. In a large study mivazerol did not reduce the incidence of perioperative cardiac complications compared to placebo |24|. However, in a planned subgroup analysis of 904 patients undergoing major vascular surgery, it reduced the incidence of perioperative myocardial infarction. This work has not been followed up by other studies and seems unlikely to be so in the near future. A recent meta-analysis addressed the potential benefits of α2-agonists (see Stevens et al. below). Neuroaxial blockade Neuroaxial blockade using either spinal or epidural anaesthesia may also produce sympatholysis. Many studies have been conducted on the impact of neuroaxial blockade on cardiac and non-cardiac outcomes, but it has proved difficult to establish unequivocal evidence of benefit. A meta-analysis appeared to show a reduction in mortality and a reduced incidence of perioperative myocardial infarction amongst patients receiving neuroaxial anaesthesia |25|. However, this finding has been challenged. On closer examination the benefits shown in the meta-analysis rest on the results of three studies, all of which had extremely high adverse event rates among patients not receiving neuroaxial blockade |26|.
Protecting the heart in non-cardiac surgery
85
Glycaemic control Tight glycaemic control reduced the morbidity and mortality in critically ill patients |27|. It remains unclear as to whether this benefit is primarily due to the glycaemic control per se or to the effects of insulin |28|. It is well established that patients who have suffered a myocardial infarction in the non-operative setting benefit from tight glycaemic control and those patients in whom this is not achieved fare less well |29|. This is an area where further work in patients undergoing non-cardiac surgery has the potential to show significant benefit. Volatile anaesthestic and ischaemic preconditioning It is now well understood that brief episodes of myocardial ischaemia can protect the heart from a subsequent ischaemic insult |30|. A more surprising but now well-established finding is that volatile agents can produce a cardiac protective effect similar to ischaemic preconditioning |31, 32|. While well established in the laboratory setting, much work remains to be done examining this potential benefit of volatile agents in man. The mechanisms by which the preconditioning by volatile agents confer benefit and the potential role of this treatment in man are examined further below in a discussion of two reviews (see Zaugg et al. below).
Association of cardiac troponin, CK-MB and post-operative myocardial ischemia with long-term survival after major vascular surgery Landesberg G, Shatz V, Akopnik I, et al. J Am Coll Cardiol 2003; 42: 1547–54 BACKGROUND. Cardiac troponins are sensitive and specific markers of myocardial necrosis and predict mortality and morbidity from cardiac causes in non-surgical patients presenting with acute coronary syndromes. The diagnosis of perioperative myocardial infarction is difficult. It is often clinically silent and the ECG changes of non-Q wave infarction may be transient. Diagnosis is often based solely on the characteristic rise and fall of biochemical markers such as creatine kinase MB fraction and cardiac troponin. In non-surgical patients presenting with chest pain, even small elevations in cardiac troponins are associated with an increased incidence of adverse cardiac outcomes. There is some evidence that this also applies in the surgical setting, although the threshold concentrations of troponin that would signify a perioperative myocardial infarction are unclear. This study aimed to determine the threshold concentrations of markers signifying a perioperative myocardial infarction and to examine the long-term prognosis of patients with infarction diagnosed according to these criteria. INTERPRETATION. A cohort of 447 patients undergoing major vascular surgery was prospectively studied. Continuous twelve-lead ECG monitoring was instituted perioperatively and serum concentrations of troponin I and/or troponin T and creatine kinase MB fraction were measured on the first 3 days post operatively Depending on the
The year in anaesthesia and critical care
86
biochemical marker thresholds chosen, between 2.9 and 23.9% of the patients had a perioperative myocardial infarction. Only 17.7–61.9% of these patients had signs and symptoms attributable to infarction and none had new Q waves. During a follow-up period of between 1 and 5 years, 82 (18.3%) patients died. All thresholds of creatine kinase MB fraction and troponin predicted long-term survival, as did ischaemia duration on an ECG of greater than 30 and 60 min (Fig. 4.1). Comment This study supports the hypothesis that even minor elevations of cardiac troponins have a similar adverse prognostic implication in the perioperative setting, as in the non-operative setting. Higher levels of marker release corresponding to a larger myocardial injury were associated with worse long-term survival. Post-operative troponin release was correlated with prolonged periods of ST-depression myocardial ischaemia. This work confirms the findings of previous studies on the prognostic value of post-operative troponin release and extends the findings to long-term survival. The
Fig. 4.1 Kaplan—Meier survival curves of all patients grouped according to their highest postoperative troponin level. cTn, cardiac troponin. Source: Landesberg et al. (2003).
Protecting the heart in non-cardiac surgery
87
creatine kinase MB fraction levels predicted long-term mortality independently of troponin although troponin was more sensitive. Patients with a combination of raised creatine kinase MB fraction and troponin had even worse long-term outcomes.
Pre-operative thallium scanning, selective coronary revascularization and long-term survival after major vascular surgery Landesberg G, Mosseri M, Wolf YG, et al. Circulation 2003; 108: 177–83 BACKGROUND. Significant CAD and cardiovascular complications leading to perioperative and late mortality are common in patients presenting for major vascular surgery. Reversible myocardial ischaemia detected by pre-operative thallium scanning predicts perioperative and late cardiac events and survival in this group of patients. A number of studies have suggested improved perioperative and long-term survival in patients undergoing successful coronary artery bypass grafting (CABG) before vascular surgery. Most of these studies take no account of the complications associated with CABG surgery and percutaneous coronary transluminal angioplasty (PCTA). Such complications may adversely affect overall outcomes in patients having undergone pre-operative coronary revascularization. This study aimed to determine whether a strategy of screening for significant reversible cardiac ischaemia with thallium scanning followed by revascularization improves the long-term survival after major vascular surgery. INTERPRETATION. Data were collected retrospectively on 502 consecutive patients presenting for major vascular surgery between 1990 and 1999. Pre-operative thallium scanning was routinely performed. Patients found to have moderate to severe reversible perfusion defects on thallium scanning were defined as having moderatesevere reversible ischaemia and referred for coronary angiography and consideration of revascularization. PCTA was performed for coronary stenosis greater than 70% where technically feasible. CABG was performed in patients with left main coronary stenosis greater than 50%, in diabetic patients with multi-vessel disease or patients with two- or three-vessel disease unsuitable for PCTA. On the basis of thallium scanning, patients were divided into four groups: group 1 contained patients with no or mild fixed or reversible defects, group 2 contained patients with moderate or severe fixed defects, group 3 contained patients with moderate-severe reversible defects who did not undergo revascularization and group 4 contained patients with moderate-severe reversible defects who underwent coronary revascularization prior to their vascular surgery. These patients were followed up to determine the long-term survival. Patients who suffered morbidity or mortality due to angiography or CABG and did not proceed to vascular surgery were included in group 4. Four hundred and seven patients underwent pre-operative thallium scanning. Seventy-four patients in group 4 with moderate-severe reversible defects underwent revascularization and 62 patients in group 3 did not. Moderate-severe reversible ischaemia predicted worse survival by Cox multivariate analysis. Of the patients with moderate-severe reversible ischaemia, those who were revascularized before their major vascular surgery had improved long-term (5-year) survival compared to those who did not. The difference at 30 days and 1 year was not significant (Fig. 4.2).
The year in anaesthesia and critical care
88
Fig. 4.2 Kaplan-Meier survival curves of all groups. ‘No PTS’ indicates patients without pre-operative thallium scanning. Group 1, patients with no or mild defects on thallium scanning; group 2, patients with only moderate— severe fixed defects; group 3, patients with moderate—severe reversible defects who did not undergo preoperative coronary revascularization; group 4, patients with moderate— severe reversible defects who underwent pre-operative coronary revascularization by CABG or PCTA. Group 3 had a significantly worse long-term survival than the other groups. Source: Landesberg et al. (2003).
Protecting the heart in non-cardiac surgery
89
Comment The 2002 update of the American College of Cardiology/American Heart Association (ACC/AHA) Guidelines for Perioperative Cardiac Evaluation in Non-cardiac Surgery recommend that CABG and percutaneous coronary intervention should only be undertaken prior to surgery if indicated under the relevant ACC/AHA guidelines for these therapies |33|. This study suggests that these interventions may have a place in the treatment of high-risk patients undergoing vascular surgery. However, the study should be interpreted with some caution. It is a retrospective analysis of a time period of over 10 years and as such examined the cardiological, anaesthetic and perioperative care of a previous decade. There have been many changes in perioperative management in this period, not least the emergence of perioperative treatment with β-blockers and statins. While the long-term outcome was improved in the revascularized group, the immediate and 1-year outcomes were identical in each group. That is to say, the main finding of this study was that revascularization improves outcome in the long term in individuals who have inducible myocardial ischaemia. This is unsurprising.
The value of pre-operative pharmacologic stress testing before vascular surgery using ACC/AHA guidelines: a prospective, randomized trial Falcone RA, Nass C, Jermyn R, et al. J Cardiothorac Vasc Anesth 2003; 17: 694–8 BACKGROUND. Debate exists as to whether additional stress testing in patients stratified using clinical predictors from the ACC/AHA Guidelines on Perioperative Evaluation Before Non-cardiac Surgery is of benefit in patients undergoing vascular surgery |34|. After stratification by ACC/AHA guideline parameters, 99 patients were randomized to pre-operative cardiac stress testing by dobutamine stress echocardiography or no testing and followed for up to 12 months post-operatively for adverse cardiac outcomes. INTERPRETATION. Of the 46 patients who were randomized to undergo preoperative stress testing, seven (15%) had inducible ischaemia. None of these seven patients suffered an adverse cardiac outcome in the immediate post-operative period. There was no significant difference in adverse cardiac outcome between the group found to have inducible ischaemia and the group without inducible ischaemia (Fig. 4.3). There was one non-fatal adverse cardiac outcome (no stress test) and one cardiac death (abnormal stress test) at 12-months follow-up in 79 (80%) patients, reflecting a 1% 12month cardiac morbidity and mortality. In this small prospective, randomized study evaluating the value of pre-operative cardiac stress ACC/AHA guidelines before major vascular surgery, pre-operative cardiac stress testing offered no incremental value for determining post-operative adverse cardiac outcomes. Comment The ACC/AHA guidelines suggest algorithms for additional pre-operative cardiac stress testing based on clinical predictors and the type of surgery for predicting adverse post-
The year in anaesthesia and critical care
90
operative cardiac risk. The authors hypothesized that the low positive predictive value of ischaemia and wall motion abnormality found on such testing would not improve the preoperative cardiac risk stratification over and above that gained by the assessment of clinical and surgical factors. The findings suggest that, in the limited subset of patients with minor and intermediate clinical predictors as defined by the ACC/AHA guidelines, the use of pre-operative stress testing does not add further to risk stratification. The study had several limitations that need to be taken into consideration before applying the findings to the general population. Patients with high-risk clinical predictors were excluded. Patients with a recent full cardiac assessment by another physician or patients having had recent coronary revascularization were also excluded from randomization. The overall adverse event rate in this study was low at around 4%, as compared to rates of 5–15% previously found. The number of patients was also low. Twenty per cent of the patients were not followed up at 12 months. It is thus
Fig. 4.3 Adverse cardiac outcomes in the immediate post-operative period after stratification by ACC/AHA clinical risk predictors and stress testing. Source: Falcone et al. (2003). possible that missed adverse outcomes in this group of 20 patients modified the results. The exclusion of patients who had seen a physician for cardiac assessment prior to vascular surgery presumably lost a group of patients with cardiac symptoms who might have brought a different but more typical risk profile to the population studied. The recommendations for pre-operative stress testing are based almost entirely on observational data. This study is important because it represents a formal trial of the utility of pre-operative stress testing. Probably the most important conclusion to be drawn from this work is that, because of the low perioperative cardiac event rate, a formal trial
Protecting the heart in non-cardiac surgery
91
of pre-operative testing will require a very large number of patients and must include high-risk patients.
Clinical outcome of patients undergoing non-cardiac surgery in the 2 months following coronary stenting Wilson SH, Fasseas P, Orford JL, et al. J Am Coll Cardiol 2003; 42: 234–40 BACKGROUND. Angioplasty followed by placement of a coronary artery stent is now a common method of myocardial revascularization. Although stent thrombosis occurs
Fig. 4.4 Adverse events within 2 months of coronary stent placement. (MI, myocardial infarction; TVR, target vessel revascularization.) Source: Wilson et al. (2003). infrequently, it is a devastating complication usually resulting in Q wave myocardial infarction or death. Thrombosis occurs mostly during the first few days after stenting. A small previous study found high mortality when surgery was undertaken within 2 weeks of stent placement |35|. This retrospective study aimed to investigate the effect of non-cardiac surgery on the risk of stent thrombosis and the optimal delay between stent placement and subsequent surgery. The authors analysed a hospital database of surgical and percutaneous coronary stent patients to identify 207 individuals who underwent intermediate and major non-cardiac surgery within 60 days of coronary stenting. The end-points used were death, Q wave and non-Q wave myocardial infarction stent thrombosis and the need for repeat
The year in anaesthesia and critical care
92
revascularization. INTERPRETATION. Eight out of 207 patients suffered an adverse cardiac ischaemic event. Six of these patients died. All stent-related complications occurred in surgical patients within 6 weeks of stent placement (Fig. 4.4). There were no adverse clinical events noted due to excessive bleeding in patients taking anticoagulant or antiplatelet medication. The findings suggest that undergoing non-cardiac surgery within 6 weeks of coronary stenting increases the risk of stent thrombosis, myocardial infarction and death. This increased risk persists for 6 weeks and is consistent with the time taken for the coronary artery to re-endothelialize. Surgery induces a stress response and subsequently a pro-thrombotic state which increases the risk of thrombosis in the recently stented coronary artery devoid of endothelium. Anti-platelet drugs reduce stent thrombosis and delaying surgery for 6 weeks allows the administration of a full course of these drugs. The option to delay surgery may be limited as one-third of the surgical procedures were for vascular repair of vessels damaged in the stenting procedure.
Cost-effectiveness of differing perioperative β blockade strategies in vascular surgery patients Fleisher LA, Corbett W, Berry C, Poldermans D. J Cardiothorac Vasc Anesth 2004; 18: 7–13 BACKGROUND. Evidence exists that the use of β-adrenergic antagonists in the perioperative period leads to a reduction in cardiac mortality and morbidity in patients undergoing major vascular surgery. A minority of patients presenting for major vascular surgery will be taking these drugs chronically and it is unclear what protocols will be effective when instituted in the perioperative period. The studies that have suggested benefit of β-blockade used a variety of drugs and protocols. The aim of this retrospective study was to determine the cost-effectiveness of different perioperative oral and intravenous β-blocker protocols during the perioperative period in high-risk vascular patients undergoing abdominal aortic aneurysm surgery. Five perioperative strategies were investigated: (1) no routine β-blockade, (2) preoperative oral bisoprolol for 7 days followed by perioperative intravenous metoprolol and oral bisoprolol based on pre-operative tit rat ion, (3) immediate preoperative atenolol with post-operative intravenous then oral atenolol, (4) intraoperative esmolol and post-operative intravenous then oral atenolol and (5) intraoperative and 18 h of post-operative esmolol then atenolol. Published mortality data were used for producing a decision analytic tree comparing the different strategies. Cost data were derived from a selection of US Medicare beneficiaries. INTERPRETATION. Perioperative death and myocardial infarction were associated with net increases of $21 909 and $15 000, respectively, in charges to Medicare. All five strategies were associated with net cost savings and an increase in efficacy, but the strategy involving pre-operative oral titration of bisoprolol for a minimum of 7 days was the most effective one (Table 4.1). The latter strategy was associated with a net hospital saving of $500 and a net increase in efficacy of 0 03 Based on a sensitivity analysis the
Protecting the heart in non-cardiac surgery
93
probability of death, the potential cost savings and the risk reduction increased, suggesting that the value of β-blockers is a function of the underlying risk in the surgical population. Comment This study was predicated on the assumption that perioperative β-blockade is an effective strategy for a reduction in cardiac mortality and morbidity in patients undergoing major vascular surgery. The work described here seeks to bolster the clinical case with an economic case for perioperative β-blockade and to determine the cost-effectiveness of different perioperative oral and intravenous protocols for β-blockade. As discussed in the introduction to this chapter, the evidence for the perioperative use of β-blockers is incomplete. This analysis, while interesting, takes no account of the costs associated with the adverse events resulting from perioperative β-blockade. Although all strategies saved money (even accounting for pharmaceutical costs), the optimal strategy for the treatment of patients who do not take β-blockers when presenting for surgery requires further study.
Table 4.1 Cost-effectiveness of the different strategies Strategy
Cost
Marginal Efficacy cost
Marginal efficacy
Cost/effect
Pre-operative 30-day bisoprolol (atenolol)
$28.4K
0.9914
No pre-operative beta-blockers— intraoperative esmolol + postoperative atenolol
$28.5K $0.1K
0.9805
–0 0109
$29.101
Immediate pre-operative atenolol
$28.6K $0.2K
0.9805
–0.0109
$29.145
No pre-operative beta-blockers— $28.6K $0.2K intra/post-operative esmolol+postoperative atenolol
0 9805
–0 0109
$29.147
No perioperative beta-blockers
0.961
–0.0304
$30.088
$28.9K $0.5K
$28.652
Source: Fleisher et al. (2004).
Pharmacologic myocardial protection in patients undergoing non-cardiac surgery: a quantitative systematic review Stevens RD, Burri H, Tramér MR. Anesth Analg 2003; 97: 623–33 BACKGROUND. A number of drugs have been tested in clinical trials for decreasing cardiac complications in patients undergoing non-cardiac surgery. To compare the results of these studies the authors conducted a quantitative systematic
The year in anaesthesia and critical care
94
review. The Medline, Embase and Cochrane databases were searched for randomized trials that assessed myocardial ischaemia, myocardial infarction, 30day cardiac mortality and adverse effects. The data were combined using a fixedeffect model and expressed as Peto odds ratios (ORs) with 95% confidence intervals (CIs) and as the numbers needed to treat/harm. Twenty-one trials involving 3646 patients were included: eleven trials using β-blockers (six drugs and 866 patients), six trials using clonidine or mivazerol (614 patients), three trials using diltiazem or verapamil (121 patients) and one trial using nitroglycerin (45 patients). All trials had an inactive control; there were no direct comparisons. INTERPRETATION. β-blockers decreased ischaemic episodes during surgery (7.6 versus 20.2% with placebo) (OR=0.32, 95% Cl=0.17–0.58 and number needed to treat— 8) and after surgery (15.2 versus 27.9% with control) (OR–0.46, 95% Cl–0.26–0.81 and number needed to treat=8). α2-agonists decreased ischaemia during surgery only (19.4 versus 32.8%) (OR=0.47, 95% Cl=0.33–0.68 and number needed to treat=7). β-blockers reduced the risk of myocardial infarction (0.9 versus 5.2%) (OR=0.19, 95% Cl=0.08– 0.48 and number needed to treat=23), but only when two trials with high-risk patients were included. The effect of α2-agonists on myocardial infarction was not significant (6.1 versus 7.3%) (OR=0.85 and 95% Cl=0.62–1.14). β-blockers significantly decreased the risk of cardiac death from 3.9 to 0.8% (OR=0.25, 95% Cl=0.09–0.73 and number needed to treat=32). α2-agonists significantly decreased the risk of cardiac death from 2.3 to 1.1% (OR=0.50, 95% Cl=0.28–0.91 and number needed to treat=83). For calcium channel blockers and nitroglycerin, evidence of any benefit was lacking. The most common adverse effect was bradycardia, which occurred in 24.5% of patients receiving a β-adrenergic blocker versus 9.1% of controls (OR=3.76, 95%=Cl 2.45–5.77 and number needed to harm=6). Comment This quantitative systematic review aimed to review the literature concerning drug therapies that might improve cardiac morbidity and mortality in non-cardiac surgery in order to gauge efficacy and risk. It concentrated on β-blockers and α2-agonists, but also found evidence related to calcium channel blockers and nitroglycerin. It dealt only with perioperative events occurring within 30 days of surgery. All of the studies included in the analysis were randomized controlled trials comparing an active drug against placebo. Meta-analysis of the β-blockade data showed a significant reduction in non-fatal myocardial infarction, but this benefit disappeared when two trials with unusually high control group event rates were excluded from the analysis (Fig. 4.5). β-blockers and α2agonists reduced intra-operative ischaemia and β-blockers continued to reduce ischaemia in the post-operative period, β-blockers and α2-agonists reduced perioperative cardiac mortality but again, when a study with an extremely high control group event rate was removed, the meta-analysis failed to show significant benefit (Fig. 4.6). There was a significantly increased incidence of bradycardia in the β-blocker group.
Protecting the heart in non-cardiac surgery
95
Fig. 4.5 Prevention of non-fatal myocardial infarction. On the event rate scatters, symbol areas are proportional to the size of the trials Trials with zero events are not shown. Source: Stevens et al. (2003).
Fig. 4.6 Prevention of perioperative cardiac mortality. On the event rate scatters, symbol areas are proportional to the size of the trials. Trials with zero events are not shown. Source: Stevens et al. (2003). One study has recently questioned the conclusion that β-blockade and α2-agonists offer significant protection against cardiac morbidity and mortality |36|. The β-blocker meta-analysis missed one non-fatal myocardial infarction in the β-blocker arm and assumed that the myocardial infarctions in another trial were non-fatal. The only β-
The year in anaesthesia and critical care
96
blocker study that showed significant improvement in 30-day cardiac mortality had an unusually high event rate in the control group, was unblinded and was stopped early after interim analysis showed a large benefit in the treatment group in preventing non-fatal myocardial infarction and cardiac death in the β-blocker group. The large relative risk reductions are implausibly large and inconsistent with the magnitude of effect seen in large, randomized controlled trials in acute myocardial infarction. Furthermore, the heterogeneity of the mechanisms by which perioperative cardiac events are likely to occur and the fact that β-blockade only affects a limited number of these mechanisms means that relative risk reductions in excess of 25% are unlikely to occur. The use of β-blockade for perioperative cardiac protection in non-cardiac surgery remains promising but is not yet worthy of the level 1 recommendation given by the AHA/ACC. Small-scale studies with few events should be interpreted cautiously. Conclusive evidence from large-scale studies is awaited.
Reduction in cardiovascular events after vascular surgery with atorvastatin: a randomized trial Durazzo AES, Machado FS, Ikeoka DT, et al. J Vasc Surg 2004; 39: 967–76 BACKGROUND. A substantial body of work exists demonstrating the efficacy of statins for the primary and secondary prevention of cardiovascular events in the non-surgical setting. The evidence that statins are efficacious in the operative setting is limited to retrospective case-control studies in vascular and cardiac surgery. This randomized prospective study was designed for evaluating whether the perioperative use of statins in patients not already taking them would reduce the incidence of fatal and non-fatal cardiovascular events in major vascular surgery. One hundred patients presenting for major vascular surgery were randomized to receive either 20 mg of atorvastatin or placebo daily for 45 days. Vascular surgery was performed within 2 weeks of starting the study medication. The patients were followed up post-operatively for 6 months. The primary end-points were cardiac death, acute non-fatal myocardial infarction, ischaemic stroke and unstable angina. INTERPRETATION. The combined rate of end-points was 26% in the placebo group compared to 8% in the atorvastatin group (P=0.031). β-blocker use was similar in both groups. The differences in individual end-points were not significant (Table 4.2). Patients in the statin group showed an increase in event-free survival after surgery, the excess events in the placebo group occurring in the immediate post-operative period (Fig. 4.7). Comment Institution of perioperative statin therapy improved cardiac outcome following vascular surgery. This confirms previous retrospective findings in aneurysm surgery
Protecting the heart in non-cardiac surgery
97
Table 4.2. Event-free survival in the 6 months after vascular surgery, according to study group Characteristic
Atorvastatin (n=50) n
%
Placebo (n=50) n
P
%
Death from cardiac causes
1
2.0
2
4.0
1.000
Non-fatal acute myocardial infarction
3
6.0
8
16.0
0.199
Unstable angina
–
1
2.0
1.000
Ischemic stroke
–
2
4.0
0.495
Combined end-point
4
13
26.0
0.031
8.0
Outcomes measures included death from cardiac causes, non-fatal acute myocardial infarction, ischaemic stroke and unstable angina. The rate of event-free survival at 6 months (180 days) was 91.4% in the atorvastatin group and 73.5% in the placebo group (P=0.018). Source: Durazzo et al. (2004).
Fig. 4.7 Event-free survival in the 6 months after vascular surgery according to study group. Source: Durazzo et al. (2004). |52|. The mechanisms by which statins improve outcome remain to be elucidated. It is unlikely to be due to a short-term improvement in the lipid profile. It may be due to
The year in anaesthesia and critical care
98
plaque-stabilizing and anti-inflammatory actions. Statins are efficacious, safe, welltolerated drugs that deserve further investigation for perioperative use.
Post-operative twelve-lead ECG predicts perioperative myocardial ischaemia associated with myocardial cell damage Böttiger BW, Motsch J, Teschendor P, et al. Anaesthesia 2004; 59: 1083–90 BACKGROUND. This study examined whether intermittent twelve-lead ECG recordings can be used as an early warning tool for identifying patients suffering from perioperative myocardial ischaemia and subsequent myocardial cell damage. Fifty-five vascular surgery patients at risk for or with a history of CAD were monitored for perioperative myocardial ischaemia using intermittent twelve-lead ECG recordings taken pre-operatively and at 15 min and 20, 48, 72 and 84 h postoperatively. The effectiveness of the twelve-lead ECG was gauged by examining concordance with continuous three-channel Holter monitoring and serial analyses of creatine kinase myocardial band isoenzyme and cardiac troponin T and I. INTERPRETATION. The incidence of perioperative myocardial ischaemia detected by twelve-lead ECG was 44% and was identifiable in most patients (88%) 15 min after surgery. The incidence of perioperative myocardial ischaemia detected by continuous monitoring was 53%, with the most severe episodes occurring intra-operatively and during emergence from anaesthesia. The concordance of the twelve-lead method with continuous monitoring was 72%. The concordance of creatine kinase myocardial band activity with the twelve-lead method was 71% and that with Holter monitoring was 57%. The concordance of mass concentration of creatine kinase myocardial band with twelvelead ECG recordings was 75% and the corresponding value for Holter monitoring was 68%. The concordances of cardiac troponin T and I concentrations with the twelve-lead method were 85 and 87%, respectively and the concordances with Holter monitoring were 72 and 66%, respectively. The post-operative twelve-lead ECG identified perioperative myocardial ischaemia associated with subsequent myocardial cell damage in most patients undergoing vascular surgery. Comment Perioperative myocardial ischaemia is the single most important predictor of an adverse cardiac outcome after non-cardiac surgery |11|. Myocardial ischaemia may produce significant myocardial injury in the perioperative period without overt clinical symptoms. As it may be amenable to therapy, the early reliable detection of this often silent phenomenon is of obvious benefit to the patient. This study has shown the value of a simple readily available test in detecting ischaemia in a timely manner to allow treatment. The finding that twelve-lead ECG was as effective in detecting myocardial
Protecting the heart in non-cardiac surgery
99
Fig. 4.8 Number of patients with perioperative myocardial ischaemia as assessed by twelve-lead ECG at different time-points (the incidence of the first detection of perioperative myocardial ischaemia in an individual patient is given in black). The overall incidence of twelve-lead ECGdocumented perioperative myocardial ischaemia was 44% (n=24). Perioperative myocardial ischaemia was detected in the immediate postoperative phase (15 min postoperatively) in 21 out of 24 patients (88%). Source: Böttiger et al. (2004). ischaemia as continuous Holter monitoring, but provided the information in a clinically useful time frame, is of considerable importance. Twelve-lead ECG also detected some episodes of ischaemia not detected by continuous recording possibly due to its ability to detect signs of ischaemia in coronary artery territories not examined by a three-channel recorder. T wave inversion (which was the only sign of ischaemia in 25% of patients with biochemical evidence) would not be detected by Holter monitoring. Also of note is the finding that a single twelve-lead ECG recorded 15 min after surgery detected 88% of patients with ischaemia detectable by ECG (Fig. 4.8). This raises the possibility that a single routine recording in the immediate post-operative period might identify the majority of patients at risk and allow efficient targeting of ischaemia surveillance and treatment.
Anaesthetics and cardiac preconditioning Part I: signalling and
The year in anaesthesia and critical care
100
cytoprotective mechanisms Zaugg M, Lucchinetti E, Uecker M, Pasch T, Schaub MC. Br J Anaesth 2003; 91: 551– 65 BACKGROUND. The phenomenon of ischaemic cardiac preconditioning has been demonstrated in vivo and in vitro in several animal species and indirect evidence exists that it occurs in humans. In brief, short periods of cardiac ischaemia followed by periods of reperfusion induce a series of myocardial changes that limit the infarct size, arrhythmias and ventricular dysfunction during subsequent prolonged ischaemia. A second phenomenon, late preconditioning, has been observed where a secondary window of protection against ischaemia is present 12– 72 h after the initial stimulus. Understanding the cellular mechanisms underlying this powerful adaptive
Fig. 4.9 Signalling for cardiac preconditioning. The signalling components depicted illustrate current knowledge regarding the mechanisms of early (left of dashed line) and delayed (right of dashed line) cardiac preconditioning. ∆ψm, inner mitochondrial membrane potential; AIRed, aldose reductase; Bcl-2, antiapoptotic protein; Ca, sarcolemmal voltage-dependent Ca2+ channels; DAG, diacylglycerol; COX-2,
Protecting the heart in non-cardiac surgery
101
cyclooxygenase type 2; eNOS, endothelial NO synthase; G-proteins, heterotrimeric G-proteins; HSP27 and HSP70, heat shock proteins; iNOS, inducible NO synthase; IP3, inositol triphosphate; |P3R, inositol triphosphate receptor; K, sarcolemmal and mitochondrial KATP channels; MnSOD, manganese superoxide dismutase; NF-KB, nuclear factor-κB; NO, nitric oxide; PIP2, phosphatidylinositol bisphosphate; PKC, protein kinase C; PLC/PLD, phospholipases C and D; ROS, reactive oxygen species; RYR, ryanodine Ca2+-release channel; SERCA2, Ca2+ pump of the SR; SR, sarcoplasmic reticulum. Source: Zaugg et al. (2003).
The year in anaesthesia and critical care
102
Fig. 4.10 Cytoprotective mechanisms of cardiac preconditioning. The sarcolemmal and mitochondrial KATP channels, the sarcolemmal voltagedependent Ca2+ channel and the mitochondrial permeability transition pore are shown. The differently marked bars in the mitochondrial
Protecting the heart in non-cardiac surgery
103
intermembrane space represent the supramolecular complex containing the proton pump, ATP synthase, the adenine nucleotide transporter and mitochondrial creatine kinase. Boldness and the direction of arrows represent the relative intensity and direction of ion flux. (a) The cytosolic Ca2+ concentration under normal conditions is mainly governed by the duration of the action potential and the resting membrane potential. The Ca2+ uniporter is dependent on the mitochondrial inner membrane potential, (b) During ischaemia ± reperfusion injury, the increased cytosolic Ca2+ concentration induces high metabolic activity with accumulation of inorganic phosphate and reactive oxygen species. This is accompanied by swelling of the intermembrane space, with subsequent shrinkage of the mitochondrial matrix and disruption of the supramolecular complex. This impairs energy production and nucleotide transport, resulting in cellular ATP depletion. The permeability transition pore opens and induces dissipation of the inner membrane potential, leading to cell death, (c) Ischaemic and pharmacological preconditioning are thought to exert cardioprotection, first by activation of the sarcolemmal KATP channel, which reduces the intracellular Ca2+ by stabilization of the resting membrane potential below ±80 mV and shortening of the action potential. Secondly, activation of the
The year in anaesthesia and critical care
104
mitochondrial KATP channel leads to an increase in the mitochondrial matrix volume and concomitantly reduces the intermembrane space, which leads to reassembly of the supramolecular complex, closure or maintenance of the closed state of the permeability transition pore and restoration of mitochondrial energy production. Source: Zaugg et al. (2003).
Fig. 4.11 Signalling pathways involved in volatile anaesthetic-induced preconditioning. The signalling cascades that are listed derive from the results of recently published studies using the indicated blockers and activators. Multiple signalling cascades prime the sarcolemmal and mitochondrial KATP channels, allowing prompt opening at the initiation of ischaemia. Arrows indicate positive activity and lines with blunted ends (=) indicate inhibition. Ad, adenosine;
Protecting the heart in non-cardiac surgery
105
CHE, chelerythrine (protein kinase C inhibitor); DAG, diacylglycerol; DIAZO, diazoxide (specific opener of the mitochondrial KATP channel); DPCPX, 8-cyclopentyl-l,3dipropylxanthine (specific adenosine-1 receptor blocker); Gas/Gai/Gaq/Gbg, different G-protein species; 5HD, 5hydroxydecanoate (selective mitochondrial KATP channel blocker); HMR-1098, specific blocker of the sarcolemmal KATP channel; L-NIL, LN6-(1-iminoethyl)lysine; IP3, inositol triphosphate; L-NAME, NG-nitro-Larginine methyl ester (L-NIL and LNAME are nitric oxide synthase inhibitors); MnTBAP, Mn(lll)tetrakis(4-benzoic acid)porphyrine chloride; MPG, N-(2mercaptopropionyl) glycine (MnTBAP and MPG are free radical scavengers); NO, nitric oxide; NOS, nitric oxide synthase; PKC, protein kinase C; PLC, phospholipase C; PTIO, 2-(4carboxyphenyl)-4,40,5,5Ctetramethylimidazole-l-oxyl-3-oxide (nitric oxide scavenger); PTX, pertussis toxin (Gai-protein inhibitor); R, receptor; ROS, reactive oxygen species; SNAP, S-nitroso-/V-acetylDL-penicillamine (nitric oxide donor); SPT=8-sulfophenyl theophylline (nonspecific adenosine receptor blocker). Source: Zaugg et al. (2003). process has the potential for unlocking numerous therapeutic approaches that may salvage myocardium at risk, including the use of pharmacological agents. Several volatile anaesthetics and opioids elicit preconditioning. INTERPRETATION This article reviews the evidence concerning the signalling
The year in anaesthesia and critical care
106
and amplification processes between the preconditioning stimulus and the proposed endeffector mechanisms. Unlike most pharmacological agents, which must be infused directly into the coronary circulation, volatile anaesthetics produce preconditioning effects when administered systemically at normal clinical and subclinical concentrations. It should be noted that most of the evidence for ischaemic and anaesthetic preconditioning has been obtained from healthy myocardial tissue in animal models and it is presently unclear how this might translate to aged, diseased, intact human myocardium where there is most potential for benefit. Both ischaemia and anaesthetics activate G-protein-coupled receptors (Fig. 4.9). This initial signal is amplified by a cascade of multiple protein kinases leading to activation of ATPsensitive potassium (KATP) channels in the sarcolemmal membrane and the mitochondrial membrane which ultimately produces cytoprotection by reducing the mitochondrial and cytosolic calcium overload (Fig. 4.10). Volatile anaesthetics act at adenosine receptors which, when stimulated, activate protein kinase C and also by producing nitric oxide and oxygen free radicals (Fig. 4.11). Opioids act via sigma and kappa opioid receptors and also activate protein kinase C. Protein kinase C stabilizes the mitochondrial KATP channel in the open state. The signalling pathway has been elucidated by the use of specific blocking agents to each step of the process in animal tissue exposed to volatile agents. Comment The discovery of pharmacological preconditioning by volatile anaesthetics is less than a decade old. In that time there have been enormous advances in our under-standing of the mechanisms underlying this process. This review gives a detailed account of the current state of knowledge and makes clear the importance of KATP channels in preconditioning. However, the relative importance of the KATP channels in the sarcolemmal and mitochondrial membranes remains unclear and is an area of ongoing research.
Anaesthetics and cardiac preconditioning. Part II: clinical implications Zaugg M, Lucchinetti E, Garcia C, Pasch T, Spahn DR, Schaub MC. Br J Anaesth 2003; 91: 566–76 Comment This article is the second part of a two-part review of the role of anaesthetics in cardiac preconditioning. It reviews both experimental studies that have implications for clinical practice and clinical studies that provide evidence that ischaemic and anaesthetic preconditioning occurs in the intact human myocardium. There is indirect evidence that ischaemic preconditioning occurs in human hearts in the clinical setting in response to both ischaemia and volatile anaesthetic agents. In patients undergoing PCTA, repeated balloon inflations led to increased tolerance to
Protecting the heart in non-cardiac surgery
107
ischaemia and reduced release of cardiac enzymes |53|. Patients who suffer anginal symptoms before a subsequent myocardial infarction suffer less cardiac dysfunction and have improved survival compared to patients who were asymptomatic before infarction |54|. Similarly, patients undergoing CABG who experienced short periods of aortic crossclamping before subsequent cross-clamping and cardioplegia demonstrated less cardiac impairment, enzyme release and metabolic derangement |55|. There is some evidence that preconditioning by volatile agents may occur during CABG with some studies showing improvement of ST segment changes and haemodynamic recovery |56|. However, the results of studies using markers of myocardial injury have been equivocal, perhaps reflecting release due to atrial cannulation, cardiac manipulation or incomplete cardioplegia distal to stenosis. The mechanisms of ischaemic preconditioning in man appear to be similar to those observed in animal studies. The involvement of α-adrenergic, bradykinin and adenosine receptors coupled to G-proteins, protein kinase C and the crucial role of ATP-sensitive calcium channel in the sarcolemmal and mitochondrial membranes have been demonstrated. In clinical studies, adenosine antagonists block the preconditioning seen with repeated balloon inflation during angioplasty, whereas intracoronary adenosine or bradykinin infusion replicates the preconditioned state |57–59|. Ecto-5′-nucleotidase, a marker of protein kinase C activity, increased after isoflurane administration during CABG |60|. Unfortunately, evidence suggests that aged or diseased hearts which have most to gain, may be less amenable to the protection of ischaemia and anaesthetics |61–63|. Prodromal angina was not protective before infarction in diabetic patients |64|, and preconditioning with isoflurane was less effective in diabetic dogs |65|. In summary, extensive laboratory work has elucidated several of the mechanisms of ischaemic- and anaesthetic-induced preconditioning. Clinical work has provided direct and indirect evidence that similar mechanisms operate in the intact human heart leading to the hope that there maybe some practical strategy using the phenomenon for improving patient outcomes. However, there are reasons for caution. Large studies would be required in order to demonstrate unequivocal long-term benefit in man. Such studies will be difficult to conduct, not the least challenge being the requirement for a large control group not receiving volatile anaesthetics. The suggestion that aged and diseased hearts may benefit less may also ultimately limit the usefulness of the concept. Conclusion Protecting the high-risk surgical patient from cardiac ischaemia, myocardial infarction and death is now a well-recognized aspect of perioperative care. The aetiology of perioperative infarction is complex and still not completely understood. However, our evolving understanding of the pathology of CAD offers useful insights into the processes that may be involved in perioperative myocardial injury |14|. It has become increasingly clear that even minor degrees of perioperative myocardial injury, manifest by low levels of cardiac troponin release, have an adverse impact on long-term prognoses (see Landesberg et al. and |10|).
The year in anaesthesia and critical care
108
Intra-operative haemodynamic stability is important. However, many perioperative events are not preceded by obvious cardiovascular instability and strategies for perioperative myocardial protection extend beyond simple haemodynamic management. Early identification of myocardial ischaemia might allow therapy to be insti-tuted for minimizing the impact of that ischaemia on perioperative outcome. Regular monitoring using twelve-lead ECGs in the immediate post-operative period seems to offer a cheap and widely available means of monitoring high-risk patients following surgery, thereby allowing treatment of those patients who demonstrate evidence of myocardial ischaemia (see Böttiger et al.). There is little evidence to support prophylactic coronary artery grafting or angioplasty prior to surgery. However, some patients will have indications for these interventions on the basis of their cardiac symptoms alone and will subsequently come to surgery |33|. Our understanding of the risks and benefits of coronary angioplasty in this setting is still evolving. However, it seems clear that, when possible, surgery should be delayed for at least 6 weeks after angioplasty and stenting (see Wilson et al.). Much work has focused on perioperative sympatholytic therapy including studies of perioperative β-blockade, of the use of α2-agonists and of neuraxial blockade |19, 20, 24, 25|. Evidence of improved outcome has been published, but many studies in this area suffer from methodological flaws or have only shown significant benefit against the background of an unusually high incidence of complications in the placebo group (see Stevens et al. and |20, 26, 36|). β-blockade remains a promising perioperative therapy for decreasing ischaemia, but evidence for improved outcomes must wait for the results of large-scale studies. At least one such study is currently under way. Another pharmacological therapy that promises great benefit is the use of statins in the perioperative period. Statins have proven value in the non-surgical setting and a favourable side effect profile (see Durazzo et al.). Although the mechanism of benefit in the operative setting is not clear, it is unlikely to be due to acute improvement in the lipid profile. The improvement in outcome needs confirmation in large-scale studies that would also be able to elucidate the optimum dosing regimen. A considerable volume of work has been published examining the mechanisms of ischaemic and pharmacological preconditioning of the myocardium (see Zaugg et al.). The potential of volatile anaesthetic agents in limiting myocardial damage during ischaemia is of great interest, but is difficult to study in the clinical setting. Much work remains to be done in this area in order to establish the extent of benefit in patients undergoing non-cardiac surgery, the best agent to use and the optimal dosing regimen. Such studies will be methodologically challenging to conduct. In conclusion, our understanding of perioperative myocardial injury has been revolutionized in the past decade. It is now clear that many perioperative events are not due simply to a myocardial oxygen supply/demand imbalance due to haemodynamic instability but have a similar aetiology to acute coronary events in nonsurgical patients. Clinical research points to roles for both well-established therapies, such as β-blockade and newer treatments, such as statins, in the management of such patients. However, much work remains to be done in order to establish the efficacy of these treatments fully and to establish which groups of patients should receive such treatments and what drug doses should be used.
Protecting the heart in non-cardiac surgery
109
References 1. Mangano DT. Perioperative cardiac morbidity. Anesthesiology 1990; 72: 153–84. 2. Hertzer NR, Beven EG, Young JR, O’Hara PJ, Ruschhaupt III WF, Graor RA, Dewolfe VG, Maljovec LC. Coronary artery disease in peripheral vascular patients. A classification of 1000 coronary angiograms and results of surgical management. Ann Surg 1984; 199: 223–33. 3. Bayly PJ, Matthews JN, Dobson PM, Price ML, Thomas DG. In-hospital mortality from abdominal aortic surgery in Great Britain and Ireland: Vascular Anaesthesia Society audit. Br J Surg 2001; 88: 687–92. 4. Eagle KA, Rihal CS, Mickel MC, Holmes DR, Foster ED, Gersh BJ. Cardiac risk of noncardiac surgery: influence of coronary disease and type of surgery in 3368 operations. CASS Investigators and University of Michigan Heart Care Program. Coronary Artery Surgery Study. Circulation 1997; 96: 1882–7. 5. Howell SJ, Sear JW. Perioperative myocardial injury—individual and population implications. Br J Anaesth 2004; 93: 3–8. 6. Antman EM, Tanasijevic MJ, Thompson B, Schactman M, McCabe CH, Cannon CP, Fischer GA, Fung AY, Thompson C, Wybenga D, Braunwald E. Cardiac-specific troponin I levels to predict the risk of mortality in patients with acute coronary syndromes. N Engl J Med 1996; 335: 1342–9. 7. Ohman EM, Armstrong PW, Christenson RH, Granger CB, Katus HA, Hamm CW, O’Hanesian MA, Wagner GS, Kleiman NS, Harrell Jr FE, Califf RM, Topol EJ. Cardiac troponin T levels for risk stratification in acute myocardial ischemia. GUSTO IIA Investi-gators. N Engl J Med 1996; 335: 1333–41. 8. Ravkilde J, Horder M, Gerhardt W, Ljungdahl L, Pettersson T, Tryding N, Moller BH, Hamfelt A, Graven T, Asberg A, et al. Diagnostic performance and prognostic value of serum troponin T in suspected acute myocardial infarction. Scand J Clin Lab Invest 1993; 53: 677–85. 9. Adams III JE. Clinical application of markers of cardiac injury: basic concepts and new considerations. Clin Chim Acta 1999; 284: 127–34. 10. Kim LJ, Martinez EA, Faraday N, Dorman T, Fleisher LA, Perler BA, Williams GM, Chan D, Pronovost PJ. Cardiac troponin I predicts short-term mortality in vascular surgery patients. Circulation 2002; 106: 2366–71. 11. Mangano DT, Browner WS, Hollenberg M, London MJ, Tubau JF, Tateo IM. Association of perioperative myocardial ischemia with cardiac morbidity and mortality in men under-going non-cardiac surgery. The Study of Perioperative Ischemia Research Group. N Engl J Med 1990; 323: 1781–8. 12. Landesberg G, Luria MH, Cotev S, Eidelman LA, Anner H, Mosseri M, Schechter D, Assaf J, Erel J, Berlatzky Y. Importance of long-duration post-operative ST-segment depression in cardiac morbidity after vascular surgery. Lancet 1993; 341: 715–19. 13. Dawood MM, Gutpa DK, Southern J, Walia A, Atkinson JB, Eagle KA. Pathology of fatal perioperative myocardial infarction: implications regarding pathophysiology and preven-tion. Int J Cardiol 1996; 57: 37–44. 14. Priebe H-J. Triggers of perioperative myocardial ischaemia and infarction. Br J Anaesth 2004; 93: 9–20. 15. Slogoff S, Keats AS. Does perioperative myocardial ischemia lead to post-operative myocardial infarction? Anesthesiology 1985; 62: 107–14. 16. Charlson ME, MacKenzie CR, Gold JP, Ales KL, Topkins M, Shires GT. Intraoperative blood pressure: what patterns identify patients at risk for post-operative complications? Ann Surg 1990; 212: 567–80. 17. Stone JG, Foex P, Sear JW, Johnson LL, Khambatta HJ, Triner L. Myocardial ischemia in untreated hypertensive patients: effect of a single small oral dose of a beta-adrenergic blocking agent. Anesthesiology 1988; 68: 495–500.
The year in anaesthesia and critical care
110
18. Pasternack PF, Imparato AM, Baumann FG, Laub G, Riles TS, Lamparello PJ, Grossi EA, Berguson P, Becker G, Bear G. The hemodynamics of beta-blockade in patients under-going abdominal aortic aneurysm repair. Circulation 1987; 76: III 1–7. 19. Mangano DT, Layug EL, Wallace A, Tateo I. Effect of atenolol on mortality and cardiovascular morbidity after non-cardiac surgery. Multicenter Study of Perioperative Ischemia Research Group. N Engl J Med 1996; 335: 1713–20. 20. Fleisher LA. Con: beta-blockers should not be used in all patients undergoing vascular surgery. J Cardiothorac Vasc Anesth 1999; 13: 496–7. 21. Poldermans D, Boersma E, Bax JJ, Thomson IR, Van de Ven LL, Blankensteijn JD, Baars HF, Yo TI, Trocino G, Vigna C, Roelandt JR, Van Urk H. The effect of bisoprolol on peri-operative mortality and myocardial infarction in high-risk patients undergoing vascular surgery. Dutch Echocardiographic Cardiac Risk Evaluation Applying Stress Echo-cardiography Study Group. N Engl J Med 1999; 341: 1789–94. 22. Wallace A, Layug B, Tateo I, Li J, Hollenberg M, Browner W, Miller D, Mangano DT. Prophylactic atenolol reduces post-operative myocardial ischemia. McSPI Research Group. Anesthesiology 1998; 88: 7–17. 23. Zaugg M, Schaub MC, Pasch T, Spahn DR. Modulation of beta-adrenergic receptor sub-type activities in perioperative medicine: mechanisms and sites of action. Br J Anaesth 2002; 88: 101–23. 24. Oliver MF, Goldman L, Julian DG, Holme I. Effect of mivazerol on perioperative cardiac complications during non-cardiac surgery in patients with coronary heart disease: the European Mivazerol Trial (EMIT). Anesthesiology 1999; 91: 951–61. 25. Rodgers A, Walker N, Schug S, McKee A, Kehlet H, Van Zundert A, Sage D, Futter M, Saville G, Clark T, MacMahon S. Reduction of post-operative mortality and morbidity with epidural or spinal anaesthesia: results from overview of randomised trials. BMJ 2000; 321: 1493. 26. Which anaesthetic technique? Bandolier 2001; 86: 5. 27. Van den Berghe G, Wouters P, Weekers F, Verwaest C, Bruyninckx F, Schetz M, Vlas-selaers D, Ferdinande P, Lauwers P, Bouillon R. Intensive insulin therapy in the critically ill patients. N Engl J Med 2001; 345: 1359–67. 28. Van den Berghe G. How does blood glucose control with insulin save lives in intensive care? J Clin Invest 2004; 114: 1187–95. 29. Wong VW, Ross DL, Park K, Boyages SC, Cheung NW. Hyperglycemia: still an important predictor of adverse outcomes following AMI in the reperfusion era. Diabetes Res Clin Prac 2004; 64: 85–91. 30. Murry CE, Jennings RB, Reimer KA. Preconditioning with ischemia: a delay of lethal cell injury in ischemic myocardium. Circulation 1986; 74: 1124–36. 31. Kersten JR, Schmeling TJ, Pagel PS, Gross GJ, Warltier DC. Isoflurane mimics ischemic preconditioning via activation of K(ATP) channels: reduction of myocardial infarct size with an acute memory phase. Anesthesiology 1997; 87: 361–70. 32. Cason BA, Gamperl AK, Slocum RE, Hickey RF. Anesthetic-induced preconditioning: previous administration of isoflurane decreases myocardial infarct size in rabbits. Anesthesiology 1997; 87: 1182–90. 33. Eagle KA, Berger PB, Calkins H, Chaitman BR, Ewy GA, Fleischmann KE, Fleisher LA, Froehlich JB, Gusberg RJ, Leppo JA, Ryan T, Schlant RC, Winters Jr WL, Gibbons RJ, Antman EM, Alpert JS, Faxon DP, Fuster V, Gregoratos G, Jacobs AK, Hiratzka LF, Russell RO, Smith Jr SC; American College of Cardiology; American Heart Association. ACC/AHA Guideline Update for Perioperative Cardiovascular Evaluation for Noncardiac Surgery—Executive Summary. A report of the American College of Cardiology/ American Heart Association Task Force on Practice Guidelines (Committee to Update the 1996 Guidelines on Perioperative Cardiovascular Evaluation for Non-cardiac Surgery), 2002. Am Coll Cardiol website. Available at http://www.acc.org/clinical/guidelines/perio/%20update/periupdate_index.htm
Protecting the heart in non-cardiac surgery
111
34. Eagle KA, Brundage BH, Chaitman BR, Ewy GA, Fleisher LA, Hertzer NR, Leppo JA, Ryan T, Schlant RC, Spencer III WH, Spittell Jr JA, Twiss RD, Ritchie JL, Cheitlin MD, Gardner TJ, Garson Jr A, Lewis RP, Gibbons RJ, O’Rourke RA, Ryan TJ. Guidelines for perioperative cardiovascular evaluation for non-cardiac surgery. Report of the American College of Cardiology/American Heart Association Task Force on Practice Guidelines. Committee on Perioperative Cardiovascular Evaluation for Non-cardiac Surgery. Circulation 1996; 93: 1278– 317. 35. Kaluza GL, Joseph J, Lee JR, Raizner ME, Raizner AE. Catastrophic outcomes of noncardiac surgery soon after coronary stenting. J Am Coll Cardiol 2000; 35: 1288–94. 36. Devereaux PJ, Yusuf S, Yang H, Choi PT, Guyatt GH. Are the recommendations to use perioperative beta-blocker therapy in patients undergoing non-cardiac surgery based on reliable evidence? CMAJ 2004; 171: 245–7. 37. Bayliff CD, Massel DR, Inculet RI, Malthaner RA, Quinton SD, Powell FS, Kennedy RS. Propranolol for the prevention of post-operative arrhythmias in general thoracic surgery. Ann Thorac Surg 1999; 67: 182–6. 38. Davies MJ, Dysart RH, Silbert BS, Scott DA, Cook RJ. Prevention of tachycardia with atenolol pretreatment for carotid endarterectomy under cervical plexus blockade. Anaesth Intensive Care 1992; 20: 161–4. 39. Jakobsen CJ, Bille S, Ahlburg P, Rybro L, Hjortholm K, Andresen EB. Perioperative metroprolol reduces the frequency of atrial fibrillation after thoracotomy for lung resection. J Cardiothorac Vasc Anesth 1997; 11: 746–51. 40. Raby KE, Brull SJ, Timimi F, Akhtar S, Rosenbaum S, Naimi C, Whittemore AD. The effect of heart rate control on myocardial ischemia among high-risk patients after vascular surgery. Anesth Analg 1999; 88: 477–82. 41. Urban MK, Markowitz SM, Gordon MA, Urquhart BL, Kligfield P. Post-operative propy-lactic administration of beta-adrenergic blockers in patients at risk for myocardial ischemia. Anesth Analg 2000; 90: 1257–61. 42. Zaugg M, Tagliente T, Lucchinetti E, Jacobs E, Krol M, Bodian C, Reich DL, Silverstein JH. Beneficial effects from beta-adrenergic blockade in elderly patients under-going non-cardiac surgery. Anesthesiology 1991; 91: 1674–86. 43. Ellis JE, Drijvers G, Pedlow S, Laff SP, Sorrentino MJ, Foss JF, Shah M, Busse JR, Mantha S, McKinsey JF, et al. Premedication with oral and transdermal clonidine provides safe and effecious post-operative sympatholysis. Anesth Analg 1994; 79: 1133–40. 44. McSPI-Europe Research Group. Perioperative sympatholysis: beneficial effects of the alpha 2adrenoceptor agonist mivazerol on hemodynamic stability and myocardial ischemia. Anesthesiology 1997; 86: 346–63. 45. Pluskwa F, Bonnet F, Saada M, Macquin-Mavier I, Becquemin JP, Catoire P. Effects of clonidine on variation of arterial blood pressure and heart rate during carotid artery surgery. J Cardiothorac Vasc Anesth 1991; 5: 431–6. 46. Stuhmeier KD, Mainzer B, Cierpka J, Sandmann W, Tarnow J. Small, oral dose of clonidine reduces the incidence of intraoperative myocardial ischemia in patients having vascular surgery. Anesthesiology 1996; 85: 706–12. 47. Caramella JP, Goursot G, Carcone B, Desmonts JM, Fargnoli JM, Marty J, Cazor JL, Elkik F. [Prevention of per- and post-operative myocardial ischemia in non-cardiac surgery by intravenous diltiazem]. Ann Fr Anesth Reanim 1988; 7: 245–50. 48. Godet G, Coriat P, Baron JF, Bertrand M, Diquet B, Sebag C, Viars P. Prevention of intraoperative myocardial ischemia during non-cardiac surgery with intravenous diltiazem: a randomized trial versus placebo. Anesthesiology 1987; 66: 241–5. 49. Dodds TM, Stone JG, Coromilas J, Weinberger M, Levy DG. Prophylactic nitroglycerin infusion during non-cardiac surgery does not reduce perioperative ischemia. Anesth Analg 1993; 76: 705–13.
The year in anaesthesia and critical care
112
50. Stone JG, Foex P, Sear JW, Johnson LL, Khambatta HJ, Triner L. Risk of myocardial ischaemia during anaesthesia in treated and untreated hypertensive patients. Br J Anaesth 1988; 61: 675–9. 51. Ghignone M, Cavillo O, Quintin L. Anesthesia and hypertension: the effect of clonidine on perioperative hemodynamics and isoflurane requirements. Anesthesiology 1987; 67: 3–10. 52. Poldermans D, Bax JJ, Kertai MD, Krenning B, Westerhout CM, Schinkel AF, Thomson IR, Lansberg PJ, Fleisher LA, Klein J, Van Urk H, Roelandt JR, Boersma E. Statins are associated with a reduced incidence of perioperative mortality in patients undergoing major non-cardiac vascular surgery. Circulation 2003; 107: 1848–51. 53. Deutsch E, Berger M, Kussmaul WG, Hirshfeld Jr JW, Herrmann HC, Laskey WK. Adaptation to ischemia during percutaneous transluminal coronary angioplasty. Clinical, hemodynamic, and metabolic features. Circulation 1990; 82: 2044–51. 54. Kloner RA, Shook T, Przyklenk K, Davis VG, Junio L, Matthews RV, Burstein S, Gibson M, Poole WK, Cannon CP, et al. Previous angina alters in-hospital outcome in TIMI 4. A clinical correlate to preconditioning? Circulation 1995; 91: 37–45. 55. Lu EX, Chen SX, Hu TH, Xui LM, Yuan MD. Preconditioning enhances myocardial protection in patients undergoing open-heart surgery. Thorac Cardiovasc Surg 1998; 46: 28–32. 56. Haroun-Bizri S, Khoury SS, Chehab IR, Kassas CM, Baraka A. Does isoflurane optimize myocardial protection during cardiopulmonary bypass? J Cardiothorac Vasc Anesth 2001; 15: 418–21. 57. Tomai F, Crea F, Gaspardone A, Versaci F, De Paulis R, Polisca P, Chiariello L, Gioffre PA. Effects of A1 adenosine receptor blockade by bamiphylline on ischaemic preconditioning during coronary angioplasty. Eur Heart J 1996; 17: 846–53. 58. Leesar MA, Stoddard M, Ahmed M, Broadbent J, Bolli R. Preconditioning of human myocardium with adenosine during coronary angioplasty. Circulation 1997; 95: 2500–7. 59. Leesar MA, Stoddard MF, Manchikalapudi S, Bolli R. Bradykinin-induced preconditioning in patients undergoing coronary angioplasty. J Am Coll Cardiol 1999; 34: 639–50. 60. Belhomme D, Peynet J, Louzy M, Launay JM, Kitakaze M, Menasche P. Evidence for preconditioning by isoflurane in coronary artery bypass graft surgery. Circulation 1999; 100: II340–4. 61. Abete P, Ferrara N, Cacciatore F, Madrid A, Bianco S, Calabrese C, Napoli C, Scognamiglio P, Bollella O, Cioppa A, Longobardi G, Rengo F. Angina-induced protection against myocardial infarction in adult and elderly patients: a loss of preconditioning mechanism in the aging heart? J Am Coll Cardiol 1997; 30: 947–54. 62. Ishihara M, Sato H, Tateishi H, Kawagoe T, Shimatani Y, Ueda K, Noma K, Yumoto A, Nishioka K. Beneficial effect of prodromal angina pectoris is lost in elderly patients with acute myocardial infarction. Am Heart J 2000; 139: 881–8. 63. Lee TM, Su SF, Chou TF, Lee YT, Tsai CH. Loss of preconditioning by attenuated activation of myocardial ATP-sensitive potassium channels in elderly patients undergoing coronary angioplasty. Circulation 2002; 105: 334–40. 64. Ishihara M, Sato H, Tateishi H, Kawagoe T, Shimatani Y, Kurisu S, Sakai K, Ueda K. Implications of prodromal angina pectoris in anterior wall acute myocardial infarction: acute angiographic findings and long-term prognosis. J Am Coll Cardiol 1997; 30: 970–5. 65. Tanaka K, Kehl F, Gu W, Krolikowski JG, Pagel PS, Warltier DC, Kersten JR. Isofluraneinduced preconditioning is attenuated by diabetes. Am J Physiol Heart Circ Physiol 2002; 282: H2018–23.
Part II Clinical pharmacology
What is new in anaesthetic pharmacology? MICHEL STRUYS Introduction Quality control in anaesthesia has become important due to the evolution of perioperative management. Changes in surgical conditions and patient populations make it more essential than ever to manage anaesthesia in a fast, simple and safe way. Thus, a wide spectrum of pharmacological actions is required for effecting full control of ‘body homeostasis’. As well as the classical drugs used for producing analgesia, hypnosis and suppression of somatic and autonomic responses to noxious stimuli, a wide range of agents are needed for optimizing the outcome in perioperative medicine. The balance between the therapeutic effect and toxic side effect governs the cost function of each drug. The ultimate goal when administering a particular dose of a drug is to obtain the desired clinical effect, for which a specific therapeutic concentration of the drug at the site of action (i.e. the receptor) is required. In addition, a minimum of side effects should occur. New drug development directly related to anaesthesia has not been very successful recently, probably due to the lack of ‘blockbuster’ medication. The small market share for anaesthesia has resulted in a change in research and development strategy and drug companies have focused on medical fields other than anaesthesia. Fortunately, many drug developments in other domains have influenced the safety of general anaesthesia (e.g. βblockers). Nevertheless, attempts at new drug development in anaesthesia have recently been described. The success of remifentanil and its specific esterase metabolism has encouraged the development of other compounds with high-speed metabolism. In the hypnotic-anaesthetic class, THRX-918661 (Theravance, San Francisco, CA, USA), a sedative-hypnotic agent, is an allosteric modulator of the γ-aminobutyric acid A receptor. It is hydrolysed by esterases to an inactive metabolite. Until now, the drug has been described only in animal studies: no human studies have been published. Two other water-soluble compounds, ORG 21465 and 25435, have been rejected after Phase I research, due to side effects such as excitation and tachycardia and disappointing pharmacokinetics leading to a slow recovery after prolonged sedation. Another benzodiazepine, Ro 48–4791, has not been developed up to the marketing stage |1|. Only zaleplon, a newer short-acting non-benzodiazepine agent, has been tested in animals |2– 4|. As the formulation of propofol has recently become ‘out of patent’, various companies have tried to optimize this successful molecule. In Chapter 5, Dr Ihmsen gives more detail about new formulations of propofol. First, research for optimizing defence against bacterial growth in the lipid emulsion has led to the use of the supplement disodium edetate or its sulphite. Changes in the lipid emulsion were studied in order to optimize the
pharmacokinetic and dynamic profile, thus minimizing side effects such as pain on injection. At the same time Aquavan™, a pro-drug of propofol, has been described. In contrast to hypnotic-anaesthetic drugs, novel therapies for pain cover a range of clinical needs and therefore constitute a continued opportunity for major pharmaceutical drug research. Novel treatments are likely to arise from better understanding of both the pathophysiology of clinical pain conditions and the pharmacology of existing therapies. Intravenous non-steroidal anti-inflammatory drugs, mostly cyclooxygenase-2 inhibitors, have been introduced into the field of acute post-operative pain despite some well-known side effects in the cardiovascular area. Despite this, the cost-benefit ratio has been seen as positive |5|. In Chapter 6 on new analgesic drugs, Dr Luginbühl describes some other results from refining existing drugs used in pain therapy, such as morphine-6-glucuronide and others. In a continuing attempt to avoid the development of opioid tolerance (acute and chronic), N-methyl-D-aspartate (NMDA) receptor-acting drugs such as ketamine have been studied. Stereo-isomer studies of ketamine revealed that the S+ketamine enantiomer has fewer side effects compared to the racemic mixture, although hard evidence is still lacking. As with other racemic drugs, including intravenous hypnotics and local anaesthetics, a dose reduction and longer duration of action of this isomer of ketamine have been shown |1|. Refinement of existing drugs provides the greatest probability of success, but there comes a time when the improvement is so small that the drug will not recoup its cost of development. Progress in molecular neurobiology has generated various new putative targets for pain therapies, but this approach has still to deliver an analgesic to the clinic. Future treatment for pain is likely to arise from genomic studies, although the challenge will be to identify those gene products with the greatest utility as drug discovery targets |5|. Nevertheless, the emergence of pharmacogenomics in new drug research has become obvious. The sequencing of the human genome represents one of the most significant advances of the 20th century that will shape the foundation of medical and drug research well into the 21st century. Due to remarkable technological advances such as throughput sequencing, increased computer power and automated methods of analysis, the Humane Genome Project offers the opportunity of shaping the face of drug discovery and development in this century. Pharmacogenetics is only at the very beginning of its use in clinical practice. It is likely that its contribution to new drug development will become a reality. Common polymorphisms in drug targets indicate that DNA sequence variations will have to be taken into account during genomic screening for drug development. Although pharmacogenetics is unlikely to change the way anaesthesia is practised today, it may help to elucidate inter-patient variability in drug response |6|. In Chapter 7, Dr Smiley illustrates the most relevant findings in this field in the last year. Concerning neuromuscular-blocking drugs, promising pre-clinical and early clinical research has been performed with fast-acting antagonists for rocuronium. Org 25969 belongs structurally to the supramolecular family of cyclodextrins, which are well known for their capacity to form an inclusion complex with various molecules. Upon encapsulation of the neuromuscular-blocking agent rocuronium, Org 25969 reduces the amount of relaxant available to bind to nicotinic receptors at the neuromuscular junction, resulting in reversal of neuromuscular block |7|.
In the field of inhaled anaesthetics, the use of xenon is still controversial. Xenon is a potent inhalation anaesthetic with many beneficial properties. Unfortunately, cost has so far mitigated the development of its use in anaesthesia. Recent research has suggested a niche for xenon, based on its pharmacokinetic, cardiac, neuroprotective and analgesic properties. Further clinical trials have to prove whether neurocognitive deficits can be reduced by the administration of xenon in cardiac surgical patients |8|. For other wellknown inhalation anaesthetics such as sevoflurane and desflurane, new therapeutic qualities have been found recently. For example, potent inhaled anaesthetics may protect against hypoxic insults to vital tissues such as the heart by ‘preconditioning’. Hence, the organ is protected against injury from transient arterial occlusion |9|. It is not in the scope of this review to describe the development of drugs primarily related to other medical disciplines and yet interesting in respect of optimizing anaesthetic outcome. It is of no doubt that drugs such as β-blockers when started preoperatively in high-risk patients have improved perioperative morbidity and mortality (see Chapter 4). Next to the development of new and improved drugs, better knowledge of drug interactions has improved the quality of drug effects. Drug interactions are important to the anaesthetist for several reasons. Patients are increasingly presenting for surgery while taking numerous medications. There may be drug interactions between the preoperative medications themselves, between the pre-operative medications and intraor postoperative medications or between drugs and agents administered only by the anaesthetist. Interactions may be unwanted because of the change in magnitude and/or time course of the clinical effect or due to the higher incidence of side effects. In contrast, anaesthetic practice may occasionally benefit from synergistic interactions. For example, when optimizing the balance between hypnotic and analgesic action, the primary concern is to ensure an appropriate contribution to the hypnotic component of anaesthesia. When this has been achieved, optimal control of analgesia during surgical anaesthesia is also required. The noxious stimulus should be identified and blocked at the spinal level. Classically, this can be done using opioids and/or locoregional anaesthetic techniques |10, 11|. The quality of anaesthesia is improved by understanding these additive or synergistic interactions during anaesthesia |12, 13|. As well as the development or optimization of drugs, optimization of drug delivery can result in a beneficial perioperative outcome. For intravenous drugs, a better knowledge of the pharmacokinetic and pharmacodynamic characteristics of drugs has led to the introduction of pharmacokinetic model-based drug administration, also called target-controlled drug delivery. A target-controlled infusion is one controlled in such a manner as to attempt to achieve a user-defined estimated drug concentration in a body compartment or tissue of interest. An anaesthetist using a target-controlled infusion system for administering an anaesthetic agent is thus able to set (and change) a desired concentration (usually referred to as the ‘target concentration’) based on their clinical observations of the patient. Multi-compartmental pharmacokinetic models, resulting in polyexponential equations, are used by target-controlled infusion systems for calculating the infusion rates required for achieving the target concentration. A computer or microprocessor is required for performing the complex calculations and controlling the infusion pump. Theoretically, a target-controlled infusion system can control the concentration in any compartment or tissue in the body. The pharmacokinetic models
used are derived from previously performed population pharmacokinetic studies. While a target-controlled infusion is in progress, no measurements of actual concentrations are made, so that these systems are referred to as ‘open-loop’ systems (as opposed to ‘closedloop’ systems where measurements of the control variable are made and the errors between the set point and the actual value are used for altering the input to the system). It is not surprising that tar get-controlled infusion systems are popular with anaesthetists, who consider them easy to use and that they provide a high level of predictability of anaesthetic effect |14|. In a study comparing manual infusion with target-controlled infusion propofol by anaesthetists unfamiliar with propofol infusion anaesthesia, it was found that anaesthetists quickly became familiar with both techniques, but expressed a clear preference for the target-controlled infusion system |15|. How does the quality of clinical control compare between target-controlled infusion and manually controlled infusions? Although many of the early target-controlled infusion systems were used for infusions of opiates, most of the evidence comes from studies comparing targetcontrolled infusions with manually controlled propofol infusions. The quality of anaesthesia is difficult to assess, but studies have used simple categorical measures where the anaesthetist rates the quality as good, adequate or poor, as well as numerical methods such as a quality of anaesthesia score. In studies comparing target-controlled infusion with manual propofol infusion regimens the quality of induction and maintenance of anaesthesia were similar, as was the incidence and severity of haemodynamic effects and the recovery times |16–18|. A large multicentre study found that the control of anaesthesia was easier in subjects anaesthetized with target-controlled than with manually controlled propofol infusions |19|. Dr Servin gives more details in Chapter 8. For the inhaled anaesthetics, new closed-circuit delivery systems were marketed recently in an attempt to optimize the time course of drug delivery. When using classic anaesthetic machines, this time course is hampered by the conflict between reducing the amount of agent used by minimizing the fresh gas flow and decreasing the difference between the desired and actual drug concentration. Accurate titration of inhaled anaesthetics requires fast and reliable adjustment without overshooting the targeted inhaled agent concentration and involves the maintenance of a stable desired drug level. In contrast, economy of consumption is only possible when minimizing the fresh gas flow towards low-flow or even closed-circuit conditions, whereby the fresh gas flow equals the uptake by the patient. With use of an out-of-circle vapourizer rapid adjustment of the inhaled agent concentration is only feasible by vastly increasing the fresh gas flow, resulting in a possible overshoot in the inhaled agent concentration and increased consumption. When an agent is directly delivered into the breathing circle system independent of the fresh gas flow, this should allow quicker adjustment of the agent concentration. By uncoupling agent delivery from the fresh gas flow, the conflict between anaesthetic agent concentration and minimal consumption will be solved |20|. It might be concluded that research continues to find new drugs in anaesthesia although future development is likely to be determined by commercial imperatives. Nevertheless, some new molecules are undergoing pre-clinical and early clinical research phase studies. In addition, research extends the scope of existing drugs and attempts to find novel applications for old drugs. Pharmacogenetics and genomics will govern most of the new drug development in the future. In addition, a better knowledge of drug interaction will improve our understanding of the advantages or disadvantages of the
agents used in anaesthesia. Optimizing drug delivery equipment will certainly improve the quality of anaesthesia. References 1. Sneyd JR. Recent advances in intravenous anaesthesia. Br J Anaesth 2004; 93: 725–36. 2. Noguchi H, Kitazumi K, Mori M, Shiba T. Electroencephalographic properties of zaleplon, a non-benzodiazepine sedative/hypnotic, in rats. J Pharmacol Sci 2004; 94: 246–51. 3. Noguchi H, Kitazumi K, Mori M, Shiobara Y, Shiba T. Effect of zaleplon, a non-benzodiazepine hypnotic, on melatonin secretion in rabbits. J Pharmacol Sci 2003; 93: 204–9. 4. Noguchi H, Kitazumi K, Mori M, Shiba T. Effect of zaleplon on learning and memory in rats. Naunyn Schmiedebergs Arch Pharmacol 2002; 366: 183–8. 5. Boyce S, Ali Z, Hill R. New developments in analgesia. Drug Discovery World 2001; 2: 31–5. 6. Iohom G, Fitzgerald D, Cunningham AJ. Principles of pharmacogenetics—implications for the anaesthetist. Br J Anaesth 2004; 93: 440–50. 7. Bom A, Thomson K. The potency and efficacy of non-steroidal neuromuscular blocking agents is not reduced when administered after reversal of rocuronium-induced neuromuscular block by Org 25969. Anesthesiology 2004; 101: A-1 144. 8. Sanders RD, Franks NP, Maze M. Xenon: no stranger to anaesthesia. Br J Anaesth 2003; 91: 709–17. 9. De Hert SG, Van der Linden PJ, Cromheecke S, Meeus R, Nelis A, Van Reeth V, ten Broecke PW, De Blier IG, Stockman BA, Rodrigus IE. Cardioprotective properties of sevoflurane in patients undergoing coronary surgery with cardiopulmonary bypass are related to the modalities of its administration. Anesthesiology 2004; 101: 299–310. 10. Glass PS, Howell S, Gan TJ, Ginsberg B. How to manage drug interactions. Eur J Anaesthesiol Suppl 1997; 15: 33–9. 11. Glass PS. Anesthetic drug interactions: an insight into general anesthesia—its mechanism and dosing strategies [editorial; comment]. Anesthesiology 1998; 88: 5–6. 12. Minto CF, Schnider TW, Short TG, Gregg KM, Gentilini A, Shafer SL. Response surface model for anesthetic drug interactions. Anesthesiology 2000; 92: 1603–16. 13. Bouillon TW, Bruhn J, Radulescu L, Andresen C, Shafer TJ, Cohane C, Shafer SL. Pharmacodynamic interaction between propofol and remifentanil regarding hypnosis, tolerance of laryngoscopy, bispectral index, and electroencephalographic approximate entropy. Anesthesiology 2004; 100: 1353–72. 14. Taylor I, White M, Kenny GN. Assessment of the value and pattern of use of a target controlled propofol infusion system. Int J Clin Monit Comput 1993; 10: 175–80. 15. Russell D, Wilkes MP, Hunter SC, Glen JB, Hutton P, Kenny GN. Manual compared with target-controlled infusion of propofol. Br J Anaesth 1995; 75: 562–6. 16. Struys M, Versichelen L, Rolly G. Influence of pre-anaesthetic medication on target propofol concentration using a ‘Diprifusor’ TCI system during ambulatory surgery. Anaesthesia 1998; 53(Suppl 1): 68–71. 17. Barvais L, Heitz D, Schmartz D, Maes V, Coussaert E, Cantraine F, d’Hollander A. Pharmacokinetic model-driven infusion of sufentanil and midazolam during cardiac surgery: assessment of the prospective predictive accuracy and the quality of anesthesia. J Cardiothorac Vasc Anesth 2000; 14: 402–8.
18. Swinhoe CF, Peacock JE, Glen JB, Reilly CS. Evaluation of the predictive performance of a ‘Diprifusor’ TCI system. Anaesthesia 1998; 53(Suppl 1): 61–7. 19. Servin FS. TCI compared with manually controlled infusion of propofol: a multicentre study. Anaesthesia 1998; 53(Suppl 1): 82–6. 20. Struys M, Kalmar A, De Baerdemaeker L, Mortier E, Rolly G, Manigel J, Buschke W. Time course of inhaled anaesthetic drug delivery using a new multifunctional closed-circuit anaesthesia ventilator; in vitro comparison with a classical anaesthesia machine. Br J Anaesth 2004; 94: 306–17.
5 New hypnotics HARALD IHMSEN Introduction Two important new compounds for anaesthesia have been introduced during the last 10– 20 years, the inhalational anaesthetic desflurane in 1987 and the opioid remifentanil in 1993. In contrast, there has been no significant development in the field of intravenous sedative-hypnotic agents since the introduction of propofol in 1979 and of midazolam in 1978. The steroid drug eltanolone was not introduced into the market because this agent was characterized by a long onset time of effect and by possible pro-convulsant properties |1|. The benzodiazepine Ro 48–6791 was also not developed until commercialization, despite showing a shorter duration of action than midazolam |2|. Two other new compounds, ORG 21465 and ORG 25435, were rejected because of undesired side effects and disappointing pharmacokineticdynamic profiles |3–5|. Thus, for general anaesthesia and increasingly also for sedation, propofol is still the most important intravenous hypnotic drug, due to its pharmacokinetic-dynamic properties, i.e. a high clearance with rapid elimination, a relatively short onset time and a short duration of action, which make propofol the best controllable intravenous hypnotic agent. However, there are some disadvantages of propofol so that there is still a need for the development of new intravenous hypnotics. As propofol is a substance with very slight solubility in water it is usually formulated as a lipid emulsion. Besides the risk of infection due to bacterial contamination |6| and the high lipid intake in the case of long-term application, for example, as sedation in the intensive care unit |7|, the lipid formulation seems to be in part responsible for pain on injection |8| and cardiovascular depressive effects |9|. Some efforts have therefore been undertaken to formulate propofol either with other lipid solvents |10–12| or without any lipid in a water-soluble formulation |13|. In the work of Trapani et al. different lipid-free formulations of propofol are investigated with respect to the solubility of propofol, to their chemical stability and to their hypnotic efficacy in rats. Whereas this paper presents possible candidates for the further development of new propofol formulations, Song et al. investigated a new lipid propofol formulation (Ampofol™) in man during anaesthesia and sedation. Paul et al. tested a similar lipid propofol formulation (AM 149) in volunteers, but the study was terminated prematurely because of the high incidence of thrombophlebitis. © Atlas Medical Publishing Ltd
Another approach is the development of a propofol pro-drug, which means an inert parent drug that releases propofol as active metabolite. GPI 15715 is a watersoluble phosphonoO-methyl pro-drug of propofol that has meanwhile passed phase II of clinical trial. In a
New hypnotics
121
volunteer study by Fechner et al., GPI 15715 was compared with Diprivan™ with respect to pharmacokinetics and pharmacodynamics. Finally, there was a paper in 2004 about the pharmacodynamics of the newer sedative, non-benzodiazepine agent zaleplon, which is mainly used for treatment of insomnia, but which may also have a potential for the use in sedation or premedication.
Evaluation of new propofol aqueous solutions for intravenous anesthesia Trapani A, Laquintana V, Lopedota A, et al. Int J Pharm 2004; 278(1): 91–8 BACKGROUND. Propofol is usually formulated as oil-in-water emulsion using soya bean oil, glycerol and egg lecithin. Some side effects such as pain on injection and a decrease in blood pressure seem to be associated with the formulation. The aim of this study was to evaluate the potential of three new aqueous formulations of propofol using (1) hydroxypropyl-gamma-cyclodextrin as a solubilizer, (2) a cosolvent mixture of propylene glycol water and (3) a pro-drug approach with the prolinate ester of propofol. The efficacy of all these formulations as intravenous anaesthetic agents was assessed in rats by the onset and duration of loss of the righting reflex and compared with that of the commercial propofol formulation Diprivan™. INTERPRETATION. For Diprivan™, the cyclodextrin and the propofol/propylene glycol formulation the onset of the effect was immediate, whereas for the prolinate ester as a pro-drug a delay of 23±3 s was observed. The duration of the effect was similar for all formulations with a trend of a shorter duration for propofol/propylene glycol and a longer duration for the cyclodextrin and the prolinate ester. It was found that, for the cyclodextrin formulation, anaesthetic effects comparable to those of Diprivan™ were still observed even when the amount of cyclodextrin was minimized. A limited stability was found for the propofol/propylene glycol formulation. Comment This paper presented possible candidates for alternative formulations of propofol. The focus was clearly more on the physico-chemical properties of the solutions. The pharmacodynamic effect of the presented propofol formulations was roughly characterized in rats. Unfortunately, no pharmacokinetic analyses were performed. The paper demonstrates that effective aqueous propofol solutions can be prepared, but as the main problems with the common lipid formulation are pain on injection and cardiovascular depression, possible new propofol formulations must be investigated further in man.
The pharmacodynamic effects of a lower lipid emulsion of propofol: a
The year in anaesthesia and critical care
122
comparison with the standard propofol emulsion Song D, Hamza M, White PF, Klein K, Recart A, Khodaparast 0. Anesth Analg 2004; 98(3): 687–91 BACKGROUND. Ampofol™ is a newer lipid emulsion of propofol that contains 50% less soya bean oil and egg lecithin than current propofol formulations. To investigate whether the formulation has an influence on the pharmacodynamics, Ampofol™ was compared with Diprivan™ with respect to their onset of action and recovery profiles as well as intra-operative efficacy when administered for induction and maintenance as part of a ‘balanced anaesthesia’ with sufentanil or fentanil in 63 healthy out-patients. The onset times to loss of the eyelash reflex were recorded. The severity of pain on injection, speed of induction, intra-operative haemodynamic variables and electroencephalographic bispectral index values were assessed. The recovery times to opening eyes and orientation were noted. INTERPRETATION. There were no significant differences between Ampofol™ and Diprivan™ with respect to onset times, speed of induction, anaesthetic dose requirements, bispectral index values, haemodynamic variables, recovery variables or patient satisfaction (Table 5.1). Therefore, Ampofol™ is apparently equipotent to Diprivan™ with respect to its anaesthetic properties. The incidence of pain on injection was more frequent in the Ampofol™ group, but no patient had a recall of this pain at the follow-up evaluation.
Table 5.1 Pharmacodynamics of Diprivan™ and Ampofol™ during anaesthesia Diprivan™ (n=32)
Ampofol™ (n=31)
Propofol infusion rate (µg/kg/min)
145±28
141±27
Duration of anaesthesia (min)
89±47
92±52
Loss of response to verbal command (s)
36±5
36±4
Loss of eyelash reflex (s)
41±6
41±9
Awakening time (min)
13±6
15±9
Orientation time (min)
19±8
22±16
Baseline bispectral index
97±3
96±5
Bispectral index 3 min after induction
32±10
32±9
Bispectral index at maintenance (60 min)
36±11
39±10
Incidence of pain on injection (%)
9
39*
Incidence (%) of decreases in mean arterial pressure >25%
31
16
* P<0.05, Source: Song et al. (2004).
New hypnotics
123
Comment This was a state-of-the-art clinical study with a double-blind randomized design. The authors also performed a power analysis prior to the study, which makes their finding that there were no significant differences with respect to the pharmacodynamics more valid. Although the propofol infusion was titrated by targeting stable haemodynamics, a similar hypnotic effect was achieved, as indicated by the bispectral index values. They did not measure propofol concentrations, but the observation that similar effects were achieved with similar doses indicates that the pharmacokinetics are probably also unaffected by the formulation. This was also seen in other studies with different lipid propofol formulations |10–12|. Another possible benefit of Ampofol™, a lower lipid load, was not evaluated, as triglycerides were not measured in this study. However, the duration of anaesthesia in this study was only approximately 90 min, which may be too short for evaluating this question. The finding of a higher incidence of pain on injection is a clear disadvantage of the new formulation.
Comparison of a lower-lipid propofol emulsion with the standard emulsion for sedation during monitored anesthesia care Song D, Hamza MA, White PF, Byerly SI, Jones SB, Macaluso AD. Anesthesiology 2004; 100(5): 1072–5 BACKGROUND. Ampofol™ is a newer lipid emulsion of propofol that contains 50% less soya bean oil and egg lecithin than current propofol formulations. The study was designed for comparing the pharmacodynamic properties of Ampofol™ with those of a standard formulation (Diprivan™) when administered for intraoperative sedation. Sixty healthy outpatients undergoing minor operations with local anaesthesia were randomly assigned to receive either Ampofol™ (n=31) or Diprivan™ (n=29) for intravenous sedation. An observer’s assessment of alertness/sedation score of 3 was targeted with a variable rate of propofol infusion. The onset times to achieve a sedation score of 3, the severity of pain on injection, intra-operative haemodynamic variables and electroencephalographic (EEG) bispectral index values were recorded. Recovery times, post-operative pain and nausea, and patient satisfaction with the sedative medication were assessed. INTERPRETATION. There were no significant differences between Ampofol™ and Diprivan™ with respect to onset times, dosage requirements, bispectral index values, haemodynamic variables, recovery times or patient satisfaction scores (Table 5.2). Ampofol™ was equipotent to Diprivan™ with respect to its sedative properties during monitored anaesthesia care. Although both groups received pre-treatment with intravenous lidocaine, Ampofol™ was associated with more pain on injection. Comment This study was very similar to the study on Ampofol™ during anaesthesia by the same authors (see above). It was designed with the same care, including prior power
The year in anaesthesia and critical care
124
Table 5.2 Pharmacodynamics of Diprivan™ and Ampofol™ during sedation Diprivan™ (n=29)
Ampofol™ (n=31)
Propofol infusion rate (µg/kg/min)
98±29
101±28
Duration of anaesthesia (min)
39±17
44±19
Time to achieve observer’s assessment of alertness/sedation score = 3 (s)
67±36
65±17
Awakening time (min)
4±2
3±6
Orientation time (min)
7±3
8±4
Baseline bispectral index
96±4
96±4
Bispeetral index at maintenance (30 min)
61±13
58±13
7
26*
Baseline mean arterial pressure (mmHg)
96±13
102±11
Mean arterial pressure (mmHg) at maintenance (30 min)
71±14
75±17
Incidence of moderate pain on injection (%)
* P<0.05. Source: Song et al. (2004).
analysis. The difference is that a lower level of hypnotic effect was targeted in the present investigation. Again, they found no differences with respect to the pharmacodynamics and haemodynamic effects. The authors did not give any information about whether the targeted level of sedation (observer’s assessment of alertness/sedation score of 3) could be maintained constant throughout the time or whether there were any fluctuations. They also did not give data on how often the infusion rate had to be adjusted in order to maintain the level of sedation. Another interesting question, which is beyond the scope of the present paper, would be whether titration between different levels of sedation can be achieved in a similar time with both formulations. Although the loading dose and the initial infusion rate were lower in the present study compared to the anaesthesia study with Ampofol™, there was a significant incidence of moderate pain on injection. Thus, this work confirms the results of the anaesthesia study on Ampofol™ by the same authors.
Pharmacological characteristics and side effects of a new galenic formulation of propofol without soya bean oil Paul M, Dueck M, Kampe S, Fruendt H, Kasper SM. Anaesthesia 2003; 58(11): 1056–62 BACKGROUND. Am149 is a newer lipid emulsion of propofol that uses mediumchain triglycerides and contains no soya bean oil In order to investigate whether the
New hypnotics
125
formulation has an influence on the pharmacokinetics and pharmacodynamics the authors compared AM149 with a standard formulation of propofol (Diprivan™). In a randomized, double-blind, crossover study, 30 healthy young volunteers received a single intravenous bolus injection of 2.5 mg/kg of propofol. Their venous plasma propofol levels were measured for 48 h following drug administration. The induction and emergence times, respiratory and cardiovascular effects and pain on injection were assessed. Patients were monitored for side effects over 48 h. INTERPRETATION. Because of a high incidence of thrombophlebitis with local inflammation of the injection site, the study was terminated prematurely and only the data of the two parallel treatment groups (15 patients in each group) were analysed. The plasma concentrations did not differ significantly between the two formulations. The anaesthesia induction and emergence times and respiratory and cardiovascular variables showed no significant differences between the two treatment groups (Table 5.3). Although both formulations had similar pharmacokinetic and pharmacodynamic profiles the new formulation is not suitable for clinical use due to the high incidence of thrombophlebitis. Comment This is a typical example of a phase I volunteer study for characterizing the pharmacokinetics, pharmacodynamics and side effects of a new drug. Unfortunately, the authors did not perform a full pharmacokinetic-pharmacodynamic modelling. The pharmacokinetics were only characterized by the area under the concentration curve, Cmax and Tmax, whereas a full pharmacokinetic model, for example a two- or threecompartment model, was not estimated. Use of venous plasma samples is a further limitation, as arterial samples are thought to be of more relevance, particularly if one wants to correlate blood concentrations and the pharmacodynamic effects. With respect to pharmacodynamics the authors investigated only clinical signs and did not apply a more quantitative measure such as, for example, the EEG. Within
Table 5.3 Pharmacokinetics and -dynamics of Diprivan™ and Am149 after a single-bolus injection Diprivan™ (n=15)
Am149 (n=15)
Area under the concentration curve (µg/ml/min)
110±27
111±24
Cmax (µg/ml)
6.1±2.5
8.3±6.2
Tmax (min)
2.0±0.3
2.0±0.3
Time to cessation of speech (s)
57±9
55±8
Time to loss of eyelash reflex (s)
80±12
76±17
Time to eye-opening (min)
9.3±4.5
11.1±5.8
The year in anaesthesia and critical care
Propofol at eye-opening (µg/ml)
126
1.4±0.5
1.2±0.5
Maximum decrease in systolic blood pressure (%)
77
78
Incidence of pain on injection (%)
20
80*
Incidence of thrombophlebitis (%)
6.6
93.3*
Cmax, maximum propofol concentration; Tmax, time at Cmax. * P<0.01. Source: Paul et al. (2003).
these limitations, the two-propofol formulation showed comparable pharmacokinetics and dynamics but, because of the high incidence of pain on injection and thrombophlebitis, further development of Am 149 seems unlikely.
Comparative pharmacokinetics and pharmacodynamics of the new propofol prodrug GPI 15715 and propofol emulsion Fechner J, Ihmsen H, Hatterscheid D, et al. Anesthesiology 2004; 101(3): 626–39 BACKGROUND. GPI 15715 is a new water-soluble phosphono-O-methyl prodrug that is hydrolysed to release propofol. The objectives were to investigate the pharmacokinetics and pharmacodynamics of GPI 15715 in comparison with propofol emulsion (Diprivan™). Nine young healthy male volunteers received GPI 15715 and Diprivan™ as a target-controlled infusion over 60 min in a crossover design. The pharmacokinetics were analysed from arterial plasma concentrations. The pharmacodynamic effect was measured by the median frequency of the power spectrum of the EEG and a sigmoid model with effect compartment was fitted to the data. The haemodynamics and side effects were assessed. INTERPRETATION. Compared with Diprivan™, propofol from GPI 15715 showed a different disposition function, particularly larger volumes of distribution and a higher potency with respect to concentration (Table 5.4). In contrast to Diprivan™, propofol from GPI 15715 did not show a hysteresis between plasma concentration and effect. Based on the estimated model, the maximum propofol concentration would occur approximately 9 min after a bolus dose of GPI 15715 and compared to Diprivan™ the maximum effect would occur with a delay of approximately 5 min. The maximum decrease in systolic blood pressure was similar (32±6% for Diprivan™ and 37±6% for GPI 15715). Pain at the site of injection was observed in three volunteers after Diprivan™ and in none after GPI 15715. All volunteers reported a transient sensation of burning heat or tingling in the anal and genital region after GPI 15715. Comment This was a state-of-the-art phase I volunteer study with complete modelling of the pharmacokinetics and pharmacodynamics and an assessment of side effects. In contrast to the other presented new propofol formulations, which differ in the composition of the
New hypnotics
127
solvent, GPI 15715 is a pro-drug of propofol. Because the active metabolite propofol must be first released, GPI 15715 has a slower onset and a longer duration of action than common propofol. This maybe a disadvantage for the induction of anaesthesia, so that the potential use may be rather maintenance of anaesthesia and particularly sedation, where the missing lipid load and the potentially lower risk of infection is a clear benefit compared to propofol emulsion. The longer duration of action might also be an advantage for short procedures of 30–40 min duration, where a single-bolus dose of GPI 15715 could probably provide sufficient
Table 5.4 Pharmacokinetics and -dynamics of lipid propofol (Diprivan™) and propofol from GPI 15715 DiprivanTM
GPI 15715
Central volume of distribution (l/kg)
0.23±0.03
0.35±0.06*
K10 (1/min)
0.11±0.02
0.07±0.02*
T1/2α(min)
4.7±0.8
2.5±1.2*
T1/2β (min)
58.2±17.9
26.3±9.3*
T1/2γ (min)
651±194
543±223
EC50 (µg/ml)
3.0±0.7
2.1±0.5*
Propofol at loss of response to verbal command (µg/ml)
4.6±0.8
1.9±0.5*
Propofol at alertness (µg/ml)
0.5±0.4
0.7±0.2
K10, elimination rate constant; T1/2α, short half-life; T1/2β, intermediate half-life; T1/2γ, terminal halflife; EC50, propofol concentration at half-maximum effect. * P<0.05. Source: Fechner et al. (2004).
hypnosis. From a more pharmacological point of view, it is interesting that the liberated propofol from GPI 15715 shows different pharmacokinetics and dynamics than propofol in lipid emulsion. Similar results were found in earlier studies with lipidfree propofol |14|. The paresthesia after GPI 15715 limits the advantage of no pain at the injection site. The potential risk of formaldehyde, which is another metabolite of GPI 15715, must be investigated further.
Electroencephalographic properties of zaleplon, a non-benzodiazepine sedative/hypnotic, in rats Noguchi H, Kitazumi K, Mori M, Shiba T. J Pharmacol Sci 2004; 94(3): 246–51 BACKGROUND. Zaleplon is a newer short-acting non-benzodiazepine sedative/hypnotic agent An EEG was used for investigating its sedative/hypnotic
The year in anaesthesia and critical care
128
effects in comparison with other sedative hypnotics in conscious rats with chronically implanted electrodes. Zaleplon (0.25–1.0 mg/kg), triazolam (0.0625–0.25 mg/kg), zopiclone (1.0–4.0 mg/kg), brotizolam (0.0625–0.25 mg/kg) and nitrazepam (0.125–0.5 mg/kg) were administered orally. The EEG effect was characterized by visual classification of the EEG, whereby four stages were differentiated: awake, slow-wave light sleep, slow-wave deep sleep and fast-wave sleep. In addition, the power spectrum of the EEG was analysed and the relative power in the four frequency bands, delta (1Z–4 Hz), theta (4–8 Hz), alpha (8–13 Hz) and beta (13–30 Hz), was determined. INTERPRETATION. With regard to the distribution of sleep-wakefulness stages, zaleplon in particular increased slow-wave deep sleep, whereas the other agents increased slow-wave light sleep in a dose-dependent manner. All tested hypnotics caused no influence on fastwave sleep at the doses tested. The onset time to sleep was more rapid for zaleplon (6.6±2.6 min for the highest dose of 1 mg/kg) compared to the other compounds tested (range 11–30 min for the highest dose). Zaleplon significantly increased the relative EEG power density in the delta frequency band. Therefore, the present findings suggest that zaleplon can be expected to exhibit high practical potential as a hypnotic. Comment This was an animal study for characterizing the pharmacodynamic effects of zaleplon. Whereas the authors give details about the onset of action and the EEG patterns during administration of the investigated compounds, there is no information about the duration of action, which is as important as the onset. Zaleplon produced an increase in the delta band of the EEG, which is also observed for typical hypnotic drugs such as, for example, propofol. In contrast, the benzodiazepine midazolam was characterized by an increase in the beta band at doses used for sedation. Further studies need to investigate the controllability of zaleplon in comparison with midazolam and its potential as a drug for sedation. Conclusion Several new efforts have been undertaken in order to formulate propofol either with other lipid solvents or without any lipid in a water-soluble formulation. From the results of the reviewed papers one has to conclude that the recently developed new lipid propofol formulations did not fulfil the expectations of similar pharmacokinetic-dynamic properties together with less side effects than the common lipid formulation. Lipid-free solutions with cyclodextrin as solvent have only been studied in animals so there may still be some potential of these propofol formulations for use in man. The pro-drug approach seems to be more promising, despite its slower onset and longer duration of action.
New hypnotics
129
Whereas all these papers deal with new formulations of the well-known agent propofol, no really new hypnotic drug was presented in 2004 with the exception of an abstract about a drug called THRX-918661 |15|. This positive allosteric modulator of the γ-aminobutyric acidA (GABA) receptor is metabolized by esterases and showed a higher clearance than Diprivan™ in rats and pigs with a very rapid decrease in concentration after discontinuing and a fast recovery. This drug may have clinical potential as a new ultra-short-acting sedative-hypnotic agent. References 1. Hering WJ, Ihmsen H, Langer H, Uhrlau C, Dinkel M, Geisslinger G, Schuttler J. Pharmacokinetic-pharmacodynamic modeling of the new steroid hypnotic eltanolone in healthy volunteers. Anesthesiology 1996; 85: 1290–9. 2. Dingemanse J, Haussler J, Hering W, Ihmsen H, Albrecht S, Zell M, Schwilden H, Schuttler J. Pharmacokinetic-Pharmacodynamic modelling of the EEG effects of Ro 48–6791, a new shortacting benzodiazepine, in young and elderly subjects. Br J Anaesth 1997; 79: 567–74. 3. Sneyd JR. Recent advances in intravenous anaesthesia. Br J Anaesth 2004; 93: 725–36. 4. Sneyd JR, Wright PM, Harris D, Taylor PA, Vijn PC, Cross M, Dale H, Voortman G, Boen P. Computer-controlled infusion of ORG 21465, a water-soluble steroid i.v. anaesthetic agent, into human volunteers. Br J Anaesth 1997; 79: 433–9. 5. Sneyd JR, Tsubokawa T, Andrews CJH, Curnow J, Cross M, Lytle J, Visser L, Lunn DV, Houwing NS, Boen P, Van Maanen R, Vijn PCM. First administration to man of ORG25435, a new intravenous anaesthetic agent. Br J Anaesth 2001; 86: 323P. 6. Kelbel I, Koch T, Weber A, Schiefer HG, Van Ackern K, Neuhof H. Alterations of bacterial clearance induced by propofol. Acta Anaesthesiol Scand 1999; 43: 71–6. 7. Lindholm M. Critically ill patients and fat emulsions. Minerva Anestesiol 1992; 58: 875–9. 8. Nakane M, Iwama H. A potential mechanism of propofol-induced pain on injection based on studies using nafamostat mesilate. Br J Anaesth 1999; 83: 397–404. 9. Ririe DG, Lundell JC, Neville MJ. Direct effects of propofol on myocardial and vascular tissue from mature and immature rats. J Cardiothorac Vasc Anesth 2001; 15: 745–9. 10. Ward DS, Norton JR, Guivarch PH, Litman RS, Bailey PL. Pharmacodynamics and pharmacokinetics of propofol in a medium-chain triglyceride emulsion. Anesthesiology 2002; 97: 1401–8. 11. Theilen HJ, Adam S, Albrecht MD, Ragaller M. Propofol in a medium- and long-chain triglyceride emulsion: pharmacological characteristics and potential beneficial effects. Anesth Analg 2002; 95: 923–9. 12. Doenicke AW, Roizen MF, Rau J, O’Connor M, Kugler J, Klotz U, Babl J. Pharmaco-kinetics and pharmacodynamics of propofol in a new solvent. Anesth Analg 1997; 85: 1399–403. 13. Egan TD, Kern SE, Johnson KB, Pace NL. The pharmacokinetics and pharmacodynamics of propofol in a modified cyclodextrin formulation (Captisol) versus propofol in a lipid formulation (Diprivan): an electroencephalographic and hemodynamic study in a porcine model. Anesth Analg 2003; 97:72–9. 14. Dutta S, Ebling WF. Formulation-dependent pharmacokinetics and pharmacodynamics of propofol in rats. J Pharm Pharmacol 1998; 50: 37–42. 15. Jenkins T, Beattie D, Jaw-Tsai S, Amagasu S, Halladay J, Vanapalli S, Kern R, Shaw JP, Egan T, Shafer S. THRX-918661, a novel, pharmacokinetically responsive sedative/ hypnotic agent. Anaesthesia 2004; 59: 100.
6 ‘New’ analgesics MARTIN LUGINBÜHL Introduction The combination of several analgesic drugs in a multi-modal approach is considered standard practice in post-operative and intra-operative analgesia. The main goal of combining analgesic drugs is to increase the efficacy and to minimize side effects, particularly opioid side effects. No principally new substances have been developed during the last 18 months, but some known analgesic drugs are being investigated for perioperative use. With the withdrawal of rofecoxib (Vioxx™) from the market the role of an important class of drugs that have been increasingly used during recent years became uncertain. Gabapentin and pregabalin, which are currently used for chronic neuropathic pain, are under investigation for perioperative use with the intention of preventing the development of chronic pain states and improving perioperative pain control. The α2-agonist dexmedetomidine has mainly been investigated as a sedative drug in intensive care. Recent studies have investigated its pharmacokinetics and dynamics for acute pain treatment. Morphine-6-glucuronide (M6G), a well-known active metabolite of morphine, has been investigated for its analgesic efficacy. It is an interesting analgesic drug because of a low incidence of respiratory depression. The question of whether these drugs will find a place in perioperative analgesia is still open. The Vioxx™ story reminds us once more that short-term outcome parameters, such as pain relief or morphine consumption or short-term side effect parameters, such as nausea and vomiting and perioperative blood loss, are not sufficient for a meaningful cost-benefit analysis. This review presents a selection of articles around the risk-benefit ratio of cyclooxygenase-2 (COX-2) inhibitors and on gabapentin, M6G and dexmedetomidine, which have recently been considered for perioperative use. The role of cyclooxygenase-2 inhibitors after the withdrawal of rofecoxib The VIOXX Gastrointestinal Outcomes Research (VIGOR) Trial showed a 53% decrease in the risk of upper gastrointestinal toxicity and a fivefold increase in the © Atlas Medical Publishing Ltd
risk of myocardial infarction for rofecoxib compared with naproxen |1|. Four years later the preliminary results of the Adenomatous Polyp Prevention on Vioxx Study revealed a fourfold increase in the incidence of serious thromboembolic adverse events in patients treated with 25 mg of rofecoxib per day as compared with placebo. The incidence of myocardial infarction and thrombotic stroke in the two groups began to diverge
'New' analgesies
131
progressively after 1 year or more of treatment |2|. These as yet unpublished data were the reason for withdrawing the drug from the market. The clinician now faces a situation where the mechanism of cardiovascular toxicity is debated and where the experts argue whether the increased cardiovascular risk is a problem of Vioxx™ or of all COX-2 inhibitors. The following articles may help the clinician in taking a personal decision. The risk—benefit ratio for rofecoxib was addressed in a reanalysis of the data from the VIGOR Study estimating the life expectancy of patients with rheumatoid arthritis and different risks for gastrointestinal and coronary events. The important question of whether the increase in myocardial infarctions is only related to rofecoxib or whether it is related to the whole group of COX-2 inhibitors is considered controversial. Solomon et al. addressed it below in their large case—control study in elderly patients.
Effects of rofecoxib and naproxen on life expectancy among patients with rheumatoid arthritis: a decision analysis Choi HK, Seeger JD, Kuntz KM. Am J Med 2004; 116: 621–9 BACKGROUND. The VIGOR Trial provided the best data for informing decision making regarding the use of naproxen or rofecoxib in patients with rheumatoid arthritis. The trial showed a 53% decrease in the risk of upper gastrointestinal toxicity and a fivefold increase in the risk of myocardial infarction for rofecoxib compared with naproxen. Clinicians need to consider the trade-offs between the potentially harmful cardiovascular effects of rofecoxib and its potentially beneficial effect of reducing gastrointestinal toxicity. The authors used decision analysis for incorporating the long-term risks of these competing adverse events and compared the life expectancy and quality-adjusted life expectancy associated with naproxen and rofecoxib therapy in patients with rheumatoid arthritis. INTERPRETATION. A complex model incorporating the age-related risk of gastrointestinal toxicity and myocardial infarction and related problems and the risk of death was developed based on the VIGOR Trial and other epidemiological studies. The calculations were referenced to 58-year-old women with rheumatoid arthritis, as these individuals were the most typical patients in the VIGOR Trial. Although the gastrointestinal complications in patients taking naproxen were far more frequent than myocardial infarctions in those taking rofecoxib (4.5 versus 0.53 events per 100 patientyears, respectively) naproxen was associated with a longer life expectancy than rofecoxib. The differences were 4.4 and 1.9 months for 58- and 75-year-old women, while those for 58- and 75-year-old men were 7.8 and 3.6 months, respectively. The estimated cost for rofecoxib was higher. The authors concluded that only patients with a high risk of upper gastrointestinal toxicity and a low risk for myocardial infarction would profit from rofecoxib compared to naproxen with respect to life expectancy.
The year in anaesthesia and critical care
132
Comment Although this study was only a post hoc analysis of previous data the results suggest that the side effects of specific or non-specific COX inhibitors cannot be addressed on a global basis but should be considered in a stratified manner. In the perioperative setting the use of COX-2 inhibitors may thus be beneficial in younger patients at low risk for acute myocardial infarction (AMI) but not in others. The reduced perioperative blood loss |3| related to the lack of thrombocyte inhibition |4| as well as the reduced opioid consumption |5, 6| of patients given COX-2 inhibitors instead of non-selective nonsteroidal anti-inflammatory drugs (NSAIDs) for post-operative analgesia may be at the expense of an increased risk of AMI at least in older patients. It was in this patient population undergoing hip or knee arthroplasty where the use of COX-2 inhibitors has been considered particularly useful. The improved mobility after total knee replacement |7| has to be considered differently because this patient population is also at risk for coronary heart disease.
Relationship between selective cyclooxygenase-2 inhibitors and acute myocardial infarction in older adults Solomon DH, Schneeweiss S, Glynn RJ, et al. Circulation 2004; 109: 2068–73 BACKGROUND. After the VIGOR Study became available it was discussed whether the difference in AMI between naproxen- and rofecoxib-treated patients could be explained by a protective effect of naproxen and/or an increased risk associated with the selective COX-2 inhibitor. The authors of this study investigated the relative risk of AMI among users of celecoxib, rofecoxib and NSAIDs in Medicare beneficiaries. INTERPRETATION. The authors enrolled 54 475 patients of 65 years of age or older (predominantly white women with a mean age greater than 85 years and frequent cardiovascular risk factors such as hypertension and diabetes and previous cardiovascular disease [CVD]). Each of the 10 895 cases of AMI was matched to four controls on the basis of age, gender and the month of hospital admission of index patients. The authors constructed matched logistic regression models including indicators for patient demographics, healthcare use, medication use and cardiovascular risk factors in order to assess the relative risk of AMI in patients who used rofecoxib compared with persons taking no NSAID, taking celecoxib or taking naproxen or ibuprofen. The use of rofecoxib was associated with an elevated relative risk of AMI compared with celecoxib (odds ratio [OR] 1.24) and no NSAID (OR 1.14), but not compared to naproxen or ibuprofen. The increased risk of AMI was detected in both dose ranges compared: ≤25 mg or >25 mg of rofecoxib compared to ≤200 mg or >200 mg of celecoxib (OR 1.21 and 1.70, respectively). At withdrawal of the drugs Merck mentioned an increased risk of AMI in patients taking rofecoxib longer than 18 months. In this study there was also an increased risk of AMI in patients taking refocoxib compared to celecoxib in the first 30 days of use (OR 1 40) and
'New' analgesies
133
in the second and third months (OR 1.38), but not when taken longer than 3 months (OR 0.96). In patients with previous AMI there was no difference in the risk of AMI between rofecoxib and celecoxib. In contrast, celecoxib was not associated with an increased relative risk of AMI compared to standard NSAIDs or no NSAIDs. Comment According to this study the use of rofecoxib compared to celecoxib or no NSAID use was associated with an increased risk of AMI in short- or intermediate-term use (up to 90 days). This implies that even very short-term perioperative use of COX-2 inhibitors may increase the cardiovascular risk of patients. Safety trials of newer COX-2 inhibitors should therefore consider the cardiac risk even in short-term use. The study might suggest differences between the different COX-2 inhibitors. However, in patients with a history of AMI representing those with a high cardiovascular risk this difference was not detectable, probably because other predisposing factors overrode the impact of the COX2 inhibitor. The data also suggest a certain dose dependence of the cardiovascular toxicity. Walter et al. proposed a non-enzymatic mechanism (i.e. not related to the inhibition of COX) as a potential explanation for the different cardiovascular risk of COX-2 inhibitors |8|. In a very recent study they compared the independent effects of COX-2 inhibitors on human low-density lipoprotein (LDL) oxidation, an important contributor to atherosclerotic CVD. Whereas rofecoxib and etoricoxid, two sulphone COX-2 inhibitors, decreased the anti-oxidative capacity of human plasma and increased the LDL oxidation rates, sulphonamide COX-2 inhibitors such as celecoxib, valdecoxib and meloxicam did not |8|. Conversely Fitzgerald assumed an enzymatic (and therefore group-specific) mechanism as being responsible for the increased cardiovascular risk. The synthesis of prostaglandin I2, which inhibits platelet aggregation, is mediated by COX-2, causes vasodilatation and prevents the proliferation of vascular smooth-muscle cells in vitro |2|. Whereas aspirin and traditional NSAIDs inhibit both thromboxane A2 and prostaglandin I2, the COX-2 inhibitors leave the generation of thromboxane A2 unaffected, reflecting the absence of COX-2 in platelets. Suppression of the COX-2-dependent formation of prostaglandin I2 by the COX-2 inhibitors might therefore predispose patients to myocardial infarction or thrombotic stroke |2|. A post hoc analysis of the data from the Celecoxib Long-term Arthritis Safety Study |9| (celecoxib versus NSAID) also revealed signs of increased cardiovascular risk for patients treated with celecoxib when used for longer than 6 months |10|. Particularly important for the perioperative setting is a recent animal study on the role of COX-2 in ischaemic preconditioning of the myocardium. Alcindor et al. demonstrated that COX-2 is an obligatory mediator of ischaemic, anaesthetic and pharmacological preconditioning |11|. Administration of celecoxib antagonized ischaemic, anaesthetic (isoflurane 1 minimal alveolar concentration) or pharmacological (2.5 mg diazoxide) preconditioning in this study, whereas aspirin and paracetamol did not (Fig. 6.1).
The year in anaesthesia and critical care
134
Fig. 6.1 Myocardial infarct size expressed as a percentage of the left ventricular area at risk in dogs receiving no preconditioning stimuli (CON), celecoxib (CELE), aspirin (ASA) or acetaminophen (ACET) in the presence or absence of anaesthetic preconditioning (APC), ischaemic preconditioning (IPC) or pharmacological preconditioning (PPC). *Significantly (P<0.05) different from no preconditioning
'New' analgesies
135
stimuli. †Significantly (P<0.05) different from anaesthetic preconditioning alone. §Significantly (P<0.05) different from ischaemic preconditioning alone. ¶Significantly (P<0.05) different from pharmacological preconditioning alone. Source: Alcindor et al. |11|. Given this uncertainty on the risk—benefit of COX-2 inhibitors the large systematic review of 18 randomized controlled studies comparing the analgesic efficacy of COX-2 inhibitors with conventional NSAIDs and oxycodone/paracetamol for perioperative analgesia emphasized the need for using meaningful and ‘hard’ end-points in clinical studies. In that sense it will be much more difficult to show positive effects than simple efficacy compared to placebo.
Systematic review of the analgesic efficacy and tolerability of COX-2 inhibitors in post-operative pain control Chen LC, Elliott RA, Ashcroft DM. J Clin Pharm Ther 2004; 29: 215–29 BACKGROUND. In this systematic review of randomized controlled trials the authors evaluated the relative analgesic efficacy and tolerability of single-dose COX2 inhibitors in post-operative pain management. INTERPRETATION. Eighteen randomized controlled trials were included with a total of 2783 patients, 17 of which were sponsored by drug companies. The proportion of patients achieving at least 50% pain relief and the proportions of patients experiencing any adverse event or specific adverse events were extracted from the publications. All COX-2 inhibitors (50 mg of rofecoxib, 200 mg of celecoxib, 20 and 40 mg of valdecoxib and 20 and 40 mg of parexocib) were more effective than placebo in all types of surgery. In trials with active controls on dental surgery (13 trials) 50 mg of oral rofecoxib was more effective than 60/600 mg of codeine/paracetamol and 40 mg of oral valdecoxib was more effective than 10/1000 mg of oxycodone/paracetamol, whereas 200 mg of celecoxib was less effective than 400 mg of ibuprofen and 50 mg of rofecoxib. There were no significant differences between other oral COX-2 inhibitors and non-selective NSAIDs. In orthopaedic surgery (three trials) 50 mg of oral rofecoxib, 550 mg of naproxen sodium, 20/40 mg of intravenous parecoxib and 30 mg of intravenous ketorolac were equally effective, whereas 20/40 mg of intravenous parecoxib was more effective than 4 mg of intravenous morphine. In gynaecology surgery (two trials) 20/40 mg of intravenous parecoxib was more effective than 4 mg of intravenous morphine but equal to 30 mg of ketorolac. Only dizziness, headache, nausea and vomiting were reported as adverse effects in these studies with no significant differences between the different treatment groups Serious adverse effects such as gastrointestinal complications surgical bleeding
The year in anaesthesia and critical care
136
renal failure or cardiovascular risks were not addressed in the trials evaluated in this review. The number of patients involved (150–500) was too small to obtain such data. The authors concluded that further studies are needed for examining the efficacy and tolerability of COX-2 inhibitors compared against active comparators over a longer duration in order to determine their ultimate risk-benefit profile. Comment The reason for presenting this paper in this review was not that this systematic review demonstrated that at least some of the COX-2 inhibitors are better analgesics than oxycodone/paracetamol or non-selective NSAIDs or a single dose of morphine. Rather, it was the fact that this review illustrated the type of end-points used in most studies for assessing the efficacy of new analgesics. Acute analgesic effects are easily identified, whereas the incidence of potential side effects cannot be reliably estimated because the sample size was limited. Only very large-scale, randomized controlled studies, for which financial support will be difficult to obtain, can answer this question. The Food and Drug Administration (FDA) has only approved valdecoxib of the second-generation COX-2 inhibitors, whereas the European authorities have approved etoricoxib. Lumiracoxib was under consideration for FDA approval until Novartis (Basle, Switzerland) retired the application in December 2004. The risk—benefit profile of these newer COX-2 inhibitors is still controversial. Valdecoxib Valdecoxib is the only COX-2 inhibitor that is also available for intravenous administration. Recent large trials have documented its efficacy in low back pain |12| and rheumatoid arthritis |13|. It was equally as effective as rofecoxib in post-operative pain |14–17|. In a meta-analysis of data on over 8000 patients, neither long-term nor shortterm use of valdecoxib at various dosages was associated with an increased risk in cardiovascular morbidity |18|. As with other COX-2 inhibitors it does not impair blood coagulation or platelet function |19, 20|. A pharmacokinetic interaction with propofol, midazolam, fentanyl and alfentanil has formally been excluded |21–23|. Etoricoxib Etoricoxib has the highest COX-2 selectivity of all COX-2 inhibitors with a COX2:COX-1 inhibition ratio of 106 |24|. As with rofecoxib the pharmacokinetics allow oncedaily administration |25|. Because it is metabolized by the liver a substantial dose reduction is necessary in patients with mild to moderate hepatic dysfunction |26|, whereas in mild to moderate renal dysfunction no dose reduction is necessary |27|. However, etoricoxib is not dialysed and should be avoided in patients with severe or end-stage renal failure. Etoricoxib was more effective in the treatment of post-operative pain after third molar extraction compared to oxycodone/ acetaminophen |28|. The potential for cardiovascular side effects cannot be excluded so far |8|.
'New' analgesies
137
Lumiracoxib Lumiracoxib was equally as effective in the treatment of osteoarthritic pain as celecoxib |29|. It was associated with a three- to fourfold risk reduction for gastrointestinal complications compared with ibuprofen and naproxen (Therapeutic Arthritis Research and Gastrointestinal Event Trial) |30|. The cardiovascular risk was similar to that of the two non-selective comparators |31|, although Fitzgerald argued that the study was underpowered for detecting a true cardiovascular risk |2|. To conclude, until the cardiovascular risk of newer COX-2 inhibitors is clear the perioperative use of these drugs in patients at risk for coronary artery disease (CAD) or with known CAD should be carefully considered based on an individual risk—benefit estimation. No general recommendations can be given at the present time. Gabapentin and pregabalin The anti-epileptic drug gabapentin is currently in clinical use for several types of neuropathic pain, e.g. neuropathic cancer pain |32|, phantom limb pain |33|, trigeminal neuralgia |34|, post-herpetic neuralgia |35| and spinal cord injury |36|.
Fig. 6.2 Structure of GABA, gabapentin (1-(aminoethyl) cyclohexane acetic acid) and pregabalin (S(+)-3-isobutyl GABA). Source: McClelland et al. |37|. Gabapentin and pregabalin are structurally related to γ-aminobutyric acid (GABA) (Fig. 6.2) |37|. Gabapentin is predominantly effective in pathological nociception, but not in physiological nociception. This was illustrated in a recent experimental volunteer study on capsaicin-induced pain, where gabapentin significantly reduced the area of brush allodynia compared with placebo and had a trend of attenuating the area of pinprick hyperalgesia, but had no significant effect on spontaneous and evoked pain intensity |38|. Because a substantial number of patients develop chronic pain syndromes after acute post-operative pain, gabapentin has also been suggested for perioperative use |39|.
The year in anaesthesia and critical care
138
Mechanism of action The controversy about the mechanism of action of gabapentin and congeners is still unsolved. Apparently gabapentin has central as well as peripheral |40| sites of action. 1. Voltage-dependent calcium (Ca2+) channels. Gabapentin and pregabalin bind to the Ca2+ channel α2δ subunit to attenuate depolarization-induced Ca2+ influx of presynaptic P/Q-type Ca2+ channels selectively. The alpha 2 delta voltage-activated Ca2+ channel is over-expressed in sensory neurones after nerve injury |41|. Binding to this channel results in decreased release of several neurotransmitters such as glutamate/aspartate, norepinephrine and substance P |42|. Animal data suggest that this mechanism mediates the anti-allodynic effect of gabapentin |43|. 2. GABAA receptors. In a rat model of secondary hyperalgesia the anti-allodynic effect of gabapentin was blocked by a selective GABAA receptor antagonist |44|. 3. GABAB receptors. In different hippocampal cell lines gabapentin was shown to bind to a specific subtype of GABAB receptors |45–47|. This could not be confirmed in a very recent study on the anti-allodynic effect in a rat model in vivo and on the effect on a periaqueductal grey cell in vitro |48|. Of particular interest for perioperative use is the potential of gabapentin in preventing morphine tolerance. A recent animal study in rats demonstrated a complete suppression of morphine tolerance developing within 4 days after treatment (Fig. 6.3) |49|. However, these results remain to be confirmed in humans. Gabapentin: clinical efficacy in perioperative use In a preliminary trial gabapentin enhanced the acute analgesic effect of morphine on cold pressure pain in healthy volunteers. This would also imply a certain effect on acute nociceptive pain. Furthermore, the plasma concentration of gabapentin was increased when morphine was administered concomitantly |50|. A single dose of 1200 mg of oral gabapentin resulted in a substantial reduction in post-operative morphine consumption and movement-related pain after radical mastectomy, without ‘significant side effects’ |51|. Subsequent trials demonstrated a morphine-sparing effect of gabapentin after spinal surgery |52|, lumbar discectomy |53|, abdominal hysterectomy |54|, vaginal hysterectomy |55| and for monitored anaesthesia care in ear, nose and throat surgery |56|. However, the opioid-sparing effect is a surrogate end-point. Reducing the morphine consumption from a dose not associated with severe side effects to a lower dose is probably not relevant for the patient. In chronic pain studies the reported side effects (2400 mg of gabapentin daily) were dizziness (33%), somnolence (21%), peripheral oedema (11%), asthenia (6%), a dry mouth (5%) and diarrhoea (5%). In the perioperative setting these side effects were similar in the treatment group to those in the placebo group because of the residual effects of the pre-medication and anaesthesia overriding the gabapentine side effects |55|.
'New' analgesies
139
Fig. 6.3 (a) Tail-flick responses and (b) paw-pressure responses (mean ± standard error [SE] of the mean error bars) under repeated daily administration of saline, gabapentin, morphine and morphine plus gabapentin (combo(1–7) group) on days 1–7. All doses of morphine and gabapentin are 15 and 300 µg, respectively. *P<0.05 compared with
The year in anaesthesia and critical care
140
morphine alone. †P<0.05 compared with saline. Source: Hansen et al. |49|. Pregabalin: clinical efficacy Whereas gabapentin has also been extensively investigated for perioperative use, only one clinical study is as yet available for pregabalin in this setting. A dose of 300 mg of pregabalin was associated with improved pain relief compared to placebo and 400 mg of ibuprofen in patients after third molar tooth removal |57|. Pregabalin demonstrates highly predictable and linear pharmacokinetics, with rapid extensive absorption proportional to the dose and a high bioavailability. The time to maximal plasma concentration is approximately 1 h and the mean elimination half-life (t1/2) is 6.3 h. Thus, steady state is achieved within 24–48 h |58|. The effect of pregabalin is stereospecific |59| and subject to pharmacogenetics |60|. In two recent studies pregabalin was effective compared to placebo in the treatment of painful diabetic neuropathy and post-herpetic neuralgia |61, 62|. The most important side effects reported were dizziness (35%), somnolence (19%), peripheral oedema (10%) and infection (upper respiratory tract) (14.5%). To conclude, gabapentin and pregabalin may also be promising drugs in the perioperative setting. However, there are no data available demonstrating an improved outcome based on hard end-points for justifying routine perioperative use. With pregabalin the increased rate of infections in the treatment group, which is completely unclear, may be a concern for large-scale and uncontrolled perioperative use. The impact of gabepentin and congenors on the development of chronic pain syndromes, e.g. after thoracotomy or herniorrhaphy, remains to be determined. Morphine-6-glucuronide M6G is a well-known active metabolite of morphine with a higher affinity for µ-receptors and poor penetration of the blood—brain barrier. Recently it was investi-gated for its therapeutic use because it is less respiratory depressant. The effect of M6G is subject to a genetic polymorphism because the adenine—guanine single-nucleotide polymorphism at position 118 (A118G) is associated with a reduction of M6G potency by 50% in heterozygous patients. The heterozygous state has a prevalence of approximately 30% in the population and implies an exchange of one amino acid in the µ-opioid receptor (see Romberg et al. 2004 below). A small study in volunteers using an electrical pain model showed that M6G is approximately half as potent as morphine |63|. M6G significantly reduced cutaneous hyperalgesia (‘freeze lesion’) and muscle hyperalgesia, but to a lesser extent than morphine. In this study- in contrast to a previous report |64|—electrical pain was not affected by M6G. The results of the study indicated that M6G has anti-hyperalgesic effects in inflammatory pain through activation of peripheral opioid receptors |65|.
'New' analgesies
141
Pharmacodynamic effect of morphine-6-glucuronide versus morphine on hypoxic and hypercapnic breathing in healthy volunteers Romberg R, Olofsen E, Sarton E, Teppema L, Dahan A. Anesthesiology 2003; 99: 788– 98 BACKGROUND. Initial clinical studies indicated excellent pain relief with intravenous M6G, with reduced respiratory depression as compared to morphine. M6G is currently undergoing phase III clinical trials and will be marketed as an analgesic for post-operative and chronic (non-) malignant pain. INTERPRETATION. The authors compared the effect of M6G against morphine on hypercapnic and hypoxic breathing in nine healthy female volunteers. The effects of 0.2 mg/kg of intravenous M6G, 0.13 mg/kg of intravenous morphine and intravenous placebo were tested on ventilation at a fixed end-tidal pressure of carbon dioxide of 45 mmHg (Vi45) and on the acute hypoxic ventilatory response (AHVR). Morphine was more potent in affecting hypoxic ventilatory control than M6G, with a potency ratio ranging from 19:1 for Vi45 to 50:1 for AHVR. At drug concentrations causing 25% depression of Vi45, M6G caused only 15% depression of AHVR, whereas morphine caused greater than 50% depression of AHVR. Furthermore, the speed of onset/offset of M6G was faster than morphine by a factor of approximately 2. Comment The pharmacokinetics and pharmacodynamics (analgesic effect) was further investigated in a subsequent study by the same investigators.
Pharmacokinetic-pharmacodynamic modeling of morphine-6glucuronide-induced analgesia in healthy volunteers: absence of sex differences Romberg R, Olofsen E, Sarton E, Den HJ, Taschner PE, Dahan A. Anesthesiology 2004; 100: 120–33 BACKGROUND. M6G is a µ-opioid receptor agonist. The authors quantified the analgesic potency and speed of onset-offset of M6G on the analgesic effect and explored putative sex dependency. INTERPRETATION. In this placebo-controlled, randomized, double-blind study ten male and ten female healthy volunteers were given 0.3 mg/kg of M6G and placebo intravenously (two-thirds of the dose as bolus and one-third as a continuous infusion over 1 h) on separate occasions. Pain tolerance was measured using gradually increasing transcutaneous electrical stimulation. A population pharmacokinetic (inhibitory sigmoid Emax)—pharmacodynamic analysis was used for quantifying M6G-induced changes in tolerated stimulus intensity The M6G analgesia data were well described by the
The year in anaesthesia and critical care
142
pharmacokinetic-pharmacodynamic model. M6G caused analgesia significantly greater than that observed with placebo (P<0.01). The M6G effect site concentration causing a 25% increase in current (C25) was 275±135 nm/l (population estimate ± SE), the blood effect site equilibration half-life was 6.2±3.3 h and the steepness parameter was 0.71±0.18. The inter-subject variability was 167% for C25 and 218% for the effect halflife. None of the model parameters showed sex dependency. Possible causes for the great intersubject response variability are genetic polymorphism of the µ-opioid receptor and placebo-related phenomena. Comment In the discussion of this study the authors reported an analgesic potency ratio of morphine:M6G of 12:1–22:1 based on a comparison of current data with a previous study |66|. This compares to a potency ratio of 19:1–50:1 for respiratory depression. Of special interest is the relation of respiratory depression and analgesia. Whereas analgesia and respiratory depression were parallel for morphine (and other µ-agonists), the effect site concentration causing a 25% increase in electrical pain tolerance was 275 nm/l, whereas the effect site concentration causing a 25% decrease in ventilation at an arterial carbon dioxide pressure (PaCO2) of 45 mmHg was 528 nm/l and that of a 25% decrease in the hypoxic ventilatory response was 872 nm/l (see the previous Romberg et al. 2003 study above). Interestingly the equilibration half-life of plasma and effect site (t1/2ke0) was shorter for respiratory depression (1.0–2.6 h) than for analgesia (6.2 h) in M6G, whereas it was similar for morphine (4.4 h). Although substantially less potent than morphine, M6G might offer an increased safety margin compared to morphine (and eventually other µ-receptor antagonists). Only large, randomized clinical trials will show whether this translates into increased safety in post-operative analgesia. An important issue will be the larger inter-individual variation of efficacy, which is at least in part related to the high prevalence of the A118G single-nucleotide polymorphism associated with a decreased sensitivity of the µ-opioid receptor to M6G. Dexmedetomidine The α2-receptor antagonist dexmedetomidine is approved for sedation in intensive care unit (ICU) patients in the USA, but not in Europe |67|. Adult ICU patients sedated with dexmedetomidine required less morphine than those sedated with propofol |68|, an opioid-sparing effect that has also been observed in children |69|. The pharmacodynamics of this promising drug have recently been investigated in view of a potential use for surgical anesthesia in the operating room. The analgesic potency of dexmedetomidine, in contrast to its sedative potency, is still controversial. At lower concentrations (up to 1.2 ng/ml) only sedation occurred, but no analgesic effect was detected with heat and electrical pain models in healthy volunteers |70|. However, a clinical study showed a significant opioid-sparing effect in postoperative analgesia |71|. It has also been successfully used for decreasing the shivering threshold compared to saline placebo. This effect was approximately half that of
'New' analgesies
143
meperidine and was additive to meperidine |72|. Adding 0.5 µg/kg of dexmedetomidine to lidocaine for intravenous regional anaesthesia improved the quality of anaesthesia and intra-operative—post-operative analgesia without causing side effects consistent with its increased selectivity for α2-adreceptors compared to clonidine |73|. Two very recent small volunteer trials provide essential information on the pharmacological profile of this drug, which may open the horizon for its use outside the ICU.
Dexmedetomidine pharmacodynamics: Part I, crossover comparison of the respiratory effects of dexmedetomidine and remifentanil in healthy volunteers Hsu YW, Cortinez LI, Robertson KM, et al. Anesthesiology 2004; 101:1066–76 BACKGROUND. Dexmedetomidine, through its actions on α2-adrenoceptors, has sedative and analgesic effects while having minimal ventilatory effects. However, there are only limited data on the ventilatory effect of dexmedetomidine. INTERPRETATION. The authors compared the respiratory effect of dexmedetomidine to that of remifentanil in six healthy male volunteers. Their respiratory responses were measured during a stepwise target-controlled infusion of remifentanil, a stepwise target-controlled infusion of dexmedetomidine and a pseudo-natural sleep session. Compared with baseline the remifentanil infusions resulted in a decrease in their respiratory rate and minute ventilation, respiratory acidosis and apnoea episodes resulting in desaturations. Remifentanil disturbed the natural pattern of breathing and flattened the distribution of ventilatory timing (inspiratory time/ventilatory cycle time). Compared with baseline the dexmedetomidine infusions increased their respiratory rate significantly and decreased the overall apnoea/hypopnoea index, whereas the distribution of the inspiratory time/ventilatory cycle time showed an increased peak. In addition, dexmedetomidine seemed to mimic some aspect of natural sleep. Hypercapnic arousal phenomena (documented by the bispectral index, an electroencephalogram and a sudden increase in their minute ventilation) were observed during the dexmedetomidine infusions while the subjects were breathing a 5% CO2 mixture. Similar phenomena during natural sleep have been reported in the literature. The authors concluded that, in contrast to remifentanil, dexmedetomitine (1) did not result in clinically significant respiratory depression, (2) decreased rather than increased the apnoea/hypopnoea index and (3) exhibited some similarity with natural sleep.
Dexmedetomidine pharmacodynamics: Part II, crossover comparison of the analgesic effect of dexmedetomidine and remifentanil in healthy volunteers Cortinez LI, Hsu YW, Sum-Ping ST, et al. Anesthesiology 2004; 101: 1077–83
The year in anaesthesia and critical care
144
BACKGROUND. Dexmedetomidine is a highly selective α2-adrenoceptor agonist that is used for short-term sedation of mechanically ventilated patients. The analgesic profile of dexmedetomidine has not been fully characterized in humans. INTERPRETATION. The analgesic response of six healthy male volunteers was compared during stepwise target-controlled infusions of remifentanil and dexmedetomidine. A computer-controlled thermode was used for delivering painful heat stimuli (between 41 and 50°C) to the volar side of the forearms and the pain intensity was recorded. The data were fitted to a sigmoid Emax model relating stimulation temperature and pain intensity. The T50 (temperature inducing a pain intensity of 50 on a visual analogue scale from 0 to 100) and the steepness of the curve were compared between the different analgesic drug concentrations and placebo. Remifentanil dose-dependently increased the T50 from 46.1°C at baseline to 48.4 and 49.1ºC without changing the steepness of the curve. Dexmedetomidine increased the T50 to 47.2°C and decreased the steepness of the curve. There was no difference in the pain responses between baseline and after recovery. As expected, dexmedetomidine was not as effective an analgesic as remifentanil. The difference in the quality of analgesia with remifentanil may be a reflection of a different mechanism of action or a consequence of the sedative effect of dexmedetomidine.
'New' analgesies
145
Fig. 6.4 (a) Observer’s assessment of alertness/sedation sum scores (mean ± standard deviation) at baseline and at different remifentanil (REMI) and dexmedetomidine (DEX) concentrations. The scale ranges from
The year in anaesthesia and critical care
146
9 to 20 (maximum sedation to maximum alertness). The target remifentanil concentrations were 1, 2, 3 and 4 ng/ml and the target dexmedetomidine concentrations were 0.6, 1.2,1.8 and 2.4 ng/ml, respectively. (b) Population predictions for all drug concentrations. The heat pain tolerance was assessed under the conditions of the different remifentanil and dexmedetomidine concentrations (see (a)). The parameters for the analgesic effect were the stimulation temperature producing a pain at a VAS score of 50 (T50) and the slope of the sigmoid temperature—pain curve (Hill coefficient γ). The temperature—pain curves for baseline and recovery (baseline—recovery) are plotted for all remifentanil steps and for all dexmedetomidine steps. The slope coeffient γ values for baseline— recovery and remifentanil were identical, but the γ value for dexmedetomidine was smaller, resulting in a flatter response during dexmedetomidine infusions when compared with remifentanil or baseline—recovery. The T50 values for baseline—recovery were significantly different for remifentanil 1 and remifentanil 2–4 and for dexmedetomidine 1 and dexmedetomidine 2. Source: Cortinez et al. (2004).
'New' analgesies
147
Comment These two studies, which were performed by the same investigators, provide essential pharmacological information on the use of dexmedetomidine in conscious sedation. Because of the strong sedative effect of dexmedetomidine only the low concentration range is accessible for measuring the analgesic effect with experimental pain models requiring cooperative subjects (Fig. 6.4a). The study illustrated the weak but detectable analgesic effect in comparison with remifentanil (Fig. 6.4b). Whereas remifentanil caused a dose-dependent respiratory depression with an increased risk of apnoea and hypopnoea and a decrease in the respiratory response to hypercapnia, dexmedetomidine only induced a small respiratory depression with a ceiling effect already seen at the lowest concentration and no risk of apnoea and hypopnoea despite strong sedation. Haemodynamic stability was maintained with both drugs with a small increase in blood pressure and a decrease in the heart rate observed with dexemedetomidine. These results may illustrate why dexmedetomidine has been successfully used for sedation during carotid endarterectomy under local anaesthesia |74| and during fibreoptic intubation |75|. They show that further clinical trials on the use of dexmedetomidine beyond sedation in ICUs are warranted. Conclusion Opioid-induced respiratory depression and other side effects are a concern in postoperative analgesia and require adequate monitoring. Multi-modal pain therapy with combinations of opioids and other analgesics reduces opioid side effects, so that the reduction in opioid requirement is frequently used as a parameter for their efficacy. Because they increase the incidence of myocardial infarction, particularly in the population at risk, COX-2 inhibitors should only be used after a cautious individual risk—benefit consideration. They cannot be recommended in the elderly population any more, such as patients undergoing hip or knee arthroplasty. Because the opioid-sparing effect of gabapentin and pregabalin is modest, the utility of these drugs in the perioperative setting may be in the prevention of chronic pain. This has not been formally proven yet. M6G may have the potential for replacing morphine in post-operative pain management. In this setting the slow equilibration between plasma and effect might be less important than in anaesthesia for surgery. The genetic polymorphism in the µ-opioid receptor may make the efficacy unpredictable in daily clinical practice because of a high number of ‘non-responders’. The interesting pharmacodynamic profile of dexmedetomidine with strong sedation, modest analgesia and only weak respiratory depression makes it a promising drug for conscious sedation or sedation in the ICU.
The year in anaesthesia and critical care
148
References 1. Bombardier C, Laine L, Reicin A, Shapiro D, Burgos-Vargas R, Davis B, Day R, Ferraz MB, Hawkey CJ, Hochberg MC, Kvien TK, Schnitzer TJ. Comparison of upper gastrointestinal toxicity of rofecoxib and naproxen in patients with rheumatoid arthritis. VIGOR Study Group. N Engl J Med 2000; 343: 1520–8. 2. Fitzgerald GA. Coxibs and cardiovascular disease. N Engl J Med 2004; 351: 1709–11. 3. Hegi TR, Bombeli T, Seifert B, Baumann PC, Haller U, Zalunardo MP, Pasch T, Spahn DR. Effect of rofecoxib on platelet aggregation and blood loss in gynaecological and breast surgery compared with diclofenac. Br J Anaesth 2004; 92: 523–31. 4. Blaicher AM, Landsteiner HT, Al-Falaki O, Zwerina J, Volf I, Gruber D, Zimpfer M, Hoerauf K. Acetylsalicylic acid, diclofenac, and lornoxicam, but not rofecoxib, affect platelet CD 62 expression. Anesth Analg 2004; 98: 1082–5. 5. Karamanlioglu B, Arar C, Alagol A, Colak A, Gemlik I, Sut N. Preoperative oral celecoxib versus preoperative oral rofecoxib for pain relief after thyroid surgery. Eur J Anaesthesiol 2003; 20: 490–5. 6. Karamanlioglu B, Turan A, Memis D, Ture M. Preoperative oral rofecoxib reduces postoperative pain and tramadol consumption in patients after abdominal hysterectomy. Anesth Analg 2004; 98: 1039–43. 7. Buvanendran A, Kroin JS, Tuman KJ, Lubenow TR, Elmofty D, Moric M, Rosenberg AG. Effects of perioperative administration of a selective cyclooxygenase 2 inhibitor on pain management and recovery of function after knee replacement: a randomized controlled trial. JAMA 2003; 290: 2411–18. 8. Walter MF, Jacob RF, Day CA, Dahlborg R, Weng Y, Mason RP. Sulfone cox-2-inhibitors increase susceptibility of human LDL and plasma to oxidative modification: comparison to sulfonamide cox-2-inhibitors and NSAIDs. Atherosclerosis 2004; 177: 235–43. 9. Silverstein FE, Faich G, Goldstein JL, Simon LS, Pincus T, Whelton A, Makuch R, Eisen G, Agrawal NM, Stenson WF, Burr AM, Zhao WW, Kent JD, Lefkowith JB, Verburg KM, Geis GS. Gastrointestinal toxicity with celecoxib vs non-steroidal anti-inflammatory drugs for osteoarthritis and rheumatoid arthritis: the CLASS study: A randomized controlled trial. Celecoxib Long-term Arthritis Safety Study. JAMA 2000; 284: 1247–55. 10. Fitzgerald GA. Cox-2 and beyond: approaches to prostaglandin inhibition in human disease. Nat Rev Drug Discov 2003; 2: 879–90. 11. Alcindor D, Krolikowski JG, Pagel PS, Warltier DC, Kersten JR. Cyclooxygenase-2 mediates ischemic, anesthetic, and pharmacologic preconditioning in vivo. Anesthesiology 2004; 100: 547–54. 12. Coats TL, Borenstein DG, Nangia NK, Brown MT. Effects of valdecoxib in the treatment of chronic low back pain: results of a randomized, placebo-controlled trial. Clin Ther 2004; 26: 1249–60. 13. Edwards JE, McQuay HJ, Moore RA. Efficacy and safety of valdecoxib for treatment of osteoarthritis and rheumatoid arthritis: systematic review of randomised controlled trials. Pain 2004; 111: 286–96. 14. Gan TJ, Joshi GP, Zhao SZ, Hanna DB, Cheung RY, Chen C. Presurgical intravenous parecoxib sodium and follow-up oral valdecoxib for pain management after laparoscopic cholecystectomy surgery reduces opioid requirements and opioid-related adverse effects. Acta Anaesthesiol Scand 2004; 48: 1194–207. 15. Gan TJ, Joshi GP, Viscusi E, Cheung RY, Dodge W, Fort JG, Chen C. Preoperative parenteral parecoxib and follow-up oral valdecoxib reduce length of stay and improve quality of patient recovery after laparoscopic cholecystectomy surgery. Anesth Analg 2004; 98: 1665–73. 16. Christensen KS, Cawkwell GD. Valdecoxib versus rofecoxib in acute post-surgical pain: results of a randomized controlled trial. J Pain Symptom Manage 2004; 27: 460–70.
'New' analgesies
149
17. Romsing J, Moiniche S. A systematic review of cox-2-inhibitors compared with traditional NSAIDs, or different cox-2-inhibitors for post-operative pain. Acta Anaesthesiol Scand 2004; 48: 525–46. 18. White WB, Strand V, Roberts R, Whelton A. Effects of the cyclooxygenase-2 specific inhibitor valdecoxib versus non-steroidal antiinflammatory agents and placebo on cardiovascular thrombotic events in patients with arthritis. Am J Ther 2004; 11: 244–50. 19. Noveck RJ, Hubbard RC. Parecoxib sodium, an injectable cox-2-specific inhibitor, does not affect unfractionated heparin-regulated blood coagulation parameters. J Clin Pharmacol 2004; 44: 474–80. 20. Leese PT, Recker DP, Kent JD. The cox-2 selective inhibitor, valdecoxib, does not impair platelet function in the elderly: results of a randomized controlled trial. J Clin Pharmacol 2003; 43: 504–13. 21. Ibrahim AE, Feldman J, Karim A, Kharasch ED. Simultaneous assessment of drug interactions with low- and high-extraction opioids: application to parecoxib effects on the pharmacokinetics and pharmacodynamics of fentanyl and alfentanil. Anesthesiology 2003; 98: 853–61. 22. Ibrahim A, Karim A, Feldman J, Kharasch E. The influence of parecoxib, a parenteral cyclooxygenase-2 specific inhibitor, on the pharmacokinetics and clinical effects of midazolam. Anesth Analg 2002; 95: 667–73. 23. Ibrahim A, Park S, Feldman J, Karim A, Kharasch ED. Effects of parecoxib, a parenteral cox-2specific inhibitor, on the pharmacokinetics and pharmacodynamics of propofol. Anesthesiology 2002; 96: 88–95. 24. Riendeau D, Percival MD, Brideau C, Charleson S, Dube D, Ethier D, Falgueyret JP, Friesen RW, Gordon R, Greig G, Guay J, Mancini J, Ouellet M, Wong E, Xu L, Boyce S, Visco D, Girard Y, Prasit P, Zamboni R, Rodger IW, Gresser M, Ford-Hutchinson AW, Young RN, Chan CC. Etoricoxib (MK-0663): preclinical profile and comparison with other agents that selectively inhibit cyclooxygenase-2. J Pharmacol Exp Ther 2001; 296: 558–66. 25. Agrawal NG, Porras AG, Matthews CZ, Rose MJ, Woolf EJ, Musser BJ, Dynder AL, Mazina KE, Lasseter KC, Hunt TL, Schwartz JI, McCrea JB, Gottesdiener KM. Single- and multipledose pharmacokinetics of etoricoxib, a selective inhibitor of cyclooxygenase-2, in man. J Clin Pharmacol 2003; 43: 268–76. 26. Agrawal NG, Rose MJ, Matthews CZ, Woolf EJ, Porras AG, Geer LA, Larson PJ, Cote J, Dilzer SC, Lasseter KC, Alam I, Petty KJ, Gottesdiener KM. Pharmacokinetics of etoricoxib in patients with hepatic impairment. J Clin Pharmacol 2003; 43: 1136–48. 27. Agrawal NG, Matthews CZ, Mazenko RS, Kline WF, Woolf EJ, Porras AG, Geer LA, Wong PH, Cho M, Cote J, Marbury TC, Moncrief JW, Alcorn Jr H, Swan S, Sack MR, Robson RA, Petty KJ, Schwartz JI, Gottesdiener KM. Pharmacokinetics of etoricoxib in patients with renal impairment. J Clin Pharmacol 2004; 44: 48–58. 28. Chang DJ, Desjardins PJ, King TR, Erb T, Geba GP. The analgesic efficacy of etoricoxib compared with oxycodone/acetaminophen in an acute postoperative pain model: a randomized, double-blind clinical trial. Anesth Analg 2004; 99: 807–15. 29. Tannenbaum H, Berenbaum F, Reginster JY, Zacher J, Robinson J, Poor G, Bliddal H, Uebelhart D, Adami S, Navarro F, Lee A, Moore A, Gimona A. Lumiracoxib is effective in the treatment of osteoarthritis of the knee: a 13-week, randomised, double-blind study versus placebo and celecoxib. Ann Rheum Dis 2004; 63: 1419–26. 30. Schnitzer TJ, Burmester GR, Mysler E, Hochberg MC, Doherty M, Ehrsam E, Gitton X, Krammer G, Mellein B, Matchaba P, Gimona A, Hawkey CJ. Comparison of lumiracoxib with naproxen and ibuprofen in the Therapeutic Arthritis Research and Gastrointestinal Event Trial (TARGET), reduction in ulcer complications: randomised controlled trial. Lancet 2004; 364: 665–74. 31. Farkouh ME, Kirshner H, Harrington RA, Ruland S, Verheugt FW, Schnitzer TJ, Burmester GR, Mysler E, Hochberg MC, Doherty M, Ehrsam E, Gitton X, Krammer G, Mellein B, Gimona A, Matchaba P, Hawkey CJ, Chesebro JH. Comparison of lumiracoxib with naproxen
The year in anaesthesia and critical care
150
and ibuprofen in the Therapeutic Arthritis Research and Gastrointestinal Event Trial (TARGET), cardiovascular outcomes: randomised controlled trial. Lancet 2004; 364: 675–84. 32. Caraceni A, Zecca E, Bonezzi C, Arcuri E, Tur RY, Maltoni M, Visentin M, Gorni G, Martini C, Tirelli W, Barbieri M, De CF. Gabapentin for neuropathic cancer pain: a randomized controlled trial from the Gabapentin Cancer Pain Study Group. J Clin Oncol 2004; 22: 2909–17. 33. Bone M, Critchley P, Buggy DJ. Gabapentin in post-amputation phantom limb pain: a randomized, double-blind, placebo-controlled, cross-over study. Reg Anesth Pain Med 2002; 27: 481–6. 34. Cheshire Jr WP. Defining the role for gabapentin in the treatment of trigeminal neuralgia: a retrospective study. J Pain 2002; 3: 137–42. 35. Rowbotham M, Harden N, Stacey B, Bernstein P, Magnus-Miller L. Gabapentin for the treatment of post-herpetic neuralgia: a randomized controlled trial. JAMA 1998; 280: 1837–42. 36. Levendoglu F, Ogun CO, Ozerbil O, Ogun TC, Ugurlu H. Gabapentin is a first line drug for the treatment of neuropathic pain in spinal cord injury. Spine 2004; 29: 743–51. 37. McClelland D, Evans RM, Barkworth L, Martin DJ, Scott RH. A study comparing the actions of gabapentin and pregabalin on the electrophysiological properties of cultured DRG neurones from neonatal rats. BMC Pharmacol 2004; 4: 14. 38. Gottrup H, Juhl G, Kristensen AD, Lai R, Chizh BA, Brown J, Bach FW, Jensen TS. Chronic oral gabapentin reduces elements of central sensitization in human experimental hyperalgesia. Anesthesiology 2004; 101: 1400–8. 39. Dahl JB, Mathiesen O, Moiniche S. ‘Protective premedication’: an option with gabapentin and related drugs? A review of gabapentin and pregabalin in the treatment of post-operative pain. Acta Anaesthesiol Scand 2004; 48: 1130–6. 40. Carlton SM, Zhou S. Attenuation of formalin-induced nociceptive behaviors following local peripheral injection of gabapentin. Pain 1998; 76: 201–7. 41. Rogawski MA, Loscher W. The neurobiology of antiepileptic drugs for the treatment of nonepileptic conditions. Nat Med 2004; 10: 685–92. 42. Fehrenbacher JC, Taylor CP, Vasko MR. Pregabalin and gabapentin reduce release of substance P and CGRP from rat spinal tissues only after inflammation or activation of protein kinase C. Pain 2003; 105: 133–41. 43. Cheng JK, Lai YJ, Chen CC, Cheng CR, Chiou LC. Magnesium chloride and ruthenium red attenuate the antiallodynic effect of intrathecal gabapentin in a rat model of post-operative pain. Anesthesiology 2003; 98: 1472–9. 44. Da Motta PG, Veiga AP, Francischi JN, Tatsuo MA. Evidence for participation of GABA(A) receptors in a rat model of secondary hyperalgesia. Eur J Pharmacol 2004; 483: 233–9. 45. Bertrand S, Ng GY, Purisai MG, Wolfe SE, Severidt MW, Nouel D, Robitaille R, Low MJ, O’Neill GP, Metters K, Lacaille JC, Chronwall BM, Morris SJ. The anticonvulsant, antihyperalgesic agent gabapentin is an agonist at brain gamma-aminobutyric acid type B receptors negatively coupled to voltage-dependent calcium channels. J Pharmacol Exp Ther 2001; 298: 15–24. 46. Bertrand S, Nouel D, Morin F, Nagy F, Lacaille JC. Gabapentin actions on Kir3 currents and N-type Ca2+ channels via GABAB receptors in hippocampal pyramidal cells. Synapse 2003; 50: 95–109. 47. Bertrand S, Morin F, Lacaille JC. Different actions of gabapentin and baclofen in hippocampus from weaver mice. Hippocampus 2003; 13: 525–8. 48. Cheng JK, Lee SZ, Yang JR, Wang CH, Liao YY, Chen CC, Chiou LC. Does gabapentin act as an agonist at native GABA(B) receptors? J Biomed Sci 2004; 11: 346–55. 49. Hansen C, Gilron I, Hong M. The effects of intrathecal gabapentin on spinal morphine tolerance in the rat tail-flick and paw pressure tests. Anesth Analg 2004; 99: 1180–4. 50. Eckhardt K, Ammon S, Hofmann U, Riebe A, Gugeler N, Mikus G. Gabapentin enhances the analgesic effect of morphine in healthy volunteers. Anesth Analg 2000; 91: 185–91.
'New' analgesies
151
51. Dirks J, Petersen KL, Rowbotham MC, Dahl JB. Gabapentin suppresses cutaneous hyperalgesia following heat-capsaicin sensitization. Anesthesiology 2002; 97: 102–7. 52. Turan A, Karamanlioglu B, Memis D, Hamamcioglu MK, Tukenmez B, Pamukcu Z, Kurt I. Analgesic effects of gabapentin after spinal surgery. Anesthesiology 2004; 100: 935–8. 53. Pandey CK, Sahay S, Gupta D, Ambesh SP, Singh RB, Raza M, Singh U, Singh PK. Preemptive gabapentin decreases post-operative pain after lumbar discoidectomy (L’administration preventive de gabapentine diminue la douleur post-opératoire d’une discectomie lombaire). Can J Anaesth 2004; 51: 986–9. 54. Turan A, Karamanlioglu B, Memis D, Usar P, Pamukcu Z, Ture M. The analgesic effects of gabapentin after total abdominal hysterectomy. Anesth Analg 2004; 98: 1370–3. 55. Rorarius MG, Mennander S, Suominen P, Rintala S, Puura A, Pirhonen R, Salmelin R, Haanpaa M, Kujansuu E, Yli-Hankala A. Gabapentin for the prevention of post-operative pain after vaginal hysterectomy. Pain 2004; 110: 175–81. 56. Turan A, Memis D, Karamanlioglu B, Yagiz R, Pamukcu Z, Yavuz E. The analgesic effects of gabapentin in monitored anesthesia care for ear-nose-throat surgery. Anesth Analg 2004; 99: 375–8. 57. Hill CM, Balkenohl M, Thomas DW, Walker R, Mathe H, Murray G. Pregabalin in patients with post-operative dental pain. Eur J Pain 2001; 5: 119–24. 58. Ben-Menachem E. Pregabalin pharmacology and its relevance to clinical practice. Epilepsia 2004; 45(Suppl 6): 13–18. 59. Chen SR, Xu Z, Pan HL. Stereospecific effect of pregabalin on ectopic afferent discharges and neuropathic pain induced by sciatic nerve ligation in rats. Anesthesiology 2001; 95: 1473–9. 60. Cunningham MO, Woodhall GL, Thompson SE, Dooley DJ, Jones RS. Dual effects of gabapentin and pregabalin on glutamate release at rat entorhinal synapses in vitro. Eur J Neurosci 2004; 20: 1566–76. 61. Rosenstock J, Tuchman M, LaMoreaux L, Sharma U. Pregabalin for the treatment of painful diabetic peripheral neuropathy: a double-blind, placebo-controlled trial. Pain 2004; 110: 628– 38. 62. Sabatowski R, Galvez R, Cherry DA, Jacquot F, Vincent E, Maisonobe P, Versavel M. Pregabalin reduces pain and improves sleep and mood disturbances in patients with postherpetic neuralgia: results of a randomised, placebo-controlled clinical trial. Pain 2004; 109: 26–35. 63. Romberg R, Olofsen E, Sarton E, Den HJ, Taschner PE, Dahan A. PharmacokineticPharmacodynamic modeling of morphine-6-glucuronide-induced analgesia in healthy volunteers: absence of sex differences. Anesthesiology 2004; 100: 120–33. 64. Skarke C, Darimont J, Schmidt H, Geisslinger G, Lotsch J. Analgesic effects of morphine and morphine-6-glucuronide in a transcutaneous electrical pain model in healthy volunteers. Clin Pharmacol Ther 2003; 73: 107–21. 65. Buetler TM, Wilder-Smith OH, Wilder-Smith CH, Aebi S, Cerny T, Brenneisen R. Analgesic action of i.v. morphine-6-glucuronide in healthy volunteers. Br J Anaesth 2000; 84: 97–9. 66. Tegeder I, Meier S, Burian M, Schmidt H, Geisslinger G, Lotsch J. Peripheral opioid analgesia in experimental human pain models. Brain 2003; 126: 1092–102. 67. Sarton E, Olofsen E, Romberg R, Den HJ, Kest B, Nieuwenhuijs D, Burm A, Teppema L, Dahan A. Sex differences in morphine analgesia: an experimental study in healthy volunteers. Anesthesiology 2000; 93: 1245–54. 68. Romberg R, Olofsen E, Sarton E, Teppema L, Dahan A. Pharmacodynamic effect of morphine6-glucuronide versus morphine on hypoxic and hypercapnic breathing in healthy volunteers. Anesthesiology 2003; 99: 788–98. 69. Martin E, Ramsay G, Mantz J, Sum-Ping ST. The role of the alpha2-adrenoceptor agonist dexmedetomidine in post-surgical sedation in the intensive care unit. J Intensive Care Med 2003; 18: 29–41.
The year in anaesthesia and critical care
152
70. Herr DL, Sum-Ping ST, England M. ICU sedation after coronary artery bypass graft surgery: dexmedetomidine-based versus propofol-based sedation regimens. J Cardiothorac Vasc Anesth 2003; 17: 576–84. 71. Tobias JD, Berkenbosch JW. Sedation during mechanical ventilation in infants and children: dexmedetomidine versus midazolam. South Med J 2004; 97: 451–5. 72. Angst MS, Ramaswamy B, Davies ME, Maze M. Comparative analgesic and mental effects of increasing plasma concentrations of dexmedetomidine and alfentanil in humans. Anesthesiology 2004; 101: 744–52. 73. Arain SR, Ruehlow RM, Uhrich TD, Ebert TJ. The efficacy of dexmedetomidine versus morphine for post-operative analgesia after major inpatient surgery. Anesth Analg 2004; 98: 153–8. 74. Doufas AG, Lin CM, Suleman MI, Liem EB, Lenhardt R, Morioka N, Akca O, Shah YM, Bjorksten AR, Sessler DI. Dexmedetomidine and meperidine additively reduce the shivering threshold in humans. Stroke 2003; 34: 1218–23. 75. Memis D, Turan A, Karamanlioglu B, Pamukcu Z, Kurt I. Adding dexmedetomidine to lidocaine for intravenous regional anesthesia. Anesth Analg 2004; 98: 835–40.
7 Pharmacogenetics and anaesthesia RICHARD SMILEY Introduction It has been understood for centuries that susceptibility to and manifestations of disease differ greatly between individuals and ethnic groups. In the modern era it has become clear that the response to and complications of drug therapy also vary from individual to individual, often with obvious familial or ethnic influences. Specific genes determine much of the variability in disease susceptibility and pharmacological response. Genetic influences are relatively easy to appreciate and evaluate in syndromes or diseases caused by a single, well-defined genetic alteration (e.g. sickle cell disease, phenylketonuria and haemophilia A), but multiple genetic influences affect the susceptibility to most diseases, modulating the effect of infectious, inflammatory, nutritional and other environmental stimuli. These more subtle genetic influences on a disease and its treatment will receive increasing attention over the next decades as information from the Human Genome Project finds its way into clinical applications. Perhaps even more promising than the study of how genetics cause or influence complex diseases is the study of how genetic variation affects the individual’s response to therapeutic agents, since these medications often work through very specific biochemical pathways that may be significantly altered by minor genetic variations. Pharmacogenetics is the field of study investigating the differences in drug responses between individuals based on genetics. With the explosion of information about the humane genome provided by the Human Genome Project in the 1990s, the term pharmacogenomics arose to describe more or less the same area of interest and investigation, referring more to genome-wide investigations rather than single-gene studies, although the terms are often used interchangeably |1|. Single-nucleotide polymorphisms Single-nucleotide polymorphisms (SNPs or ‘snips’) are the most common and most studied genetic alterations with implications for drug responses. SNPs are defined as natural mutations of one nucleotide occurring at a frequency of greater than 1% in © Atlas Medical Publishing Ltd
the human population. SNPs may (and often do) change the amino acid sequence of the coded protein and such protein changes may or may not have functional consequences. There are well over 1 million such SNPs known in the genome |2|. The notation usually used for SNPs designates the two possible nucleotides (or amino acids) surrounding the
The year in anaesthesia and critical care
154
number of the base pair or amino acid position in the protein. Thus, the β1-adrenergic receptor has a common polymorphism at amino acid number 389, where either arginine or glycine may exist. This would be designated Arg389Gly. Much pharmacogenetic research over the past decade has been aimed at determining which of these SNPs have significant effects on the efficacy or side effects of the medications used for treating common diseases. Drug effects maybe altered by differences in metabolism that have a genetic basis, leading to vastly differing concentrations of active drug in patients treated with the same dose. Genetic differences may also result in alterations in the targets (receptors, enzymes, etc.) leading to vastly different sensitivities to the same concentrations of drug. Similar genetic influences are seen in the side effects and complications of drug administration. Depending on the specific drug, it has been estimated that between 20 and 95% of pharmacological variability between patients has a genetic basis |1|. Anaesthesiology and pharmacogenetics While anaesthetists are obviously involved daily in drug administration and the monitoring of drug responses and effects, the field of pharmacogenetics may appear rather remote from the daily practice and professional interest of anaesthetists in 2004. In fact, however, anaesthesiology is the field in which one of the first pharmacogenetic ‘diseases’ was discovered. In the 1950s Kalow described patients who had delayed recovery of neuromuscular function after administration of succinylcholine |3|. The issue of plasma cholinesterase activity is so ingrained in anaesthesiology training that it probably does not even appear to be part of the ‘new-fangled’ discipline of pharmacogenetics to most anaesthetists, but this finding in the 1950s, along with the discovery of variability in the hepatic metabolism of some common drugs |4|, were the seeds of the modern concern with specific genetic variation and drug responses. Soon after, Denborough described what was to be called malignant hyperthermia, another early ‘pharmacogenetic disease’ |5, 6|. The ‘pseudo-cholinesterase deficiency’ spoken of in the clinical anaesthesiology environment is actually due to a variety of alternative genes coding for the plasma cholinesterase enzyme with altered levels and limited or almost absent activity in the hydrolysis of succinylcholine (see Girard and Kindler below). Many of the genetic variants leading to altered, decreased or almost absent succinylcholine hydrolysis have been described |7|, demonstrating the power of genetic testing for predicting a response, at least with certain drugs. The issue of whether such testing is worth performing is of course open to question, but certainly for families with an index case one could make a good argument for knowing the genotypes in the other family members. It would appear that over the next decade genotyping will (or should) replace the classic ‘dibucaine number’ and other phenotypic tests in determining a likely response or explaining an atypical response to succinylcholine or mivacurium |8|.
Pharmacogenetics and anaesthesia
155
Single-nucleotide polymorphisms and responses during anaesthesia There are multiple SNPs currently undergoing investigation in order to evaluate their role in drug responses or metabolism. One common drug target of interest to anaesthetists and intensivists is the β2-adrenoceptor. The human β2-adrenoceptor is the most well studied of the many G-protein-coupled cell surface receptors and is of course a target used in the therapy of hypertension (and hypotension), ischaemic, valvular and other cardiac diseases, asthma and preterm labour. The β2-adrenoceptor is encoded on chromosome 5 and its genetic variability has been well characterized with ten different point mutations identified. Several SNPs with functional consequences have been identified. Polymorphisms of the β2-adrenoceptor gene encoding amino acid substitutions 16 and 27 have received much attention due to their high prevalence and have been evaluated as causal factors for or modifiers of several pathological conditions including asthma, allergy, hypertension, heart failure and cystic fibrosis, all diseases in which β2adrenoceptor or β2-adrenoceptor agonist or antagonist therapy is commonly used. These two SNPs occur with high prevalence and affect the rate at which the receptor downregulates in response to agonists. The substitution of a glycine residue for an arginine residue at amino acid position 16 in the receptor protein results in an enhanced rate of receptor downregulation, The substitution of a glutamine residue for a glutamic acid at amino acid position 27 results in a decreased rate of receptor downregulation (Table 7.1) |9|. The incidence and effects of the haplotypes (combinations of SNPs) of the receptor involving the major functional polymorphisms have recently been described, including their approximate distributions within various ethnic groups |10|.
Table 7.1 Polymorphisms of the β2-adrenoceptor Nucleotide position
Amino acid position
‘Wild-type’ amino acid
‘Mutant’ amino acid (%)
Mutant phenotype
−47
19 (5′−LC)
Arg
Cys (63%)
increased expression
46
16
Arg
Gly (61%)
increased desensitization
79
27
Gln
Glu (43%)
decreased desensitization
100
34
Val
Met (<1%)
same as wildtype
491
164
Thr
lle (2−5%)
decreased coupling
Source: Liggett (1999) |11|.
These β2-adrenoceptor variants have been shown to influence drug responses to β2adrenoceptor agonists in asthma |12|, cardiovascular disease (CVD) |13, 14| and preterm delivery |15, 16|. Theoretically, these pharmacological differences result from different rates of or extent of desensitization or tachyphylaxis during therapy, β-agonists are frequently used in clinical anaesthesia and critical care and one may expect many studies
The year in anaesthesia and critical care
156
in the near future to look at the effects of β2-adrenoceptor genetics on responses to inotropic and similar drugs. Few studies have actually appeared thus far in the anaesthesia literature. Kim et al. examined whether β2-adrenoceptor gene polymorphisms were associated with the pressor response to laryngoscopy and tracheal intubation, i.e. did patients with a particular genotype show more of a tendency to raise their blood pressure or heart rate |17|. The arterial systolic pressure, heart rate and rate pressure product of 92 patients undergoing intubation under standardized induction conditions (thiopental-succinylcholine) were measured before induction of anaesthesia and 1 min following laryngoscopy. Patients who possessed the glutamic acid homozygote of the β2adrenoceptor at amino acid 27 produced significantly greater changes in their mean arterial pressure and rate pressure products than patients with the glutamine homozygote of the β2-adrenoceptor at amino acid 27. These findings suggest that genetic variability in the human β2-adrenoceptor gene polymorphisms maybe associated with the pressor response to laryngoscopy and tracheal intubation and presumably the cardiovascular response to other anaesthetic and surgical interventions. We have presented preliminary data in abstract form investigating whether polymorphisms of the β2-adrenoceptor or the nitric oxide synthase enzyme might affect the cardiovascular response to spinal anaesthesia for Caesarean section and/or the response to the usual drugs used for treating hypotension during the anaesthetic |18|. The variant endothelial nitric oxide synthase enzyme with an aspartate for glutamate substitution at amino acid position 298 has been associated with lower nitric oxide levels and hypertension and related diseases |19|. Our preliminary data suggest that its presence may be associated with less hypotension and/or a better response to pressor treatment than the ‘normal’ variant. Results such as the above genetic effects on the blood pressure and heart rate must be viewed as preliminary and their effect on clinical care is questionable unless and until the effects are shown to be of a clinically significant magnitude and genetic testing becomes routine: as anyone who reads newspapers knows, that day may be closer than many perhaps once thought. Current areas and direction There is continuing interest in the two classic pharmacogenetic ‘anaesthesia diseases’, malignant hyperthermia and plasma cholinesterase ‘deficiency’. In the area of malignant hyperthermia, defining the specific genetics of the syndrome may aid in a more specific prediction and diagnosis, but may also help us to understand better the underlying cause, since a genetic change altering a protein should yield hints as to mechanism. Anaesthetists have always been interested in genetically determined alterations in drug metabolism, and genetic variations in the cytochrome P-450 family of enzymes continue and there will no doubt be other drug-metabolising enzymes found to possess polymorphic sites with functional consequences. Over the past decade multiple polymorphisms have been detected and studied in drug ‘targets’, receptors and enzymes, mostly those involved in the action of cardiovascular drugs. These drugs are of course also used in perioperative management, so investigation of SNPs in the adrenergic receptors, other G-protein-coupled receptors, including the opioid receptors and enzymes such as endothelial nitric oxide synthase and angiotensin-converting enzymes (ACE) are receiving some interest and may have clinical implications for anaesthetists in the years
Pharmacogenetics and anaesthesia
157
ahead. Finally, although not exactly fitting the classic definition of pharmacogenetics, the field of ‘surgical genomics’ or ‘perioperative genomics’ is emerging |20–22| and is concerned with how genetics affect surgical risk and outcome. Genetic screening may ultimately play a significant role in recommendations for surgery versus medical (versus no) treatment for some conditions and genetic profiles will almost certainly be used for making recommendations for perioperative therapy (e.g. β-blockade and anti-thrombotic treatment). As befits a field still very much in development, in addition to original research studies, this section also includes a few review articles that survey the current state of relevance of genetics to anaesthesia, pain and perioperative medicine.
Pharmacogenetics and anaesthesiology Girard T, Kindler CH. Curr Pharmacogenom 2004; 2: 119–35 BACKGROUND. Anaesthesiology has been concerned with pharmacogenetics since the 1950s and 1960s, starting with the recognition of plasma cholinesterase genetic variants and continuing with the recognition of malignant hyperthermia as a genetically determined syndrome. Today, pharmacogenetics in anaesthesiology includes interest in and investigation of genetic influences on drug metabolism and responses to a wide variety of medications, including benzodiazepines, volatile anaesthetics and adrenergic agents. In addition, the response to pain and its treatment may be significantly affected by genetics. INTERPRETATION. In this review the authors discuss the genetics of butyrylcholinesterase or ‘plasma cholinesterase’ or ‘pseudo-cholinesterase’ and the response to suxamethonium (succinylcholine), summarizing the history of the development of this field from clinical observation to detailed genetic analysis of the many variants of the enzyme that result in altered clinical responses. There is a somewhat shorter discussion of the importance of genetic variability in the cytochrome P-450 family of drug-metabolising enzymes. The role of the ryanodine receptor as the site of the ‘defect’ in malignant hyperthermia and the complexity of its genetic variability is discussed. More speculatively, the possible importance to anaesthesiology of known functional genetic variants in drug targets including β-adrenergic receptors, ion channels and ACE is presented. The possibility that genetic effects on inflammation and other perioperative physiology may have significant effects on surgical outcomes is another present and future area of research. Comment This review is an excellent starting place for the clinician who would like to become more familiar with the pharmacogenetics field and how it may impact on anaesthetic practice over the next two decades, The authors point out that phenotypic (biochemical) methods such as dibucaine or fluoride inhibition may often be imprecise in identifying the genotype in ‘pseudo-cholinesterase deficiency’ |7|. Twenty-three SNPs of butyrylcholinesterase are listed, seven of which are associated with altered activity. The authors suggest that, as genetic testing becomes more practical (i.e. inexpensive and
The year in anaesthesia and critical care
158
available), it will start to become routine for the families of suspected butyrylcholinesterase-deficiency patients or perhaps for everyone. This seems unlikely if the only reason is to screen for abnormal responses to succinylcholine, which are relatively rarely clinically significant, particularly since moderate prolongation of succinylcholine action has few clinical consequences (unless, of course, one cannot intubate or ventilate the patient). Alternatively, if a new drug, perhaps a new muscle relaxant, is developed with beneficial properties making it popular, but which is metabolised by butyrylcholinesterase (such as mivacurium) and the genetic effects on its metabolism are clinically significant, it could spur the need for genetic testing. Forty-four genetic variants of the ryanadine receptor associated with malignant hyperthermia or central core disease are listed, illustrating the potential complexity of this syndrome and the difficulties in using genetics in this area. SNPs in the µ-opioid receptor are mentioned: this will be discussed in relation to several specific studies below (vide infra). SNPs of the β1-adrenoceptor and β2-adrenoceptor are discussed. As mentioned above in relation to the β2-adrenoceptor, several of these variants have effects on the blood pressure and vascular |23| and perhaps heart rate responses. The decrease in blood pressure in response to β-blockade is influenced by genetic variants of the β1adrenoceptor. Patients homozygous for Arg389 had a much greater decrease in blood pressure in response to metoprolol and those homozygous for serine at amino acid 49 and arginine at amino acid 389 had increased responses compared to patients who were heterozygous at the sites. Overall, Gly389 homozygotes show a decreased response to βblockade |24, 25|. With evidence mounting of differential β-blockade effects on the basis of genetics and with the increasing widespread recommendation and use of β-blockade perioperatively, anaesthetists and/or surgeons may need to investigate whether some patients benefit more from perioperative β-blockade than others on the basis of the genetics of the receptor. It appears a bit early to tell how much effect these genetic variants will have on acute responses during anaesthesia or in the critical care unit.
Watson and Crick 50 years on; from double helix to pharmacogenomics Sweeney BP. Anaesthesia 2004; 59: 150–65 BACKGROUND. It has been only half a century since the description of the structure of DNA. In the past decade, genome-related science has begun to impact on all areas of medical care and anaesthesiology and critical care are and will be included in this transformation. Increasing information and understanding of the effects of inter-individual genetic variability in disease risk, presentation and drug response will create opportunities, risks and ethical dilemmas over the next decades. INTERPRETATION. The authors provide an excellent basic introduction to molecular genetics for the anaesthetist, starting with the structure of DNA and progressing to a discussion of the organization of genes on chromosomes, how genes are mapped and the definitions of SNPs and pharmacogenomics. The most specific example of genetic variability relevant to anaesthesia that is discussed is the classic variability in
Pharmacogenetics and anaesthesia
159
P-450 enzyme subtypes. The possibility that variation in the genes for drug transporters and receptors will have clinical relevance is discussed, but few examples with clinical relevance are given, reflecting the current state of knowledge. The remainder of the review discussed polygenic illnesses such as diabetes and hypertension, in which genetics play an influential although perhaps not determinative role, as well as the potential uses of pharmacogenetics in drug development. The review is a bit short of specific examples of pharmacogenetic research and examples relevant to anaesthetists, but does provide a good introduction to the field. Comment The review is a bit short of specific examples of pharmacogenetic research and examples relevant to anaesthetists, but does provide a good introduction to the field for the clinician with an interest.
Anesthetic requirement is increased in redheads Liem EB, Lin C-M, Suleman M-I, et al. Anesthesiology 2004; 101: 279–83 BACKGROUND. Several variables are known to affect anaesthetic requirements in humans (e.g. age, body temperature and thyroid status). There is no defined human genotype that influences anaesthetic depth or effect. There have been anecdotal reports that people with red hair require more anaesthesia. Most people with red hair have a specific mutation of the melanocortin1-receptor. This study examined desflurane requirements in redheads versus non-redheads using a modification of the classic minimal alveolar concentration studies for determining the end-tidal concentration of desflurane required for preventing movement to a noxious electrical stimulus (100 Hz using 60–70 mA via needles inserted intradermally in both thighs). Ten redheads were compared to ten people with dark brown or black hair. INTERPRETATION. People with red hair did indeed require 19% more desflurane (6.2%) (95% confidence interval [Cl] 5.9–6.5) than the ‘controls’ (5.2%) (95% Cl 4.9– 5.5) for preventing movement in response to the stimulus. Comment This was a simple and very impressive study, demonstrating for the first time that clear differences in the response to inhalation anaesthesia in humans are genetically determined. Eight of the ten redheads studied had a desflurane requirement higher than the highest of the ten dark-haired controls, i.e. there was not that much overlap between groups. The clinical implications of these results would appear moderate, as most anaesthetists titrate the anaesthetic depth to patient response (or to some cerebral monitor) and would probably just end up giving a bit more ‘gas’ to redheads. However, the difference (19%) is enough to make clinicians at least keep this finding in mind when
The year in anaesthesia and critical care
160
taking care of redheads. Perhaps a bit more concern about awareness or light anaesthesia is warranted. More interesting than simply how much anaesthesia to give a patient, the work may have implications for the anaesthetic mechanism, since a difference in drug response based on genetics can be further investigated in order to determine the cause of the differential effect, which should shed some light on the overall mechanism of anaesthesia. As expected, nine of the ten redheads were found to have a variant melanocortin1-receptor gene, with the tenth being heterozygous for the R151C mutation and the consensus (non-redhead-type) gene. Five of the ten dark-haired patients were heterozygous, with one variant melanocortin1 -receptor gene and one ‘normal’ gene, with no obvious effect on their anaesthetic requirement of having one mutant allele. The melanocortin1-receptor gene is not heavily expressed in the central nervous system (CNS), so the mechanism by which it could affect anaesthetic action is unclear. The authors cited a study suggesting that some melanocortinreceptor variants increase the analgesic effect of κ-opioids. However, the increased κ-opioid effect was only seen in women and would be expected to work in the opposite direction, with decreased anaesthetic requirement in melanocortin1-receptor variants if endogenous κ-opioids worked better. The authors suggest a complex feedback system where a ‘dysfunctional’ melanocortin1 -receptor results in increased pituitary secretion of α-melanocytestimulating hormone). Cytochrome P-450 genetics Genetic variation plays a well-established role in drug metabolism, with obvious implications for drug effect. The cytochrome P-450 enzymes are responsible for the initial metabolism of many common anaesthetic, sedative and adjuvant drugs. Genetic polymorphisms are common in the cytochrome P-450 family and may affect the metabolism of midazolam by cytochrome P-450 3A5 and diazepam (cytochrome P-450 2C19), propofol (cytochrome P-450 2B6), ketamine (cytochrome P-450 2B6), warfarin (cytochrome P-450 2C9) and opioids (cytochrome P-450 2D6) (see Girard and Kindler above). Cytochrome P-450 2D6 is the most studied polymorphic cytochrome P-450 enzyme, with more than 70 alleles described. ‘Poor metabolisers’ have two mutant alleles, while ‘extensive metabolisers’ have two normal alleles and ‘ultra-rapid metabolisers’ have extra gene copies. It is well established that the clinical effect of codeine (which needs to be metabolised to morphine to be active) is closely tied to metaboliser status and that poor metabolisers have little or no response to codeine. Almost half the children in an American urban centre were found to have genotypes associated with reduced codeine bio availability |26|. At the other end of the metabolism spectrum, ultra-rapid metabolisers may not achieve appropriate drug levels of medications metabolised by cytochrome P-450 2D6 when administered the typical doses. This has been reported for the anti-emetics ondansetron and tropisetron, where chemotherapy patients who were ultra-rapid metabolisers suffered significantly more nausea and vomiting due to lower blood levels of the anti-emetics |27|. Although the percentage of ultra-rapid metabolisers is only 2–4% among European Caucasians, the frequency of ultra-rapid metabolisers is different in other ethnic groups, for example 7– 12% in the Mediterranean, 29% in Ethiopia and 21% in Saudi Arabia |26|. Cytochrome P-
Pharmacogenetics and anaesthesia
161
450 2D6 is also the main enzyme responsible for the metabolism of tramadol, a fairly new analgesic that acts via opioid receptors and other non-opioid mechanisms.
Genetic testing for enzymes of drug metabolism: does it have clinical utility for pain medicine at the present time? A structured review Fishbain DA, Fishbain D, Lewis J, et al. Pain Med 2004; 5: 81–93 BACKGROUND. Genetic tests for polymorphisms of the enzymes of drug metabolism (such as cytochrome P-450 2D6 above and cytochrome P-450 2D9) are now available and are sometimes being promoted by the drug and testing industry. However, it is not clear what the role of such testing is or should be in pain medicine or other clinical fields. For example, ‘poor metabolisers’ metabolise nortriptyline poorly, so the average daily dose should be approximately 50 mg/day. ‘Ultra-rapid metabolisers’ would need 500 mg/day to achieve the same blood levels. Although information about the genetics of drug-metabolising enzymes is extensive, similar information about drug transporter proteins is not and it is probably too early to say much about the possible role of genetic variability in transport to the drugs’ proposed sites of actions. INTERPRETATION. The major portion of the review consists of an extensive table listing over 100 drugs in 30 categories, with information about the clinical consequences of polymorphisms of the relevant metabolising enzyme. It is instructive to note that, of the 109 drugs listed, only 18 have a specific ‘consequence’ of genetic polymorphisms listed. This reflects the fact that we still do not know which enzymes are involved in the metabolism of many drugs and, when we do, the actual clinical consequences of even functional genetic variation in the enzyme have not been determined in most cases. The authors conclude with the arguments for and against the use of genetic testing now for the rational treatment of chronic pain. Comment Some arguments for the benefits of routine genetic testing include the presence of significant amounts of cytochrome P-450 2D6 in the CNS, the ease of genotyping compared to the use of therapeutic ‘cocktails’ or drug level monitoring for detecting patients with an unusual metabolism and the potential of genetic testing for prevent-ing some occasional severe drug interactions where medications that are cytochrome P-450 2D6 inhibitors are given to patients who are already poor metabolisers. There is some evidence suggesting that side effects could be decreased with genetic testing. There is an estimate that the yearly total healthcare cost from poor metaboliser adverse reactions to antidepressants in the USA is $420 million. The arguments against the routine use of genotyping include the fact that the metabolic pathways of many of the drugs of interest are not known and the effects or presence of many polymorphisms are not well described. There are no data to suggest that outcomes are improved with genetic testing and genetic testing is only likely to help with drugs that have a low therapeutic index. The lack (so far) of a known functional polymorphism of cytochrome P-450 3A4, which is involved in
The year in anaesthesia and critical care
162
the metabolism of many psychoactive and analgesic drugs, limits the potential of any genotyping programme. The authors’ final conclusion is not surprising: that the time has not come for genetic testing to be the routine but, in cases of adverse reactions or lack of treatment efficacy, consideration should be given to genotyping. This rather obvious ‘compromise’ is probably the current state of the field with regard to genotyping for chronic pain management and may reflect the general unsatisfactory nature of chronic pain therapy as much as the early stage we are at in genotyping for planning drug therapy.
Impact of CYP2D6 genotype on post-operative tramadol analgesia Stamer UM, Lehnen K, Hothker F, et al. Pain 2003; 105: 231–8 BACKGROUND. Genetic polymorphisms result in absent enzyme activity of cytochrome P-450 2D6 (poor metabolisers) in approximately 10% of the Caucasian population. Tramadol is a synthetic opioid. Its major mechanism of analgesia appears to be via inhibition of the reuptake of serotonin and noradrenaline. In addition, an O-desmethylation to (+) O-desmethyltramadol is catalysed by the cytochrome P-450 2D6 enzyme. This has significant activity at the µ-opioid receptor and contributes to the overall analgesic efficacy of the drug. This study investigates whether the poor metaboliser genotype has an impact on the response to tramadol analgesia in post-operative patients. Three hundred patients recovering from abdominal surgery were studied. Patients self-administered tramadol plus dipyrone for post-operative pain. The use of tramadol, need for rescue medication (piritramide), pain scores and patient satisfaction with pain treatment were assessed and compared between genotypes. INTERPRETATION. The hypothesis of this study was that the (+) Odesmethyltramadol metabolite contributes significantly to tramadol analgesia and, therefore, that patients who do not metabolise tramadol well (poor metabolisers) would receive less analgesia (or use more drug). The percentage of ‘non-responders’, defined as the need for rescue piritramide after the initial tramadol/dipyrone loading dose or a negative response (unsatisfied with analgesia) on the final questionnaire, was significantly higher in the poor metaboliser group (46.7%) compared with the ‘extensive metabolisers’ group (21.6%) (P=0.005). This does suggest that the metabolism of tramadol, on a genetic basis, has a significant effect on analgesia post-operatively. Comment Overall 241 subjects completed the study and were analysed, 35 of whom were poor metabolisers, which is approximately the expected percentage. There were nine patients with gene duplication, i.e. ultra-rapid metabolisers. These patients did not appear to respond differently to the extensive metabolisers and their results were combined with the extensive metabolisers for analysis. The subjects were classified as ‘responders’ or ‘non-
Pharmacogenetics and anaesthesia
163
responders’ by the need for rescue medication and patients’ satisfaction at the final interview was compared between the groups. Most outcome measures related to pain and analgesic use were significantly different between the groups, in the expected direction (e.g. more tramadol used in the poor metabolisers). The tramadol dose necessary for getting the patient comfortable initially post-surgery amounted to 108.2±56.9 and 144.7±22.6 mg (P<0.001) in extensive metabolisers and poor metabolisers, respectively. There were two interesting (but statistically non-significant) trends: after the initial recovery room period, 27% of the poor metabolisers and 12% of extensive metabolisers or ultra-rapid metabolisers required piritramide rescue and, in the final questionnaire, 10% of the poor metabolisers were not satisfied with the analgesia versus only 4% of the extensive metabolisers/ultra-rapid metabolisers. It should be noted that most of the differences that were seen between the groups pertain to the initial 3 h post-operatively: drug usage and rescue were fairly similar after the initial period, suggesting that further studies are needed to confirm this result and determine the mechanism and timing of any difference in analgesia further. This is the first study in surgical patients to demonstrate a difference in the analgesic efficacy of tramadol based on the cytochrome P-450 2D6 genotype. The degree of difference between the groups that was seen in the postoperative context of this study may not be of tremendous clinical significance, since most post-operative patients control at least part of their analgesic regimen, so will titrate the drugs to the effect. However, even with a patient-controlled analgesia protocol, poor metaboliser patients may be less satisfied. This suggests that, particularly if tramadol is being given on a fixed-dose schedule, the cytochrome P-450 2D6 genetics may have a real impact on analgesic response. µ-opioid receptor polymorphism As discussed in the Introduction, genetic variation leading to functional consequences occurs in the targets of drug therapy as well as in metabolic pathways. Inter-individual variability in pain perception and sensitivity to analgesic therapy has been long noted. Opioids are among the most widely used drugs for the management of acute and chronic pain, yet they display large inter-individual variability in efficacy, side effects and tolerance profiles. The major target of action of opioid analgesics such as morphine and fentanyl and the endogenous opiate peptides β-endorphin and enkephalin is the µ-opioid receptor, which is encoded by genetic locus OPRM1. The µ-opioid receptor is a member of the super-family of transmembrane G-protein-coupled receptors (along with adrenergic receptors and others) and a number of SNPs have been described for OPRM1. At nucleotide position 118, an adenine substitution by a guanine (A118G) results in the asparagine residue at amino acid position 40 being changed into an aspartate residue (Asn40Asp), which has been reported to occur at an allelic frequency of 10–25% in Caucasian populations |28–30|. In vitro, the A118G polymorphism appears to increase the binding affinity and potency of β-endorphin |28|. Thus, individuals carrying the variant receptor gene may show differences in some of the functions mediated by β-endorphin action at the altered µ-opioid receptor, with the possibility that G118 carriers would be better responders to opioid therapy |28|. There is significant inter-ethnic variability in the incidence of this variant receptor. In Asian populations, genotype distributions of several
The year in anaesthesia and critical care
164
µ-opioid receptor SNPs frequencies have been recently reported, with a G118 frequency in control subjects ranging between 35 and 47% (Chinese 35%, Thais 44%, Malays 45% and Indians 47%) |31| and the reported frequency among the Japanese being 49% |29|. Most of the studies of this polymorphism thus far have been epidemiological in nature, looking for associations of the variant receptor with opioid addiction and alcohol abuse |26, 32, 33|, rather than studies of analgesia response. There is limited evidence that this polymorphism may have functional, clinical consequences for pain tolerance or pain treatment response. The influence of the A118G polymorphism on the clinical response to morphine and morphine-6-glucuronide (M6G), an active metabolite of morphine, has been evaluated in vivo. Pupillary constriction, a measure of the central effects of opioids, was reduced in subjects with the A118G variant after the administration of M6G, but not with morphine administration |34|. These same authors investigated the risk for M6G accumulation and opioid overdose and toxicity in two patients with severe kidney failure genotyped for SNPs of the µ-opioid receptor |35|. One patient tolerated morphine treatment despite high plasma M6G and was found to be homozygous for the G118 allele. The other patient, homozygous for the wild-type allele A118, was found to have severely impaired vigilance during the first day of morphine treatment. The authors postulated that the G118 allele might be one of several genes or factors conferring protection against M6Grelated opioid side effects and perhaps toxicity. Recently, two studies have reported that individuals with an 118G allele (five heterozygotes and one homozygote) had an increased cortisol response to the administration of naloxone compared to A118A homozygotes, suggesting that there may indeed be functional consequences to this polymorphism |36, 37|. Lötsch et al. have recently published an extensive review of the possible roles of genetics in opioid analgesia |38|.
The 118 A→G polymorphism in the human µ-opioid receptor gene may increase morphine requirements in patients with pain caused by malignant disease Klepstad P, Rakvåg Π, Kaasa S, et al. Acta Anaesthiol Scand 2004; 48: 1232–9 BACKGROUND. The authors investigated whether the genotype of the µ-opioid receptor influenced the use of oral morphine in cancer patients. INTERPRETATION. Patients homozygous for the 118G polymorphism of the receptor used more morphine than those with the 118A genotype. No other polymorphism of the receptor affected morphine use. This suggests a lower efficacy of morphine in patients who were homozygous for the 118G allele, although it should be noted that there were only four subjects homozygous for 118G, so the sample was small. Comment The results of this study are somewhat surprising in that previous data have suggested that the protein with the 118G polymorphism binds M6G and β-endorphm better than the 118A receptor, so one might have expected morphine use to be less in 118G patients. The findings for morphine itself have generally been that there was no difference in its
Pharmacogenetics and anaesthesia
165
binding to the two variants of the receptor. The authors screened 207 Caucasian cancer patients receiving chronic oral morphine therapy. Only those ‘adequately treated’, as determined by a score of 4 or less on the Brief Pain Inventory, were included in the analysis of the effect of the 118G polymorphism on drug use. The authors reasoning for this study structure was that they wanted to study patients who were adequately treated and see how their doses compared between genotypes, rather than include patients who were simply receiving the ‘wrong’ dose and were therefore still in pain. Of the 207 patients, 99 were ‘adequately treated’ by the above criterion. Of these 78 patients were homozygous 118A, 17 were heterozygous and four were homozygous 118G. Although there were only four homozygous 118G patients, their morphine use was strikingly higher than the patients homozygous for the wild-type receptor (Table 7.2). This is the first study of clinical pain to demonstrate a difference in morphine efficacy based on receptor genotype. The study had several limitations, most of which were acknowledged by the authors. The number of 118G homozygotes was very small, which is a function of the gene incidence in Caucasian populations. As noted above, this variant is much more prevalent in nonCaucasian populations |29|. The selection of only 99 patients from 207 screened could have introduced bias. One would ideally like to see a ‘gene—dose’ effect, where the morphine use in the heterozygous patients was between those of the 118A homozygotes and the 118G homozygotes and this was not the case. The authors noted that the heterozygous patients had significantly higher pain scores than the others, so perhaps were under-treated, which could explain the lack of a ‘gene—dose’ effect.
Table 7.2 Pharmacological observations for 118A→118G genotype Wild-type (AA) (n=78)
Heterozygous (AG) (n=17)
Variant (GG) (n=4)
Morphine dose (mg 24h−1)
97±89
66±50
225±143*
Morphine serum conc. (nmol l−1)
71±67
52±46
117±92
M6G serum conc. (nmol l−1)
404±449
267±237
711±517
M3G serum conc. (nmol l−1)
2300±2166
1666±1462
3815±2729
All values are mean ± standard deviation. * P=0.006 for differences in morphine dose between groups (one way ANOVA). No other statistically significant differences. M6G, morphine 6-glucoronide; M3G, morphine 3-glucoronide. Source: Klepstad et al. (2004).
These patients were receiving chronic oral morphine therapy so, even if confirmed by other investigators, these results may not apply to intravenous administration in the acute post-operative setting. Despite the above limitations, this paper is still quite intriguing, since most experimental pain models and in vitro studies have not suggested that such a
The year in anaesthesia and critical care
166
clinically significant difference would be found in morphine efficacy in this kind of clinical scenario |28, 34, 39, 40|. Further work investigating the effective clinical doses of morphine and other opioids in various clinical scenarios over the next few years will clarify whether this polymorphism is really of any clinical significance. Conclusion The field of pharmacogenetics, in many ways founded by anaesthesiology, has been around for 50 years, but has clearly entered a new phase, with the explosion of information coming from the Human Genome Project and the techniques of modern genetic analysis. Some of the current claims for the potential of ‘genetic medicine’ in explaining all differences in drug response and in making predictions for selection and outcome are almost certainly too optimistic and much of the unpredictability in pharmacological responses may well remain, even after this next generation of pharmacogenetic investigations is completed. Examples of specific genetic variants with significant effects on drug responses during acute perioperative care are still very few. However, there will undoubtedly be some SNPs that have significant influences on specific drug responses and, once genetic testing becomes cheaper and routine, we may well find ourselves noting that individual patients might be better treated with drug ‘A’ for nausea (or pain or anticoagulation) than drug ‘B’. Perhaps more importantly for overall public health, we will likely find that specific genetic variants correlate with perioperative outcomes and complications, influencing the risk assessment for undergoing surgical procedures in the first place and for the strategies of intensive and post-operative care. Genetic factors as well as patient history and anatomy may well be part of routine pre-operative assessment in a decade or two. References 1. Evans WE, McLeod HL. Pharmacogenomics—drug disposition, drug targets, and side effects. N Engl J Med 2003; 348: 538–49. 2. Sachidanandam R, Weissman D, Schmidt SC, Kakol JM, Stein LD, Marth G, Sherry S, Mullikin JC, Mortimore BJ, Willey DL, Hunt SE, Cole CG, Coggill PC, Rice CM, Ning Z, Rogers J, Bentley DR, Kwok PY, Mardis ER, Yeh RT, Schultz B, Cook L, Davenport R, Dante M, Fulton L, Hillier L, Waterston RH, McPherson JD, Gilman B, Schaffner S, Van Etten WJ, Reich D, Higgins J, Daly MJ, Blumenstiel B, Baldwin J, Stange-Thomann N, Zody MC, Linton L, Lander ES, Altshuler D. A map of human genome sequence vari-ation containing 1.42 million single nucleotide polymorphisms. Nature 2001; 409: 928–33. 3. Kalow W. Familial incidence of low pseudocholinesterase level. Lancet 1956; 2: 576–7. 4. Evans DAP, Manley KA, McKusick VA. Genetic control of isoniazid metabolism in man. BMJ 1960; 5197: 485–91. 5. Denborough MA, Forster JF, Lovell RR, Maplestone PA, Villiers JD. Anaesthetic deaths in a family. Br J Anaesth 1962; 34: 395–6. 6. Nelson TE. Malignant hyperthermia: a pharmacogenetic disease of Ca++ regulating pro-teins. Curr Mol Med 2002; 2: 347–69. 7. Pantuck EJ. Plasma cholinesterase: gene and variations. Anesth Analg 1993; 77: 380–6.
Pharmacogenetics and anaesthesia
167
8. Ceppa F, Gidenne S, Benois A, Fontan E, Burnat P. Rapid identification of atypical variant of plasma butyrylcholinesterase by PCR. Clin Chem Lab Med 2002; 40: 799–801. 9. Liggett SB. β2-adrenergic receptor pharmacogenetics. Am J Respirat Crit Care Med 2000; 161: S197–201. 10. Drysdale CM, McGraw DW, Stack CB, Stephens JC, Judson RS, Nandabalan K, Arnold K, Ruano G, Liggett SB. Complex promoter and coding region β2-adrenergic receptor haplotypes alter receptor expression and predict in vivo responsiveness. Proc Natl Acad Sci USA 2000; 97: 10 483–8. 11. Liggett SB. Molecular and genetic basis of β2-adrenergic receptor function. J Allergy Clin Immunol 1999; 104: S42–6. 12. Liggett SB. The pharmacogenetics of β2-adrenergic receptors: relevance to asthma. J Allergy Clin Immunol 2000; 105: S487–92. 13. Cockcroft JR, Gazis AG, Cross DJ, Wheatley A, Dewar J, Hall IP, Noon JP. β2-adreno-ceptor polymorphism determines vascular reactivity in humans. Hypertension 2000; 36: 371–5. 14. McNamara DM, MacGowan GA, London B. Clinical importance of beta-adrenoceptor polymorphisms in cardiovascular disease. Am J Pharmacogenom 2002; 2: 73–8. 15. Landau R, Xie H-G, Dishy V, Stein CM, Wood AJJ, Emala CW, Smiley RM. β2 adrenergic receptor genotype and preterm delivery. Am J Obstet Gynecol 2002; 187: 1294–8. 16. Ozkur M, Dogulu F, Ozkur A, Gokmen B, Inaloz S, Aynacioglu A. Association of the Gln27Glu polymorphism of the beta-2-adrenergic receptor with preterm labor. Int J Gynecol Obstet 2002; 77: 209–15. 17. Kim NS, Lee IO, Lee MK, Lim SH, Choi YS, Kong MH. The effects of β2 adrenoceptor gene polymorphisms on pressor response during laryngoscopy and tracheal intubation. Anaesthesia 2002; 57: 227–32. 18. Landau R, Negron M, Blouin J, Scott JA, Smiley RM. Effect of eNOS genetic polymorphism on hypotension and treatment response during spinal anesthesia for cesarean section (abstract). Anesthesiology 2004; 100(Suppl 1): 33. 19. Benjafield AV, Morris BJ. Association analyses of endothelial nitric oxide synthase gene polymorphisms in essential hypertension. Am J Hypertens 2000; 13: 994–8. 20. Donahue BS, Balser JR. Perioperative genomics. Venturing into uncharted seas. Anesthesiology 2003; 99: 7–8. 21. Faraday N, Martinez EA, Scharpf RB, Kasch-Semenza L, Dorman T, Pronovost PJ, Perler B, Gerstenblith G, Bray PF, Fleisher LA. Platelet gene polymorphisms and cardiac risk assessment in vascular surgical patients. Anesthesiology 2004; 101: 1291–7. 22. Ziegeler S, Tsusaki BE, Collard CD. Influence of genotype on perioperative risk and outcome. Anesthesiology 2003; 99: 212–19. 23. Dishy V, Sofowora GG, Xie HG, Kim RB, Byrne DW, Stein CM, Wood AJ. The effect of common polymorphisms of the beta2-adrenergic receptor on agonist-mediated vascular desensitization. N Engl J Med 2001; 345: 1030–5. 24. Johnson JA, Zineh I, Puckett BJ, McGorray SP, Yarandi HN, Pauly DF. Beta 1-adrenergic receptor polymorphisms and antihypertensive response to metoprolol. Clin Pharmacol Ther 2003; 74: 44–52. 25. Sofowora GG, Dishy V, Muszkat M, Xie HG, Kim RB, Harris PA, Prasad HC, Byrne DW, Nair UB, Wood AJ, Stein CM. A common beta 1-adrenergic receptor polymorphism (Arg389Gly) affects blood pressure response to beta-blockade. Clin Pharmacol Ther 2003; 73: 366–71. 26. Williams DG, Patel A, Howard RF. Pharmacogenetics of codeine metabolism in an urban population of children and its implications for analgesic reliability. Br J Anaesth 2002; 89: 839– 45. 27. Kaiser R, Sezer O, Papies A, Bauer S, Schelenz C, Tremblay PB, Possinger K, Roots I, Brockmoller J. Patient-tailored antiemetic treatment with 5-hydroxytryptamine type 3 receptor antagonists according to cytochrome P-450 2D6 genotypes. J Clin Oncol 2002; 20: 2805–11.
The year in anaesthesia and critical care
168
28. Bond C, LaForge KS, Tian M, Melia D, Zhang S, Borg L, Gong J, Schluger J, Strong JA, Leal SM, Tischfield JA, Kreek MJ, Yu L. Single-nucleotide polymorphism in the human mu opioid receptor gene alters beta-endorphin binding and activity: possible implications for opiate addiction. Proc Natl Acad Sci USA 1998; 95: 9608–13. 29. Gelernter J, Kranzler H, Cubells J. Genetics of two mu opioid receptor gene (OPRM1) exon I polymorphisms: population studies, and allele frequencies in alcohol- and drug-dependent subjects. Mol Psychiatry 1999; 4: 476–83. 30. Landau R, Cahana A, Smiley RM, Antonarakis SE, Blouin JL. Genetic variability of mu-opioid receptor in an obstetric population. Anesthesiology 2004; 100: 1030–3. 31. Tan EC, Tan CH, Karupathivan U, Yap EP. Mu opioid receptor gene polymorphisms and heroin dependence in Asian populations. Neuroreport 2003; 14: 569–72. 32. Herz A. Endogenous opioid systems and alcohol addiction. Psychopharmacology (Berl) 1997; 129: 99–111. 33. Herz A. Opioid reward mechanisms: a key role in drug abuse? Can J Physiol Pharmacol 1998; 76: 252–8. 34. Lötsch J, Skarke C, Grosch S, Darimont J, Schmidt H, Geisslinger G. The polymorphism A118G of the human mu-opioid receptor gene decreases the pupil constrictory effect of morphine-6-glucuronide but not that of morphine. Pharmacogenetics 2002; 12: 3–9. 35. Lotsch J, Zimmermann M, Darimont J, Marx C, Dudziak R, Skarke C, Geisslinger G. Does the A118G polymorphism at the mu-opioid receptor gene protect against morphine-6-glucuronide toxicity? Anesthesiology 2002; 97: 814–19. 36. Hernandez-Avila CA, Wand G, Luo X, Gelernter J, Kranzler HR. Association between the cortisol response to opioid blockade and the Asn40Asp polymorphism at the mu-opioid receptor locus (OPRM1). Am J Med Genet 2003; 118B: 60–5. 37. Wand GS, McCaul M, Yang X, Reynolds J, Gotjen D, Lee S, Ali A. The mu-opioid receptor gene polymorphism (A118G) alters HPA axis activation induced by opioid receptor blockade. Neuropsychopharmacology 2002; 26: 106–14. 38. Lötsch J, Skarke C, Leifold J, Giesslinger G. Genetic predictors of the clinical response to opioid analgesics. Clin Pharmacokinet 2004; 43: 983–1013. 39. Beyer A, Koch T, Schroder H, Schulz S, Hollt V. Effect of the A118G polymorphism on binding affinity, potency and agonist-mediated endocytosis, desensitization, and resensitization of the human mu-opioid receptor. J Neurochem 2004; 89: 553–60. 40. Skarke C, Darimont J, Schmidt H, Geisslinger G, Lotsch J. Analgesic effects of morphine and morphine-6-glucuronide in a transcutaneous electrical pain model in healthy volunteers. Clin Pharmacol Ther 2003; 73: 107–21.
8 Clinical applications of pharmacokinetics and pharmacodynamics FRÉDÉRIQUE SERVIN Introduction Anaesthesia is usually described as the association of unconsciousness, absence of response to adrenergic stimuli and immobility generated by the action of drugs, most of which have unwanted side effects with consequences the anaesthetist has to minimize. Moreover, the time course of action of those drugs should be carefully controlled in order to ensure adequate surgical conditions and rapid recovery. Only a clear understanding of the pharmacology of anaesthetic drugs allows their optimal use. Nowadays, anaesthesia is very seldom achieved with a single drug. Unfolding the mechanisms of drug interactions has therefore been at the forefront of pharmacological research in anaesthesia during the last decade. In order to achieve this goal thinking in concentrations and not in doses has been mandatory. Owing to the usually complex pharmacokinetics of anaesthetic drugs the administration of a bolus dose generates a concentration that is highly variable over time and therefore the effect and drug interactions are also variable over time. During the last few years the corpus of knowledge on the pharmacokinetics and concentration—effect relationships of anaesthetic drugs has increased tremendously as the clinical importance of these works unfolded. Nowadays, in most countries in the world, propofol, sufentanil and remifentanil can be administered using target-controlled infusion devices that include pharmacokinetic simulation modules, thereby allowing the user to apply the concentration—effect relationship directly, rather than bothering with (tedious) pharmacokinetics anymore. Nevertheless, much remains to be done to improve both the knowledge and the devices and to broaden their application field. Thus, two great axes in clinical research can be distinguished: improving the understanding of the mode and time course of actions of anaesthetic drugs, including drug interactions, and studying specific groups of patients in whom the pharmacokinetic-pharmacodynamic relationships maybe modified. The works presented in this review illustrate the current research in those two main areas. © Atlas Medical Publishing Ltd
The year in anaesthesia and critical care
170
Clinical pharmacology of pharmacokinetic-pharmacodynamic relationships: improving the basic knowledge of the underlying mechanisms This year works in this domain are in a direct continuum with previous research works. Bouillon et al. collected data from volunteers from which they had already published pharmacokinetic interactions between propofol and remifentanil (remi-fentanil has no influence on propofol’s pharmacokinetics, whereas propofol impairs remifentanil’s early distribution thus reducing the initial bolus dose to achieve a given initial concentration without altering the maintenance requirements or recovery times) |1|. They were then ready to analyse the pharmacodynamic part of their work and provide us with two papers: one on the respiratory depressant effect of propofol alone and one on propofol/remifentanil pharmacodynamic interactions. The same issue was addressed by Kern et al. using a slightly different methodology, namely applying the response surface models described a few years ago by Minto et al. |2|. Their mechanism of action being slightly different from that of intravenous hypnotics, propofol/opioid interaction studies cannot be directly applied to volatile/ opioid combinations which therefore require specific attention. The response to surgical incision has been extensively studied |3|, but has in fact little use in clinical practice in paralysed patients. Blunting the haemodynamic response to painful stimuli is of more immediate clinical use |4|. The ultimate practical use of pharmacokinetic—pharmacodynamic studies is to improve anaesthetic drug administration. Target-controlled infusion has been an important milestone in this domain, with the launching nearly 10 years ago of the Diprifusor™ (AstraZeneca, London, UK) |5|. Nevertheless, there remains a wide margin for improvement of target-controlled infusion devices. Propofol’s induction of anaesthesia is not yet fully understood. The influence of non-steady-state blood flow changes has been addressed |6| and one of the most puzzling issues for research today is the time course of propofol induction. In order to study the transfer rate of propofol to the effect site properly, the independence of the concentration—effect relationship towards the speed of induction had to be assessed first.
Mixed effects modelling of the intrinsic ventilatory depressant potency of propofol in the non-steady state Bouillon T, Bruhn J, Radu-Radulescu L, Andresen C, Cohane C, Shafer SL Anesthesiology 2004; 100: 240–50 BACKGROUND. Propofol is widely used for sedation in spontaneously breathing patients. This study was designed for investigating its ventilatory depressant properties in the clinically important range. After determination of their carbon dioxide responsiveness by a rebreathing design ten volunteers received propofol in
Clinical applications of pharmacokinetics
171
a stepwise ascending pattern until significant ventilatory depression occurred (arterial carbon dioxide partial pressure [PaCO2] superior to 65 mmHg and/or apnoea periods of more than 60 s). Thereafter the concentration was allowed to drop to 1 µg/ml. Their arterial propofol blood concentrations and PaCO2 values were measured frequently. An indirect response model was used for describing the PaCO2 time course in order to account for the non-steady-state environment in which carbon dioxide accumulates during drug-induced ventilatory depression. A correction factor was introduced for the decreased carbon dioxide production during propofol administration. Pharmacodynamic parameters were estimated with the NONMEM software. INTERPRETATION. At a propofol concentration of 3 µg/ml, all the subjects were anaesthetized but spontaneous ventilation was maintained. The propofol concentration causing 50% depression of minute ventilation under a constant PaCO2 and 50% of the maximum decrease in carbon dioxide production was 1.33±0.23 µg/ml (mean ± standard error [SE]) with a coefficient of variation of 50%. The important inter-individual variability of the parameters suggests that the ventilatory response in individuals may be poorly predicted by population estimates. A bolus of propofol will produce a more pronounced decrease in ventilation than a continuous infusion yielding the same maximum effect site concentration because, after a bolus, the onset of drug effect is very fast and does not allow the slower increase in PaCO2 to counteract effectively the drug effect. Comment It is very difficult to quantify precisely the ventilatory depression induced by a given drug in non-steady-state conditions, which are those observed in clinical practice. End tidal carbon dioxide pressure (PETCO2) is a function of the minute ventilation, with a hysteresis that can be modelled using an effect compartment for carbon dioxide. It is also the balance between carbon dioxide production and elimination. The basal carbon dioxide production can be estimated through the baseline minute ventilation in the absence of the study drug. The introduction of the study drug, here propofol, induces several changes: a respiratory depression assumed to be from central action on the ventilatory drive, a reduction in metabolism and carbon dioxide production, which remains modest and a resulting increase in PaCO2, which in turn stimulates ventilation. High PaCO2 levels also have sedative properties (carbon dioxide narcosis). The main interest of the complex methodology described in this paper is to propose an illustration of these phenomena that explains the time course of action of ventilatory depression in clinical practice. The reason why large propofol boluses should be avoided in spontaneously breathing patients is clearly illustrated. This study also outlines the importance of controlling the level of PETCO2 in clinical research during recovery for comparing different anaesthetic protocols. The launching of anaesthesia ventilators including some pressure support during spontaneous ventilation and recovery may clinically alleviate this problem.
The year in anaesthesia and critical care
172
Pharmacodynamic interaction between propofol and remifentanil regarding hypnosis, tolerance of laryngoscopy, bispectral index and electroencephalographic approximate entropy Bouillon TW, Bruhn J, Radulescu L, et al. Anesthesiology 2004; 100: 1353–72 BACKGROUND. The purpose of this investigation was to quantify the Pharmacodynamic interaction between propofol and remifentanil for the probability of no response to various stimuli and to link that relation with the interaction on two electroencephalograph (EEG) end-points, the bispectral index and EEG approximate entropy. Propofol (pharmacokinetic model of Schnider et al. |7|) and remifentanil (pharmacokinetic model of Minto et al. |8|) target-controlled infusions were applied to 20 young healthy volunteers who were studied in two consecutive phases, namely a single drug or drug combination. During the combination phase one drug concentration was titrated and the second one was kept constant at a target of 0–4 ng/ml for remifentanil and 0–4 µg/ml for propofol. Different levels of stimulation were applied from calling the subject’s name in a low voice to laryngoscopy. The data were analysed with NONMEM software. Two response surface models were used for characterizing the interaction between propofol and remifentanil: the empirical response surface described by Minto et al. |2| and a novel hierarchical model based on the concept that opioids attenuate the pain signal subsequently projected to the cortex, where hypnotics modulate the probability of response. INTERPRETATION. The authors found that the interaction between propofol and remifentanil was synergistic for loss of response to shaking and shouting and to laryngoscopy, that remifentanil was not hypnotic in clinically relevant concentrations and that the greatest reduction in propofol needs was observed with remifentanil concentrations up to 4 ng/ml. In general, the two models yield similar predictions, although the hierarchical model predicted higher propofol concentrations in the absence of opioids and provided a better prediction of the likelihood of response both for hypnosis and for laryngoscopy. Propofol was equipotent in its effect on the bispectral index and approximate entropy with or without remifentanil and the two drugs had additive effects on the two EEG measures of drug effect considered. Comment The results described in this paper confirm already well-established notions and its main interest probably lies elsewhere, mainly in the description of a new so-called hierarchical model for describing propofol/opioid interactions. This model, based on a physiological assumption about the mechanism of action of the two drugs, provided a better fit to the data than the response surface model. If the results concerning laryngoscopy are clear-cut and convey robust information, it is perhaps not the case for the ‘hypnotic’ state, which the authors properly refer to as ‘response to shaking and shouting’. In their methods section they stated that the volunteers were exposed to a series of stimuli with increasing
Clinical applications of pharmacokinetics
173
intensity, namely calling their name in a low voice, shouting their name, shaking and shouting, insertion of a laryngeal mask and laryngoscopy. It is difficult to believe that the propofol and remifentanil concentrations required to blunt the response to calling the subject’s name in a low voice are similar to those required to blunt the response to shaking the subject actively while shouting his or her name. Nevertheless, during the rest of the paper, only one response is referred to for ‘hypnosis’. The interest for interaction models during the next few years will probably grow since such models can be implemented into infusion devices for prescribing help or closed loop systems. In that respect the description of this hierarchical model is quite interesting. Its ability for describing volatile/opioid interactions remains to be demonstrated since the mechanisms of action of the volatile agents are not exactly the same as those of propofol.
Opioid-hypnotic synergy: a response surface analysis of propofolremifentanil pharmacodynamic interaction in volunteers Kern SE, Xie G, White JL, Egan TD. Anesthesiology 2004; 100: 1373–81 BACKGROUND. Response surface models are useful for characterizing drug interactions over a complete range of clinically relevant concentrations |2|. This volunteer study investigated the interaction of propofol and remifentanil using four surrogate measures of drug effect: observer’s assessment of alertness/sedation score, tibial pressure algometry, electrical tetany and response to laryngoscopy. The observer’s assessment of alertness/sedation score and response to laryngoscopy were considered as binary variables. Three different drug delivery profiles (targetcontrolled infusion) were administered to the 24 subjects, one with a single agent (0.5–12 µg/ml of propofol or 0.5–80 ng/ml of remifentanil) and two with combinations of propofol and remifentanil. The arterial blood concentrations of both drugs were measured. The data from the single-drug administrations were fitted using sigmoid Emax models (models describing the relationship between concentration and effect as a sigmoid up to a maximum effect) and the propofol remifentanil interactions were mapped using response surface models. INTERPRETATION. The mean (SE) values of the concentration associated with 50% of the effect (EC50) for propofol and remifentanil alone are given in Table 8.1. Nonlinear analysis of the concentration—response data indicated a strong degree of synergistic interaction between propofol and remifentanil. The fitting of the models was good with an absolute error on the surface of approximately 25%. The predicted concentration—response relation for sedation and laryngoscopy approximated the concentrations needed clinically for loss of consciousness and intubation, whereas the models for algometry and tetanic stimulation approximated concentrations needed clinically for skin closure and surgical incision. The surfaces clearly showed a large plateau area on top wherein further increases in either drug concentration do not result in more effect. The response surface for laryngoscopy illustrated the difficulty in achieving absence of response to this stimulus in the complete absence of a hypnotic.
The year in anaesthesia and critical care
174
Table 8.1 Mean (SE) values of the EC50 for propofol and remifentanil alone Propofol EC50 (µg/ml)
Remifentanil EC50 (ng/ml)
OAA/S
1.8 (0.06)
12.5 (0.53)
Algometry
4.16 (0.65)
8.84 (1.48)
Tetanic stimulus
4.56 (0.42)
21.3 (3.55)
Laryngoscopy
5.6 (0.42)
48.9 (0.53)
Source: Kern et al. (2004).
Comment This study confirmed the profound synergy between propofol and remifentanil and quantified the interaction for various stimuli. What is the interest for the clinician of such models describing phenomena of which he already is well aware with more precision? These models might be displayed in future anaesthesia ‘expert systems’ and help in analysing clinical settings by showing what is the probability of no response in front of predicted drug concentrations. For example, if a patient receives propofol and remifentanil target-controlled infusions, the anaesthetist may be able to predict the probability of movement to surgical incision or of haemodynamic answer to tracheal intubation from such a model and therefore modify his/her drug targets in accordance. Similarly, they may help in analysing a situation where a patient does not react as expected. Another interest is to cover the whole range of clinically relevant concentrations and therefore provide means of identifying target concentration pairs that optimize some outcomes of interest (e.g. recovery time, drug costs, analgesic state on emergence, etc.).
Effects of two target-controlled concentrations (1 and 3 ng/ml) of remifentanil on MAC-BAR of sevoflurane Albertin A, Casati A, Bergonzi P, Fano G, Torri G. Anesthesiology 2004; 100: 255–9 BACKGROUND. Haemodynamic sympathetic responses to adrenergic stimuli in paralysed patients remain the main monitoring of anaesthesia adequacy. Establishing concentration—effect relationships and drug interactions at clinically relevant drug concentrations in this context may allow the description of dosage regimens for preventing a response to surgical stimuli in the majority of patients. Seventy-four healthy adult patients scheduled for abdominal surgery were included in this randomized double-blind study and received a remifentanil target-controlled infusion with a target of 0,1 or 3 ng/ml associated with sevoflurane to a stable endtidal concentration chosen according to an up and down sequential allocation technique starting at 3% All patients received 60% nitrous oxide and
Clinical applications of pharmacokinetics
175
cisatracurium. The sevoflurane alveolar concentration, which blunted the haemodynamic response to skin incision in 50% of patients (MAC-BAR), was calculated. Inter-group comparisons were performed by analysis of variance and multiple comparisons by a t test with Bonferroni correction. INTERPRETATION. The sevoflurane MAC-BAR value in patients receiving no remifentanil was 2.8% (95% confidence interval [Cl] 2.5–3.0%), compared with 1.1% (95% Cl 0.9–1.3%) in patients receiving remifentanil at a predicted concentration of 1 ng/ml and 0.2% (95% Cl 0.1–0.3%) in patients receiving remifentanil at a predicted concentration of 3 ng/ml. Based on the estimation of the sevoflurane MAC in a similar population, the MAC-BAR values represented respectively 1.95 (1.8–2.1), 1.1 (1.0–1.2) and 0.68 (0.6–0.75) MAC. The ED50 and ED95 (dose associated with respectively 50% and 95% of effect) values for blockade of response to a skin incision (logistic regression analysis) were 2.8±0.4% and 3.3±0.5%, 1.57±0.2% and 1.98%±0.3%, and 0.27±0.05 and 0.29±0.2% in the three groups of patients, respectively. Comment This study confirmed that the administration of low doses of remifentanil (a 1 ng/ml concentration of remifentanil is achieved at steady state with an infusion rate of 0.03 µg/kg/min) dramatically reduced the concentration of sevoflurane required for blunting the haemodynamic response to a skin incision. When first launched, the recommended doses for remifentanil were much higher and led to unwanted effects in some patients. One of the drawbacks of MAC studies is that they provide indications only for a specific stimulus, namely skin incision. Nevertheless, it gives a ‘starting point’ from which to titrate the drugs further. In patients receiving 3 ng/ml of remifentanil, the MAC-BAR of sevoflurane was very low, lower than the MAC-AWAKE (the concentration associated with consciousness in 50% of patients) published for this drug, thus putting the patient at risk of intra-operative awareness with no haemodynamic stigmata of inadequate anaesthesia. The influence of nitrous oxide was probably significant in these results. Nevertheless it should be advocated not to use such low sevoflurane concentrations in the absence of EEG monitoring.
Induction speed is not a determinant of propofol pharmacodynamics Doufas AG, Bakhshandeh M, Bjorksten AR, Shafer SL, Sessler DI. Anesthesiology 2004; 101: 112–21 BACKGROUND. Evidence suggests that the rate at which intravenous anaesthetics are infused may influence their plasma effect site equilibration. The authors studied five different rates of increases in propofol predicted effect site concentration by way of a target-controlled infusion system implemented with Schnider et al ’s |7| pharmacokinetic model in 18 healthy volunteers in order to
The year in anaesthesia and critical care
176
assess whether different sedation end-points occurred at the same effect site propofol concentration whatever the infusion rate. Arterial blood samplings were drawn at significant times and measures of effect included clinical observer’s assessment of alertness/sedation score and drop in syringe time, response to the automated responsiveness monitor prompt and continuous recording of the bispectral index. The pharmacokinetic-pharmacodynamic analysis was performed with NONMEM software. The authors concurrently evaluated an automated responsiveness monitor against other sedation measures and propofol effect site concentrations. INTERPRETATION. The pharmacokinetic model of Schnider et al. |7| performed well with a bias less than 2% and a very good accuracy. When using individual estimates of the rate constant for the plasma effect site equilibration (ke0) values, the clinical endpoints were reached at the same effect site propofol concentration independent of the rate of infusion. Population-based pharmacokinetics combined with real-time EEG measures of drug effect may provide a means of individualizing pharmacodynamic modelling during target-controlled infusion. The automated responsiveness monitor appears a useful tool for accurate estimations of propofol-induced sedation. Comment Target-controlled infusion is gaining popularity as an accurate way of administering intravenous drugs. Nevertheless, owing to the important intra- and inter-individual variability and the numerous factors that may interfere with its performance in a given individual, much work is yet to be done to provide the anaesthetist with an optimal infusion device. Targeting the effect site rather than the plasma provides quicker clinical control, but is based on the assumption that the concentration—effect relationship at the effect site is independent of the drug administration and dependent on the accurate estimation of ke0. This study confirmed |9| that the predicted effect site concentration is a robust parameter, but that the ke0 value displays a great inter-individual variability, which is not yet properly explained even if some hints are available (early distribution kinetics dependent on the speed of administration, EEG analysis hysteresis, etc.). Rather than population approaches yielding incompressible inaccuracy, the solution maybe in individual estimations of ke0, i.e. during induction of anaesthesia for feeding the system for the rest of the administration |10|. Future studies need to address whether the rate of blood—brain propofol equilibration is faster after a bolus dose when compared to conventional infusion rates and how to implement this information in complex infusion devices. Unfortunately, even if their volunteers all received the different sedation ramps at random during the study, the authors did not provide us with estimations of the intraindividual variability in ke0 nor the influence of the speed of induction on this parameter.
Clinical applications of pharmacokinetics
177
Clinical applications of pharmacokinetic-pharmacodynamic relationships: special patients groups Many pharmacokinetic—pharmacodynamic studies requiring extensive data collection in the absence of ‘clinical noise’ (i.e. surgery) are performed in volunteers, who are usually young and healthy. Unfortunately, this seldom reflects the anaesthetist’s everyday life and the direct application of the results of volunteer studies to routine patients may lead to unpleasant surprises. Pharmacological studies in special patient groups are therefore of the utmost importance to the clinician, specifically if they present continuous data that can be readily translated into recommendations. This is the case for the study of Shibutani et al. on fentanyl in morbidly obese patients. The increasing number of such patients presented for surgery has generated several studies. In particular, this is because the pharmacokinetics of (anaesthetic) drugs are frequently altered in the obese. The drugs should unfortunately be studied one by one, since no class conclusion can be drawn and the results depend on the specific drug pharmacokinetics. Data are already available, for example, for remifentanil |11|, sufentanil |12| or propofol |13|. Muscle relaxants have been the target of two studies by Leykin et al., one of which is presented here. Another clinical application of pharmacology is the administration of drugs by noninvasive routes. These routes add several parameters to the pharmacokinetics that are not always easy to control. The study by Wheeler et al. is a good illustration of this fact. Those non-invasive routes are often proposed for pre-medication, analgesia or sedation in children, another specific patient group where adult pharmacology cannot be directly applied. This explains the need for specific studies that can raise safety issues. Some important clinical situations are not easy to implement in clinical research. Yet the information obtained would be very relevant clinically. Such is, for example, the case of emergency situations. In such cases, the only way is to use animal models. The group from the University of Utah have already published several meaningful works on the pharmacology of anaesthetic agents in haemorrhagic shock in a porcine model |14|. They have now addressed the issue of the influence of crystalloid resuscitation on propofol pharmacology. Their work is accompanied by an editorial by Shafer |15|.
Accuracy of pharmacokinetic models for predicting plasma fentanyl concentrations in lean and obese surgical patients; derivation of dosing weight (‘pharmacokinetic mass’) Shibutani K, Inchiosa MA, Sawada K, Bairamian M. Anesthesiology 2004; 101: 603–13 BACKGROUND. The currently available pharmacokinetic models for fentanyl were derived from normal weight patients and were not scaled to body weight. Their application to obese patients may cause over-prediction of the plasma concentration of fentanyl. This study was performed on 70 normal weight patients (mean body mass index [BMI]=24) and 39 obese patients (mean BMI=44) receiving fentanyl infusions both intra- and post-operatively in order to examine the
The year in anaesthesia and critical care
178
Fig. 8.1 The correlation between predicted and measured concentrations of plasma fentanyl in the lean group (group L) and the obese group (group O). The regression for group L is y=0.89x+0.80 (r=0.794; P<0.001) and that for group O is y=1.12x+1.38 (r=0.866; P<0.001). Source: Shibutani et al. (2004). accuracy of predicting plasma fentanyl concentrations with currently available pharmacokinetic models The extent of departure from linearity was further used to
Clinical applications of pharmacokinetics
179
attempt to derive suggested dosing weights for fentanyl. To do so, the blood samples were randomly divided into two groups: the first was used for establishing the nonlinear regression equation for the performance error versus total body weight whereas the second group was used for estimating the correction of bias thus achieved. INTERPRETATION. Shafer’s |15| pharmacokinetic model performed well in lean patients, but consistently over-predicted measured plasma concentrations in obese patients (Fig. 8.1). The bias between predicted and measured fentanyl plasma concentrations increased with
Fig. 8.2 Pharmacokinetic mass derived from the non-linear regression analysis of the performance error with the Shafer model versus the total body weight (TBW) (solid line). Pharmacokinetic mass=52/correction factor. This nomogram provides suggested dosing weights to be used for correcting the patient total body weight to a more appropriate Pharmacokinetic mass. Pharmacokinetic mass derived from the non-linear regression analysis of the performance error with the Scott model versus the total body weight (dashed line). Source: Shibutani et al. (2004).
The year in anaesthesia and critical care
180
body weight, reaching over 50%. A correction factor may then be calculated. In clinical practice where fentanyl is seldom used by target-controlled infusion, it gives an indication of the virtual weight (‘pharmacokinetic mass’) to be used for fentanyl dosing in obese patients (Fig. 8.2). Comment The relationships between anaesthetic drug doses and clinical markers of obesity (total body weight, BMI, etc.) are not simple, with an increased proportion of body fat to total body mass, but also increases in cardiac output and intravascular volumes. The main interest of this paper, apart from demonstrating the overestimation of plasma fentanyl concentrations with the currently available pharmacokinetic models, is to provide an empirical nomogram of suggested dosing weights to be used for correcting the patient total body weight in usual dosing formulae. The use of inaccurate pharmacokinetic models for target-controlled infusion in this population is a rare occurrence which leads to underdosing of the patients and this can always be corrected by titration, but the use of the actual total body weight in super-obese patients as boluses during induction of anaesthesia might well lead to excess doses and therefore unwanted effects.
The pharmacodynamic effects of rocuronium when dosed according to real body weight or ideal body weight in morbidly obese patients Leykin Y, Pellis T, Lucca M, Lomangino G, Marzano B, Gullo A. Anesth Analg 2004; 99: 1086–9 BACKGROUND. Obesity is now considered to be a worldwide concern and the introduction of new surgical treatments such as laparoscopic gastric banding or bypass has dramatically increased the number of morbidly obese patients submitted to anaesthesia care. The adaptation of drug dosages to obese patients is a major issue. The pharmacokinetic changes in this population are numerous and obese patients have larger fat masses but also increased absolute lean body masses when muscle mass and body water are in a decreased proportion of their total body weight. Muscle relaxants are molecules mainly distributed in lean tissues and poorly in the fat. It is therefore advocated to base their dosage on ideal rather than actual body weight. The aim of this prospective parallel groups study was to investigate whether rocuronium onset time, duration of action to 25% recovery and recovery time were modified if the total dose was calculated according to ideal body weight or real body weight. Twelve obese patients (six receiving 0.6 mg/kg of rocuronium based on their ideal body weight and six receiving 0.6 mg/kg of rocuronium based on their real body weight) admitted for laparoscopic gastric banding and six normal weight controls (gynaecological laparoscopic surgery) were recruited. INTERPRETATION. When rocuronium was administered based on the subjects’ real body weight the duration of action to 25% recovery of the initial twitch of the train of four was more than doubled whereas the onset time and recovery index were not
Clinical applications of pharmacokinetics
181
statistically significantly different (Table 8.2). The results of this study support the assumption that rocuronium dosing in morbidly obese patients should be based on their ideal body weight and not their real body weight. Comment The literature on the pharmacokinetics and pharmacodynamics of anaesthetic drugs in morbidly obese patients remains sparse and in this context this prospective study demonstrated that dosing rocuronium according to real body weight leads to a prolonged duration of action, i.e. the patients are overdosed. The authors acknowledge one of the weaknesses of their paper, which is the small number of patients (only six per group). They expected a statistically shorter onset time when over-dosing their patients, which they did not get. Similarly the recovery time was not modified. On another point, the authors stated that they obtained statistical significance on the duration of action to 25% recovery which other authors had failed to demonstrate because their patients were morbidly obese (BMI approximately 43, with a mean weight of approximately 115 kg) whereas previously studied patients were only
Table 8.2 Onset time, duration of action to 25% recovery, spontaneous recovery index and bolus dose of rocuronium in the three groups of patients Variable
Onset (s)
Duration of action to 25% recovery (min)
Recovery index (min)
Dose (mg)
Real body weight
77 (37– 92)
56 (44–60)
17 (11–24)
67 (59– 82)
Ideal body weight
88 (54– 99)
22 (21–25)
14 (8–16)
33 (28– 38)
Normal body weight
67 (50– 85)
25 (18–31)
11 (5–18)
38 (31– 43)
P value
0.201
0.003
0.102
0.03
Values are given as medians (ranges) (Kruskal—Wallis test). Source: Leykin et al. (2004).
moderately obese. One can argue that nowadays more and more patients are superobese, with BMIs over 60 and total body weights over 150 kg. In this population the lean body mass and extracellular volumes are significantly increased and dosing water-soluble drugs according to the ideal body weight may well lead to underdosage. No formulae are currently available for calculating the lean body mass in morbidly obese patients: with the standard formula, at a given height, the lean body mass increases to a peak, but body weights beyond that peak value yield decreasing values of lean body mass. An elegant way of designing a dosing scheme adequate for all patients from normal to super-obese might be to use a target-controlled infusion of rocuronium for determining the pharmacokinetic model that might best describe the time course of action of
The year in anaesthesia and critical care
182
rocuronium or any other muscle relaxant in this population |16|. Dosing schemes might be proposed from that modified pharmacokinetic model for non-target-controlled infusion administration.
The pharmacokinetics of the intravenous formulation of fentanyl citrate administered orally in children undergoing general anaesthesia Wheeler M, Birmingham PK, Lugo RA, Heffner CL, Coté CJ. Anesth Analg 2004; 99: 1347–51 BACKGROUND. Fentanyl Oralet™ (oral transmucosal fentanyl citrate [OTFC]) has recently been proposed for treating breakthrough cancer pain. It is absorbed by both the oral mucosa and the digestive tract in adults, leading to an overall bioavailability of approximately 50%. Nevertheless, in children the bioavailability seems less, around 33–36%, perhaps due to a predominant digestive absorption. To address this issue, a single dose of the intravenous fentanyl formulation (10–15 µg/kg, with upper dose limit 400 µg) was administered orally to ten healthy children of 5–10 years old presented for surgery and a non-compartmental pharmacokinetic analysis was performed from venous blood samples taken at regular intervals of time up to 600 min after fentanyl administration. The authors expected similar plasma concentrations and pharmacokinetic variables with this formulation, which bypasses the oral mucosal absorption as an equivalent dose of OTFC. Thus, the historical data of OTFC administration in children were re-examined using the same non-compartmental analysis as the current liquid formulation data to be used as controls. INTERPRETATION. The time to reach the maximum concentration (Tmax) was statistically longer with the liquid (1.74±1.64 h) than with the OTFC (0.96±0.6 h) formulation, but the time to achieve a minimum effective concentration of 0.6 ng/ml was similar (0.64±0.37 versus 0.55±0.36 h). The areas under the concentration curve were similar, corresponding to the same bioavailability. The use of OTFC in children has a number of disadvantages, including a large time variability for complete consumption of the lozenge. This study has demonstrated that the bioavailability of the liquid formulation is similar to that of OTFC. Nevertheless, the great inter-individual variability in the maximum concentration (Cmax) (median 1.88 ng/ml with a range of 0.34–3.96 ng/ml) (Fig. 8.3), leading to potentially dangerous blood levels in some children and to small blood levels inadequate for analgesia in others, suggest that this method of administration be used with caution.
Clinical applications of pharmacokinetics
183
Fig. 8.3 Measured plasma fentanyl concentrations versus time for ten children administered the intravenous formulation of fentanyl citrate orally. Source: Wheeler et al. (2004). Comment The treatment of severe pain in children remains a challenge and non-invasive administration routes are very appealing. One of the greatest problems in this field is that these routes preclude the ability for titrating the dose to the effect when the interindividual variability is very high. The authors have demonstrated variable plasma concentrations, but the even bigger pharmacodynamic variability should further increase the overall range in the effects observed. It would have been interesting to have some sort of pharmacodynamic assessment of the efficacy and safety of the study drug. An obvious limitation in this study is the choice of historical data for the OTFC groups. Although the difficulties in dealing with pharmacokinetic studies in children explain this choice, the differences in the children’s ages and doses weaken the results. Similarly, the liquid fentanyl was administered pre-operatively, which is necessary considering the onset time, but the use of sevoflurane and isoflurane for general anaesthesia may have influenced the fentanyl blood levels.
The year in anaesthesia and critical care
184
Determination of the pharmacodynamic interaction of propofol and remifentanil during esophago-gastro-duodenoscopy in children Drover DR, Litalien C, Wellis V, Shafer SL, Hammer GB. Anesthesiology 2004; 100: 1382–6 BACKGROUND. The combination of propofol and remifentanil, two drugs with a rapid onset and offset, may be particularly useful for procedures of short duration, including oesophago-gastro-duodenoscopy in children. This study describes the pharmacodynamic interaction of those two drugs in this setting. Propofol was administered by plasma target-controlled infusion (paediatric model) to 32 children aged 3–10 years, associated to remifentanil at a randomized infusion rate of 0, 25, 50 or 100 ng/kg/min. The response to the procedure was defined as gross movement and/or increases in heart rate and blood pressure greater than 120% of baseline. A sigmoid Emax model was developed for describing the interaction. INTERPRETATION. All observed responses were movements, with haemodynamic variables remaining within 120% of baseline. Three patients in the 100 ng/ml group developed oxygen desaturation. The pharmacodynamic analysis demonstrated a positive interaction between remifentanil and propofol when used in combination (Fig. 8.4). Doses of remifentanil above 25 ng/ml did not improve the interaction and should not be recommended in this setting. Comment This study demonstrated a propofol-sparing effect of remifentanil on the response to stimulation when administered in combination, which was expected. The results are slightly blunted by the fact that the stimulus was applied only 5 min after initiating the administrations, at a time when the concentration of neither the propofol nor remifentanil was stable and no blood sampling was performed. From a clinical point of view, the combination of propofol and remifentanil in adult patients leads to a strikingly synergistic depressant effect on ventilation with an important risk of severe respiratory depression |17|. It is probably the same in children. In this study, the reduction in the propofol EC50 was less than 25% when remifentanil was used and oxygen desaturation was observed in two patients receiving 100 ng/kg/min of remifentanil. It is therefore doubtful that the risk—benefit balance of using remifentanil in combination with propofol for oesophago-gastro-duodenoscopy in children justifies this association.
Clinical applications of pharmacokinetics
185
Fig. 8.4 The probability of response curves for propofol when used alone and when used with remifentanil. Source: Drover et al. (2004).
Influence of hemorrhagic shock followed by crystalloid resuscitation on propofol; a pharmacokinetic and pharmacodynamic analysis Johnson KB, Egan TD, Kern SE, McJames SW, Cluff ML, Pace ML Anesthesiology 2004; 101: 647–59 BACKGROUND. Previous work has demonstrated that ongoing haemorrhagic shock dramatically alters the distribution, clearance and potency of propofol |14|. Patients suffering from haemorrhagic shock are typically resuscitated before surgery. It is therefore particularly relevant to assess whether volume resuscitation after haemorrhagic shock restores pharmacokinetic and pharmacodynamic drug behaviour to baseline. Thus, in this study a swine model was used for investigating the influence of an isobaric bleed followed by crystalloid resuscitation on the pharmacokinetics and pharmacodynamics of propofol. The animals were randomly assigned to control or shock followed by resuscitation groups. All animals received 750 µg/kg/min of propofol for 10 min after the real or sham resuscitation and a nonlinear mixed effect population pharmacokinetic analysis was derived from propofol
The year in anaesthesia and critical care
186
arterial blood concentrations up to 180 min after the infusion began. The recorded EEG bispectral index scale (BIS™, Aspect Medical, Newton, USA) was used as a measure of effect. INTERPRETATION. The pharmacokinetic analysis demonstrated that resuscitation had approximately restored the propofol pharmacokinetic parameters to their baseline values. On the contrary, the pharmacodynamic effects of propofol were enhanced: the haemodynamic consequences of propofol administration were of a greater magnitude in the shock—resuscitation group and in this group the bispectral index dropped to a lower score and displayed a slower return to baseline than in the control group. The baseline values of the bispectral index prior to propofol administration were similar in both groups and the authors had checked in another study that the sequence shock—resuscitation had no effect on the bispectral index. The pharmacokinetic—pharmacodynamic modelling estimated an effective concentration to obtain 50% of the effect (C50) of 5.0±0.4 µg/ml in the control group and 3.4±0.4 (µg/ml in the shock—resuscitation one. The concentration—effect relationship in each animal as characterized by the pharmacodynamic model is displayed in Fig. 8.5. Comment This study demonstrated that, in patients presented for surgery after some sort of resuscitation of a haemorrhagic shock, the risk of excessive blood concentrations and delayed recovery due to altered pharmacokinetics are minimized. Nevertheless, both the haemodynamic and the hypnotic effects of propofol remain increased. This might be due to an increase in the unbound propofol concentration (which was not measured in this study). Plasma propofol is highly bound to albumin (around 98%) and readily penetrates red blood cells and the hypoalbuminaemia frequently observed after an important fluid loading may well account for this result. In this respect this study represents an extreme in haemorrhagic shock—resuscitation in that it used only lactated Ringer’s solution. These results may probably be extended to all situations of extreme anaemia and hypoalbuminaemia. This paper was associated with an editorial |15| where the author outlined the consistency of all the pharmacokinetic—pharmacodynamic studies done so far on anaesthetic drugs in shock: in shock the body becomes a blood—brain circuit, at the expense of circulation to the gut, liver and muscles. This results in higher brain concentrations, more rapid onset and a more profound effect. The complex relationship between propofol, its lipid vehicle and blood proteins increases the brain’s (and blood vessels’) sensitivity to propofol in shock and this remains true after fluid loading. The authors have performed a simulation of what these drug dosages should be in haemorrhagic shock from all published pharmacokinetic—pharmacodynamic data of anaesthetic drugs in shock (Fig. 8.6). The ‘take home’ message to the clinician is that propofol is a particularly poor choice for induction of anaesthesia in patients with shock even after adequate fluid resuscitation. In contrast, etomidate may be the drug of choice in this situation.
Clinical applications of pharmacokinetics
187
Fig. 8.5 The concentration—effect relationship for each animal as characterized by the pharmacodynamic model. The solid lines represent control animals over a propofol plasma concentration range of 1 to 20 µg/ml. The dashed lines represent shockresuscitation animals over the same range. The bold lines portray the mean pharmacodynamic model for each group. The horizontal axis is on the log scale. Source: Johnson et al. (2004).
The year in anaesthesia and critical care
188
Fig. 8.6 The reduction in the dose for achieving a given drug effect in animals with haemorrhage, compared to control animals, based on simulations using pharmacokineticpharmacodynamic models published in the literature. Source: Johnson et al. (2004). Conclusion For many years pharmacokinetic—pharmacodynamic studies have been considered by most clinicians as the hobby of a small group of people and their clinical applications remained difficult to foresee. The introduction of target-controlled infusion in clinical practice has modified this vision, but anaesthesiologists often complained that the models did not correspond to their everyday patients. Pharmacokinetic—pharmacodynamic research is now reaching a new maturity through a better under-standing of the underlying mechanisms and a broadening of the populations studied. Thus, the results of the studies can be adapted to clinical practice more readily. This year publications in the field illustrate this evolution. In the near future, one can already gather from the abstracts presented in scientific congresses that basic knowledge will continue to grow, specifically focused on the induction of anaesthesia and non-steady-state situations and propofol/opioids and volatile/opioids interactions. The direct applications of this will certainly be an improvement in target-controlled infusion devices and a growing interest in closed loop systems. The launching of an anaesthesia machine (Zeus™ from Dräger) |18| which allows a rapid titration of end-tidal volatile concentrations with minimal drug consumption will renew the interest in clinical research on volatile/opioids titration.
Clinical applications of pharmacokinetics
189
References 1. Bouillon T, Bruhn J, Radu-Radulescu L, Bertaccini E, Park S, Shafer S. Non-steady state analysis of the pharmacokinetic interaction between propofol and remifentanil. Anesthesiology 2002; 97: 1350–62. 2. Minto C, Schnider T, Short T, Gregg K, Gentilini A, Shafer S. Response surface model for anesthetic drug interactions. Anesthesiology 2000; 92: 1603–16. 3. Sonner JM, Antognini JF, Dutton RC, Flood P, Gray AT, Harris RA, Homanics GE, Kendig J, Orser B, Raines DE, Rampil IJ, Trudell J, Vissel B, Eger EL Inhaled anesthetics and immobility: mechanisms, mysteries, and minimum alveolar anesthetic concentration. Anesth Analg 2003; 97: 718–40. 4. Daniel M, Weiskopf RB, Noorani M, Eger EL. Fentanyl augments the blockade of the sympathetic response to incision (MAC-BAR) produced by desflurane and isoflurane: desflurane and isoflurane MAC-BAR without and with fentanyl. Anesthesiology 1998; 88: 43– 9. 5. Glen JB. The development of ‘Diprifusor’: a TCI system for propofol. Anaesthesia 1998; 53(Suppl 1): 13–21. 6. Avram M, Krejcie T. Using front-end kinetics to optimize target-controlled drug infusions. Anesthesiology 2003; 99: 1078–86. 7. Schnider TW, Minto CF, Gambus PL, Andresen C, Goodale DB, Shafer SL, Youngs EJ. The influence of method of administration and covariates on the pharmacokinetics of propofol in adult volunteers. Anesthesiology 1998; 88: 1170–82. 8. Minto CF, Schnider TW, Egan TD, Young E, Lemmens HJM, Gambus PL, Billard V, Hoke JF, Moore KHP, Hermann DJ, Muir KT, Mandema JW, Shafer SL. Influence of age and gender on the pharmacokinetics and pharmacodynamics of remifentanil. I. Model development. Anesthesiology 1997; 86: 10–23. 9. Wakeling HG, Zimmerman JB, Howell S, Glass PS. Targeting effect compartment or central compartment concentration of propofol: what predicts loss of consciousness? Anesthesiology 1999; 90: 92–7. 10. Struys MM, De Smet T, Versichelen LF, Van De Velde S, Van den Broecke R, Mortier EP. Comparison of closed-loop controlled administration of propofol using bispectral index as the controlled variable versus ‘standard practice’ controlled administration. Anesthesiology 2001; 95: 6–17. 11. Egan T, Huizinga B, Gupta S, Jaarsma R, Sperry R, Yee J, Muir K. Remifentanil pharmacokinetics in obese versus lean patients. Anesthesiology 1998; 89: 562–73. 12. Slepchenko G, Simon N, Goubaux B, Levron JC, Le Moing JP, Raucoules-Aime M. Performance of target-controlled sufentanil infusion in obese patients. Anesthesiology 2003; 98: 65–73. 13. Servin F, Farinotti R, Haberer JP, Desmonts JM. Propofol infusion for maintenance of anesthesia in morbidly obese patients receiving nitrous oxide: a clinical and pharmaco-kinetic study. Anesthesiology 1993; 78: 657–65. 14. Johnson KB, Egan TD, Kern SE, White JL, McJames SW, Syrold N, Whiddon D, Church T. The influence of hemorragic shock on propofol: a pharmacokinetic and pharmaco-dynamic analysis. Anesthesiology 2003; 99: 409–20.
The year in anaesthesia and critical care
190
15. Shafer S. Shock values. Anesthesiology 2004; 101: 567–8. 16. Vermeyen KM, Hoffmann VL, Saldien V. Target-controlled infusion of rocuronium: analysis of effect data to select a pharmacokinetic model. Br J Anaesth 2003; 90: 183–8. 17. Nieuwenhuijs DJF, Olofsen E, Romberg RR, Sarton E, Ward D, Engbers F, Vuyk J, Mooren R, Teppema LJ, Dahan A. Response surface modeling of remifentanil-propofol interaction on cardiorespiratory control and bispectral index. Anesthesiology 2003; 98: 312–22. 18. Struys M, Kalmar A, De Baerdemaeker L, Mortier E, Rolly G, Manigel J, Buschke W. Time course of inhaled anaesthetic drug delivery using a new multifunctional closed-circuit anaesthesia ventilator: in vitro comparison with a classical anaesthesia machine. Br J Anaesth 2004; 94: 306–17.
Part III Monitoring and equipment
New technology in anaesthesia and critical care TIM COOK Introduction The following four chapters critically examine developments in the fields of new technology in the areas of anaesthesia and critical care. The clinical utility and implementation of this technology raises controversies in all these fields. In Chapter 9 Professor Marx reports on advances in non-invasive (or more accurately ‘less invasive’) methods of cardiac output measurement in 2004. The pulmonary artery catheter (PAG) has been the clinical standard for cardiac output measurement for four decades |1|, yet it is widely accepted that the PAG is an imperfect tool. It is invasive, the data it produces are not always reliable and cardiac output estimation with the PAG may be affected by user variability and device inconsistency. One illustration of this is the PAC’s repeatability coefficient. This measures the degree of consistency between repeated measures of cardiac output in the same patient (intraindividual variation). It is defined as the variation (±2 standard deviations [SDs]) in consecutive measures expressed as a percentage of the mean value. The repeatability coefficient has been reported as 26% with double injections and 13% with triple injections |2|. The optimal reliability is reduced to 6% when four injections, all timed to a fixed phase of the respiratory cycle, are used |3|. Despite this, PAG thermodilution remains the clinical ‘gold standard’ against which other techniques are benchmarked. It has long been argued by some that the PAG may put patients at risk through both direct complications (e.g. insertion-related complications, pulmonary complications and sepsis) and indirect complications. Indirect complications involve misleading data derived from the PAG, misinterpretation or misuse of derived data and inappropriate treatment goals made in response to that data |4|. Potential misapplication of data was involved in the vogue for ‘supranormal’ goal-directed management of oxygen delivery in the critically ill in the late 1980s |5|. This was not beneficial to all patients and in some was clearly deleterious |6, 7|. In 1987 Robin called for a moratorium on the use of the PAG |8| after Gore et al. reported increased mortality associated with its use after myocardial infarction |9|. In 1996 Connors et al. reported increased costs and mortality in patients who had a © Atlas Medical Publishing Ltd
PAG inserted during their first 24 h in an intensive care unit (ICU) |10|. This US study was accompanied by editorials on both sides of the Atlantic |11, 12|. The studies by both Gore et al. and Connors et al. found increased mortality with use of the PAC, but both
were rightly criticized for being non-randomized and retrospective cohort comparisons |11|. Despite this, both studies have influenced practice and fuelled controversy. Firstly, as is common in these circumstances, some individuals adopted polarized positions, either calling for a moratorium on use of the PAC or taking the view that the increased risk did not apply to their practice and their patients. Arguments were raised over whether it was use of the PAC itself, poor case selection, poor data collection or misinterpretation of data that led to poor outcomes. Marx refers to one recent paper |13| suggesting an outcome benefit when using the PAC, but only when used in the sickest quartile of critically ill patients, with harm associated with use in patients in the lowest quartile of critical illness. This graphically illustrates the concept of risk-benefit applied to interventional procedures in sick patients. However, the argument over risk and benefit is not resolved. In a US case—control prospective cohort study of over 1000 patients with severe sepsis no difference was observed in mortality or resource use between patients treated with or without a PAC |14|. More recently in the UK the Pulmonary Artery Catheters in Patient Management (PAC-man) Study attempted to determine whether the PAC is associated with benefit or harm in the UK’s ICU population. This important study prospectively randomized over 1000 patients to use or withholding use of a PAC at a point when the clinician would normally have considered its use were the patient not in the study. Those patients who did not have a PAC inserted were able to have their cardiac output measured by other methods. This pragmatic approach should lead to clinically relevant information about the potential benefits and risks to patients in the UK setting. Full results are awaited, but preliminary reports indicate no significant benefit (or harm) to the patient whether a PAC was used or not. Use of a PAC was not associated with a change in hospital mortality, ICU or hospital length of stay or organdays of support (Sheila Harvey, ICNARC [Intensive Care National Audit and Research Centre], personal communication). The second development was to accelerate the development of alternative methods of cardiac output measurement and this is the focus of Marx’s chapter. Marx comments on papers covering the following monitors: oesophageal Doppler, transoesophageal echo with colour Doppler, transoesophageal echo for the measurement of aortic flow velocity, arterial thermodilution with pulse contour analysis, indocyanine green dye dilution and peripheral dye estimation by densitometry, pulmonary carbon dioxide elimination, whole body electroimpedence and, finally, a modification of the PAC where cardiac output is measured continuously and almost instantly by determining the energy required for maintaining the temperature gradient between two distally positioned thermistors. Other methods not discussed in the chapter are already commercially available and include arterial lithium dilution with pulse contour analysis and an alternative oesophageal Doppler technique. New developments in cardiac output monitoring provide the possibility of acquiring data on cardiac output and systemic circulation vascular resistance as well as other novel measures of cardiac performance, vascular filling, lung water and its distribution. Some allow continuous and near instantaneous cardiac output measurement. Most devices, but not all, are markedly less invasive to the patient than the PAC and therefore offer the possibility of reducing or eliminating direct complications. However, they will not necessarily reduce the indirect complications and morbidity associated with the
misinterpretation and misuse of data. They also offer the possibility of the development of novel targets for goal-directed therapy. It is widely (but not universally) accepted that, in certain circumstances and in certain patients, there is benefit from measuring cardiac output and related haemodynamic indices and using changes in these for optimizing fluid loading and guiding cardiovascular management. The circumstances include high-risk surgery in high-risk patients and the sicker group of patients admitted to critical care units |15–19|. This approach in high-risk surgical patients has demonstrated both improved clinical outcome |15–17| and economic benefit |18, 19|. Similarly goal-directed haemodynamic manipulation, albeit without recourse to cardiac output monitoring, has shown benefit in severe sepsis |20|. However, there is much work still required in order to define exactly which patients do benefit from goal-directed therapy and which therapeutic goals should be applied. It is likely that cardiac output or a surrogate marker of this will be among the goals. The first requirement would therefore appear to be the development of a cardiac output monitor that is accurate and reliable, both between different patients and when measuring trends in an individual patient. In determining what is a useful, reliable and accurate cardiac output monitor some explanation of the terms used in Marx’s chapter may be helpful. Accuracy measures how closely the measured value represents the true value. This is recorded as bias, which is the mean difference between two monitors’ readings. The limits of agreement are bias ±2 SDs (i.e. the limits within which 95% of the measured differences lie). The r2 value is the square of the correlation coefficient and is a measure of how much variation is explained by a linear relationship between the data derived from two monitors. The r2 value does not examine non-linear relationships and Bland-Altman plots may be more appropriate where this is required |21|. The reliability of consecutive measurements in the same patient is reported as the repeatability coefficient, which is explained above. It is clear that, since 1996, there has been a general re-evaluation of the use of the PAC. While indications for and the frequency of use of the PAC 10 years ago probably differed considerably between the UK, mainland Europe and North America, there is no doubt that PAC use has been reduced in all these regions. It is possible that, in the near future, measures of regional blood flow and perfusion may become increasingly important, but at present this is some way off outside specialist centres and global measures of flow and cardiovascular performance remain a more realistic goal. A reliable, accurate, less invasive monitor of cardiac output will potentially benefit many chronically and critically ill patients. Professor Marx’s chapter illuminates the recent developments in the search for such a device. In Chapter 10, Drs Vereeke and Struys examine an area of similar controversy. Depth of anaesthesia monitoring, like cardiac output monitoring, is another area where a ‘gold standard’ does not exist, which makes comparisons between devices difficult. A further problem exists as ‘anaesthesia’ or ‘loss of consciousness’ is difficult to define and identify accurately in practice. The point at which deep sedation becomes unconsciousness or loss of responsiveness is important, but imprecise. Defining this depends on a variety of factors, importantly including the type of stimulus applied to the patient in order to induce a response. Measurement of the depth of anaesthesia may serve three roles: (1) avoiding inadequate dosing of anaesthetic agents (preventing ‘awareness’), (2) avoiding excessive
dosing of anaesthetic agents (reducing side effects, complications and recovery times) and (3) allowing insights into pharmacokinetic and pharmacodynamic aspects of anaesthetic drug behaviour. The first two roles have received most focus in the anaesthetic literature. It is clear that awareness during anaesthesia is more common than many anaesthetists would be happy to admit: it is notable that several large recent trials enrolling in excess of 40 000 patients have all identified explicit awareness in 500–1000 patients |22–25|. This fact is concerning to both patients and anaesthetists as avoidance of awareness is a prerequisite of quality anaesthesia and complaints of awareness maybe responsible for up to 12% of anaesthesia-related malpractice claims |26|. These studies have also identified three further important issues. Firstly, that awareness may not be detected immediately post-operatively. Three interviews; when leaving the recovery area; at 24–72 h; and at 1– 2 weeks after anaesthesia, are now accepted as necessary to identify all cases of awareness |22, 27|. Secondly, a significant proportion of these unfortunate patients may suffer significant long-term psychological distress and sequelae |28|. Thirdly, that ‘near misses’ maybe even more common. Sebel et al. reported awareness in 0.13%, ‘possible awareness’ in 0.25% and ‘inira-operative dreaming’ in 6% |25|, while the figures from the unmonitored group in the B-Aware trial were 1.3% for possible awareness and 6.9% for intra-operative dreaming |24|. With these figures in mind, several authors have raised the concern that the use of depth of anaesthesia monitors for reducing anaesthesia to the lowest level possible or ‘skating on thin ice’, as described by Sneyd, might paradoxically raise the incidence of awareness |29, 30|. Two major papers recently offered some reassurance in this respect: depth of anaesthesia monitoring in studies of several thousand patients being associated with a fivefold decrease in awareness during anaesthesia |24, 31|. Ekman et al.’s paper studied awareness in 4945 relatively unselected patients who received bispectral index monitoring during routine anaesthesia |31|. This cohort was then compared to an unmonitored historical cohort of 7826 patients. Awareness was reduced by 78% in the monitored group. Clearly the study design has flaws as several years elapsed between cohorts and the monitoring was not blinded. However, the result is very similar to the result of the B-Aware study discussed in the chapter |24|. So 2004 saw two studies suggesting that depth of anaesthesia monitoring can reduce anaesthetic underdosing (awareness). A different challenge is to use such monitoring for reducing the incidence of anaesthetic overdosage. This might be expected to lead to a reduction in anaesthetic drug-related side effects. Several previous studies have reported improvements in minor outcomes such as reduced drug use, faster recovery from anaesthesia and reduced post-operative nausea and vomiting when depth of anaesthesia monitoring is used |32, 33|, but these findings are far from consistent |24|. Of potentially more importance in higher risk patients, the avoidance of overdosage might be anticipated to reduce perioperative morbidity or mortality. At present there is little evidence to support this theory. However, one recent paper did report an increase in the long-term mortality after non-cardiac surgery in patients in whom prolonged deep anaesthesia was used (as recorded by bispectral index monitoring), the mortality risk increasing 25% for each hour bispectral index score remained below 45 |34|. Further studies are required in order to determine whether an increased depth of anaesthesia is more generally associated with worsened outcomes, but if this were confirmed it would
prove a major stimulus to the widespread use of intra-operative depth of anaesthesia monitoring |35|. A good depth of anaesthesia monitor detects change from the ‘state of wakefulness’ to ‘unconsciousness’, but should also have the ability to distinguish an increasing depth of sedation before anaesthesia is reached (and different levels of anaesthesia). Such monitors might have a role in critical care as well as anaesthesia. As with cardiac output monitors, once an anaesthetist has decided that monitoring the depth of anaesthesia is a desirable, the solution is by no means obvious. Several monitors based on quite different principles exist with variable amounts of basic and clinical research available for determining which monitor is best. Once more the absence of a gold standard is problematic. Monitors must either be compared head to head during increasing sedation and anaesthesia or be compared to the very imperfect ‘isolated forearm technique’. Neither solution is ideal. Clinical utility requires a rapid response, minimal inter-individual variability (consistency), a gradation of response from deep anaesthesia through to light anaesthesia to sedation and wakefulness (monotonicity) and good trending of the data for one individual (reliability). In addition, ease of use (a high use success rate), the generation of clinically sensible data (low inappropriate data rate) and resistance to external electrical sources (low overlap with electromyograph [EMG] activity and rapid recovery from electrodiathermy interference) are desirable features. Finally, Vereecke and Struys report on several papers examining the use of depth of anaesthesia monitors in studying the interaction between different anaesthetic agents offering an insight into their pharmacodynamics, pharmacokinetics and drug interactions. Such studies offer the possibility of creating better population and individualized pharmacokinetic and pharmacodynamic models of drug action. This is a complex but rewarding section. It becomes clear from this and other sections that ‘depth of anaesthesia monitors’ do not measure the depth of anaesthesia as a clinician would regard it. A complex synthesis of observed and assumed drug effects is what an experienced anaesthetist summates when estimating the ‘depth of anaesthesia’. However, electroencephalograph (EEG)-based monitors only measure electrical activity from the cerebral cortex. Subcortical, midbrain and hindbrain activity is not detected. Opioids and other drugs that have greater effects in the non-cortical areas of the brain are therefore less likely to induce changes in cortical/brain surface electrical activity and will be less likely to influence EEG-based depth of anaesthesia monitors. Current depth of anaesthesia monitors measure the degree of hypnosis caused by anaesthetic (hypnotic) drugs. They do not measure analgesia, nor amnesia and drugs such as opioids and nitrous oxide in clinical doses have little effect on these monitors. As such, the accurate term for these EEG-based monitors is ‘hypnotic component of anaesthesia’ (HCOA) monitors. This year has brought important advances in both the basic understanding of EEGbased HCOA monitors and evidence of their clinical usefulness. Data allowing the clinician to decide with greater certainty when to use a HCOA monitor and which monitor is suitable is developing fast. In Chapter 11, Dr Asai addresses another area of controversy: progress in supraglottic airway development. This area is bedevilled by a dearth of products, some poor research, shifting goalposts and inadequate statutory control of the introduction of new products |36|.
In 1988 the laryngeal mask airway (LMA), subsequently called the classic Laryngeal Mask Airway™ (cLMA™, Intavent Orthofix Ltd, Maidenhead, Berkshire, UK), was introduced and this gained primacy over the next decade to the extent that, in 2003, twothirds of general anaesthetics in the UK were delivered via a laryngeal mask. There are currently 13 supraglottic airways and six single-use LMAs available. The drive behind the development of this large number of devices is not entirely clear, but commercial interest is one factor. Statutory control over release of these devices is somewhat limited. There is no distinct ‘standard’ for supraglottic airways in the UK or USA and as a result these devices either come under the standards applicable to oropharyngeal airways or no standards apply at all. Instead manufacturers are required to comply with more generic ‘directives’. Statutory control through the ‘directives’ appears to focus more on material performance and manufacturing quality control rather than assessing whether the device is ‘fit for purpose’ (i.e. it performs its intended purpose effectively and safely). For several devices there has been a cycle of introduction of a new device, informal clinical evaluation, the appearance of a few clinical reports of the use of the device, subsequent modification of the device (often without clear identification by the manufacturer of the changes that are made) and continued marketing of the device. This cycle may happen several times. Time delays between device evaluation and the publication of reports may mean that a current report relates to a device that no longer actually exists in that form. Confusion results. Devices that may illustrate this problem are the Pharyngeal Airway xpress™ (PAx™, Vital Signs, Littlehampton, Sussex, UK) and the Laryngeal Tube Sonda™ (LTS™, VBM GmbH, Sulz, Germany). The PAx™ has been modified three times (i.e. there have been four versions) since its introduction in 2001. There are nine publications listed under the Medline search ‘PAxpress or PAx’, but in only one is there an indication of which device was evaluated. The LTS™ is a modification of the Laryngeal Tube™, itself introduced in 2000 and modified at least twice since. The LTS™ was introduced in 2002 and modified in late 2004. In this case the manufacturers are to be congratulated as the device has been renamed as the LTS II™ to differentiate it from the LTS™. However, the LTS™ is no longer available and all published work on LTS™/LTS II™ is in fact performed on the LTS™. It is such changes that lead to the ‘shifting goalposts’ problem with supraglottic airways. One cannot blame the manufacturers for trying to improve a device, but it is surprising that changes that alter performance (for why else make the changes?) can be introduced without a statutory requirement to inform users. As a result much of the published literature on many supraglottic airway devices is either now difficult to interpret or redundant. Confusion reigns. A further area of current interest and confusion is in the provision, marketing and purchase of single-use LMAs. Despite their number non-LMA supraglottic airways have achieved very limited penetration of the anaesthesia market, but there is considerable interest in single-use LMAs. The use of disposable equipment has increased in the last few years and national advice has been given recommending that ‘Where appropriate single use disposable equipment will remove the difficulties of reuse and decontamination procedures. The use of such equipment is to be encouraged’ |37|. The particular stimulus for an apparent shift towards single-use airway equipment might be attributed to concerns over cross-infection and in particular concern over prion transmission of new variant Creutzfeldt—Jakob disease (vCJD). In 2001 some estimates
of the likely deaths from vCJD in the UK exceeded 100 000. Lymphoid tissue has been cited as particularly high risk and tonsillectomy is specifically singled out in Government and Royal College guidance as an area in which reusable LMAs must not be reused and single-use LMAs are a realistic alternative |38|. However, since 2000 the number of deaths from vCJD, rather than rising, has halved (http://www.cjd.ed.ac.uk/figures.htm) and, even before this fall, estimates of the peak of the epidemic in the UK had been reduced dramatically from >100 000 deaths to between 10 and 7000 |39| and even further to between 40 and 600 (over the next 70 years) |40|. A recent histological study of 2000 consecutive surgically removed tonsils found no evidence of prion protein in any specimen |41|. It is true that the recent observation of iatrogenic vCJD in a patient, without the usual genotytpe for vCJD susceptibility, has raised concerns that estimates may rise again |42|, but most evidence appears to indicate strongly that initial calculations of the epidemic were overly pessimistic. vCJD is a particularly awful and emotive disease, but this fact should not deflect from a cool analysis of risk and benefit. Although concern over cross-infection is legitimate the evidence to support the view that reusable airways add infective risk to patients is largely theoretical. Data on the transfer of infection from one patient to another through the use of a sterilized supraglottic airway do not currently exist. While it would be rash to ignore the possibility of risk, this should be defined and balanced against the risk of introduction of single-use equipment. It has been noted that even a small deterioration in safety as a result of using a single-use device of poorer quality in place of a reuseable device would increase the overall risk to patients and go against the recommendation of the Spongiform Encephalopathy Advisory Committee |43|. So how safe are single-use LMAs at present? The patent for the general design of the cLMA™ ended in 2003 (though not for the epiglottic bars, the patent for which runs to 2007). Several companies have introduced single-use competitors for the cLMA™ this year. There are currently six different manufacturers of single-use LMAs in the UK. Five of these devices are made of polyvinyl chloride (PVC) or polyurethane, rather than silicone. Each of the LMAs, with the exception of that manufactured by Intavent Orthofix, are of a different design to the reusable LMA, not least in the absence of epiglottic bars, for patent-related reasons. It would be wrong to assume that these devices (of somewhat different design and made of a different material) behave in the same manner as the reusable cLMA™. At the time of writing there are no publications relating to any of these new devices except that made by Portex Ltd (Hythe, Kent, UK). There are four significant studies, studying 890 patients, comparing this airway to traditionally designed LMAs (cLMA™ and Unique LMA™) |44–47|. Three of these find in favour of the traditionally designed devices |45–47| and one, as described by Asai |44|, modestly in favour of the Portex device. The point is not to decry a particular device (after all there are no publications at all with which to judge the other devices). However, the reusable cLMA™ has a considerable pedigree (150 million uses and >2000 publications). In contrast, there is limited, mostly negative evidence or no evidence on which to judge the efficacy and safety of single-use devices. It may be that their performance and safety will exceed that of the traditionally designed cLMA™, but this is conjecture. What is open to criticism is the ease with which devices, with limited evidence of efficacy or safety, enter the marketplace. The use of single-use products, particularly those made of PVC, raises other practical questions of cost, storage and the environmental dangers of disposal. So is
the use of a single-use LMA in place of a reusable LMA currently appropriate? It is a far from easy question to answer. Dr Asai commented recently, in an editorial, that ‘pulmonary aspiration is the key factor that will determine the future of supraglottic airways’ |48|. Supraglottic airways are generally regarded as contraindicated where there is an increased risk of regurgitation or aspiration. However, identifying risk (and balancing this against the often ignored risks of neuromuscular paralysis and tracheal intubation and extubation), as Asai infers in his editorial, is a very inexact science |48|. Therefore it would seem logical to focus the development of supraglottic airways on those devices with design features or performance characteristics that reduce the likelihood of aspiration. Proof of aspiration risk reduction is difficult to demonstrate as aspiration with the cLMA™ is estimated to occur in only one in 4000–10000 elective cases |49, 50|. This makes comparative trials that are designed for detecting an actual difference in aspiration too large to be practical |51|. Therefore one is left with design features, benchwork, case reports and controlled trails with surrogate markers of aspiration risk to determine which devices afford more safety. Amongst currently available supraglottic airways those devices that fulfil the criteria of suitable design include the LTS II™, the Proseal LMA™ and the Combitube™ (Kendall Sheridan, Argyle, NY, USA). It is the ProSeal LMA™ and the Combitube™ about which there is most accrued evidence and both would appear to offer increased airway protection. Whether the Combitube™, which was recently modified, is sufficiently atraumatic to allow routine use during anaesthesia is yet to be agreed |52, 53|. It would seem logical to demand that future developments in supraglottic airway devices should include design features that aim to reduce the risk of regurgitation and aspiration specifically, while ideally enabling both spontaneous and controlled ventilation. This is particularly so if airway use is extended to areas outside the operating theatres, such as during cardiopulmonary resuscitation and airway rescue. At present there is too little statutory control, too little pre-launch testing, too much marketing based on efficacy and too little consideration for safety. When one considers the confusion in this area it is salutary to compare the statutory process for the introduction of equipment to that for drugs. Were any of these products drugs, there would be no question of their introduction. In the USA the Food and Drug Administration is in the process of creating standards for supraglottic airways. It is hoped that this will take this field forward and establish requirements for evaluating new airway devices by laboratory testing including tests in mannequins and in controlled cohort and comparative clinical trials before they are marketed. This model could then be reproduced around the world. To finish on a high note, it must be acknowledged that little progress in anaesthesia technology would take place without enthusiastic innovation by individuals and the investment by medical equipment companies. The challenge is to ensure that the enthusiasm, innovation and investment are responsibly directed to benefit both patients and the profession. In Chapter 12 Dr Chapman and Professor Bodenham examine a topic that is perhaps less overtly controversial, but is equally topical to the previous three: the increasing use of ultrasound for aiding vascular access and nerve blockade. One difference in the technology discussed in this chapter is that it has been inherited from other specialties. Radiology, obstetrics and cardiology have all embraced ultrasound techniques in their practice in the last 20 years and the technology has therefore come to
anaesthesia having already overcome the development challenges that newer technology must encounter. As a result there is less discussion of the technology itself (although some changes have been required to fit the equipment to the needs of anaesthetists) and more interest in whether this technology has a useful and necessary role in anaesthesia and critical care. The publication of the National Institute for Clinical Excellence (NICE) guidelines on the use of ultrasound guidance for central venous catheterization |54| created a degree of revolution in an evolving practice. The advice was published at a time when few departments of anaesthesia owned an ultrasound machine and few anaesthetists had the skills to use one. Suddenly ultrasound machines were required in all departments and their use was positively encouraged. How much these devices are now used for routine practice remains unclear. Chapman and Bodenham are both expert and clear enthusiasts for the use of ultrasound in anaesthetic and critical care practice. The section on vascular access indicates the widening applications of ultrasound for vascular access, including conventional central venous access, novel routes of access to central veins, peripherally placed central catheters and even assistance with difficult peripheral venous access. In this section the emphasis therefore appears to be on the increasing breadth of potential indications for ultrasound use in assisting vascular access, rather than stringent identification of the benefits over conventional techniques. However, benefits in terms of increased speed and reduced complications |55| and reduced overall costs with moderate intensity of machine use |56| have been demonstrated and are the basis of existing NICE guidance. Importantly, these reports do not apply to all routes of access. One of several questions unanswered in this field is that of whether there is a downside to the use of ultrasound. In a recent editorial Hall and Russell highlighted areas of required good practice in central line management and also reported an increase in difficulty during the ‘learning curve’ of ultrasound use |57|. This is an area for concern, particularly for experienced clinicians with low complication rates for landmark-based techniques. Research into methods of training that might minimize this and its consequences is needed. Another area of concern is in the use of ultrasound in assisting central venous cannulation of infants and children, where there is limited evidence on which to base advice. Chapman and Bodenham address this and indicate that, contrary to NICE guidance, the use of ultrasound may increase the risk |58|, though important methodological problems with the reported study exist. The use of ultrasound for assisting in nerve location is even more novel than its use for vascular access. However, the increasing availability of ultrasound machines to anaesthetists and no doubt the increasing skill with which they are wielded by certain anaesthetists is inevitably leading to wider use of this technology. Chapman and Bodenham’s chapter makes clear that, in experienced hands, ultrasound can enable a surprisingly wide variety of nerve blocks. What is less clear is whether the use of ultrasound at present offers genuine and important benefits to patients and/or to the anaesthetist. In particular is there benefit over nerve stimulator-guided techniques? In another recent editorial, Denney and Harr op-Griffiths point out that nerves are, due their structure, rather difficult to identify by ultrasound |59|. They further suggested that the use of ultrasound for neural blockade remains the preserve of enthusiasts and that
excellent results are often achieved by enthusiastic experts in regional anaesthesia, whether with landmark-based or nerve stimulation-based techniques. One challenge for these experts is to identify whether the use of ultrasound does achieve reproducible, clinically substantial benefit in neural blockade, first by experts and then in routine practice. If both are indeed shown a final challenge will be to ensure that adequate investment of time and money is applied in order to purchase equipment and enable anaesthetists to learn these new and currently rather unfamiliar skills. At present these challenges all lie ahead of the proponents of routine use of ultrasound for nerve location. Chapman and Bodenham’s chapter introduces several studies addressing such questions. To conclude, anaesthesia and critical care medicine are technology-based specialties. Improving the service we deliver to our patients is frequently dependent on the development and appropriate use of new technologies. However, it is interesting to note that few if any technologies used in anaesthesia and critical care medicine have been shown to improve patient outcomes reliably or reduce costs |27|. Once a technology is introduced and embedded into anaesthesia or critical care practice, it is difficult to reverse the process. This is one of the lessons to learn from studies with the PAC over the last 25 years. It is therefore even more important that new equip-ment and technologies are robustly evaluated and the benefit or risk to the patient fully established before they are widely introduced. All four of the chapters that follow illustrate this in their particular areas. Progress lies in working with industries developing these technologies to ensure that correct evaluations and choices are made so that poorly performing and unreliable technologies and devices are excluded and appropriate new technologies are embraced in critical care and anaesthetic practice. References 1. Swan HJ, Ganz W, Forester J, Marcus H, Diamond G, Chonette D. Catheterization of the heart in man with use of a flow-directed balloon-tipped catheter. N Engl J Med 1970; 283: 447–51. 2. Stetz CW, Miller RG, Kelly GE, Raffin TA. Reliability of the thermodilution method in the determination of cardiac output in clinical practice. Am Rev Respirat Dis 1982; 126: 1001–4. 3. Jansen JR, Versprille A. Improvement of cardiac output estimation by the thermodilution method during mechanical ventilation. Intensive Care Med 1986; 12: 71–9. 4. Walker MB, Waldmann CS. The use of pulmonary artery catheters in intensive care: time for reappraisal? Clin Intensive Care 1994; 5: 15–19. 5. Shoemaker WC, Appel PL, Kram HB, Waxman K, Lee TS. Prospective trial of supra-normal values of survivors as therapeutic goals in high-risk surgical patients. Chest 1988; 94: 1176–86. 6. Hayes MA, Timmins AC, Yau EH, Palazzo M, Hinds CJ, Watson D. Elevation of systemic oxygen delivery in the treatment of critically ill patients. N Engl J Med 1994; 330: 1717–22. 7. Velmahos GC, Demetriades D, Shoemaker WC, Chan LS, Tatevossian R, Wo CC, Vassiliu P, Cornwell III EE, Murray JA, Roth B, Belzberg H, Asensio JA, Berne TV. End-points of resuscitation of critically injured patients: normal or supranormal? A prospective randomized trial. Ann Surg 2000; 232: 409–18. 8. Robin ED. Death by pulmonary artery flow-directed catheter: time for a moratorium. Chest 1987; 92: 727–31. 9. Gore JM, Goldberg TJ, Spodick DH, Alpert JS, Dalen JE. A community-wide assessment of the use of pulmonary artery catheters in patients with acute myocardial infarction. Chest 1987; 92(4): 721–7.
10. Connors Jr AF, Speroff T, Dawson NV, Thomas C, Harrell Jr FE, Wagner D, Desbiens N, Goldman L, Wu AW, Califf RM, Fulkerson Jr WJ, Vidaillet H, Broste S, Bellamy P, Lynn J, Knaus WA. The effectiveness of right heart catheterization in the initial care of critically ill patients. JAMA 1996; 276: 889–97. 11. Soni N. Swan song for the Swan-Ganz catheter? BMJ 1996; 313: 763–4. 12. Dalen J, Bone R. Is it time to pull the pulmonary artery catheter? JAMA 1996; 16: 916–18. 13. Chittock DR, Dhingra VK, Ronco JJ, Russell JA, Forrest DM, Tweeddale M, Fenwick JC. Severity of illness and risk of death associated with pulmonary artery catheter use. Crit Care Med 2004; 32: 911–15. 14. Yu DT, Platt R, Lanken PN, Black E, Sands KE, Schwartz JS, Hibberd PL, Graman PS, Kahn KL, Snydman DR, Parsonnet J, Moore R, Bates DW; AMCC Sepsis Project Working Group. Relationship of pulmonary artery catheter use to mortality and resource utilization in patients with severe sepsis. Crit Care Med 2003; 31: 2734–41. 15. Boyd O, Grounds RM, Bennett ED. A randomized clinical trial of the effect of deliberate perioperative increase of oxygen delivery on mortality in high-risk surgical patients. JAMA 1993; 270: 2699–707. 16. Wilson J, Woods I, Fawcett J, Whall R, Dibb W, Morris C, McManus E. Reducing the risk of major elective surgery: randomised controlled trial of preoperative optimisation of oxygen delivery. BMJ 1999; 318: 1099–103. 17. Sinclair S, James S, Singer M. Intraoperative intravascular volume optimisation and length of hospital stay after repair of proximal femoral fracture: randomised controlled trial. BMJ 1997; 315: 909–12. 18. Guest JF, Boyd O, Hart WM, Grounds RM, Bennett ED. A cost analysis of a treatment policy of a deliberate perioperative increase in oxygen delivery in high risk surgical patients. Intensive Care Med 1997; 23: 85–90. 19. McFall MR, Woods WG, Wakeling HG. The use of oesophageal Doppler cardiac output measurement to optimize fluid management during colorectal surgery. Eur J Anaesthesiol 2004; 21: 581–3. 20. Rivers EP, Nguyen HB, Huang DT, Donnino M. Early goal-directed therapy in the treatment of severe sepsis and septic shock. N Engl J Med 2001; 345: 1368–77. 21. Bland M, Altman D. Statistical methods for assessing agreement between two methods of clinical measurement. Lancet 1986; 1: 307–10. 22. Sandin RH, Enlund G, Samuelsson P, Lennmarken C. Awareness during anaesthesia: a prospective study. Lancet 2000; 355: 707–11. 23. Myles PS, Williams DL, Hendrata M, Anderson H, Weeks AM. Patient satisfaction after anaesthesia and surgery: results of a prospective survey of 10 811 patients. Br J Anaesth 2000; 84: 6–10. 24. Myles PS, Leslie K, McNeil J, Forbes A, Chan MT. Bispectral index monitoring to prevent awareness during anaesthesia: the B-Aware randomised controlled trial. Lancet 2004; 363: 1757–63. 25. Sebel PS, Bowdle TA, Ghoneim MM, Rampil IJ, Padilla RE, Gan TJ, Domino KB. The incidence of awareness during anesthesia: a multicenter United States study. Anesth Analg 2004; 99: 833–9. 26. Aitkenhead AR. The pattern of litigation against anaesthetists. Br J Anaesth 1994; 73: 10–21. 27. Sebel PS. Comfortably numb? Acta Anaesthesiol Scand 2004; 48:1–3. 28. Lennmarken C, Bildfors K, Enlund G, Samuelsson P, Sandin R. Victims of awareness. Acta Anaesthesiol Scand 2002; 46: 229–31. 29. Kalkman CJ, Drummond JC. Monitors of depth of anesthesia, quo vadis? Anesthesiology 2002; 96: 784–7. 30. Sneyd JR. How low can we go? Br J Anaesth 2003; 91: 771–2. 31. Ekman A, Lindholm ML, Lennmarken C, Sandin R. Reduction in the incidence of awareness using BIS monitoring. Acta Anaesthesiol Scand 2004; 48: 20–6.
32. Song D, Joshi GP, White PF. Titration of volatile anaesthetics using bispectral index facilitates recovery after ambulatory anesthesia. Anesthesiology 1997; 87: 842–8. 33. Can TJ, Glass PS, Windsor A, Payne F, Rosow C, Sebel P, Manberg P. Bispectral index monitoring allows faster emergence and improved recovery from propofol, alfentanil and nitrous oxide anaesthesia. Anesthesiology 1997; 87: 808–15. 34. Monk TG, Saini V, Weldon BC, Sigl JC. Anesthetic management and one-year mortality after non-cardiac surgery Anesth Analg 2005; 100(1): 4–10. 35. Cohen NH. Anesthetic depth is not (yet) a predictor of mortality! Anesth Analg 2005; 100: 1–3. 36. TM Cook. Spoilt for choice? New supraglottic airways. Anaesthesia 2003; 58: 107–10. 37. Association of Anaesthetists of Great Britain and Ireland. Infection Control in Anaesthesia. London: AAGBI, November 2002. 38. G Smith. Variant CJD: What You Need to Know at Present. Royal College of Anaesthetists Bulletin May 2001. London: Royal College of Anaesthetists, 2001. 39. Ghani AC, Ferguson NM, Donnelly CA, Anderson RM. Factors determining the pattern of the variant Creutzfeldt—Jakob disease (vCJD) epidemic in the UK. Proc R Soc Lond B Biol Sci 2003; 270: 689–98. 40. Ghani AC, Donnelly CA, Ferguson NM, Anderson RM. Updated projections of future vCJD deaths in the UK. BMC Infect Dis 2003; 3: 4. 41. Frosh A, Smith LC, Jackson CJ, Linehan JM, Brandner S, Wadsworth JD, Collinge J. Analysis of 2000 consecutive UK tonsillectomy specimens for disease-related prion protein. Lancet 2004; 364: 1260–2. 42. Peden AH, Head MW, Ritchie DL, Bell JE, Ironside JW. Preclinical vCJD after blood transfusion in a PRNP codon 129 heterozygous patient. Lancet 2004; 364: 527–9. 43. Blunt MC, Burchett KR. Variant Creutzfeldt-Jakob disease and disposable anaesthetic equipment—balancing the risks. Br J Anaesth 2003; 90: 1–3. 44. Van Zundert AA, Fonck K, Al-Shaikh B, Mortier E. Comparison of the LMA-classic with the new disposable soft seal laryngeal mask in spontaneously breathing adult patients. Anesthesiology 2003; 99: 1066–71. 45. Orlikowski CE. An audit of the single use Portex laryngeal mask. Anaesth Intensive Care 2004; 32: 693–6. 46. Paech MJ, Lain J, Garrett WR, Gillespie G, Stannard KJ, Doherty DA. Randomized evaluation of the single-use Soft Seal and the re-useable LMA Classic laryngeal mask. Anaesth Intensive Care 2004; 32: 66–72. 47. Brimacombe J, Von Goedecke A, Keller C, Brimacombe L, Brimacombe M. The laryngeal mask airway Unique versus the Soft Seal laryngeal mask: a randomized, crossover study in paralyzed, anesthetized patients. Anesth Analg 2004; 99: 1560–3. 48. Asai T. Who is at increased risk of pulmonary aspiration. Br J Anaesth 2004; 93: 497–500. 49. Verghese C, Brimacombe J. Survey of laryngeal mask airway usage in 11910 patients: safety and efficiency for conventional and non-conventional usage. Anesth Analg 1996; 82: 129–33. 50. Brimacombe JR, Berry A. The incidence of aspiration associated with the laryngeal mask airway; a meta-analysis of the published literature. J Clin Anesth 1995; 7: 297–305. 51. Cook TM, Nolan JP. The Pro-Seal laryngeal mask airway. Anaesthesia 2002; 57: 288–9. 52. Keller C, Brimacombe J, Boehler M, Loeckinger A, Puehringer F. The influence of cuff volume and anatomic location on pharyngeal, esophageal, and tracheal mucosal pressures with the esophageal tracheal Combitube. Anesthesiology 2002; 96: 1074–7. 53. Krafft P, Hartmann T, Agro F, Gaitini LA, Vaida SJ. Is it unethical to use the Combitube in elective surgery patients? Anesthesiology 2003; 98: 1022. 54. National Institute for Clinical Excellence. Guidance on the Use of Ultrasound Locating Devices for Placing Central Venous Catheters. London: NICE Technology Appraisal No 49, 2002. 55. Hind D, Calvert N, McWilliams R, Davidson A, Paisley S, Beverley C, Thomas S. Ultrasonic locating devices for central venous cannulation: metanalysis. BMJ 2003; 327: 361–4.
56. Calvert N, Hind D, McWilliams R, Davidson A, Beverly CA, Thomas SM. Ultrasound for central venous cannulation: economic evaluation of cost-effectiveness. Anaesthesia 2004; 59: 1116–20. 57. Hall AP, Russell AP. Toward safer central venous access: ultrasound guidance and sound advice. Anaesthesia 2005; 60: 1–4. 58. Grebenik CR, Boyce A, Sinclair ME, Evans RD, Mason DG, Martin B. NICE guidelines for central venous catheterization in children. Is the evidence base sufficient? Br J Anaesth 2004; 92: 827–30. 59. Denney NM, Harrop-Griffiths W. Location, location, location! Ultrasound imaging in regional anaesthesia. Br J Anaesth 2005; 94: 1–3.
Declaration of interest Dr Cook has received honoraria from Intavent Orthofix and the LMA Company, distributors of laryngeal mask airways (Classic LMA, Unique LMA, Intubating LMA and ProSeal LMA) for lecturing.
9 Less invasive cardiac output monitoring GERNOT MARX Introduction Cardiac output measurement is still regarded as one of the key haemodynamic variables in the assessment of cardiac function and as an essential measure for guiding the therapy of critically ill patients |1|. In order to achieve adequate oxygen delivery, haemodynamic monitoring is necessary for providing appropriate cardiovascular support |2|. Although measurements of regional perfusion and oxygenation are gaining increasing interest, cardiac output measurement is still regarded as the key haemodynamic variable in the assessment of cardiac function and guidance of therapy |1|. Flow, as gauged by cardiac output, is the key measure of how well the circulation is delivering oxygen and nutrients to the vital organs and is the focus of all resuscitation efforts. Traditionally a pulmonary artery catheter (PAG) has been used for measuring pulmonary arterial pressures and cardiac output by thermodilution techniques in order to guide fluid therapy and haemodynamic manipulation. There have been serious concerns about the risk—benefit ratio of using pulmonary artery occlusion pressure as a measure of cardiac pre-load and recent studies have suggested a lack of benefit in some patient groups using the PAG |3– 5|. Recently, in a cohort of patients undergoing major non-cardiac surgery, it was demonstrated that there was no benefit associated with the use of the PAG |6|. Among patients with severe sepsis, PAG placement was not associated with a change in their mortality rate or resource use, although small non-significant trends towards a lower resource use were present in the PAG group |7|. In a consensus statement of the National Heart, Lung, and Blood Institute and the Food and Drug Administration its use was not recommended for elective major surgery |8|. Thus, there is considerable controversy surrounding the use of a PAG for measuring cardiac output, since its use may be associated with an increase in morbidity and mortality |3, 9|. As a result, there is increasing interest in less invasive cardiac output measurement techniques, such as applications of the Pick principle, oesophageal Doppler, thoracic bioimpedance and pulse contour analysis, since none of these techniques require the placement of a PAG. Pulse contour analysis provides a continuous display of cardiac output according to a modified version of the algorithm of Wesseling et al. |10|. This algorithm calcu© Atlas Medical Publishing Ltd
lates the stroke volume on a beat-to-beat basis by dividing the area under the systolic time curve by aortic impedance. Continuous arterial waveform analysis has been validated within recent years as a continuous cardiac output measurement method while other measured and derived parameters enable the simultaneous estimation of the cardiac pre-load |11|. Before clinicians accept or even adopt a new monitoring device they need
The year in anaesthesia and critical care
206
to be convinced that it compares favourably with the existing ‘gold’ standard. However, as there is no true clinical gold standard in measuring cardiac output, the accuracy of new cardiac output measuring devices is often determined by comparison with the PAG. It is important to realize that the thermodilution technique itself has some technical limitations resulting in sources of inaccuracy. The accuracy of PAC cardiac output measurement by thermodilution is affected to an extent by the transient decrease in the heart rate at the time of cold injection. The right ventricular cardiac output measured by thermodilution is unequal to the left ventricular cardiac output in the presence of intra- or extra-cardiac shunts. Valvular regurgitation gives inaccurate results as well. Thus, it is possible that, as newer devices are validated, some of these will become more appropriate reference methods than thermodilution. This chapter reviews recent studies of less invasive cardiac output monitoring devices and examines the impact of these techniques on clinical decision making in the critically ill patient.
Measurement of cardiac output by transpulmonary arterial thermodilution using a long radial artery catheter: a comparison with intermittent pulmonary artery thermodilution L’E Orme RM, Pigott DW, Mihm FG. Anaesthesia 2004; 59: 590–4 BACKGROUND. Cardiac output can be measured accurately by transpulmonary arterial thermodilution using the PiCCO™ (Pulsion Medical Systems, Munich, Germany) system with a femoral artery catheter. The authors investigated the accuracy of a new 50-cm four-French gauge radial artery catheter and the ability to use the system with a shorter radial catheter. They studied 18 patients who had undergone coronary artery surgery and made three simultaneous measurements of cardiac output using arterial thermodilution and a PAC. INTERPRETATION. Cardiac output may be measured accurately by arterial thermodilution using a 50-cm four-French gauge radial artery catheter. In this study the authors demonstrated the need for central catheterization for measuring cardiac output accurately using arterial thermodilution. In view of the potential hazards of using a long radial artery catheter it may be preferable to use a shorter catheter in the femoral artery or the axillary artery, since this approach allows accurate cardiac output monitoring |12, 13|. Comment This was an interesting study addressing the question of whether a peripherally placed arterial catheter could be used for determining cardiac output by arterial thermodilution. The authors compared the cardiac output measurements in 18 patients undergoing coronary artery bypass surgery by arterial thermodilution using this 50-cm four-French gauge radial artery catheter and a PAC. They found an agreement with a mean ± standard deviation (SD) bias of 0.38±0.771/min. It has to be stressed that, after withdrawal of more than 5 cm of the 50-cm catheter they were no longer able to measure cardiac output in all patients and, furthermore, the arterial thermodilution measurements became
Less invasive cardiac output monitoring
207
unreliable (Table 9.1). The authors suggested a low complication rate using long radial catheters. In one recent study Clementi reported no complications in a report of 6303 procedures over a 22-year period. This study included 74 patients with a 50-cm fourFrench gauge radial artery catheter. However, it is important to take care during flushing of the catheter since there is a theoretical risk of embolization into the carotid artery circulation due to the position of the tip. Since there is no large body of data available it seems inadvisable to leave this catheter in situ for more than 24 h. There were 54 comparative measurements of cardiac output at full insertion of the catheter. The ranges of cardiac output measurements were 3.43–8.93 1/min for arterial thermodilution and 3.47–9.08 1/min for pulmonary artery thermodilution. Comparing transpulmonary arterial thermodilution with pulmonary artery thermodilution demonstrated a bias of 0.38 1/min, an SD of 0.77 1/min and 95% limits of agreement of1.16 to +1.92 1/min. After a single 5-cm withdrawal of the radial catheter, cardiac output could be determined by arterial thermodilution in all subjects. However, there was an increase in both the bias and SD. Following a further 5-cm withdrawal, the cardiac output was non-recordable in one of the 18 subjects. With a catheter length of 35 cm, the cardiac output was measurable in only twelve subjects (36 measurements), whilst at 30 cm this fell to nine subjects (27 measurements) (Table 9.1).
Table 9.1 Measurement of cardiac output by transpulmonary arterial thermodilution using a long radial artery catheter Catheter length (cm)
Bias (I/min−1)
SD (I/min−1)
95% agreement limits (I/min−1)
50 (n=54)
0.38
0.77
−1.16 to 1.92
45 (n=54)
0.45
1.18
−1.91 to 2.81
40 (n=51)
0.64
1.37
−2.07 to 3.38
35 (n=36)
1.03
1.30
−1.57 to 3.63
30 (n=27)
1.29
1.56
−1.83 to 4.41
A comparison with intermittent pulmonary artery thermodilution. Bias, SD and 95% limits of agreement for the comparisons of arterial thermodilution with intermittent pulmonary artery thermodilution are given. The numbers of measurements obtained at each catheter length are shown in parentheses. Source: L’E Orme et al. (2004).
The authors concluded that cardiac output could be measured accurately by arterial thermodilution with a 50-cm four-French gauge radial artery catheter. This study also demonstrated the need for a centrally placed arterial catheter for reliable measurement of cardiac output with the PiCCO™ system. As well as eliminating the requirement for a PAG and providing the potential for continuous measurement of cardiac output by pulse contour analysis, this catheter permits monitoring of central arterial pressure, which may be preferable under such conditions as weaning from cardiopulmonary bypass and during high-dose vasopressor infusion.
The year in anaesthesia and critical care
208
Comparison of cardiac output measurements by arterial transcardiopulmonary as well as pulmonary arterial thermodilution with direct Pick Marx G, Schuerholz T, Sumpelmann R, Simon T, Leuwer M. Eur J Anaesthesiol 2005; 22: 129–34 BACKGROUND. Cardiac output measurement is still regarded as one of the key haemodynamic variables in the assessment of cardiac function and as an essential measure for guiding the therapy of critically ill patients. Currently, the most common method for the measurement of cardiac output is based on the thermodilution technique requiring insertion of a PAC. The accuracy of PACderived cardiac output values in septic patients has been questioned. INTERPRETATION. The results suggest that, even during haemodynamic instability in experimental septic shock, the correlation of cardiac output derived from arterial thermodilution and pulmonary arterial thermodilution with the direct Pick method was good. As the cardiac output measurement obtained by arterial thermodilution is less invasive than by pulmonary arterial thermodilution, it may offer practical advantages. Comment Marx et al. studied the cardiac output measurements derived from arterial thermodilution and pulmonary arterial thermodilution in a porcine septic shock model using faecal peritonitis. Twenty anaesthetized and ventilated pigs were investigated. In this experimental septic shock model, which mirrors the clinical situation of human intensive care patients as closely as possible, both the cardiac output derived from the arterial thermodilution and pulmonary arterial thermodilution methods correlated well with the cardiac output values calculated by the Pick method. The Pick method is regarded as the physiological gold standard. The authors discussed the methodological limitations and specific sources of inaccuracy of all the cardiac output measurement methods used. This study demonstrated that cardiac output could be accurately measured in septic pigs despite substantial haemodynamic disturbances. The cardiac output estimation derived from the Pick method and the cardiac output obtained by pulmonary arterial thermodilution were correlated (r=0.94; r2=0.87; P<0.001) (Fig. 9.1a). The mean cardiac output measured by pulmonary arterial thermodilution was 94.3±40.1 ml/min/kg (range 25.2–218.7 ml/min/kg). The bias between the cardiac output derived from the Pick method and the cardiac output obtained by pulmonary arterial thermodilution was 10.1 ml/min/kg (95% confidence interval [CI] 6.0–14.2 ml/min/kg) with limits of agreement of −26.8 to 47.0 ml/min/kg and a precision of 14.5 ml/min/kg (Fig. 9.1b). The analysis of the influence of the mean value on the bias was cardiac output Pick method—cardiac output pulmonary arterial thermodilution (ml/min/kg)=0.2 (cardiac output Pick method+cardiac output pulmonary arterial thermodilution)/2 (ml/min/kg) −9.8 (r=0.48; r2=0.23; P<0.05). These data suggest that
Less invasive cardiac output monitoring
209
pulmonary arterial thermo-dilution tends to overestimate lower cardiac output values and underestimate higher cardiac output values.
Fig. 9.1 (a) Linear regression diagram of the cardiac output derived from pulmonary arterial thermodilution versus the cardiac output estimated from the Pick equation via a metabolic monitor for 80 measurements in 20
The year in anaesthesia and critical care
210
septic animals, (b) Bland—Altman plot showing the mean bias and limits of agreement for the cardiac output estimated from the Pick equation via a metabolic monitor and the cardiac output derived from pulmonary arterial thermodilution. There were 80 measurements in 20 septic animals. Source: Marx et al. (2005).
Less invasive cardiac output monitoring
211
Fig. 9.2 (a) Linear regression diagram of the cardiac output derived from arterial transcardiopulmonary thermodilution versus the cardiac output estimated from the Pick equation via a metabolic monitor for 80 measurements in 20 septic animals, (b) Bland—Altmann plot showing the mean bias and limits of agreement for the cardiac output estimated from the Pick equation via a metabolic monitor and the cardiac output derived from arterial transcardiopulmonary thermodilution. There were 80 measurements in 20 septic animals. Source: Marx et al. (2005). The correlation between the cardiac output estimation derived from the Pick method and the cardiac output obtained by arterial thermodilution was similar (r=0.91; r2=0.83; P<0.001) (Fig. 9.2a). The mean cardiac output measured by arterial thermodilution was 104.3±43.2 ml/min/kg (range 27.0–235.0 ml/min/kg). The bias between the cardiac output derived from the Pick method and the cardiac output obtained by arterial thermodilution was 0.75 ml/min/kg (95% CI −3.8 to 5.3 ml/min/kg) with limits of agreement of –39.7 to 41.2 ml/min/kg and a precision of 17.8 ml/l/min/kg (Fig. 9.2b). The analysis of the influence of the mean value on the bias was cardiac output Fick method—cardiac output arterial thermodilution (ml/min/kg)=0.1 (cardiac output Fick method+cardiac output arterial thermodilution)/2 (ml/min/kg) −12.9 (r=0.29; r2=0.08; P<0.05). Thus, arterial thermodilution tended to overestimate lower cardiac output values and underestimate higher cardiac output values. However, this tendency was more pronounced for pulmonary arterial thermodilution than for arterial thermodilution in comparison with the cardiac output measurements derived from the Fick method. The authors concluded that, even during haemodynamic instability in septic shock, the correlation of the cardiac output derived from arterial thermodilution and pulmonary arterial thermodilution with the cardiac output derived from the direct Fick method output was good. As the cardiac output measurement obtained by arterial thermodilution is less invasive than by pulmonary arterial thermodilution it may offer practical advantages.
A comparison of CardioQ and thermodilution cardiac output during offpump coronary artery surgery Hullett B Gibbs N Weightman W Thackray M Newman M J Cardiothorac Vasc
The year in anaesthesia and critical care
212
Anesth 2003; 17: 728–32 BACKGROUND. Patients undergoing coronary artery surgery often have coexisting artherosclerotic disease and may have considerable haemodynamic instability during their procedures. Most commonly, cardiac output is monitored intermittently using pulmonary arterial thermodilution. The oesophageal Doppler technique measures cardiac output continuously and non-invasively by estimating the blood flow velocity in the descending aorta. This is achieved by measuring the Doppler frequency shift from a 4-MHz continuous ultrasound beam transmitted from the tip of a 6-mm diameter, flexible oesophageal probe. The device uses a nomogram for estimating the aortic cross-section, dependent on patient size and age. Several studies have shown close agreement between cardiac output measurements using the CardioQ device and thermodilution in surgical and critically ill patients. This was a prospective clinical study in 20 patients undergoing elective coronary bypass surgery without cardiopulmonary bypass. INTERPRETATION. The authors suggest that, due to the wide limits of agreement and the large inter-patient differences in both bias and precision, the CardioQ oesophageal Doppler cardiac output cannot currently be recommended as an alternative to cardiac output measured using thermodilution during off-pump coronary artery bypass (OPCAB) surgery. Comment This was yet another study demonstrating problems using the oesophageal Doppler technique for measuring cardiac output. The authors analysed the underlying pitfalls and limitations of the technique very clearly, which may be the reason for the negative result of this study. The measurement of cardiac output by this technique is highly dependent on the position of the probe. Even minor axial rotation may affect the signal. Thus, a consistent feature of this technique is the need for frequent repositioning in order to obtain an optimal signal. Another problem is the underlying assumption inherent in this technique that is used for calculating the cardiac output. The nomogram estimating the aortic cross-sectional area maybe inaccurate and a further assumption, that the amount of blood flow through the descending aorta is a fixed percentage of cardiac output, may also be incorrect. Another source of variability may be the angle of incidence between the ultrasound beam and the direction of blood flow. This angle is assumed to be 45° and the cosine of this angle (0.70) is entered into the equation in order to calculate the aortic blood velocity. In practice, the actual angle may vary slightly from patient to patient, making the estimation less accurate. For example, if the actual angle is 40° (cosine=0.76), the velocity will be under-estimated by approximately 8%. Similarly, if the actual angle is 50° (cosine=0.64), the velocity will be overestimated by approximately 9%. It was not possible to obtain a satisfactory signal in two patients despite extensive manipulation of the probe over a period of 15 min. These patients were excluded from the study. Three hundred and thirty-one comparisons of simultaneous oesophageal Doppler and thermodilution cardiac outputs were made. The Pearson correlation coefficient for the pooled data was 0.62. Using Bland—Altman analysis, the overall bias was −0.56
Less invasive cardiac output monitoring
213
1/min with a precision of 0.64 1/min. The 95% limits of agreement (bias±2 SD) were −0.56±1.28 1/min. For individual patients, the bias ranged from −1.35 to 0.271/min and the precision from 0.24 to 0.741/min. Although the operators were trained to obtain an optimal signal before commencement of the study and were considered proficient in the use of the device, they found wide limits of agreement within individual patients suggesting that the ability of the CardioQ for tracking changes in cardiac output is poor. Despite the criticism of this study that only a small number of patients were investigated, it adds evidence that a specific oesophageal Doppler cardiac output monitoring device, the CardioQ, cannot currently be recommended as an alternative to cardiac output measured by thermodilution during a specific type of surgery, i.e. OPCAB surgery.
Cardiac output monitoring during off-pump coronary artery bypass grafting Grow MP, Singh A, Fleming NW, Young N, Watnik M. J Cardiothorac Vasc Anesth 2004; 18: 43–6 BACKGROUND. OPCAB has become a more common surgical procedure. Monitoring cardiac output during procedure-related abrupt manipulations occurring in the beating heart challenges the anaesthetist. The aortic flow velocity using transoesophageal echo may offer surrogate markers for changes in cardiac output during OPCAB surgery. Nineteen consecutive patients were investigated in this prospective, observational clinical study. INTERPRETATION. This suggests that a mathematical formula, velocity—time integral (VTI)×heart rate and mixed venous oxygen saturation (SvO2) may offer the possibility of predicting changes in cardiac output during OPCAB surgery. The estimation of cardiac output trends could be helpful in guiding therapy in this clinical setting. Comment Similar to other studies Grow et al. investigated ways of improving haemodynamic monitoring in patients undergoing OPCAB surgery. They studied 19 consecutive patients and measured cardiac output using thermodilution via a PAG and changes in the descending aortic flow velocity using transoesophageal echo combined with the heart rate and SvO2. The intention of this study was to evaluate some of the commonly used continuous monitors and determine their potential utility for tracking the acute changes in cardiac output during this portion of the procedure. No single variable appeared to give reliable, simple, real-time measurements of the acute changes in cardiac output that can occur during repositioning and stabilization for OPCAB surgery. On initial inspection, the mathematical formula appears even more complex and tedious than repeated cardiac output measurements using thermodilution. However, the authors found that observing the qualitative changes that occur in VTI×heart rate and SvO2 at the time of cardiac manipulation maybe helpful in predicting cardiac output trends.
The year in anaesthesia and critical care
214
After positioning and stabilization for OPCAB surgery, the changes in VTI×heart rate and SvO2 could be used for predicting the changes in cardiac output measurements using thermodilution using the following model: cardiac output by thermodilution=−13.15+0.35(∆VTI×heart rate)+0.61(∆SvO2), where ∆VTI×heart rate and ∆SvO2 represent the percentage change from baseline values in these variables. The changes in mean arterial pressure, mean pulmonary artery pressure and continuous cardiac output did not correlate with the changes in cardiac output measurements using thermodilution. The authors suggested that VTI×heart rate, as determined by transoesophageal echo, in addition to SvO2 can strengthen clinical decision making during repositioning and stabilization of the heart during OPCAB surgery. Changes in VTI×heart rate and SvO2 maybe used as surrogate markers for changes in cardiac output during OPCAB surgery.
Automated cardiac output measurement by transesophageal color Doppler echocardiography Akamatsu S, Oda A, Terazawa E, et al. Anesth Analg 2004; 98: 1232–8 BACKGROUND. Transoesophageal echocardiography is frequently used during surgery for evaluation of regional wall motion abnormalities and ventricular volume status. The aim of this study was to evaluate a new colour Doppler technique for automated cardiac output measurement. This new method uses spatial and temporal integration of colour Doppler velocity profiles. Thus, it may provide an accurate and quick automated flow measurement without assuming a flat flow velocity profile or colinearity with the scan line. The study was undertaken in 36 patients undergoing cardiac or vascular surgery. INTERPRETATION. These data suggest that automated cardiac output measurement by transoesophageal echocardiography at the mitral annulus is a practical and accurate method for measuring cardiac output. It needs to be stressed that four patients were excluded due to the presence of ventricular arrhythmia. More information needs to be gained in different groups of patients before we can tell whether this new approach fulfils its promise. Comment In this study Akamatsu et al. employed yet another method of cardiac output measurement. Automated cardiac output integrates digital colour Doppler velocities in space and time and has been validated using transthoracic echocardiography, but has not been tested using transoesophageal echocardiography. The method relies on spatial integration assuming hemiaxial symmetry and temporal integration during selected time intervals of the instantaneous flow components (Fig. 9.3). In addition, an adequate colour gain setting is essential for obtaining accurate measurement by this method. The authors determined its feasibility in 36 patients undergoing cardiovascular surgery. Regions of interest for automated cardiac output measurement were placed within a colour sector, i.e. across the main pulmonary artery, the mitral annulus and the left ventricular outflow
Less invasive cardiac output monitoring
215
tract. Cardiac output was determined from the pulmonary artery flow, the mitral flow and the left ventricular ejection fraction flow at each view using the automated cardiac output measurement and compared to simultaneously obtained thermodilution-derived cardiac output measurements. In the mitral flow analysis the values derived from automated cardiac output measurement correlated well with those from thermodilution (r2=0.85; mean difference=0.01±0.58 l/min in the two-chamber view and r2=0.78; mean difference = −0.10±0.68 l/min in the four-chamber view). In the pulmonary artery flow analysis the values derived from automated cardiac output measurement did not correlate with those from thermodilution (r2=0.30). It was very difficult to obtain
Fig. 9.3 Schematic representation of the calculation of stroke volume by the automated cardiac output measurement method. The automated cardiac output measurement method uses digital colour Doppler velocities within a box positioned over a region of interest in the two-dimensional images with colour Doppler signals and relies on the spatial integration assuming hemiaxial symmetry and temporal integration during selected time intervals of the instantaneous flow components. Source: Akamatsu et al. (2004). the optimal view in which adequate colour Doppler flow signals appeared in the left ventricular outflow tract analysis and this was possible in only 44% of patients. Using the
The year in anaesthesia and critical care
216
automated cardiac output measurement method, the authors obtained good correlation and agreement for cardiac output measurements in the mitral flow analysis compared with thermodilution. These data suggest that automated cardiac output measurement by transoesophageal echocardiography at the mitral annulus can measure cardiac output accurately in cardiac surgical patients. It is of note that this method only calculates forward flow during the systolic and diastolic periods. Thus, it is not applicable in the presence of valvular regurgitation. The authors excluded four patients from their analysis because of ventricular arrhythmias.
Accurate, non-invasive continuous monitoring of cardiac output by wholebody electrical bioimpedance Cotter G, Moshkovitz Y, Kaluski E, et al. Chest 2004; 125: 1431–40 BACKGROUND. The authors stated the need for a simple, reliable, non-invasive and continuous method for cardiac output measurement in order to enable its use in patients with different degrees of cardiac dysfunction and in diverse settings. They examined the use of a new non-invasive, continuous whole-body electrical bioimpedance (WBEB) system for measuring cardiac output. Bioimpedance is based on the concept that, after transmitting a small electrical current through the body, an impedance to its transmission (resistivity or R) can be measured. This resistivity is called bioimpedance. The aim of this study was to test the validity of this noninvasive cardiac output system/monitor in a cohort of cardiac patients. In a prospective, double-blind study the cardiac output determinations measured by the WBEB system and thermodilution were compared in 122 patients in three different groups: during cardiac catheterization (n=40), before, during and after coronary bypass surgery (n=51) and while being treated for acute exacerbation of congestive heart failure (CHF) (n=31). INTERPRETATION. The data suggest that the WBEB system offers a simple, noninvasive, reliable and continuous measurement of the cardiac index in cardiac patients, with particular emphasis on CHF. This observation needs further confirmation in view of the long list of limitations. Comment This was a study showing the accuracy of a continuous whole-body bioimpedance system for measuring cardiac output. It is important to note that it is not possible to use the WBEB system in patients after major abdominal and thoracic surgical resections, particularly those that include major rapid shifts in fluid distribution between the intravascular and extravascular space or in the presence of significant oedema because, in such haemodynamic situations, the baseline impedance will routinely become distorted, preventing accurate measurement of the systolic impedance variation. Further limitations include diseases of the aorta and aortic valve, shunts, restlessness or arrhythmias. In other
Less invasive cardiac output monitoring
217
words, due to the limitations of the technique, the WBEB system cannot be recommended in many surgical patients in whom the anaesthetist might want to measure cardiac output in the perioperative situation. Four hundred and eighteen paired cardiac output measurements were obtained. The overall correlation between the WBEB cardiac index and the thermodilution cardiac index was r=0.886 and r2=0.79, with a bias of 0.0009±0.684 l/min and this finding was consistent within each group of patients (Table 9.2). The thermodilution readings were 15% higher than the WBEB readings when the cardiac index was <1.5 l/min/m2 and 5% lower than the WBEB readings when the cardiac index was >3 l/min/m2. Of note, 26 out of 51 (>50%) coronary artery bypass graft (CABG) patients had incomplete data (less than five paired measurements for each patient). These data suggest that the WBEB system offers a simple, non-invasive, reliable and continuous measurement of cardiac output in cardiac patients, with particular emphasis on CHF. In view of the numerous likely limitations of the WBEB system in the perioperative period its usefulness needs to be carefully considered.
Table 9.2 Results of using the new non-invasive, continuous whole-body bioimpedance system (NICaS; NI Medical, Hod-Hasharon, Israel) for measuring cardiac output during cardiac catheterization (n=40), before, during and after coronary bypass surgery (n=51) and while being treated for exacerbation of acute CHF (n=31): linear regression analysis, bias and precision in the three clinical groups and in the whole cohort Sample
Intercept Slope r2
Whole
0.18
0.92
0.79 −0.0009
−0.6849±0.6831
Catheterization 0.45
0.84
0.77 0.0040
−0.7134±0.6393
CABG
0,08
0.96
0.81 −0.0247
−0.6789±0.7331
CHF
0.29
0.89
0.72 0.0271
−0.6840±0.6920
Bias (mean betweenmethod difference) (l/min)
Precision (mean±SD) (l/min)
Source: Cotter et al. 2004
Evaluation of a new continuous cardiac output monitor in off-pump coronary artery surgery Leather HA, Vuylsteke A, Bert C, et al. Anaesthesia 2004; 59: 385–9 BACKGROUND A new system that allows the continuous measurement of
The year in anaesthesia and critical care
218
cardiac output (TruCath™ PAC and TruCCOMS™ monitor, Aortech International plc, UK) was recently introduced. The authors evaluated this new, ultra-fast response, continuous cardiac output monitor in 34 adult patients undergoing OPCAB graft surgery. The patients’ cardiac output was measured using triplicate cold bolus PAC thermodilution as the comparative standard at fixed time-points during surgery and during dobutamine infusion. INTERPRETATION. These data show that, in the clinical setting of cardiac surgery, the new continuous cardiac output monitor underestimates cardiac output at higher values and overestimates low cardiac outputs. Most importantly, this new continuous cardiac output monitor failed to detect dobutamine-induced changes in cardiac output accurately. This is a major limitation of the device as a clinical monitoring tool and it cannot be recommended in its current form. Comment In this study Leather et al. validated the new continuous cardiac output monitoring system by comparison with the cold bolus thermodilution technique. The TruCath™/ TruCCOMS™ system derives cardiac output continuously and instantaneously from the amount of energy required for heating a filament to maintain a fixed blood temperature gradient between two thermistors located distally in a PAC. The response time to changes in cardiac output is, in theory, a matter of seconds. In addition, the principle of measurement is not affected by tricuspid regurgitation because mass flow is measured at the level of the pulmonary trunk. This new technology has not yet been extensively validated in vivo. OPCAB surgery is associated with acute changes in cardiac output and some form of invasive cardiac output monitoring is part of standard care in many centres. Because of the rapid nature of the changes in cardiac output in this type of surgery, a fastresponse cardiac output monitor would offer significant advantages to the anaesthetist over the intermittent bolus technique, providing it is accurate. The authors suggested that the equation used for the calculation in the current device is based on in vitro calibration data and assumes an almost linear relationship between energy input and cardiac output. In the present study, they found that this relationship was inappropriate at values above and below 4 l/min. Regression analysis of the cardiac output data suggested no significant correlation between the cardiac output derived from the TruCath™/TruCCOMS™ continuous cardiac output monitor and that of conventional cold bolus thermodilution (y=0.14x+3.34; r2=0.03). Bland—Altman analysis of the TruCath™/ TruCCOMS™ continuous cardiac output monitor versus cold bolus thermodilution data showed a bias of 0.4 l/min with limits of agreement of 2.5 and 1.7 l/min. Dobutamine caused an increase in the mean (±SD) cardiac output from 3.1±0.6 to 4.2 ± 1.1 l/min as recorded by cold bolus thermodilution. Although the mean cardiac output value recorded by the study TruCath™/TruCCOMS™ device also increased, regression analysis showed that the accuracy in detecting individual changes in cardiac output was very poor (y=0.18x+0.45; r2=0.13). Bland—Altman analysis of the dobutamine-induced change in cardiac output, as recorded by the TruCath™/TruCCOMS™ device versus cold bolus thermodilution, showed a bias of 0.4 l/min, with limits of agreement of 1.2–2 l/min. Regression analysis
Less invasive cardiac output monitoring
219
of the Bland-Altman plot suggested that the degree of inaccuracy of the study device was linearly related to the magnitude of the change in cardiac output. These data suggest that that this new continuous cardiac output monitor system proved inaccurate in the determination of cardiac output in cardiac surgical patients when using cold bolus thermodilution as the comparator. The study device tended to overestimate low values and underestimate high cardiac output values. Indeed, the system did not adequately detect the changes in cardiac output during dobutamine infusion, with the degree of error being linearly related to the magnitude of the cardiac output change.
analysis, and real-time Comparison of esophageal Doppler, pulse contour pulmonary artery thermodilution for the continuous measurement of cardiac output Bein B, Worthmann F, Tonner PH, et al. J Cardiothorac Vasc Anesth 2004; 18: 185–9 BACKGROUND. Continuous measurement of cardiac output is of great importance in the critically ill. However the use of pulmonary artery thermodilution has been questioned because of possible complications associated with right heart catheterization. Furthermore, measurements are delayed in the continuous mode during rapid haemodynamic changes. The authors compared the above-mentioned new PAC continuous cardiac output device (Aortech, UK) enabling real-time updating of cardiac output with two different, less invasive methods of continuous cardiac output determination: oesophageal Doppler and pulse contour analysis. This prospective observational study was undertaken in ten ASA IV (American Society of Anesthesiologists classification grade IV) patients undergoing cardiovascular surgery. INTERPRETATION. The data suggest an overestimation of low cardiac outputs and an underestimation of high cardiac outputs with the new PAC continuous cardiac output device compared to oesophageal Doppler and pulse contour analysis. Unfortunately, the authors did not use pulmonary arterial thermodilution as the reference for these measurements, but instead compared the continuous methods with each other. Thus, their observation needs further confirmation. Comment In this study, Bein et al. compared three different methods of continuous cardiac output measurement. It is of concern that they did not compare those with the pulmonary arterial thermodilution as the reference for these measurements. The authors concluded that the agreement between the continuous cardiac output method and both less invasive measurements was clinically acceptable. These results are in contrast to the data reported by Hullet et al. (comparing thermodilution-derived cardiac output measurements with oesophageal Doppler cardiac output) and by Leather et al. (comparing PAC thermodilution-derived continuous cardiac output with continuous cardiac output derived
The year in anaesthesia and critical care
220
from the amount of energy required to maintain a temperature gradient between two thermistors within a PAC). The continuous cardiac output measurements were analysed using a Bland—Altman plot. The bias between the PAC continuous cardiac output and pulse contour cardiac output and Doppler-derived cardiac output was (mean±SD) −0.71±1 l/min versus −0.15±1.09 l/min and that between Doppler-derived cardiac output and pulse contour cardiac output was −0.58±1.06 l/min. The bias was not significantly different between methods, nor were the comparative values before and after cardiopulmonary bypass (P>0.05). Linear regression analysis of the PAC continuous cardiac output/pulse contour cardiac output and PAC continuous cardiac output/ Doppler-derived cardiac output Bland—Altman plot yielded a negative slope representing an overestimation of low cardiac outputs and an underestimation of high cardiac outputs by PAC continuous cardiac output compared with pulse contour cardiac output and Doppler-derived cardiac output (r2=0.48 and 0.27, respectively). The authors concluded that there was a clinically acceptable agreement between the PAC-based continuous cardiac output method and both less invasive measurements. In addition, there were no adverse events associated with the use of either device.
A comparative evaluation of thermodilution and partial CO2 rebreathing techniques for cardiac output assessment in critically ill patients during assisted ventilation Rocco M, Spadetta G, Morelli A, et al. Intensive Care Med 2004; 30: 82–7 BACKGROUND. The NICO2 (partial non-invasive carbon dioxide rebreathing technique) device (Novametrix Medical System, CT, USA), which uses the Fick principle and partial carbon dioxide (CO2) rebreathing, may have a role in the haemodynamic monitoring of critically ill patients following endotracheal intubation. The system consists of a disposable device inserted between the ventilator circuit and the endotracheal tube, a computer and a pulse oximeter. The disposable device consists of an adjustable rebreathing loop and a pneumatically controlled valve that can direct flow through this loop, an infrared light absorption CO2 sensor and an airflow sensor. On a breath-by-breath basis CO2 elimination is calculated from the flow and CO2 concentration at the airway opening. The arterial CO2 content is estimated from the PETCO2 (ET=endtidal) and CO2 dissociation curve. The Fick principle then allows estimation of the blood flow through the lungs (cardiac output). The device has been validated in surgical patients receiving controlled mechanical ventilation during sedation and paralysis. The purpose of this study was to evaluate the reliability and clinical value of the partial CO2 rebreathing technique for measurement of cardiac output (NIC02 cardiac output) compared with the thermodilution-derived cardiac output technique in twelve mechanically ventilated critically ill patients with high (n=6) and low (n=6) pulmonary shunt fractions. INTERPRETATION. These data suggest that the partial CO2 rebreathing technique is reliable in measuring cardiac output in non surgical critically ill patients affected by
Less invasive cardiac output monitoring
221
diseases causing low levels of pulmonary shunt, but underestimates it in patients with a shunt higher than 35%. This is an important limitation, because it is not possible to assess a patient’s degree of shunt clinically. Comment In this study Rocco et al. investigated another method of non-invasive cardiac output measurement allowing, theoretically, a real-time determination of shunt fraction using pulse oximetry, FIO2 (fraction of inspired oxygen) and Nunn isoshunt plots. The system is completely non-invasive and easy to use, placing a sensor between the endotracheal tube and the breathing circuit Y piece. Although the accuracy of the partial CO2 rebreathing technique has been established in surgical patients mechanically ventilated during anaesthesia and the post-operative phase, its use in critically ill patients is somewhat limited due to a number of theoretical assumptions that may not apply in these patients. Specifically, the partial CO2 rebreathing technique uses a non-invasive approach for estimating the shunt fraction by adapting Nunn isoshunt plots. The lack of knowledge regarding haemoglobin (Hb) P50 could be a major problem in calculating shunts because it affects the real value of Hb saturation. The leftward or rightward shift of the Hb dissociation curve changes the difference in capillary to venous oxygen content (CcO2– CvO2) and subsequently the shunt computation. CcO2 is considered to be equal to PaO2 (arterial oxygen pressure), assuming a particular shape of the Hb dissociation curve. Unfortunately, this assumption is not true in the presence of significant Hb P50 alterations. The mean cardiac output values of the overall population of twelve patients were as follows: thermodilution 7.27±2.42 l/min and NICO2 6.10±1.66 l/min, with a moderate correlation (r2=0.62) and bias on the Bland—Altman test of±1.5 l/min. The values in the low-shunt group were 5.22±1.28 and 5.23±1.32 l/min, respectively, with a high correlation (r2=0.90) and bias of 0.01±0.4 l/min and those in the elevated-shunt group were 9.32±1.23 and 6.97±1.53 l/min, respectively, with a poor correlation (r2=0.38) and a bias of −2.3±1.2 l/min. These data suggest that the partial CO2 rebreathing technique for the measurement of cardiac output offers a simple, non-invasive, reliable and continuous measurement of cardiac output only in patients with a low shunt. In view of the numerous limitations of the partial CO2 rebreathing technique relevant to the critical care area and the difficulty in identifying the correct group of patients with a low shunt, its use needs to be treated with caution.
Pulsed dye densitometry with two different sensor types for cardiac output measurement after cardiac surgery: a comparison with the thermodilution technique Hofer CK, Buhlmann S, Klaghofer R, Genoni M, Zollinger A. Acta Anaesthesiol Scand 2004; 48: 653–7
The year in anaesthesia and critical care
222
BACKGROUND. Assessment of cardiac output by the indocyanine green dye dilution (IDD) technique with transcutaneous signal detection may be a less invasive alternative to the PAC. The aim of this study was to determine the accuracy and reliability of the DDG2001 analyser (Nihon Kohden Corp, Tokyo, Japan) using a finger and a nose sensor as compared with the thermodilution technique by a PAC. INTERPRETATION. Cardiac output measured by the IDD technique with transcutaneous signal detection has been suggested as yet another less invasive alternative to the PAC. Indocyanine green-sensitive optical probes can be attached to a finger or the nose. Following injection of indocyanine green, the non-invasive test instrument could record the arterial concentration of the dye optically. Its performance was limited due to signal detection failure and this study showed a systematic bias using this less invasive technique for cardiac output measurement compared with the thermodilution technique. Comment The authors showed a minimal underestimation of cardiac output in 31 patients following elective cardiac surgery by the IDD technique (i.e. negative mean bias) for the nose sensor, but considerable imprecision using the finger sensor. They also reported that the cardiac output measurement using the finger probe was too variable during haemodynamically stable conditions for clinical use. One hundred and twenty-four sets of data were obtained from the 31 patients (offpump cardiac surgery=61%, ΛSΛ=3, female/male=9/22, age=62±10 years and body mass index=29±4 kg/m2). The failure rate of signal detection was 33% for the finger sensor and 9% for the nose sensor, resulting in a total of 54 complete sets of pulsed dye dilution measurements using the finger sensor and 48 sets using the nose sensor available for statistical analysis. Bland-Altman analysis revealed a mean bias for the finger sensor of 0.5 l/min/m2 and a mean bias for the nose sensor of 0.1 l/min/m2 compared with the iced water bolus technique. The linear correlation between the finger sensor and iced water bolus technique (r=0.20) was found to be inferior to the correlation between the nose sensor and iced water bolus technique (r=0.45). These data showed a systematic bias using the pulsed dye dilution technique for cardiac output measurement compared with the thermodilution technique. Furthermore, performance was limited due to signal detection failure. Therefore, the authors concluded that the pulsed dye dilution technique by the DDG2001 analyser could not be recommended as a substitute for the thermodilution technique in routine monitoring of cardiac output after cardiac surgery.
Severity of illness and risk of death associated with pulmonary artery catheter use Chittock DR, Dhingra VK, Ronco JJ, et al. Crit Care Med 2004; 32: 911–15 BACKGROUND Controversy exists over whether critically ill patients benefit
Less invasive cardiac output monitoring
223
from or are harmed by the use of a PAC. The question arises, however, as to what effect the severity of illness has on the efficacy of the PAC. Could a patient with a lower severity of illness benefit less from the catheter, as opposed to a patient who is sicker? The role that the severity of illness plays in the outcomes of patients exposed to the PAC has not been formally examined. INTERPRETATION. This study is an important step forward, not because it investigates the validity of a cardiac output monitoring system, but because it indicates a potential benefit of using the PAC in a specific group of patients, namely a decreasing mortality rate in the most severely ill, while increasing it in a population with a lower severity of illness. These findings underscore the necessity of careful consideration of who should be managed with a cardiac output monitoring system. Comment The authors of this landmark observational cohort study examined the association between the use of a PAC and mortality rate in 7310 critically ill patients, dividing the cohort into critically ill patients with a higher versus a lower severity of illness. The
Fig. 9.4 Mortality by PAC status in critically ill patients for 25th percentiles of the APACHE II scores. *P<0.05. Source: Chittock et al. (2004). results of the effect of use of a PAC on the in-hospital mortality rate, when stratified for severity of illness in critically ill adults, are presented. Data were extracted for all intensive care unit (ICU) admissions spanning the 10 years from March 1988 to March 1998 in the authors’ institution. A single, trained data collector collected all data
The year in anaesthesia and critical care
224
prospectively. The study was carefully conducted and is one of the most important studies of the last 5 years in this field. The main outcome measure was the hospital mortality rate, controlled by multivariable logistic regression within four patient groups based on the severity of illness. Cut-offs for the severity of illness were chosen based on 25th percentiles of the Acute Physiology and Chronic Health Evaluation (APACHE) II score. Logistic regression analysis demonstrated no increased risk of death associated with exposure to the PAC in the population as a whole. The associated odds ratio (OR) of hospital death for the entire cohort was 1.05 (95% CI 0.92–1.21). Using the PAC in the group of patients with the highest APACHE II score quartile was associated with a decreasing mortality rate in the most severely ill, while increasing it in the patients with the lower APACHE II score quartiles (APACHE II <18, OR 2.47; 95% CI 1.27–4.81; APACHE II 18–24, OR 1.64; 95% CI 1.24–2.17; APACHE II 25–31, OR 1.00; 95% CI 0.80–1.24 and APACHE II >31, OR 0.80; 95% CI 0.64–1.00) (Fig. 9.4). These data suggest that the use of the PAC may decrease the mortality rate in the most severely ill while increasing it in a population with a lower severity of illness. The authors concluded that these findings underscore the necessity of examining the effect of the severity of illness in future randomized controlled trials. Conclusion There remains little doubt that measuring cardiac output in critically ill patients is important, but it is still unclear how this can be best achieved. An ideal cardiac output monitor should be non-invasive, accurate, reliable, easy to use, provide data continuously and offer additional variables. Currently, the most common method for measurement of cardiac output is based on the thermodilution technique, requiring the insertion of a PAC, although routine use of a PAC for determining cardiac output has become the subject of controversial debate |3–5|. There are several non-invasive or less invasive methods available. Their operational principles, advantages and limitations have been demonstrated in the studies presented. Limited accuracy in critically ill patients is one of the remaining hurdles for the broader use of some less invasive cardiac output measurement techniques. Based on the available evidence some of the less invasive cardiac output measurement techniques, e.g. partial CO2 rebreathing, whole-body bioimpedance and pulsed dye dilution, cannot currently be recommended as a substitute for the PAC in routine monitoring of cardiac output in the ICU or operating theatre. The newly developed PAC TruCath™ has serious shortcomings because it failed to detect dobutamine-induced changes in cardiac output accurately. Transoesophageal echocardiography and arterial thermodilution and derived pulse contour analysis may represent alternative, less invasive techniques for cardiac output measurement. Both devices offer additional variables and arterial thermodilution and derived pulse contour analysis provide cardiac output data continuously. These devices show promise in the search for a device both robust and practical for use in the ICU and operating theatres. As is so often the case, further research is needed in order to elucidate the ability of these techniques to improve outcomes.
Less invasive cardiac output monitoring
225
References 1. Reinhart K, Sakka SG, Meier-Hellmann A. Haemodynamic management of a patient with septic shock. Eur J Anaesthesiol 2000; 17: 6–17. 2. Pinsky MR. Hemodynamic monitoring in the intensive care unit. Clin Chest Med 2003; 24: 549– 60. 3. Connors Jr AF, Speroff T, Dawson NV, Thomas C, Harrell Jr FE, Wagner D, Desbiens N, Goldman L, Wu AW, Califf RM, Fulkerson Jr WJ, Vidaillet H, Broste S, Bellamy P, Lynn J, Knaus WA. The effectiveness of right heart catheterization in the initial care of critically ill patients. SUPPORT Investigators. JAMA 1996; 276: 889–97. 4. Weil MH. The assault on the Swan-Ganz catheter: a case history of constrained technology, constrained bedside clinicians, and constrained monetary expenditures. Chest 1998; 113: 1379– 86. 5. McLean AS. Echocardiography assessment of left ventricular function in the critically ill. Anaesth Intensive Care 1996; 24: 60–5. 6. Polanczyk CA, Rohde LE, Goldman L, Cook EF, Thomas EJ, Marcantonio ER, Mangione CM, Lee TH. Right heart catheterization and cardiac complications in patients undergoing noncardiac surgery: an observational study. JAMA 2001; 286: 309–14. 7. Yu DT, Platt R, Lanken PN, Black E, Sands KE, Schwartz JS, Hibberd PL, Graman PS, Kahn KL, Snydman DR, Parsonnet J, Moore R, Bates DW; AMCC Sepsis Project Working Group. Relationship of pulmonary artery catheter use to mortality and resource utilization in patients with severe sepsis. Crit Care Med 2003; 31: 2734–41. 8. Bernard GR, Sopko G, Cerra F, Demling R, Edmunds H, Kaplan S, Kessler L, Masur H, Parsons P, Shure D, Webb C, Weidemann H, Weinmann G, Williams D. Pulmonary artery catheterization and clinical outcomes: National Heart, Lung, and Blood Institute and Food and Drug Administration Workshop Report. Consensus Statement. JAMA 2000; 283: 2568–72. 9. Sandham JD, Hull RD, Brant RF, Knox L, Pineo GF, Doig CJ, Laporta DP, Viner S, Passerini L, Devitt H, Kirby A, Jacka M; Canadian Critical Care Clinical Trials Group. A randomized, controlled trial of the use of pulmonary-artery catheters in high-risk surgical patients. N Engl J Med 2003; 348: 5–14. 10. Wesseling K, De Wit B, Weber J, Smith N. A simple device for the continuous measurement of cardiac output. Adv Cardiovasc Physiol 1983; 5: 16–52. 11. Goedje O, Hoeke K, Lichtwarck-Aschoff M, Faltchauser A, Lamm P, Reichart B. Continuous cardiac output by femoral arterial thermodilution calibrated pulse contour analysis: comparison with pulmonary arterial thermodilution. Crit Care Med 1999; 27: 2407–12. 12. Sakka SG, Reinhart K, Meier-Hellmann A. Comparison of pulmonary artery and arterial thermodilution cardiac output in critically ill patients. Intensive Care Med 1999; 25: 843–6. 13. Segal E, Katzenelson R, Berkenstadt H, Perel A. Transpulmonary thermodilution cardiac output measurement using the axillary artery in critically ill patients. J Clin Anesth 2002; 14: 210–13.
10 Depth of anaesthesia monitoring HUGO VEREECKE, MICHEL STRUYS Introduction The use of electroencephalography (EEG) or mid-latency auditory evoked potentials (MLAEPs) as a source of information for measuring the hypnotic component of anaesthesia (HCOA) is becoming more and more widespread. In recent years this has led to a vast number of publications investigating the applicability of commercial devices in anaesthetic practice. Although the results are promising, particularly in pharmacology research, certain pitfalls and limitations remain a problem for any monitor using raw EEG or MLAEP monitoring. An important pitfall to consider is the difference in the ‘depth of anaesthesia’ and ‘HCOA’. There is not yet a consensus on the exact interpretation of these concepts. Therefore different articles may use terminology with a slightly different interpretation, causing problems when comparing results. There are many definitions for describing the clinical state of ‘anaesthesia’. In this chapter we prefer using ‘depth of the HCOA’ because recent technology does focus on the quantification of the hypnotic effects of anaesthetic drugs. ‘Depth of anaesthesia’ may also imply other factors that are needed for enabling planned surgery (e.g. immobility, muscle relaxant effects and analgesic effects). The HCOA does not include these factors, but relates only to hypnosis. During 2004 a large proportion of publications focused on the fundamental questions that might explain the unexpected behaviour of an HCOA monitor in certain clinical conditions. Some of these articles will be reviewed in the first section. The subjects range from ‘the effects of age on the raw EEG’ to ‘the correlation between the power calculation of small frequency domains in the EEG and a clinical end-point (detection of awareness)’ or ‘the interpretation of electromyography (EMG) as a surrogate end-point of anaesthetic drug effect’. In a review article by Jensen et al. a new terminology is introduced for clarifying possible causes of bias in depth of anaesthesia research. This might be used for optimizing the methodology in research protocols and for avoiding pitfalls in the clinical and pharmacological end-points chosen for evaluating an HCOA monitor’s performance. A second field of interest is the avoidance of ‘awareness’. In an editorial in Anesthesiology, Kalkman and Drummond raised the concern that titration of anaesthetic drugs © Atlas Medical Publishing Ltd
Depth of anaesthesia monitoring
227
to the limits of unconsciousness might lead to a potential increase in the incidence of awareness |1|. The B-Aware Trial is the first large multicentre trial investigating this hypothesis. The results were published this year. Finally, the implementation of EEG- or MLAEP-based monitors in pharmacology research has led to an accelerated progress in pharmacokinetic-pharmacodynamic and drug interaction knowledge |2, 3|. By using EEG- or MLAEP-extracted information from a specific patient, the pharmacological compartmental models become more individualized and more practical for clinical use. This information can be implemented in ‘closed loop systems’ or in the development of more accurate interaction theories involving hypnotics and opioids |4|. Although this research is very promising for future developments in anaesthesia, the clinical implementation of the results will probably take some more years. When reviewing publications on depth of anaesthesia monitors, a wide variety of topics are studied each year. However, many of these studies are characterized by poor methodology or limited statistical significance. As the topics in this area of research are diverse this chapter can never be considered to be complete. The selected papers are chosen because of their informative value or the high quality of the methods used. Fundamental research In fundamental research on depth of hypnosis monitoring, we recognise two approaches. One group of scientists focuses on the molecular, anatomical and/or physiological background of the clinical end-points of depth of hypnosis. The expanding field of neuroscience reveals a progressive understanding of consciousness and memory formation. Although it is crucial to explore these clinical features in as detailed a way as possible, this approach does not always provide information on the individual response of a patient to the administration of a hypnotic drug. Another approach in fundamental research does focus on the optimization of pharmacokinetic—pharmacodynamic population models, combined with data input from patient-specific EEG or MLAEPderived parameters. The ultimate goal of these models is to obtain a more individualized titration of hypnotic drugs without the need to understand fully the molecular processes evoked by the hypnotics in a particular patient. In this view, it is mandatory to explore the limits of EEG- and MLAEP-derived indices as a monitor for hypnosis, before they can fully act as an input for anaesthesia control. In the following section, some studies on this subject are reviewed.
Age-related effects in the EEG during propofol anaesthesia Schultz A, Grouven U, Zander I, Beger FA, Siedenberg M, Schultz B. Acta Anaesthesiol Scand 2004; 48: 27–34 BACKGROUND. This trial investigated age-related differences in the spectral composition of an EEG during an intravenous bolus of 2 mg/kg of propofol. A changing EEG spectrum with age can have important implications for the accuracy
The year in anaesthesia and critical care
228
of EEG-based HCOA monitors. Sixty female patients of between 22 and 85 years of age were monitored continuously with raw EEG analysis extracted from the EEG monitor Narcotrend™ (MT Monitor Technik, Bad Bramstedt, Germany). In addition, a retrospective evaluation of the propofol consumption was performed using a target-controlled infusion based on data of a subset of 546 patients included in a multicentre study with 4630 patients monitored by the Narcotrend™. INTERPRETATION. The EEG was visually assessed in epochs of 20 s according to a scale ranging from A (awake) to F (very deep hypnosis) as described by Schultz et al. |5|. Visual EEG classification, classical spectral parameters and induction times were compared between different age groups. Patients older than 70 years reached deeper EEG stages than younger patients during induction and needed more time to reach a light EEG stage. The total power in patients older than 70 years during deep EEG stages was significantly smaller due to distinctly smaller absolute power in the delta frequency band. In the awake patient a shift in EEG frequencies towards more slow waves was detected in the older age group. The authors concluded that an efficient automatic assessment of the EEG during anaesthesia needs a multivariable approach accounting for age effects on the EEG. The retrospective analysis of patients monitored with Narcotrend™ revealed that, as age increases, a decreasing blood concentration of propofol is required for obtaining a comparable predefined level of hypnosis (stage D2/E0). Comment This study was selected for this review because of its potential importance in the interpretation of EEG-derived indices in older patients. By combining knowledge of the mathematical approach to EEG-derived HCOA indices and the behaviour of the raw EEG in older patients, it becomes possible to define the limitations of EEG interpretation for measuring the HCOA in the overall patient population. The most prominent change with increasing age in the EEG is a decrease in the total power and a shift from faster (alpha and beta) towards slower (theta and delta) in the frequency content of the power spectrum. This might explain the occurrence of surprisingly low EEG-derived indices in the awake geriatric patient. However, it is not clear from these data whether the ability of HCOA monitors using EEG-derived indices for discriminating between levels of hypnotic drug effect becomes less accurate with advancing age. Minto et al. |6| and Schnider et al. |7| demonstrated increased pharmacodynamic sensitivity to the cerebral effects of anaesthetic drugs with advancing age. Therefore, it remains unclear whether the limits of the EEG characteristics for an adequate level of hypnosis should be adapted for the geriatric patient as is suggested by the authors. It might well be that a difference in EEG power exists in the awake state between older and younger patients. However, this does not imply that the correlation between EEG effects and clinical hypnotic effects has changed with advancing age. This trial indeed concluded that there is a difference in EEG effects in both the awake and anaesthetized states, but the propofol dose used was not equipotent for both populations. The standard induction dose of propofol caused a larger change in the EEG response in the elderly compared to the younger subjects, but this only reflects the deeper clinical level of hypnosis achieved (reflected in the longer recovery
Depth of anaesthesia monitoring
229
times, etc.). In fact, this confirms that an EEG is a good quantifier for the HCOA in both younger and older populations except in the awake state. As a secondary result, this trial confirmed the increased sensitivity to propofol effects in older patients, as has been more thoroughly investigated by Schnider et al. |7|. The results of the retrospective pharmacological analysis in this study should be interpreted cautiously because it involved a variety of surgical procedures and a non-steady-state titration method.
Does bispectral analysis of the electroencephalogram add anything but complexity? Miller A, Sleigh JW, Barnard J, Steyn-Ross DA. Br J Anaesth 2004; 92: 8–13 BACKGROUND. This article questioned the potential benefit of using bispectral analysis, as used in the bispectral index (BIS, Aspect Medical Systems Inc., Newton, MA, USA), compared to a classical power spectrum analysis, as a basis for calculating a logarithmic ratio of high-frequency components (40–47 Hz) of the EEG versus the total power (1–47 Hz). This ratio of fast versus slow (or total) power is used for index calculation algorithms in several HCOA monitors, which is different from the bispectral index. A decrease in conscious level is accompanied by an increase in the low-frequency power, causing a shift in this ratio. To investigate the influence of contamination by the effects of signal amplitude in the bispectrum, a third parameter was analysed using bicoherence as an indicator of phase coupling. By doing so, the authors attempted to detect the amount of information based on phase coupling alone. This study was performed on raw EEG data obtained during routine induction of anaesthesia in 39 American Society of Anesthesiologists (ASA) Class I and II patients. INTERPRETATION. The authors focused on an important component of the bispectral index, the SynchFastSlow. They compared the SynchFastSlow with a new ratio called the PowerFastSlow, which is based on the classically used approach for EEG analysis, the Fourier power analysis. The main conclusion in this study was that thirdorder statistics (bispectral analysis) do not add any extra information compared to a second-order statistical approach (Fourier analysis) when used as a discriminating variable for the clinical states ‘awake’ versus ‘anaesthetized’. Despite the large number of studies showing promising results for the bispectral index, other EEG-based parameters might be as informative without implementing the bispectrum as such. Comment This article contains some crucial information on EEG interpretation during anaesthesia. The mathematical approach for rejecting the hypothesis is high ranked, but one might argue that the raw data acquisition is not performed in a standardized way. The patients received different amounts and/or types of hypnotic and analgesic drugs for induction, thereby creating a heterogeneous study population with regard to the level of hypnotic drug effect. Moreover, the definition of ‘awake’ versus ‘anaesthetized’ is not based on a
The year in anaesthesia and critical care
230
direct clinical end-point, but only on the timing of EEG acquisition, that being at the beginning or the end of an EEG analysis period of 160 s during induction. However, if the conclusions of this study prove to be consistent across the complete range of hypnotic drugs, it might be important for the development of new EEG-based hypnotic drug effect monitors. The ‘bispectral’ input in the bispectral index algorithm might not be mandatory for the development of an appropriate clinically useful HCOA monitor. Despite the shortfalls in the protocol this basic research improves the understanding of the technology used for automatic EEG interpretation. This enables anaesthetists to optimize their interpretation of the behaviour of hypnotic drug effect monitors.
Awareness and EEG power spectrum: analysis of frequencies Dressler 0, Schneider G, Stockmans G, Kochs EF. Br J Anaesth 2004; 93(6): 806–9 BACKGROUND. Power spectral analysis is used in many commercially available EEG monitors for quantifying the pharmacological effects of anaesthetics on the brain and for detecting the level of sedation. This study evaluated the performance of a subset of frequency domains, each with an interval of 1 Hz, extracted from the overall EEG spectral analysis, for their ability to detect awareness defined by a positive response during the isolated forearm technique. EEG data were obtained from 39 patients receiving propofol—remifentanil- or sevoflurane—remifentanilbased anaesthesia by means of a Physiometrix™ PSA-4000 EEG monitor (Baxter Healthcare Corporation, IL, USA). While the continuous infusion of remifentanil was maintained at a constant rate, the hypnotic drug was repeatedly increased and decreased in order to obtain sequential periods of responsiveness and unresponsiveness. Forty-five EEG segments of the awake patient state were compared with 45 segments of the unresponsive state in this trial. To compare these segments a modified prediction probability (rPK) was used for obtaining performance values independent of the negative or positive nature of the correlation. Normally, a high prediction probability (PK) indicates that the measured value will increase (decrease) every time the hypnotic level changes towards a lighter (deeper) level. The authors considered it not to be important for a certain frequency bin to decrease or increase as a detector of awareness as long as the response itself was consistent in the majority of data couples. Therefore they modified the PK to rPK, which enables a variable to have a high performance when values change, independent of the direction of that change. INTERPRETATION. The authors concluded that EEG frequencies close to 20 Hz have the lowest ability for discriminating between the two clinical end-points ‘squeezing hand on demand’ versus ‘unresponsive’. Power calculations on the frequency domains below this minimum had a negative correlation with the detection of awareness. The highest rPK values were seen in the frequency domain between 36 and 127 Hz. This included electrical activity evoked by both the EEG and EMG activity. At 50 Hz and on the harmonic at 100 Hz a decrease was seen in predictability The same was seen at 78
Depth of anaesthesia monitoring
231
Hz, which was probably caused by the impedance check performed by the Physiometrix™ PSA-4000 device. This suggests that a poor signal quality severely decreases the predictability. The high performance of the high-frequency domains can be ascribed to both EEG and EMG activity as they had a large overlap in the frequency domain. No neuromuscular blocking agents were given and patients with a positive response had higher levels of EMG activity detected than unresponsive patients. However, it was impossible to determine which factor had the largest impact on this result. The authors concluded that high frequencies might be useful as a component for detecting awareness. Comment Although this study attempted to increase the understanding of EEG behaviour during anaesthesia, there are some major remarks to be made about the methodology and interpretation of the results. Firstly, the anaesthetic regimens used were not truly comparable. It is possible that the accuracy of EEG detection of awareness varies when different hypnotic agents are used. It is not clearly stated how many data couples were extracted from the sevoflurane group compared to the propofol group. Moreover, as more than one data couple was extracted from the same patient, the data were not truly independent. Secondly, the use of the isolated forearm technique as a detector of awareness is not 100% sensitive. This implies that false-negative results in the detection of awareness may have occurred and, thus, influenced the results of this study. Finally, one could question whether the poor results in certain frequency domains would be smoothed out when larger frequency domains are used for index calculation, which is the case in the commercialized devices. On the other hand, some information obtained with this study might have important implications when the analysis of the EEG beta band is included in an EEG index algorithm. An increase in the lower part of this domain (<21 Hz) indicates an increased probability of unresponsiveness, whereas an increase of the higher part (>21 Hz) indicates an increased probability of awareness. Due to the opposite responses to frequencies within the same bandwidth it might be reasonable to split the beta band at 21 Hz for further power calculations. In most monitors EMG activity ranges (40–80 Hz) are omitted for further analysis due to EMG interference. However, in this study no difference was seen in performance between power calculations in the 40–80 Hz and 80–127 Hz domains. The question remains whether we should consider EMG as noise in the data or whether it could be used as a surrogate end-point of hypnotic drug effect.
Pitfalls and challenges when assessing the depth of hypnosis during general anaesthesia by clinical signs and electronic indices Jensen EW, Litvan H, Struys M, Martinez Vazquez P. Acta Anaesthesiol Scand 2004; 48: 1260–7
The year in anaesthesia and critical care
232
BACKGROUND. In this article the authors described the fundamental problems inherent in the clinical and electronic methods used for quantifying the HCOA. The end-points used in clinical and pharmacological studies are often difficult to compare or insufficiently validated, making it impossible to compare results. INTERPRETATION. Three concepts were defined in order to facilitate a description of the major pitfalls. Firstly, the real depth of hypnosis (DHreal), which is a theoretical concept that cannot be measured directly, was defined. It has to be estimated in a subjective way by the observed depth of hypnosis (DHobs) or in a more objective way by the electronic index of the depth of hypnosis (DHel-ind). This DHel-ind has an objective meaning and is reproducible with low dispersion, although it may be prone to bias. The anaesthetist uses a number of skills for assessing the DHobs. To generalize these skills, clinical scales such as the observers’ assessment of alertness and sedation scale have been developed and validated. In the ideal world those three concepts should describe hypnosis in a comparable way, but in clinical practice this is rarely the case. This article demonstrated the uncertainty of the DHobs due to the ambiguous correlation between the real hypnotic state (DHreal) and the clinical signs used. Some articles conclude that the bispectral index and an auditory evoked potentials (AEP)-derived index do not correlate well with the HCOA when using opioids. Such results are probably biased because of the use of clinical end-points in a DHobs such as the observers’ assessment of alertness and sedation scale. In addition, opioids might blunt the response to increasing stimuli given to the patient, thereby disturbing the detection accuracy of a DHel-ind. Finally the influence of muscular activity on the DHel-ind is extensively covered. However, as some of these phenomena are not fully understood there is still a need for further research. Comment By keeping in mind some of these pitfalls before performing a comparative trial on measuring the HCOA, the results might become more reproducible and comparable. However, because of the unknown nature of the DHreal it remains unclear how to quantify the deviation between the three variables defined by the authors. The lack of a real gold standard for hypnotic cerebral drug effects does lead to a multitude of surrogate endpoints that are compared with each other. A consensus for a more standardized approach to the comparison of HCOA monitors is necessary if real progress is to be made. Awareness This year a major breakthrough has been realized in the scientific value of using a HCOA monitor in clinical practice. The avoidance of awareness during an anaesthesia is difficult to investigate because of its rare incidence and the large numbers of patients needed to perform conclusive trials. In this section, the first sufficiently large randomized controlled multicentre trial investigating this subject is discussed.
Depth of anaesthesia monitoring
233
Bispectral index monitoring to prevent awareness during anaesthesia: the B-Aware randomised controlled trial Myles PS, Leslie K, McNeil J, Forbes A, Chan MT. Lancet 2004; 363: 1757–63 BACKGROUND. This was the first large, multicentre, randomized controlled trial investigating the ability of the bispectral index as an end-point for anaesthetic drug titration for decreasing the incidence of awareness in a high-risk population. The studied population consisted of surgical patients undergoing relaxant general anaesthesia with at least one risk factor for awareness: Caesarean section, high-risk cardiac surgery, acute trauma with hypovolaemia, rigid bronchoscopy, significant impairment of cardiovascular status, etc. A bispectral index-guided group of 1225 patients was compared to a routine practice group of 1238 patients. A blinded observer assessed the patients for awareness at three time-points post-operatively. A blinded independent committee assessed every report of awareness and coded it as being ‘awareness’, ‘possible awareness’ or ‘no awareness’. A variety of anaesthetic drugs were used, reflecting daily clinical practice. INTERPRETATION. The main result of this trial was a 74% absolute risk reduction in explicit awareness, with two (0.17%) confirmed cases in the bispectral index-guided group compared to eleven (0.91%) confirmed cases in the control group. The statistical significance was high with a P-value of 0.022. The results were less pronounced, but still significant (P=0.49) when analysing the ‘possible awareness’ cases, with 22 (1.8%) reported cases in the bispectral index group versus 27 (2.2%) reported cases in the control group. The number needed to treat in order to avoid an episode of awareness in this highrisk group was 138. This means a total cost of US$2200 at current costs of bispectral index monitoring at US$16 per patient. Comment Titration of anaesthetics to the limits of unconsciousness might increase the incidence of awareness |1|. The incidence of awareness remains low |8|. Therefore, large population studies are necessary for demonstrating a reduction in the incidence of awareness. The importance of this trial is linked to its magnitude. This blinded, controlled trial showed the safety of using the bispectral index as a preventive tool for awareness. One patient who was aware in the bispectral index group had high bispec-tral index values during a reasonable time period and awareness could probably have been avoided if anaesthetic management had been adjusted more rigorously in response to the bispectral index score. The other case showed bispectral index values in the range of 55–60. This suggests that awareness can arise when the bispectral index is at the upper limit of the recommended 40–60 range. It is consistent with the underlying derivation of the bispectral index, whereby several EEG variables are used for calculating a probability of patient wakefulness. This also explains why most patients with a bispectral index score of 60–70 will not be aware. One might speculate that a threshold of 55 could be even more efficient in preventing awareness. However, this cannot be concluded with certainty from
The year in anaesthesia and critical care
234
these data, as this was not the target score in the prospective setting of the trial. The threshold of 55 can only be advocated with a level of evidence comparable with the results from a retrospective trial. The pharmacological approach This section mainly discusses the use of HCOA monitors as a measure for a calculated anaesthetic drug concentration. By combining a known pharmacokinetic— pharmacodynamic model and multiple EEG or MLAEP based monitors as an endpoint, a research tool for comparing monitors with different units is created. This is important for depicting the accuracy of all monitors on a wide range of anaesthetic drug effects, using a variety of anaesthetic drugs. In a next step, the accuracy to detect clinical features, such as loss of consciousness or loss of response to noxious stimulus, can be evaluated. Finally, interaction models between opioids and hypnotics can be defined based on both clinical end-points and EEG-derived indices. This research provides important information for defining the limits of the progressive evolution towards secure closedloop systems.
Spectral entropy as an electroencephalographic measure of anaesthetic drug effect: a comparison with bispectral index and processed midlatency auditory evoked response Vanluchene A, Vereecke H, Thas 0, Mortier E, Shafer S, Struys M. Anesthesiology 2004; 101: 34–42 BACKGROUND. This study compared the state entropy, response entropy, bispectral index and A-line auto regressive model with exogenous input (ARX) index with regard to their ability to detect and quantify a wide range of anaesthetic cerebral drug effects accurately. Cerebral effects were measured by all HCOA parameters simultaneously during a constant propofol infusion of 50 mg/min administered to ten spontaneously breathing patients. The full range of depth of hypnosis—from fully awake to total burst suppression—was monitored. This methodology enabled investigation of the ‘baseline variability’ in the awake patient, consistency of response during increasing effect site concentrations of propofol (CePROP), ‘prediction probability’ calculations for the detection of the CePROP value and ‘accurate burst suppression detection’ for all measures. A pharmacodynamic population model was developed with these data. INTERPRETATION. The smallest baseline variability was seen with the entropy measures (state plus response), followed by the bispectral index and A-line ARX index. A consistent decrease was seen in the state entropy and response entropy progression over the full scale of depth of hypnosis until 100% burst suppression occurred. This was not the case for the bispectral index where a plateau phase was seen at burst suppression levels of between 0 and 40% At higher levels of burst suppression an inverse linear
Depth of anaesthesia monitoring
235
correlation existed between burst suppression and the bispectral index. The A-line ARX index (version 1.5) had no correlation with burst suppression and would never have reached zero even at high levels of burst suppression. The relations between the CePROP value and the state entropy, response entropy, bispectral index and A-line ARX index were reflected in the calculations of PK values and individualized Spearman rank correlations, which are given in Table 10.1. The pharmacodynamic modelling demonstrated that the bispectral index and A-line ARX index had a more graded response to CePROP compared to the response entropy and state entropy. This means that the response entropy and state entropy had a steeper decrease in most patients compared to the bispectral index and A-line ARX index. The intra-individual variability in the bispectral index and A-line ARX index was smaller than for the response entropy and state entropy. Comment It remains a challenge to compare the performances of EEG- and MLAEP-based HCOA monitors with a variable physiological and mathematical background. In this study, a systematic statistical approach was used for obtaining a high level of relevant information. The CePROP value was used as an end-point because it reflects the ‘hypnotic condition’ of the patient in a continuous, reproducible way. However, one should be aware that the correlation between the CePROP value and the clinical state is not a linear one and depends on the pharmacokinetic—pharmacodynamic model (e.g. Schnider) used. Even with a constant CePROP value the clinical pharmacodynamic effects can vary by up to 400% between patients. However, this error is the same for all monitors and the ideal monitor will still be able to track the changes as calculated by the pharmacokinetic— pharmacodynamic model. As such, this methodology enables comparison of several monitors in a reproducible way, without large groups of patients. Moreover, clinical endpoints are less accurate end-points for comparison of HCOA monitor performances because they are often dichotomous (absent versus present) and have a low sensitivity/specificity. Due to these limitations most clinical studies comparing HCOA monitors lack power. A low baseline variability improves the correlation between the awake state and the value shown by the monitor. Although the state entropy and response entropy have the lowest baseline variability, the steep decrease in these parameters at a certain CePROP level indicates a relative inability of entropy in detecting a sedative level of hypnosis compared to the bispectral index and A-line ARX index. Monotonicity means that a value on the monitor represents a single level of cerebral drug effect. When the value of the HCOA monitor decreases in a monotonic way during increasing CePROP levels, the interpretation of an individual value becomes more accurate. The bispectral index loses monotonicity once the burst suppression reaches a value of between 0 and 40%. As such, the anaesthetist would be advised to use the suppression rate rather than the bispectral index value as the discriminating factor of the HCOA at that level of cerebral drug effect. The smaller the variability in the population, the better the performance for detecting the CePROP value. This is reflected in the PK analysis and individualized Spearman rank
The year in anaesthesia and critical care
236
correlation coefficients in Table 10.1. The authors chose to use both a parametric and non-parametric approach for confirming the higher ‘all round’ performance of the bispectral index versus the A-line ARX index, response entropy and state entropy for detecting the level of the CePROP value. In order to depict the inter-individual versus intra-individual variability in the population studied, a pharmacodynamic sigmoidal maximal effect (Emax) model was fitted using non-linear mixed effect modelling (NONMEM, GloboMax LLC, Hanover, MD, USA). The results of this evaluation might help to decide which monitor suits the needs of a certain anaesthetist best. The choice will be based on the targeted level of hypnosis (e.g. sedation level in intensive care versus anaesthesia during surgery) and the accuracy of each monitor in that range. Although the power of this study was sufficiently high, the results are only based on ten patients. One can never exclude the existence of a more pronounced outlier in hypnotic drug effect responses when using an HCOA monitor in the general population. Secondly, this trial was performed using one anaesthetic agent without surgical stimulus or interaction effects of opioids. These would certainly have influenced the results as discussed elsewhere in this review.
Table 10.1 Prediction probability and individualized Spearman rank correlation coefficients for each EEG measure versus the effect site concentration of propofol SE
RE
AAI
BIS
PK, median
0.86
0.89
0.87
0.91
(minimum—maxium)
(0.60–0.96)
(0.67–0.96)
(0.81–0.99)
(0.72–0.94)
Individualized Spearman
−0.841±0.014
−0.860±0.013
−0.869±0.010
−0.891±0.011
rank correlation Mean±SD (95% CI)
(−0.864, −0.808)*
(−0.882, −0.831) (−0.883, −0.844) (−0.907, −0.865)
* P<0.05 between bispectral index (BIS) and the other measures. AAI, A-Line® ARX index; Cl, confidence interval; PK, prediction probability; RE, response entropy of the spectral entropy; SE, state entropy of the spectral entropy. Source: Vanluchene et al. (2004).
Spectral entropy measurement of patient responsiveness during propofol and remifentanil: a comparison with the bispectral index Vanluchene A, Struys M, Heyse B, Mortier E. Br J Anaesth 2004; 93(5): 645–54 BACKGROUND. In this study a comparison was made between the ability of the bispectral index and the state entropy and response entropy with regard to measuring the loss of response to a verbal command (LOR ) and the loss of
Depth of anaesthesia monitoring
237
response to a noxious stimulus (LORnoxious) during propofol and remifentanil administration. The LORverbal was defined by a transition from level 3 to level 2 of the observer’s assessment of alertness and sedation scale. The noxious stimulus was a tetanic electrical stimulus (100 Hz and 50 mA) for 2 s on the volar forearm. Three groups of 20 patients received an effect site-controlled propofol infusion started at a CePROP value of 1 µg/ml, which was increased stepwise by 0.5 (µg/ml at intervals of 4 min. An effect site-controlled remifentanil infusion was also maintained at a CeREMI value of 0, 2 or 4 ng/ml according to group randomization. The observer’s assessment of alertness and sedation scale was measured before every increase in the CePROP value and the response to a noxious stimulus was evaluated for the bispectral index, state entropy and response entropy. The 4-minute interval allowed measurements to be performed in a ‘pseudo’-steady-state anaesthetic environment. INTERPRETATION. This study showed a progressive decrease in the bispectral index, state entropy and response entropy with decreasing score on the observer’s assessment of alertness and sedation scale. Observer’s assessment of alertness and sedation scores are reached at higher bispectral index, state entropy and response entropy values with the addition of remifentanil. There was a significant decrease in the median value of all measures at the LORverbal and LORnoxious, but a large overlap in individual values was observed. Table 10.2 shows the effective bispectral index, state entropy and response entropy values at which 50% (ED50) or 95% (ED95) of patients reached the LORverbal or LORnoxious. The overall accuracy for detection of the LORnoxious tended to be lower than the LORverbal. The sensitivity/specificity analysis showed that the cut-off values for the bispectral index, state entropy and response entropy at which the sum of sensitivity and specificity was highest (representing the monitoring value at which the overall ‘errors’ in detecting the LORverbal or LORnoxious was minimized) were similar and did not differ between remifentanil groups. This finding indicates that the addition of remifentanil did not influence the performance accuracy of the bispectral index, state entropy and response entropy in detecting the HCOA. Comment This study investigated the performance of an HCOA monitor during interactions between hypnotics and opioids. In order to optimize the applicability of an HCOA monitor in clinical practice, the accuracy of the monitors must be high in both mono anaesthesia and balanced anaesthesia conditions. In pharmacology research the behaviour of HCOA monitors should first be evaluated in a validated and reproducible setting. In this article, the application of a standardized noxious stimulus was performed during ‘pseudo’-steady-state anaesthesia with propofol and remifentanil administered according to the pharmacodynamic—pharmacokinetic models of Schnider et al. |7, 9| and Minto et al. |6, 10|. Although this approach only examined a small portion of the interaction range between opioids and hypnotics, these results do reveal some useful information for further exploration. Apparently, the accuracy of the monitors studied remained intact when adding remifentanil although the cut-off values for detecting the LORverbal had to be adjusted according to the amount of remifentanil given. These results are comparable with the results of the next article included in this chapter.
The year in anaesthesia and critical care
238
Table 10.2 The effective values of the bispectral index, state entropy and response entropy on which 50% or 95% (ED50 [95% CI]/ED95) of the population reaches loss of response to verbal comment or loss of response to noxious stimulus Group Remi 0 ng/ml
Remi 2 ng/ml
Remi 4 ng/ml
61 (59–62)*/52*
68 (67–70)*/65#
71 (70–73)*/64*
#
#
70 (68–72)*/54#
LORverbal BIS SE
64 (62–66)*/53
68 (67–69)*/60
RE
70 (68–72)*/57$
74 (73–75)*/69$
81 (80–83)*/68$
BIS
32 (31–34)*/17*
53 (52–54)*/45*
61 (60–62)*/50*
SE
37 (34–40)$/22$
50 (49–51)$/39$
53 (51–55)$/32$
RE
39 (36–42)*/22*
53 (52–54)*/41*
59 (57–61)*/34*
LORnoxious
* P<0.0167 between all groups, $ P<0.0167 for remi 0 ng/ml vs remi 2 ng/ml and remi 0 ng/ml vs remi 4 ng/ml; # P<0.0167 for remi 0 ng/ml vs remi 2 ng/ml and remi 0 ng/ml vs remi 4 ng/ml. BIS, bispectral index; SE, state entropy; RE, response entropy; LORverbal, loss of response to verbal command; LORnoxious, loss of response to electrical tetanic stimulus. Source: Vanluehene et al. (2004).
Pharmacodynamic interaction between propofol and remifentanil regarding hypnosis, tolerance of laryngoscopy, bispectral index, and electroencephalographic approximate entropy Bouillon T, Bruhn J, Radulescu L, et al. Anesthesiology 2004; 100: 1353–72 BACKGROUND. The purpose of this study was to describe the pharmacodynamic interaction between propofol and remifentanil for three end-
Depth of anaesthesia monitoring
239
points: firstly, the probability of no response to shaking and shouting, which is an estimation of the moment of loss of consciousness (LOC), secondly, the probability of no response to laryngoscopy (LORL), which is a measure of the loss of response to a noxious stimulus and, finally, the relation between this interaction and the behaviour on the bispectral index and approximate entropy. The primary goal of this study was to compare the accuracy of a new interaction model (the hierarchical model of opioid—hypnotic drug interaction) with the empirical response surface model illustrated by Minto et al. for describing a clinically obtained data set |11|. However, for this review the results on the EEG monitoring system were highlighted. The ability of the bispectral index and approximate entropy for describing both a Bayesian prediction of a measured drug concentration and a target-controlled infusion prediction of administered drugs were included in the analysis. INTERPRETATION. The new hierarchical interaction model illustrated the clinically obtained data set more accurately than the empirical model. The interaction of propofol and remifentanil on both EEG measures was additive. The bispectral index and approximate entropy were equally accurate in describing a propofol mono administration as reflected in the typical values of the concentration—effect relationship. The values for the propofol/remifentanil concentration associated with 50% of the maximal propofol/remifentanil-induced EEG suppression were similar for the bispectral index and approximate entropy in both the interaction models and the mono administration models. The presence of remifentanil did not significantly alter the estimates of the propofol concentration associated with 50% of the maximal propofol-induced EEG suppression (C50propofol). The remifentanil concentration associated with 50% of the maximal remifentanil-induced EEG suppression (C50remifentanil) for the bispectral index response was high for remifentanil mono administration, nearing the top of the concentration range measured and had a high standard error of approximately 40% indicating the modest effect of remifentanil on the EEG measure. The approximate entropy had a significantly smaller C50remifentanil value suggesting that the approximate entropy was more sensitive to opioid cerebral drug effects. Response prediction models were developed for displaying the range of the 5–95% probability of LOC and LORL based on the hierarchical model and the approximate entropy or bispectral index (Fig. 10.1). In the clinically relevant concentration range (propofol >1 µg/ml), the approximate entropy and bispectral index values associated with a 95% probability of LOC ranged from 0.35 to 1.2 and from 27 to 72, respectively, depending on the combination of propofol and remifentanil chosen. For the 95% probability of loss of response to laryngoscopy these values ranged from 0.14 to 0.80 and from 15 to 54, respectively. The graph shows that some propofol is required even to have only a 5% chance of non-response. It also shows that modest doses of remifentanil greatly decrease the propofol requirement and concurrently increase the bispectral index for a given probability of response to each stimulus. Comment This is a key article in pharmacology research concerning the interaction between propofol and remifentanil and the subsequent influences on the bispectral index and
The year in anaesthesia and critical care
240
approximate entropy. The additive interaction is not a new finding. However, by using this new hierarchical interaction model it becomes possible to define a probability of LOC and LORL only by knowing the target-controlled infusion concentration of propofol and remifentanil and the value of a cerebral drug effect monitor. This information might be of major importance for the future development of closed loop systems, using EEGextracted information as a surrogate measure for the
Depth of anaesthesia monitoring
241
Fig. 10.1 The additive interaction surface for the effect of propofol and remifentanil on the bispectral index, with the propofol effect being considerably more profound than the remifentanil effect. The trajectory lines are simulations of the bispectral index for equipotent combinations of propofol and remifentanil for (a) 5– 95% probabilities of no response to shouting and shaking and (b) 5–95% probabilities of no response to laryngoscopy based on the hierarchical model, showing that some propofol is required even to have only a 5% chance of non-response and that modest doses of remifentanil greatly decrease the propofol requirement and concurrently increase the the bispectral index for the same probability of response to each stimulus. Source: Bouillon et al. (2004). cerebral drug effect. These results should caution the reader to consider the relative contributions of hypnotic versus analgesic drugs when using an EEG-based index as an HCOA monitor. It is of note that the intensity of the noxious stimulus can also alter the probability ranges described with this model. This article indicates that, when guiding anaesthetic practice with information on cerebral drug effect from an EEG-based monitor, both pharmacokinetic and pharmacodynamic information about the drugs in use must be considered. Miscellaneous research on commercialized devices A vast amount of publications focus on the specific behaviour of a commercialized HCOA monitor in clinical practice. Some of these studies are constructed in such a way that positive characteristics are highlighted in an exaggerated way whilst comparisons with other devices are not included in the protocol. Other studies focus on interfering environmental or physiological factors that might lead to potentially harmful anaesthetic conditions. However, the scientific value of these publications is often limited due to a lack of statistical significance, the presence of a non-steady state condition at the time of measurement or the absence of comparison between monitors. Nevertheless, these
The year in anaesthesia and critical care
242
publications can be very informative on potentially conflicting clinical conditions that must be explored in depth in well-performed clinical trials or experimental settings. In this section, a study on the interfering effects of EMG on two versions of the bispectral index is discussed. It demonstrates that no anaesthetist should work with an HCOA monitor without having some basic knowledge of the methodology used with the machine.
The effect of different stages of neuromuscular block on the bispectral index and the bispectral index-XP under remifentanil/propofol anesthesia Dahaba AA, Mattweber M, Fuchs A, et al. Anesth Analg 2004, 99: 781–7 BACKGROUND. The goal of this study was to assess the effects of a standard dose of a neuromuscular blocking agent (twice the 95% effective dose of mivacurium) on two versions of bispectral index monitoring (A-2000BIS™ versus BIS-XP™, Aspect, Newton, MA, USA) during a deep level of propofol/remifentanil anaesthesia in a surgical setting. Both indices were simultaneously recorded from 38 ASA I-II patients. Both indices were stored in a time-synchronized way, together with EMG detection in the frequency domain of 70–110 Hz. The level of neuromuscular blocking activity was monitored with a train-of-four stimulus at intervals of 12 s. For induction remifentanil ranged from 0.15 to 0.3 µg/kg/min. Propofol was administered by a Diprifusor™ target-controlled infusion system with a target concentration set at 4 µg/ml and afterwards propofol was titrated to a bispectral index value of 40. Once a steady state in the bispectral index value was obtained mivacurium was administered. INTERPRETATION. The response to mivacurium administration was a transient bispectral index A2000 increase from 43±4 to 49±7 (P=0.007) and a bispectral index XP decline from 41±3 to 35±3 (P=0.003) at 1±0.2 min. This change returned to baseline in both monitors after 2.5±1 min. The mean duration of complete neuromuscular blockade (measured by T1% and the train-of-four ratio) was 7.4±3.9 min. The authors claimed the ‘afferentation theory’ as a possible cause for this phenomenon. It states that afferent signals from muscle stretch receptors produce certain EEG stimulation. As the patients tested had a high level of tonic facial muscle activity, indicated by the EMG detection on screen, the administration of mivacurium might have affected the muscle stretch receptor activity. This change was detected by both bispectral index versions and classified as an artefact that evokes an adjustment in the filtered information used for bispectral index calculation. Once the basal facial muscle tone had returned, the bispectral index returned to baseline. The different artefact rejection algorithms between monitors caused the opposite direction of change. The authors concluded that the two versions of bispectral index could not be considered interchangeable.
Depth of anaesthesia monitoring
243
Comment As noted before, the interaction between EEG and EMG needs further exploration in order to explain some unexpected behaviour of the bispectral index in clinical conditions. The results in this study are very interesting as an observation, particularly as they are the basis of a tempting hypothesis for explaining the interaction between EEG and EMG. However, by testing this hypothesis in a surgical setting, as was done in this study, some interfering factors cannot be excluded. In particular the level of remifentanil used varied during the measurements. As propofol was titrated to a constant bispec-tral index value, we must be aware that this value represents a variable level of depth of anaesthesia when used with a variable level of remifentanil (see pharmacological section, Bouillon et al.). It is possible that sensitivity to a neuromuscular blocking effect varies at different levels of anaesthesia. Studies investigating an EMG inter-action with EEG effects should therefore be performed in pharmacological Steady-state’ situations as much as possible before devices are tested in more challenging clinical conditions. However, even with variable levels of HCOA in the data, the two monitors were evaluating the same population. As such, the conclusion of the authors remains correct. The two versions are not interchangeable. When using an HCOA monitor, the anaesthetist must have some basic knowledge of the methods used with the machine to avoid a wrong interpretation in clinical practice. An HCOA monitor should never be a ‘black box’ for the anaesthetist. Conclusion At the present time there is no single HCOA monitor available that delivers a perfect performance in the variable clinical circumstances of everyday anaesthetic practice. This is partially caused by remaining fundamental questions about the correlation between EEG/MLAEPs and the effects of anaesthetic drugs at the cerebral and spinal receptor site. However, as already mentioned in this chapter, the number of publications addressing fundamental questions about the role of EEG and MLAEPs in measuring the HCOA is increasing fast. This evolution is important for optimizing the interpretation of the currently available devices and for increasing the general knowledge of anaesthetists regarding EEG and MLAEP behaviour. The ultimate goal is to achieve more reliable anaesthesia. This terminology covers the optimization of hypnotic and analgesic drug titration, the avoidance of awareness, the avoidance of sudden movement during surgery and the maintenance of haemodynamic stability. It becomes clear that this ideal anaesthetic effect cannot be reached by titration to one number. At the present time pharmacological research suggests that increased reliability will only be reached if the EEG/MLAEP-derived information is correlated with information on blood or effect site concentrations of both hypnotic and analgesic drugs. As pharmacokinetic and pharmacodynamic interaction models are optimized, it becomes possible to calculate the predictability for a variety of responses. This information will only be clinically applicable by using computer systems interpreting the patient’s derived data in a continuous way, if necessary linked to a closed loop anaesthetic drug delivery system. Such closed loop technology will certainly need some more years to develop.
The year in anaesthesia and critical care
244
In the meantime some remarkable results do suggest a beneficial effect when using an HCOA monitor. The B-Aware Trial clearly showed a positive result for the bispectral index as a detector of awareness in patients with a high risk of awareness. Other clinical claims for measuring the HCOA, such as drug saving, improved haemodynamic stability, faster recovery and reduced post-operative nausea and vomiting, are often less evidence based. The number of commercial EEG- and MLAEP-based monitors has never been greater than now. The development of accurate and reproducible methods for comparing these new devices is essential. By avoiding some major pitfalls when investigating or comparing HCOA monitors in clinical practice the evidence on which to base practice may become clearer and more definite in years to come. In conclusion, there is no such thing as an ‘ideal HCOA monitor’. The clinical anaesthetist should choose a device according to the available evidence at this time. He/she should be aware that the greatest accuracy of any device is in the group in whom it was tested. Most research in HCOA monitoring is performed in ASA I or II patients. Using these devices in other more demanding populations should therefore be conducted cautiously. The debate about the role of EEG- and MLAEP-derived HCOA variables has already revealed new insights into anaesthetic phenomena, but much has still to be clarified before the optimal means of measuring the depth of anaesthesia is defined. References 1. Kalkman CJ, Drummond JC. Monitors of depth of anesthesia, quo vadis? Anesthesiology 2002; 96: 784–7. 2. Struys MM, Vereecke H, Moerman A, Jensen EW, Verhaeghen D, De Neve N, Dumortier FJ, Mortier EP. Ability of the bispectral index, autoregressive modelling with exogenous inputderived auditory evoked potentials, and predicted propofol concentrations to measure patient responsiveness during anesthesia with propofol and remifentanil. Anesthesiology 2003; 99: 802–12. 3. Vuyk J. Pharmacokinetic and pharmacodynamic interactions between opioids and propo-fol. J Clin Anesth 1997; 9: 23S-6S. 4. Struys MM, De Smet T, Greenwald S, Absalom AR, Binge S, Mortier EP. Performance evaluation of two published closed-loop control systems using bispectral index monitor-ing: a simulation study. Anesthesiology 2004; 100: 640–7. 5. Schultz B, Grouven U, Schultz A. Automatic classification algorithms of the EEG monitor Narcotrend for routinely recorded EEG data from general anaesthesia: a validation study. Biomed Tech (Berl) 2002; 47: 9–13. 6. Minto CF, Schnider TW, Egan TD, Youngs E, Lemmens HJ, Gambus PL, Billard V, Hoke JF, Moore KH, Hermann DJ, Muir KT, Mandema JW, Shafer SL. Influence of age and gender on the pharmacokinetics and pharmacodynamics of remifentanil. I. Model development. Anesthesiology 1997; 86: 10–23. 7. Schnider TW, Minto CF, Shafer SL, Gambus PL, Andresen C, Goodale DB, Youngs EJ. The influence of age on propofol pharmacodynamics. Anesthesiology 1999; 90: 1502–16. 8. Ekman A, Lindholm ML, Lennmarken C, Sandin R. Reduction in the incidence of aware-ness using BIS monitoring. Acta Anaesthesiol Scand 2004; 48: 20–6. 9. Schnider TW, Minto CF, Gambus PL, Andresen C, Goodale DB, Shafer SL, Youngs EJ. The influence of method of administration and covariates on the pharmacokinetics of propofol in adult volunteers. Anesthesiology 1998; 88: 1170–82.
Depth of anaesthesia monitoring
245
10. Minto CF, Schnider TW, Shafer SL. Pharmacokinetics and pharmacodynamics of remifentanil. II. Model application. Anesthesiology 1997; 86: 24–33. 11. Minto CF, Schnider TW, Short TG, Gregg KM, Gentilini A, Shafer SL. Response surface model for anesthetic drug interactions. Anesthesiology 2000; 92: 1603–16.
11 New airway equipment TAKASHI ASAI Introduction A major development in airway management for anaesthesia in the last decade has been the use of a supraglottic airway. The laryngeal mask airway (LMA) was invented by Dr Brain as a totally new concept and was made available to clinicians in 1988 |1|. This device has forced anaesthetists to reconsider the routine practice of airway management as a whole |2|. In many circumstances in which tracheal intubation was once considered mandatory, the LMA has been used safely and may even offer specific advantages. For example, the LMA can be used during adenotonsillectomy and is associated with less complications than tracheal intubation, such as aspiration of blood |3|. Since the introduction of the LMA several new supraglottic airways, including modified LMAs, have become available. Each device may have theoretical advantages (or disadvantages) over other devices, but it is necessary to study the efficacy and safety of each of these in a clinical setting in order to establish its place in clinical practice. Laryngeal mask airway The LMA has gained a firm place in anaesthetic practice and in some countries its use is more frequent than tracheal intubation. After introduction of the original LMA (the classic Laryngeal Mask Airway, LMA-Classic™ or cLMA™), four major modified devices have become available: an LMA with a flexible tube (LMA-Flexible™), an intubating LMA (LMA-Fastrach™), a single-use LMA (LMA-Unique™) and the LMAProSeal™. Efficacy of the LMA-ProSeal™ The LMA-ProSeal™ is a new laryngeal mask device with a modified cuff and a drain tube for gastric access. Design changes are intended to improve the seal with the airway and minimize gastric insufflation and pulmonary aspiration |4|. The airway pressure at which gas leaks around the LMA-ProSeal™ is considerably higher than that for the LMAClassic™ |5|. The drain tube, which is parallel to the ventilation tube, © Atlas Medical Publishing Ltd
New airway equipment
247
passes through the bowl of the mask to its tip, which sits at the entrance to the upper oesophagus. This should allow insertion of a gastric tube into the stomach in order to remove gas or liquid contents and should direct passively regurgitated gastric fluid away from the airway |4|. These features suggest that the LMA-ProSeal™ maybe more appropriate than the LMA-Classic™ when patient or operative factors lead to reduced pulmonary compliance, thereby necessitating higher airway pressures for adequate ventilation, such as in obese patients, patients with lung disease and during operations with raised intra-abdominal pressure.
LMA-Classic™ and LMA-ProSeal™ are effective alternatives to endotracheal intubation for gynecologic laparoscopy Maltby JR, Beriault MT, Watson NC, Liepert DJ, Fick GH. Can J Anaesth 2003; 50: 71– 7 BACKGROUND. The objective of this study was to compare the LMA (LMAClassic™ in non-obese patients and LMA-ProSeal™ in obese patients) with the tracheal tube in terms of pulmonary ventilation and gastric distension during gynaecologic laparoscopy. INTERPRETATION. The LMA (LMA-Classic™ in non-obese patients and LMAProSeal™ in obese patients) was as effective as tracheal intubation for positive pressure ventilation without clinically important gastric distension during gynaecological laparoscopy. Comment In this study Maltby et al. stratified 209 adult women by body mass index as nonobese or obese and randomized them to receive either an LMA (a size 4 LMA-Classic™ in nonobese patients and a size 4 or 5 LMA-ProSeal™ in obese patients) or a tracheal tube (7.0 mm internal diameter [ID]). Anaesthesia was induced with propofol and fentanyl and neuromuscular blockade was achieved with either suxamethonium or rocuronium. Anaesthesia was maintained with isoflurane and nitrous oxide in oxygen. Ventilation was controlled, with a tidal volume of 10 ml/kg. The following factors were studied for short (15 min or less) or long (>15 min) duration of peritoneal insufflation. 1. The oropharyngeal leak pressure. 2. The surgeon-scored stomach size, with the surgeon blinded to the type of airway, on an ordinal scale of 0–10 (0=empty stomach and 10=distension of stomach that interfered with surgery). 3. Respiratory variables: primary outcome the end-tidal carbon dioxide concentration (PETCO2) and secondary outcome the arterial haemoglobin oxygen saturation (SpO2). 4. Airway complications at removal of the device. 5. Post-operative airway complications. There were no significant differences between the laryngeal mask and tracheal tube groups for the PETCO2, SpO2 or airway pressure before or during peritoneal insufflation.
The year in anaesthesia and critical care
248
There was also no difference in the stomach size between the groups. The drain tube of the LMA-ProSeal™ filled with gastric fluid in one patient immediately before deflation of the pneumoperitoneum. There was no visual contamination of the anterior surface of the LMA-ProSeal™ or the pharyngeal wall and respiration was not affected immediately or post-operatively. There was a tenfold greater incidence of coughing at removal of a tracheal tube (91 out of 105 patients [87%]) than at removal of a laryngeal mask (eight out of 104 patients [8%]). Sore throat 24 h post-operatively was more common after tracheal extubation than removal of the laryngeal mask (28 versus 17%). From this and other studies we can conclude that the LMA is as effective as a tracheal tube for maintaining a clear airway during laparoscopic gynaecological surgery |6, 7| and during cholecystectomy |8|. This study has shown that the LMA-ProSeal™ is effective in obese patients undergoing laparoscopic surgery. Nevertheless, the number of patients studied may be too small for determining the safety of the laryngeal mask (both the LMAClassic™ and LMA-ProSeal™) with respect to pulmonary aspiration. In fact, it should be remembered that there have been several reports of pulmonary aspiration during the use of the LMA-ProSeal™ (see Gaitini et al. below and |9, 10|). Pulmonary aspiration during the use of the laryngeal mask airway
Aspiration and the laryngeal mask airway: three cases and a review of the literature Keller C, Brimacombe J, Bittersohl P, Lirk P, Von Goedecke A. Br J Anaesth 2004; 93: 579–82 BACKGROUND. In case 1 a 73-year-old, morbidly obese woman with a history of diabetes mellitus, chronic obstructive pulmonary disease and gastric surgery was scheduled for varicose vein surgery. After induction of anaesthesia and neuromuscular blockade, an LMA-Classic™ was inserted. In the middle of the procedure the patient aspirated gastric contents. Post-operatively she developed severe adult respiratory distress and had a complete left-sided hemiplegia from a cerebral infarct. In case 2 a 78-year-old obese woman with a history of a hiatus hernia (without symptoms of reflux) was scheduled for a total hip replacement. After induction of anaesthesia and neuromuscular blockade, an LMA-Fastrach™ was inserted. Before tracheal intubation through the LMA-Fastrach™ the patient aspirated. Post-operatively the patient developed adult respiratory distress syndrome and died from multiple organ failure. INTERPRETATION. Pulmonary aspiration can occur during the use of the LMA.
Comment
New airway equipment
249
Keller et al. reported three cases (two cases are shown above) of pulmonary aspir -ation associated with LMAs. This is the first report of brain injury and death after the use of an LMA. All three patients in this report were at increased risk of aspiration. They summarized a total of 20 specifically described case reports of pulmonary aspiration during the use of an LMA. Of these 20 cases 14 were in adults and six were in children and 19 were with the LMA-Classic™ and one with the LMA-ProSeal™. There is only one report of aspiration where the patient had no risk factors. In an accompanying editorial to Keller et al.’s case reports, Asai claimed that it is apparent from these reports that the incidence of aspiration can be reduced by not using supraglottic airways in anaesthetized patients who are at increased risk of aspiration |11|. He pointed out that there are four major categories of factors that will theoretically predispose patients to risk: patient factors, operation factors, anaesthesia factors and device factors (Table 11.1). He also pointed out that the
Table 11.1 Predisposing factors increasing the risk of pulmonary aspiration Patient factors Full stomach (e.g. emergency sw3w) Diseases and symptoms known to delay gastric emptying Diabetes mellitus Increased intracranial pressure Hiatus hernia Gastrointestinal obstruction Recurrent regurgitation Dyspeptic symptoms Oesophageal disease (e.g. achalasia, pouches, etc) History of upper gastrointestinal surgery Injured or receiving opioids or other drugs delaying gastric emptying Women in labour Morbid obesity Operation factors Upper abdominal surgery Lithotomy or the head-down position Laparoscopic cholecystectomy Anaesthesia factors Inadequate anaesthesia Intermittent positive pressure ventilation particularly with high airway pressures Prolonged anaesthesia Removal of the airway before spontaneous recovery from anaesthesia Device factors Presence of a supraglottic airway inserted into the hypopharynx Incorrectly inserted airway Inadequate seal with oropharynx Source: Asai (2004) |11|.
nature and amount of material aspirated affects the consequences of aspiration. For example, bile damages the lungs more severely than gastric acid does and, thus, greater precaution is required if the trachea is not intubated during laparoscopic cholecystectomy.
The year in anaesthesia and critical care
250
It seems simple to decide whether or not a supraglottic airway is indicated by weighing its advantages over tracheal intubation against the risk of pulmonary aspiration. In reality, we may often face a difficult decision, mainly because there are so many uncertainties in estimating the risk of aspiration. For example, it is not clear whether all obese patients, all patients in the lithotomy position or all those undergoing laparoscopic surgery should be regarded as being at high risk. The reported incidence of regurgitation during use of a laryngeal mask in patients without predisposing factors varies considerably from 0 to 80% (into the oesophagus) and from 0 to 28% (into the oropharynx) |11|. We know almost nothing about the reasons for these marked differences in incidence between studies and, thus, we know little about how to reduce the incidence. A method that facilitates the insertion of the LMA-ProSeal™ One difficulty with the use of the LMA-ProSeal™ is that its correct insertion may be more difficult than the LMA-Classic™. The success rate of insertion at the first attempt is approximately 95% for the LMA-Classic™ and approximately 85% for the LMAProSeal™ |5|. The main reason for the difficulty in insertion of the LMA-ProS-eal™ is impaction of the device at the back of the mouth. Even when the LMA-ProSeal™ is inserted successfully it may be placed in a suboptimal position: the tip of the mask becomes folded back on itself or occasionally the tip is directed into the glottic inlet rather than the hypopharynx. If the device is placed in a suboptimal position the drain tube may not function properly, thereby failing to prevent pulmonary aspiration of gastric contents. Therefore, any method that would facilitate insertion and correct positioning is useful. One method has been suggested by Howath et al. |12|, who inserted a gum elastic bougie in the oesophagus to guide the LMA-ProSeal™ around the back of the mouth and into its correct position in the hypopharynx.
Gum elastic bougie-guided insertion of the ProSeal™ laryngeal mask airway is superior to the digital and introducer tool techniques Brimacombe J, Keller C, Dana J. Anesthesiology 2004; 100: 25–9 BACKGROUND. The objective of this study was to assess whether insertion of the LMA-ProSeal™ using a gum elastic bougie is more frequently successful than insertion using the fingers or the introducer tool provided by the manufacturer. INTERPRETATION. Insertion of the LMA-ProSeal™ guided by a gum elastic bougie was more frequently successful than insertion using the fingers or the introducer tool. Comment In this study Brimacombe et al. randomized 240 patients to insertion of the LMAProSeal™ either digitally, with the introducer tool (provided by the manufacturer of the LMA-ProSeal™) or with a gum elastic bougie. For the bougie group the drain tube of the
New airway equipment
251
LMA-ProSeal™ was primed with the bougie, the bougie placed in the oesophagus under conventional laryngoscopy and the LMA-ProSeal™ inserted digitally with the bougie as a guide. The following factors were studied: (1) the success rate of insertion (up to three attempts), (2) the presence or absence of blood on the device after removal and (3) postoperative complications. Insertion was more frequently successful at the first attempt with the bougie technique than the digital or insertion tool techniques (both significant), but the success rates after three attempts were similar (Table 11.2). The time taken for successful insertion was similar between groups at the first attempt, but was shorter for the bougie technique after three attempts (both significant). Insertion was successful in all patients with the bougie technique, failed in one with the digital technique and failed in two with the insertion tool technique. No patient had injury to the mouth or tongue, but two patients had minor cuts on the lips. Although there were no significant differences in the incidence of visible blood on the device between groups, occult blood was detected less frequently with the bougie-guided technique than the
Table 11.2 Success rate and time for insertion of the LMA-ProSeal™ between the digital, insertion tool or bougie-guided techniques Digital (n=80)
Insertion tool (n=80)
Bougie guided (n=80)
Successful insertion at the first attempt
70*
67‡
80*†
Successful insertion within three attempts
79
78
80
Insertion time (s)
33#
37§
25#§
* P<0.001; † P<0.001; # P<0.003; § P<0.003. Source: Brimacombe et al. (2004).
Table 11.3 Post-operative airway complications Digital (n=80) Insertion tool (n=80) Bougie guided (n=80) Visible blood on the device
3
4
2
Occult blood on the device
23*
25†
10*†
Sore throat
12
7
9
Dysphagia
9
7
9
Dysphonia
5
7
5
* P<0.02, † P<0.02. Source: Brimacombe et al. (2004).
The year in anaesthesia and critical care
252
other two techniques (Table 11.3). There were no differences in post-operative airway complications between the groups (Table 11.3). Insertion of the LMA-ProSeal™ with the aid of a gum elastic bougie was more frequently successful than the conventional method (digital or with the aid of the insertion tool). The bougie guide may be a useful back-up technique when conventional insertion has failed. Nevertheless, caution is required with the use of the bougie, because the bougie is not designed for oesophageal intubation and could injure the oesophagus. Single-use laryngeal masks Because of concern about transmission of infection, several different single-use laryngeal masks are available from different companies. The structure of these devices is basically similar to the LMAs (LMA-Classic™ or the single-use device LMA-Unique™) invented by Brain, but most are made of polyvinyl chloride (PVC) rather than silicone. It is necessary to evaluate whether these new devices are as effective and safe as the LMAClassic™. The Soft Seal LM (Portex Ltd, Hythe Kent, UK) is fabricated from latex-free medicalgrade plasticized PVC and is translucent, thereby allowing easier visualization of any secretions inside it. This device has several minor differences from the LMA-Classic™. One notable difference is that the LMA-Classic™ has two bars at the distal aperture of the tube (to prevent the epiglottis migrating into the tube), whereas the Soft Seal LM does not have bars. Another difference is that the Soft Seal LM allows a passage of a 7.0 mm ID tracheal tube whereas the LMA-Classic™ allows only a 6.0 mm ID tube.
Comparison of the LMA-Classic™ with the new disposable Soft Seal laryngeal mask in spontaneously breathing adult patients van Zundert AA, Fonck K, Al-Shaikh B, Mortier E. Anesthesiology 2003: 99: 1066– 71 BACKGROUND. The primary aim of this study was to compare the efficacy of the Soft Seal LM with the LMA-Classic™ in terms of ease of insertion, cuff pressure changes during nitrous oxide anaesthesia, position by fibreoptic evaluation and possible complications. INTERPRETATION. The efficacy of the Portex Soft Seal LM was similar to the LMA-Classic™. In addition, the incidence of post-operative sore throat was significantly less for the Soft Seal LM than the LMA-Classic™.
Comment Two hundred patients were randomly allocated to receive either the LMA-Classic™ (i.e. the reusable device) or the single-use Soft Seal LM. Anaesthesia was induced with
New airway equipment
253
propofol and fentanyl and maintained with sevoflurane and nitrous oxide in oxygen. The following factors were studied. 1. The success rate of insertion (up to two attempts). 2. The ease of insertion. 3. The position of the device assessed by fibrescope. 4. The intra-cuff pressure. 5. The presence or absence of blood on the device at removal. 6. The presence or absence of sore throat at 2 and 24 h after anaesthesia. Insertion was successful in all patients in both groups. There was no significant difference in the ease of insertion between the groups (Table 11.4). Fibreoptic evaluation showed that the position of the device was similar in the groups. The vocal cords were not seen through the device in six patients (6%) in the LMA-Classic™ group and four patients (4%) in the Soft Seal LM group. The mean intra-cuff pressure markedly increased from 45 to 100 mmHg in the LMA-Classic™ group, whereas it remained stable (from 45 to 47 mmHg) in the Soft Seal LM group (a significant difference between groups). Blood was detected on the device at removal in four patients in the LMA-Classic™ group and none in the Soft Seal LM group. A clear airway was maintained in all patients in both groups. The incidence of sore throat was significantly higher in the LMA-Classic™ group (20.5%) than the Soft Seal LM group (10.2%), 2 h but not 24 h after surgery. The Portex Soft Seal LM is as effective as the LMA-Classic™ in anaesthetized patients breathing spontaneously and the intra-cuff pressure is less likely to increase during anaesthesia including nitrous oxide. However, it is not known whether the Soft Seal LM is useful during controlled ventilation. It is also not known whether the absence of the increase in the intra-cuff pressure is clinically advantageous, because there maybe no correlation between the intra-cuff pressure and the pressure exerted by the cuff on the oropharyngeal tissues. The success rate of tracheal intubation through the Soft Seal LM is also not known.
Table 11.4 Success rate and ease of insertion of the LMA-Classic™ and Soft Seal LM LMA-Classic™ (n=100)
Soft Seal LM (n=100)
Successful insertion at the first attempt
97
95
Successful insertion within two attempts
100
100
Easy insertion at the first attempt
70
67
Moderately difficult insertion at the first attempt
27
28
Difficult insertion at the first attempt
3
5
No significant differences between groups. Source: van Zundert et al. (2003).
The year in anaesthesia and critical care
254
Laryngeal Tube The Laryngeal Tube (VBM, Medizintechnik, Sulz, Germany), which is a supraglottic airway, consists of an airway tube with a small balloon cuff attached at the tip (distal cuff) and a larger balloon cuff at the middle part of the tube (proximal cuff). Since its introduction into clinical practice the design of the Laryngeal Tube has been modified on several occasions |13|: the prototype had two pilot tubes whereas the current device has one, a softer silicone has been used for the tip of the device (in the distal cuff) in order to minimize oropharyngeal injury, the distal aperture consists of two orifices (instead of one in the prototype), an indentation has been made to the proximal cuff around the second distal orifice and two side holes have been added near the distal orifice. The cuffs are inflated through a single pilot tube and balloon. The proximal cuff provides a seal in the upper pharynx and the distal cuff seals the oesophageal inlet. The distal orifice should face the glottic aperture. The device is reusable for up to 50 times. Efficacy of the Laryngeal Tube Several studies have shown that the efficacy of the Laryngeal Tube is generally similar to that of the LMA-Classic™ during controlled ventilation |13–16|.
Randomized crossover comparison of the ProSeal™ laryngeal mask airway with the Laryngeal Tube™ during anaesthesia with controlled ventilation Cook TM, McKinstry C, Hardy R, Twigg S. Br J Anaesth 2003; 91: 678–83 BACKGROUND. The aim of this study was to compare the seal pressure of the Laryngeal Tube with that of the LMA-ProSeal™ in paralysed patients during the entire course of anaesthesia. Secondary outcome measures were designed for examining the relative clinical utility of each device. INTERPRETATION. Although the ease of insertion and the success rate of ventilation between the LMA-ProSeal™ and the Laryngeal Tube were similar, the time to insertion was shorter and the adequacy of ventilation was better for the LMA-ProSeal™ than the Laryngeal Tube. Comment In a randomized, crossover study design Cook et al. allocated 32 anaesthetized and paralysed patients to receive either the LMA-ProSeal™ (size 4 in females and size 5 in males) or the Laryngeal Tube (size 4 for those between 155 and 180 cm in height and size 5 for those above 180 cm in height). The following factors were studied. 1. The ease of insertion. 2. The insertion time. 3. The minimum airway pressure at which gas leaked around the airway device (oropharyngeal leak pressure).
New airway equipment
255
4. The position of the device as assessed by fibrescope. 5. The peak and plateau airway pressure. 6. The adequacy of ventilation (good represents optimal ventilation without complications during anaesthesia, fair represents optimal ventilation with complications or suboptimal ventilation without complications and failure represents failed insertion or abandonment of use). 7. The airway manipulations required for achieving a patent airway. 8. The presence or absence of secretions interfering with airway management. 9. The presence or absence of blood on the device at removal. 10. Airway complications (during induction of, maintenance of and emergence from anaesthesia). There was no significant difference in the oropharyngeal leak pressure between the LMAProSeal™ (median 26.5 cm H2O) and the Laryngeal Tube (24 cm H2O). Ventilation failed in two patients after insertion of the Laryngeal Tube and in none in the LMAProSeal™ group. The peak airway pressure with the LMA-ProSeal™ (mean 16 cm H2O) was lower than with the Laryngeal Tube (18 cm H2O) (95% confidence interval [CI] of the difference –3.1 to –0.3 cm H2O). The LMA-ProSeal™ took significantly less time (median 18.5 s) to insert than the Laryngeal Tube (22 s). The view of the glottis through the LMA-ProSeal™ was significantly better than the view through the Laryngeal Tube. Ventilation was good in 15 out of 16 patients, fair in none and failed in one for the LMAProSeal™, whereas it was good in nine out of 16 patients, fair in six and failed in one for the Laryngeal Tube (P=0.009). There was no significant difference in the plateau airway pressure, ease of insertion of the device, number of manipulations required to achieve or maintain an airway or overall complications. From this and other studies it can be concluded that the Laryngeal Tube is as effective as the LMA-Classic™ during controlled ventilation, whereas it may be less effective than the LMA-ProSeal™. Ease of insertion in patients with restricted neck movement In patients with an unstable cervical spine the head and neck may need to be stabilized manually (manual in-line stabilization). There have been several reports studying the ease of insertion of various forms of the LMA |17–21|. For the LMA-Classic™ its insertion may become more difficult when the patient’s head and neck are stabilized by the manual in-line method, but it is often possible to ventilate the lungs through it |17–19|. The insertion of the intubating laryngeal mask or the LMA-ProSeal™ during manual in-line stabilization has been shown to be significantly easier than insertion of the LMAClassic™ |20, 21|. It is not known whether this procedure affects the ease of insertion of the Laryn-geal Tube.
Ease of insertion of the Laryngeal Tube during manual in line neck
The year in anaesthesia and critical care
256
stabilization Asai T, Marfin AG, Thompson J, Popat M, Shingu K. Anaesthesia 2004; 59: 1163–6 BACKGROUND. The aim of this study was to compare the success rate of ventilation through the Laryngeal Tube between the Magill position (a pillow under the occiput and the head extended) or the manual in-line position of the head and neck (without a pillow under the occiput). INTERPRETATION. Stabilization of the patient’s head and neck by the manual inline method made insertion of the Laryngeal Tube either difficult or impossible. Comment In a randomized, crossover study design Asai et al. allocated 21 anaesthetized and paralysed patients to compare the success rate of ventilation through the Laryngeal Tube between the Magill position or the manual in-line position. After induction of anaesthesia and neuromuscular blockade the Laryngeal Tube was inserted in turn in the two positions. The ease of insertion was scored with four categories (easy, moderately difficult, difficult and impossible) and adequacy of ventilation through the device was assessed. Ventilation through the Laryngeal Tube was adequate in all 21 patients (100%) with the patient’s head and neck in the Magill position, whereas it was adequate in only two out of 21 patients (9.5%) during manual in-line neck stabilization (P<0.01; 95% CI for difference 68–94%). In the Magill position insertion was easy in 16 patients and moderately difficult in the remaining five patients, whereas insertion was moderately difficult in two patients and impossible in the remaining 19 patients during manual in-line stabilization. Although the Laryngeal Tube is potentially useful during cardiopulmonary resuscitation it may have a lesser role than the LMAs (classic, intubating or LMAProSeal™) when the patient’s head and neck are stabilized by the manual in-line method. The curve of the Laryngeal Tube is less similar to the curve of the oropharynx than the curve of the intubating laryngeal mask or LMA-ProSeal™ so that it is in theory more difficult to slide the Laryngeal Tube along the oropharyngeal wall. The tip of the Laryngeal Tube impacted upon the posterior pharyngeal wall in all patients in this study and it was not possible to advance the device beyond this. Laryngeal Tube-Suction The Laryngeal Tube-Suction (Laryngeal Tube Sonda™ or LTS™ [VBM Medizintechnik, Sulz, Germany]) is a further development of the Laryngeal Tube that aims to separate the respiratory tract from the alimentary tract. This device has two lumens with one used for ventilation and the other for passage of a gastric tube.
A randomized controlled trial comparing the ProSeal™ laryngeal mask
New airway equipment
257
airway with the Laryngeal Tube suction in mechanically ventilated patients Gaitini LA, Vaida SJ, Somri M, Yanovski B, Ben-David B, Hagberg CA. Anesthesiology 2004; 101: 316–20 BACKGROUND. The objective of this study was to compare the efficacy of the LMA-ProSeal™ and the LTS™ for ventilation during anaesthesia. INTERPRETATION. With respect to both physiological and clinical function, the LMA-ProSeal™ and LTS™ were similar and either device could be used for establishing a safe and effective airway. Comment In this study Gaitini et al. randomly allocated 150 anaesthetized and paralysed patients to receive either the LMA-ProSeal™ or the LTS™. The following factors were assessed. 1. The success rate of insertion (up to two attempts). 2. The insertion time. 3. The oropharyngeal leak pressure. 4. The position of the device assessed by fibrescope. 5. The SpO2 and PETCO2 values. 6. The ease of insertion of a gastric tube (18 gauge for the LMA-ProSeal™ and 14 gauge for the LTS™). 7. The presence or absence of blood on the device at removal. 8. Airway complications 24 h after anaesthesia. The LMA-ProSeal™ was successfully inserted in 72 out of 75 (96%) patients (first attempt in 57 patients and second attempt in 15 patients), whereas the LTS™ was successfully inserted in 71 out of 75 (94%) patients (first attempt in 60 patients and second attempt in eleven patients). The mean time to achieving an effective airway was 36 s with the LMA-ProSeal™ and 34 s with the LTS™. The mean oropharyngeal leak pressure was 28 cm H2O for the LMA-ProSeal™ and 34 cm H2O for the LTS™. There was no significant difference in the position of the device (in relation to the vocal cords) between the two groups. Airway intervention (thrusting the jaw forward or adjustment of the patient’s head and neck or of the device) was required in 32 patients for the LMAProSeal™ and in 35 patients for the LTS™. Of the patients in whom insertion was successful, insertion of a gastric tube was successful in 97% with the LMA-ProSeal™ and 96% with the LTS™. The SpO2 and PETCO2 values were similar in each group. Airway obstruction occurred in three patients and pulmonary aspiration in one patient for the LMA-ProSeal™, while airway obstruction occurred in four patients and laryngospasm in one patient for the LTS™. The incidence of blood on the device at removal and postoperative respiratory complications are shown in Table 11.5
Table 11.5 The incidence of blood on the device at removal and post-operative respiratory complications LMA-ProSeal™
LTS™
The year in anaesthesia and critical care
Blood on the device at removal
258
12
6
Sore throat
9
4
Pain at deglutition
3
4
Source: Gaitini et al. (2004).
. It seems that the LTS™ is as effective as the LMA-ProSeal™ in maintaining a clear airway and that both devices facilitate the passage of a gastric tube. It is not known whether the LTS™ truly reduces the incidence of regurgitation of gastric contents and of pulmonary aspiration when compared with the standard Laryngeal Tube or LMA-Classic laryngeal mask. (Since publication of this study the LTS™ has been modified and the performance of the LTS II™ is unknown.) Airway Management Device The Airway Management Device (AMD™, Nagor, Douglas, Isle of Man; manufactured by Biosil, Cumbernauld, UK) is also a new supraglottic airway. It consists of a semitranslucent silicone tube with two cuffs. The cuffs and the tube are asymmetrical in crosssection and it is claimed that this feature minimizes rotation of the device in the oropharynx. The cuffs are inflated through independent pilot tubes. The distal cuff occludes the oesophageal inlet and the pharyngeal cuff pushes the base of the tongue forward and lifts the epiglottis. There is an aperture between the two cuffs for ventilation. There is a second distal aperture at the tip of the device. When the distal cuff is inflated, the cuff occludes this aperture. When the balloon is partially deflated the distal orifice opens and it is possible to pass a gastric tube through the device. The distal cuff is inflated and the proximal cuff is fully deflated for the insertion of the device. The manufacturer claims that it is possible to intubate the trachea through the AMD™ with the aid of a gum elastic bougie. The device is reusable up to 40 times. After a few conflicting reports on its efficacy |22|, modifications were made to the original device and the modified AMD™ was reintroduced for clinical use in January 2002. Modifications have been made to the proximal cuff, which has a midline ‘expansion band’ that is meant to offer increased cuff expansion in relation to the volume inflated and facilitate a gas-tight seal with reduced intra-cuff pressure. The new device is available in three sizes (3, 4 and 5) compared with the two sizes in the earlier version.
An evaluation of the modified Airway Management Device Sivasankar R, Bahlmann UB, Stacey MR, Sehgal A, Hughes RC, Hall JE. Anaesthesia 2003; 58: 558–61 BACKGROUND. The aim of this study was to assess the efficacy of the modified AMD™ in anaesthetized patients breathing spontaneously.
New airway equipment
259
INTERPRETATION. The efficacy of the modified AMD™ in anaesthetized patients breathing spontaneously was less than ideal. Comment In this observational study Sivasankar et al. assessed the efficacy of the new AMD™ in 60 anaesthetized patients who were breathing spontaneously. Anaesthesia was induced with 1.0 µg/kg of fentanyl and propofol until there was no motor response to anterior jaw thrusting and the device was inserted. Anaesthesia was maintained with isoflurane or propofol and the patient was allowed to breathe spontaneously. The following factors were studied. 1. The number of attempts at insertion (up to three attempts). 2. The ease of insertion (easy, moderately difficult, difficult or impossible). 3. The intra-cuffpressure. 4. Airway complications during anaesthesia. 5. The ease of removal (easy, moderately difficult or difficult). 6. The presence or absence of blood on the device at removal. 7. The presence or absence of post-operative sore throat. Insertion of the AMD™ was generally easy (easy in 54 patients, moderately difficult in five patients, difficult in no patients and impossible in one patient) and the airway was secured at the first attempt in 41 patients (70%). The median intra-cuff pressure during maintenance of anaesthesia was 28 cm H2O. The most important problem during anaesthesia was loss of airway, which occurred in eleven patients (19%). Lifting the chin or thrusting the jaw forward relieved the obstruction in only two of the eleven patients. Severe laryngospasm with airway obstruction developed in another patient after transfer into theatre. This was thought to be due to dislodgement of the device. Removal of the device was easy in all the patients. In total, the AMD™ was not effective in twelve out of 60 (20%) patients. Blood was found on the device in three patients (6%) and nine patients (18%) had a sore throat. Another concern about the device was that herniation of the pilot balloon occurred frequently (in six out of 60 occasions) on testing the device before insertion. Judged from this observational report, even the modified AMD™ frequently fails to maintain a clear airway during anaesthesia.
Randomised comparison of the classic Laryngeal Mask Airway™ with the Airway Management Device™ during anaesthesia Cook TM, Porter MV. Br J Anaesth 2003; 91: 672–7 BACKGROUND. The objective of this study was to compare the efficacy of the AMD™ with the LMA-Classic™ during anaesthesia. INTERPRETATION. The AMD™ caused a greater number of complications and the efficacy of the AMD™ was less than that of the LMA-Classic™.
The year in anaesthesia and critical care
260
Comment In this randomized study Cook et al. planned to study 300 patients due to receive either the AMD™ (size 4 for patients 45–75 kg in weight and size 5 for patients >75 kg in weight) or the LMA-Classic™ (size 4 in females and size 5 in males). The primary outcome measure was the success of airway placement. The following factors were studied. 1. The success rate of insertion (up to three attempts). 2. The insertion time. 3. The oropharyngeal leak pressure. 4. The adequacy of ventilation (good represents optimal ventilation without complications during anaesthesia, fair represents optimal ventilation with complications or suboptimal ventilation and failure represents failed insertion or abandonment of use). 5. The position of the device assessed by fibrescope. 6. The numbers of manipulations during insertion and maintenance of anaesthesia. 7. The incidence of secretions interfering with airway management. 8. The incidence of blood on the device at removal. 9. Airway complications (during induction, maintenance and emergence from anaesthesia). For the AMD™ an attempt was made to pass a gastric tube through the device. Air was removed from the distal cuff and a gastric tube was inserted through the access port. If the gastric tube could not be advanced, a fibrescope was inserted and under direct vision air was removed or added until the distal cuff orifice was seen to open. Passage of the gastric tube was then attempted again. An interim analysis after 100 patients indicated that the study should be stopped because there was a marked difference in the efficacy between the two devices. The AMD™ was inserted on the first attempt significantly less frequently than the LMAClassic™ (Table 11.6). Eight AMDs™ and one LMA-Classic™ could not be placed within three attempts (P=0.03). The AMD™ required significantly more attempts, significantly more manipulations and caused significantly more complications during insertion. The AMD™ had to be removed in three patients during maintenance of anaesthesia as a result of complications, whereas this was not necessary in any patient with the LMA-Classic™. The oropharyngeal leak pressure was significantly higher with the AMD™ (25 cm H2O) than with the LMA-Classic™ (20 cm H2O). The efficacy of ventilation and the position of the device over the larynx were significantly better with the LMA-Classic™ than with the AMD™. It was only possible to pass a gastric tube via the AMD™ in two out of 32 attempts (even with fibreoptic assistance). During emergence from anaesthesia two AMDs™ but no LMA-Classic™ required premature removal. There was no significant difference in the incidence of blood on the devices. Nor was there a significant difference in the incidence of post-operative sore throat between the devices (Table 11.7). The overall failure rate of the AMD™ was significantly greater than that of the LMA-Classic™: the use of the AMD™ had to be abandoned in a total of 13 of 50 patients. Insertion of the AMD™ was significantly more difficult than the LMA-Classic™. The AMD™ was associated with a higher incidence of complications than the LMA-Classic™
New airway equipment
261
during insertion of, maintenance of and emergence from anaesthesia. The overall performance of the AMD™ was poorer than with the LMA-Classic™. From these and other studies it can be concluded that even the modified AMD™ is not an effective supraglottic airway.
Table 11.6 Success rate of insertion of the AMD™ or LMA-Classic™ LMA-Classic™* First attempt
AMD™* 45
35
Second attempt
4
9
Third attempt
1
6
* P=0.03. Source: Cook and Porter (2003).
Table 11.7 The incidence of blood on the device at removal and post-operative respiratory complications LMA-Classic™
AMD™
Blood on the device at removal
4/49
1/38
Sore throat
1/49
6/38
No significant differences between the groups. Source: Cook and Porter (2003).
Elisha Airway Device The Elisha Airway Device (Elisha Medical Technologies Ltd, Katzrin, Israel), a new reusable supraglottic airway, was designed based on spiral computed tomographs of the oropharyngeal cavity in subjects weighing 50–70 kg, followed by post-processing with multi-planar reconstruction. The device is constructed of latex-free, medical-grade silicone and has three separate channels for ventilation, tracheal intubation and gastric tube insertion. One of the most important features of the Elisha Airway Device is that the ventilation channel and the intubation channel are independent, existing side by side and this unique design allows tracheal intubation without inter-ruption of ventilation. The ventilation channel and the intubation channel have a partitioning wall between them, but they join at the ventilation outlet situated in front of the laryngeal inlet. The intubation channel allows passage of a tracheal tube (up to 8.0 mm ID) for blind or fibreoptic intubation. The gastric tube channel has an outlet located in the distal end of the device. None of the existing supraglottic ventilatory devices combines all three functions of ventilation, tracheal intubation and passage of a gastric tube in one device. There are two high-volume, low-pressure balloons. The proximal balloon seals the oropharynx and
The year in anaesthesia and critical care
262
nasopharynx, whereas the distal balloon occludes the oesophageal inlet. These balloons can be inflated through a single pilot tube.
A new supraglottic airway, the Elisha Airway Device: a preliminary study Vaida SJ, Gaitini D, Ben-David B, Somri M, Hagberg CA, Gaitini LA. Anesth Analg 2004; 99: 124–7 BACKGROUND. The objective of this study was to assess the success rate of insertion of the Elisha Airway Device, adequacy of ventilation, the success rate of tracheal intubation through the device and the incidence of post-operative airway complications. INTERPRETATION. Insertion of the Elisha Airway Device was generally easy and the success rate of blind tracheal intubation though the device was high. The incidence of post-operative airway complications was acceptably low. Comment In this observational study Vaida et al. studied the efficacy of the Elisha Airway Device in 70 anaesthetized and paralysed patients. The following factors were studied. 1. The success rate of insertion. 2. The insertion time. 3. The oropharyngeal leak pressure. 4. The success rate of insertion of a gastric tube (18 Fr). 5. The success rate of blind tracheal intubation (when blind inubation failed after two attempts fibreoptic intubation was attempted). 6. The presence or absence of blood on the device at removal. 7. Post-operative airway complications. Insertion of the Elisha Airway Device was successful in 67 out of 70 (96%) patients (first attempt in 76% and second attempt in 20%), with a mean insertion time of 20 s. The average oropharyngeal leak pressure was 27 cm H2O. In these patients it was always possible to maintain oxygenation and ventilation throughout the surgical procedure. Insertion of a gastric tube through the device was successful in all cases. Tracheal intubation through the Elisha Airway Device was attempted in 20 patients. Blind intubation was successful in 17 out of the 20 patients (15 patients at the first attempt and two patients at the second attempt). In the remaining three patients fibreoptic intubation was successful in two patients. Blood was detected on the device after its removal in six patients (8%). Post-operatively four patients (6%) complained of a sore throat and three patients (4%) of hoarseness. This observational study suggests that the Elisha Airway Device can function as an effective airway, as a conduit to tracheal intubation and for insertion of a gastric tube. Pharyngeal Airway xpress™
New airway equipment
263
The Pharyngeal Airway xpress™ (PAxpress™, Vital Signs Inc., NJ, USA), a single-use supraglottic airway device, is made of PVC. It consists of an anatomically curved tube with a cuff in its midsection and a non-inflatable gilled conical tip at the distal end. It has been modified three times since its original description. The cuff in the midsection forms an airtight seal in the oropharynx and the distal end forms a no-pressure seal in the oesophageal inlet. There is a rectangular hooded vent between the cuff and tip, facing anteriorly towards the glottis. The distal half of the vent has three vertical gills to prevent airway obstruction. It is possible to insert a tracheal tube (up to 8.0 mm ID) through the PAxpress ™. Currently, there is only one size for adults.
A comparison of the PAxpress™ and facemask plus Guedel airway by inexperienced personnel after mannequin-only training Dimitriou V, Voyagis GS, latrou C, Brimacombe J. Anesth Analg 2003; 96: 1214–17 BACKGROUND. The objective of this study was to compare the ability of nurses with no previous experience of airway management for ventilating through the PAxpress™ and a facemask (with the aid of a Guedel airway). INTERPRETATION. Inexperienced nurses after mannequin-only training could produce a target tidal volume of 7 ml/kg in anesthetized, paralysed adults, either through the PAxpress™ or facemask (with a Guedel airway). The maximal tidal volume was larger, peak airway pressure higher and airway trauma more common with the PAxpress™. Comment In this non-randomized, comparative study, Dimitriou et al. compared the ability of nurses with no previous experience of airway management for ventilating the lungs of 45 anaesthetized and paralysed patients using either the PAxpress™ or a facemask (with a Guedel airway). Each nurse performed 15 insertions with each device with mannequins. Patients were then studied and nurses were asked to ventilate to an expired target tidal volume of 7 ml/kg and then to the maximum tidal volume achievable. The facemask was used first and then the PAxpress™. The following factors were studied. 1. The number of insertion attempts (up to three attempts). 2. The ease of insertion (easy or difficult). 3. The insertion time. 4. The maximum tidal volume. 5. The peak airway pressure. 6. The presence or absence of insufflation of gas into the oesophagus (by listening over the epigastrium with a stethoscope). 7. The oesophageal leak pressure. 8. The presence or absence of blood on the device at removal.
The year in anaesthesia and critical care
264
There were no significant differences in the time taken (PAxpress™ 41s and facemask 39 s) or the number of insertion attempts to achieve the target tidal volume. The target tidal volume was achieved in all patients with both devices. There were no significant differences in the frequency of oesophageal leaks at the target (PAxpress™ 9% and facemask 4%) and maximum tidal volume (PAxpress™ 51% and facemask 49%). The maximal tidal volume (1261 versus 958 ml) and peak airway pressure (37 versus 28 cm H2O) was significantly larger for the PAxpress™, but blood was detected significantly more frequently (22 versus 0%) (P=0.001). Ventilation to a target tidal volume of 7 ml/kg in anaesthetized, paralysed adults was equally successful with the PAxpress™ and facemask by inexperienced nurses after mannequin-only training. The maximal tidal volume and peak airway pressure were higher with the PAxpress™. However, airway trauma was more common with the PAxpress™. The limitations of the study are that the study was not randomized, and that only three nurses participated in the study, making the results less applicable to the general nursing (or anaesthetic) population.
The PAxpress™ is an effective ventilatory device but has an 18% failure rate for flexible lightwand-guided trachea! intubation in anesthetized paralyzed patients Dimitriou V, Voyagis GS, latrou C, Brimacombe J. Can J Anesth 2003; 50: 495–500 BACKGROUND. The main objective of this study was to determine the success rate for tracheal intubation through the PAxpress™ with the aid of a light wand. INTERPRETATION. The success rate of light wand-guided tracheal intubation through the PAxpress™ was 82%, but the success rate at the first attempt was 60%.
Comment In this observational study, Dimitriou et al. studied the success rate of tracheal intubation (7.0 or 7.5 mm ID in females and 8.0 mm ID in males) through the PAxpress™ in 90 anaesthetized and paralysed patients. The following factors were studied. 1. The success rate of intubation through the PAxpress™. 2. The intra-cuff pressures in vitro at 30–60 ml cuff inflation volume. 3. The oropharyngeal leak pressure. 4. The expired tidal volume at 30–60 ml and at airway sealing pressure. 5. The calculated mucosal pressures (ex vivo-in vivo pressure difference). 6. The oesophageal leak pressure (by listening over the epigastrium with a stethoscope). 7. The presence or absence of blood on the PAxpress™ at removal.
New airway equipment
265
Insertion of the PAxpress™ was successful at the first attempt in 86 out of 90 (96%) patients and at the second attempt in the remaining four patients. The mean oropharyngeal leak pressures at cuff volumes of 30, 40, 50 and 60 ml were 27, 29, 32 and 35 cm H2O, respectively, the expired tidal volumes at the leak pressure were 16, 18, 19 and 19 ml/kg, respectively and the calculated mucosal pressures were 38, 55, 56 and 57 cm H2O, respectively. The oropharyngeal leak pressure, expired tidal volume at the leak pressure and calculated mucosal pressures significantly increased with cuff inflation volume. Oesophageal leak was detected in eight out of 90 (9%) patients, but only at peak pressures >35 cm H2O and cuff volumes ≥40 ml. Light wand-guided tracheal intubation was successful in 74 out of 90 (82%) patients. Of these, 54 (60%) patients were intubated at the first attempt without any adjusting manoeuvres. Mild, moderate and severe bloodstaining was detected in 36 (40%) patients, 13 (15%) patients and one (1%) patient, respectively. Bloodstaining was more frequent after adjusting manoeuvres (22 out of 54 patients versus 32 out of 36 patients). The PAxpress™ has a high insertion success rate and is an effective ventilatory device with a low risk of gastric insufflation, but has a moderately high failure rate for light wand-guided intubation and is associated with a relatively high incidence of trauma to the mucosa. The pressure exerted by the device on the oropharynx may exceed pharyngeal perfusion pressure. CobraPLA airway The perilaryngeal airway (CobraPLA, Engineered Medical Systems, Indianapolis, IN, USA), a supraglottic airway invented by Dr David Alfery, consists of a breathing tube with a wide distal end and a cuff attached in the mid-portion of the breathing tube. The tip of the device is designed to occupy the hypopharynx and the cuff seals off the distal end from the upper airway. The widened distal end has a smooth posterior surface and soft grilles covering the airway orifice. The distal aperture of the breathing tube is inside the distal end. Once in place the distal aperture should face directly against the glottis with the anterior wall holding the epiglottis out of the way. The anterio-posterior width of the distal end is smaller than the LMA-Classic™. Therefore, it may require a lesser mouth opening for insertion and it might be easier to insert than the LMA-Classic™.
The new perilaryngeal airway (CobraPLA) is as efficient as the laryngeal mask airway (LMA) but provides better airway sealing pressures Akca 0, Wadhwa A, Sengupta P, et al. Anesth Analg 2004; 99: 272–8 BACKGROUND. The objective of this study was to compare the efficacy of the CobraPLA with the LMA-Classic™ during anaesthesia. INTERPRETATION. The CobraPLA was as easy to insert as the LMA-Classic™, was as effective as the LMA-Classic™ in maintaining a clear airway and had a better sealing effect.
The year in anaesthesia and critical care
266
Comment In this study Akca et al. randomly allocated 81 patients to receive either the CobraPLA or LMA-Classic™. The test device was inserted after induction of anaesthesia with 2–3 ml/kg of propofol (without injection of a neuromuscular blocking agent). Anaesthesia was maintained with fentanyl, and sevoflurane and nitrous oxide in oxygen. Patients in whom the estimated duration of anesthesia was less than 1 h were allowed to breathe spontaneously. Otherwise, rocuronium was given and the lungs were mechanically ventilated. The following factors were studied. 1. The success rate of insertion (up to two attempts). 2. The insertion time.
Table 11.8 Post-operative airway complications CobraPLA (n=40)
LMA-Classic (n=41)
Visible blood on the device
19
14
Sore throat
16
9
Dysphagia
7
5
Dysphonia
8
6
Source: Akca et al. (2004).
3. The oropharyngeal leak pressure. 4. The number of repositioning attempts. 5. The position of the device assessed by fibrescope (in 35 patients). 6. The oropharyngeal leak pressure. 7. The adequacy of the airway (no leak at 15 cm H2O peak pressure or tidal volume of 5 ml/kg). 8. The airway sealing quality (perfect means no leak detected). 9. The presence or absence of gastric insufflation. 10. Post-operative airway complications. Overall, the airways were successfully managed in 39 out of 40 patients in the CobraPLA group and in all 41 cases in the LMA-Classic™ group. The oropharyngeal leak pressure was significantly higher for the CobraPLA group (mean 23 cm H2O and range 8–36 cm H2O) than for the LMA-Classic™ group (mean 18 cm H2O and range 8–32 cm H2O). The insertion times, number of attempts and positions of the devices were similar in the groups. Perfect airway sealing was obtained significantly more frequently in the CobraPLA group (21 out of 40 patients) than the LMA-Classic™ group (13 out of 40 patients). Gastric insufflation occurred in six patients in the CobraPLA group and four patients in the LMA-Classic™ group. There were no significant differences in postoperative airway complications between the two groups (Table 11.8). It appears that the CobraPLA is a useful airway device during anaesthesia. It is not known whether tracheal intubation through the CobraPLA is as easy as intubation through the LMA-Classic™.
New airway equipment
267
Conclusion It is likely that the use of supraglottic airways will increase and that of tracheal intubation will decrease. The LMA has certainly established the main position among supraglottic airways, but it seems that some new supraglottic airways may also have potential roles during anaesthesia. It is necessary to decide which devices are more suitable than others by carrying out comparative studies and by using each device in a large number of patients in different circumstances. References 1. Brain AI. The development of the Laryngeal Mask—a brief history of the invention, early clinical studies and experimental work from which the Laryngeal Mask evolved. Eur J Anaesthesiol Suppl 1991; 4: 5–17. 2. Asai T, Brimacombe J. Cuff volume and size selection with the laryngeal mask. Anaesthesia 2000; 55: 1179–84. 3. Williams PJ, Bailey PM. Comparison of the reinforced laryngeal mask airway and tracheal intubation for adenotonsillectomy. Br J Anaesth 1993; 70: 30–3. 4. Brain AIJ, Verghese C, Strube PJ. The LMA-ProSeal—a laryngeal mask with an oesophageal vent. Br J Anaesth 2000; 84: 650–4. 5. Brimacombe J, Keller C. The ProSeal laryngeal mask airway: a randomized, crossover study with the standard laryngeal mask airway in paralyzed, anesthetized patients. Anesthesiology 2000; 93: 104–9. 6. Goodwin AP, Rowe WL, Ogg TW. Day case laparoscopy. A comparison of two anaesthetic techniques using the laryngeal mask during spontaneous breathing. Anaesthesia 1992; 47: 892– 5. 7. Swann DG, Spens H, Edwards SA, Chestnut RJ. Anaesthesia for gynaecological laparoscopy—a comparison between the laryngeal mask airway and tracheal intubation. Anaesthesia 1993; 48: 431–4. 8. Maltby JR, Beriault MT, Watson NC, Pick GH. Gastric distension and ventilation during laparoscopic cholecystectomy: LMA-Classic versus tracheal intubation. Can J Anaesth 2000; 47: 622–6. 9. Brimacombe J, Keller C. Aspiration of gastric contents during use of a ProSeal laryngeal mask airway secondary to unidentified foldover malposition. Anesth Analg 2003; 97: 1192–4. 10. Koay KC. A case of aspiration using the ProSeal LMA. Anaesth Intensive Care 2003; 31: 123. 11. Asai T. Who is at increased risk of pulmonary aspiration? (editorial). Br J Anaesth 2004; 93: 497–500. 12. Howath A, Brimacombe J, Keller C. Gum-elastic bougie-guided insertion of the ProSeal laryngeal mask airway: a new technique. Anaesth Intensive Care 2002; 30: 624–7. 13. Cook TM, McCormick B, Asai T. Randomized comparison of the Laryngeal Tube and the laryngeal mask during anaesthesia with controlled ventilation. Br J Anaesth 2003; 91: 373–8. 14. Asai T, Kawashima A, Hidaka I, Kawachi S. The Laryngeal Tube compared with the laryngeal mask: insertion, gasleak pressure and gastric insufflation. Br J Anaesth 2002; 89: 729–32. 15. Wrobel M, Grundmann U, Wilhelm W, Wagner S, Larsen R. Laryngeal Tube versus laryngeal mask airway in anaesthetised non-paralysed patients. A comparison of handling and postoperative morbidity. Anaesthesist 2004; 53: 702–8. 16. Ocker H, Wenzel V, Schmucker P, Steinfath M, Dorges V. A comparison of the Laryngeal Tube with the laryngeal mask airway during routine surgical procedures. Anesth Analg 2002; 95: 1094–7.
The year in anaesthesia and critical care
268
17. Asai T, Neil J, Stacey M. Ease of placement of the laryngeal mask during manual in-line neck stabilization. Br J Anaesth 1998; 80: 617–20. 18. Brimacombe J, Berry A. Laryngeal mask airway insertion. A comparison of the standard versus neutral position in normal patients with a view to its use in cervical spine instability. Anaesthesia 1993; 48: 670–1. 19. Pennant JH, Pace NA, Gajraj NM. Role of the laryngeal mask airway in the immobile cervical spine. J Clin Anesth 1993; 5: 226–30. 20. Asai T, Wagle AU, Stacey M. Placement of the intubating laryngeal mask is easier than the laryngeal mask during manual in-line stabilization. Br J Anaesth 1999; 82: 712–14. 21. Asai T, Murao K, Shingu K. Efficacy of the ProSeal® laryngeal mask airway during manual in-line stabilisation of the neck. Anaesthesia 2002; 57: 918–20. 22. Cook TM, Gupta K, Gabbott DA, Nolan JP. An evaluation of the Airway Management Device. Anaesthesia 2001; 56: 660–4.
12 Ultrasound techniques for central venous access and regional anaesthesia GORDON CHAPMAN, ANDREW BO DENHAM HAM Introduction Anaesthetists and intensivists spend a considerable proportion of their working time inserting needles and catheters into patients. In order to access deeper structures such as central veins and nerves, they have traditionally relied on surface markings for guiding the needle into the correct position. Problems may arise when relying on surface landmarks due to anatomical abnormalities, obesity, the size of small children and anticoagulation status. Irrespective of the skill of the operator, there is the ever-present risk of needle misplacement with the potential for damage to structures such as veins, arteries, nerve bundles and the pleura. Occasionally such damage may have devastating implications for both the patient and practitioner. For all these reasons practitioners have tried many different techniques in order to try and improve the safety and efficacy of such interventions. In recent years the introduction of portable and affordable high-resolution ultrasound scanners has accelerated interest in the use of ultrasound for interventional procedures. In the UK the National Institute for Clinical Excellence (NICE) has recommended the routine use of ultrasound for elective central venous catheterization |1|. This should have led to increased access to this imaging modality in the UK. Increased availability of ultrasound equipment and skills has been instrumental in the development and use of ultrasound for other applications, notably nerve blocks and the diagnosis or drainage of pleural collections. In this chapter we will discuss recent publications relating to the use of ultrasound for vascular access and nerve blockade. Vascular access In the UK two-dimensional ultrasound guidance is recommended by the NICE as the preferred method for the insertion of central venous catheters (CVCs) into the © Atlas Medical Publishing Ltd
internal jugular vein in adults and children in elective situations. It recommends that its use should be considered in most clinical circumstances where CVC insertion is necessary either electively or in an emergency. However, due to a lack of evidence NICE has not recommended that ultrasound guidance be used for all subclavian or femoral CVC placements, rather that it should be considered.
The year in anaesthesia and critical care
270
The advantages of ultrasound guidance for central venous cannulation include a greater likelihood of success, fewer complications and less time spent on the procedure |2|. This may result in increasing patient satisfaction and, despite the costs of equipment and training, may favour a modest financial benefit in the long term |3–5|. Yet, despite the potential benefits that ultrasound guidance offers over the traditional landmark-based techniques, there has been resistance to routine use of the technique |6|. As it is impossible to identify reliably all patients who may present difficulties with central venous cannulation and as there are no patients in whom complications are not unpleasant and potentially serious, this is difficult to comprehend. Ultrasound allows for the visualization of deeper vessels not traditionally favoured for venous access. This includes the cannulation of the infraclavicular axillary vein and the mid-arm approach to the basilic and cephalic veins. Cannulation of the femoral vein may also be performed further distally in the upper thigh using ultrasound guidance. Ultrasound has its limitations in that it will not easily show a central position of the catheter tip, which requires verification by a chest X-ray, electrocardiograph or transoesophageal echo. However, even this has been challenged |7|. The use of ultrasound for identifying a suitable target vessel and visualizing the surrounding anatomy allows a planned approach. The pathway to and the site of vessel puncture can be determined and viewed in real-time to avoid specific surrounding structures. This was illustrated by the anatomical study of the axillary vein using ultrasound by Galloway and Bodenham |8|. The authors proposed that the axillary vein is a suitable alternative to the subclavian vein for central venous access. The vessel can be easily imaged using ultrasound. The margin for safety is potentially increased, as there is less arterio-venous overlap, a greater distance between the artery and vein and more space between the vein, rib cage and pleura. A clinical report followed (see below).
Ultrasound-guided infraclavicular axillary vein cannulation for central venous access Sharma A, Bodenham AR, Mallick A. Br J Anaesth 2004; 93: 188–92 BACKGROUND. It is commonly believed that ultrasound guidance cannot easily be used in the subclavian area due to the physical presence of the collarbone. Whilst this is true for the medial section of the subclavian vein, the axillary vein (from which the subclavian vein originates) can be easily imaged from an infraclavicular site (see Figs. 12.1 and 12.2 on p. 287). INTERPRETATION. The authors provided a formal description of this technique based on 200 consecutive patients having Hickman lines inserted under local anaesthesia with
Ultrasound techniques for central venous access
271
Fig. 12.1 Approach to the right axillary vein. The ultrasound probe (U) in a sterile sheath has been used for showing the axillary vessels in crosssection. Note the steep angle of approach and insertion depth of the introducing needle (overall needle length 7 cm). Blood has been aspirated into the syringe after the needle enters the vein. Source: Sharma et al. (2004). intravenous sedation. The procedure was successful in 96% of cases, but there was guide wire and catheter malposition, usually to the neck, in 15% of cases. In this series there were three cases of inadvertent axillary artery puncture and no pneumothoraces reported. Comment This description provides an alternative route for central venous access in those patients where subclavian cannulation is indicated. The use of ultrasound would appear to have the potential for reducing many procedural complications related to blind landmark-based subclavian puncture, e.g. arterial puncture, pneumothorax and potential nerve damage. Interestingly, the approach is very similar to that described by Nickalls |9| for the original blind, landmark-based puncture of the vein and ultrasound-guided infraclavicular brachial plexus block by Sandhu and Capan |10|.
The year in anaesthesia and critical care
272
NICE guidelines for central venous catheterisation in children. Is the evidence base sufficient? Grebenik CR, Boyce A, Sinclair ME, Evans RD, Mason DG, Martin B. Br J Anaesth 2004; 92(6): 827–30 BACKGROUND. Central venous cannulation in infants can be particularly challenging and the risk of complications is increased |11|. Whilst the NICE guidelines recommend the use of ultrasound guidance for central venous catheterization in children the evidence on which this is based is small (180 patients).
Fig. 12.2 This cross-sectional ultrasound image of the right infraclavicular area shows the right axillary vein (V) in the centre of the screen and the adjacent artery (A) The path of the advancing
Ultrasound techniques for central venous access
273
needle is seen above the vein. The white arrow shows the approximate position of the needle tip. This is easier to see when it is being advanced during real-time imaging. Source: Sharma et al. (2004). INTERPRETATION. In this study 134 infants and children aged between 1 day and 8 years presenting for heart surgery were randomized to either ultrasound-guided or traditional landmark-based central venous catheterization. All procedures were carried out by one of three experienced consultant paediatric cardiac anaesthetists. On ten occasions the ultrasound device was unavailable and these patients were excluded from further analysis. There were seven failures at cannulating the internal jugular vein (10.8%) in the landmark group, with three associated with arterial puncture and haematoma formation. There were 13 failures (22%) in the ultrasound group, with seven associated with carotid artery puncture and four patients who developed a haematoma. All failures except one occurred in patients <10 kg or under 1 year in age. The success rates were generally greater with children older than 2 years. The success rates for cannulation were significantly greater in the landmark group (P <0.002) and the arterial puncture rates were also significantly lower (P <0.03). There was no significant time difference noted between the two groups. The most common problem noted in both groups was the inability to pass the guide wire despite successfully aspirating venous blood. Comment These results are in contrast with the published data on which the NICE guidelines are based. The authors cited many reasons including difficulties with the bulky ultrasound probe and needle guide (a guiding device built in or clipped onto the probe head) in these small infants. However, many clinicians do not find it necessary to use needle guides and use a freehand technique. This method allows more flexibility in that the needle may be inserted at a more convenient and accessible site, 1 cm or so away from the probe and then be directed into the ultrasound beam before being advanced towards the target vein. It is difficult to understand how ultrasound guidance can lead to an increased number of carotid artery punctures. The artery and vein should be easy to differentiate and, if not, a puncture should not be undertaken. Most modern ultrasound machines have the facility of colour Doppler, which makes this easier. All three operators were apparently able to identify the internal jugular vein, but then punctured the carotid artery and cited unsuitability of the equipment as the cause. The equipment used, a Bard Site-Rite 3 ultrasound probe with attached needle guide (Dymax Corp., Pittsburgh, PA, USA), may be considered outdated and suboptimal in this group of patients. The image quality is poor and it lacks colour Doppler. Whilst the operators accepted that possibly they were inadequately experienced in this technique, there was no trend towards improvement during the study. The study population represented a potentially difficult paediatric patient group and we would
The year in anaesthesia and critical care
274
suggest that considerable experience in using ultrasound for vascular access may be required prior to its use in this setting. The study highlighted the difficulties experienced using ultrasound guidance for CVC placement in a paediatric population. However, as a minimum, ultrasound should be able to identify the 6% of children whose veins can be expected to be difficult to cannulate |12|. Blind, landmark-based techniques are likely to be dangerous in these circumstances. The authors suggested that the NICE guidelines on central venous access should be amended to remove references to infants and children until the results of further assessments are available. We feel that this study needs to be repeated by operators experienced in both ultrasound and landmark-based techniques, using state-of-the-art, high-frequency small footprint probes before any such conclusions should be drawn.
Mid-arm approach to basilica and cephalic vein cannulation using ultrasound guidance Sandhu NPS, Sidhu DS. Br J Anaesth 2004; 93(2): 292–4 BACKGROUND. Obtaining peripheral intravenous access can occasionally be a challenge even to experienced physicians, particularly in infants, obese adults, intravenous drug abusers and oedematous patients. Difficulty may be caused by thrombosis or scarring of normally visible veins or an inability to visualize or palpate veins as a result of subcutaneous fat or oedema. After multiple attempts at peripheral cannulation, physicians often resort to central venous catheterization with its associated risks. This can be frustrating for the clinicians involved and unpleasant and risky for the patient. The basilic and cephalic veins offer an alternative site. The basilic vein lies beneath the deep fascia in the upper two-thirds of the arm and is almost always patent, even in long-term intravenous drug abusers. The cephalic vein is more superficial and is usually well preserved in most obese patients, being deeply buried in adipose tissue. INTERPRETATION. This ultrasound-guided technique was described with four case reports illustrating the usefulness of this site. In the authors’ experience of over 120 cases, by ensuring a sufficient length of catheter inside the vein (>2.5 cm), they reported no case of extravasation. They did not experience any nerve injury in the series. The midarm location reduces the chance of dislodgement of the catheter by movement at the elbow. Real-time ultrasonography made cannulation possible in <1 min with catheterover-needle devices and <5 min using the Seldinger technique, timed from the first needle puncture until suturing of the catheter. The authors advocated the use of these vessels for short-term venous access in the operating room or immediate perioperative period. The use of ultrasound guidance for venous access in the mid-arm region appears a practical and safe alternative to central venous catheterization in situations where it is otherwise difficult to gain intravenous access.
Ultrasound techniques for central venous access
275
Comment This approach lies between peripheral and central venous catheterization and has similarities to previous techniques of long lines and midline catheters. Its use as an alternative site for the administration of vasoactive drugs requires further evaluation. However, this site could be used for facilitating CVC placement with a longer catheter. Similar approaches have been used for peripherally inserted central catheters inserted from the antecubital fossa, with vasoactive and chemotherapy drugs given via this route. The safety of such catheters with regards to the administration of these drugs relies not on the site of insertion, but rather on their length and the position of the catheter tip. Clinicians need to be more aware of the potential benefits that ultrasound can offer with regard to peripheral venous and arterial cannulation. Section summary The use of ultrasound is an example where an established technology from one speciality has been adopted and applied for uses in another. In our opinion the role of ultrasound in vascular access is now well established. Ultrasound is particularly useful in locating an optimal site and pathway for vessel puncture and avoids multiple needle passes in cases of anatomical variation. Ultrasound will not ensure the central passage of the guide wire and catheter or avoid the damage caused by the dilator or passage and presence of the guide wire and catheter. A small number of patients who have had multiple, long-term, central venous catheterizations will develop central vessel occlusions. These patients may present with difficulty in passing the guide wire despite aspiration of venous blood. The site of the occlusion may not be apparent with ultrasound although this should be suspected from the history, the presence of venous collaterals and the difficulty in passing the guide wire. In such cases venography may be useful in localizing the occlusion and determining further possible access points (see Fig. 12.3). There have been calls in previous studies for establishing the effectiveness of ultrasound for vascular access further. We see it as unethical for practitioners with ultrasound skills and access to devices to have to revert to blind techniques for the sake of performing larger scale controlled trials with full randomization simply to supply an evidence base for all routes of access. The NICE guidelines are clear. Despite the fact that anaesthetists are responsible as a group for a major proportion of CVC insertions, we suspect that the majority have as yet restricted access to ultrasound and limited training (if any) in using this imaging modality for facilitating central venous catheterization. Clearly the challenge now is to ensure the training of individuals and departments in the use of ultrasound and its applications in anaesthesia and intensive care. Ultrasound guidance for central venous catheterization and for assisting peripheral venous cannulation is likely to become firmly established as standard practice at most anatomical sites and for all ages of patients.
The year in anaesthesia and critical care
276
Fig. 12.3 Left venogram in a patient with failed attempts to pass a left-sided CVC centrally after ultrasound-guided needle puncture. Intravenous contrast has been injected via a peripheral arm vein. An occluded left innominate vein can be seen, with venous drainage via the azygous and hemiazygous system. This followed multiple, long-term central venous catheterizations. Training and accreditation in the use of ultrasound Information regarding training and accreditation in the use of ultrasound is available. These documents may be more applicable to radiologists and ultrasonographers than anaesthetists at the present time. In the UK there is a national shortage of qualified ultrasonographers, with many departments struggling with limited resources and increasing demand for services. Training of anaesthetists in the use of ultra-sonography may place additional burdens on the clinical departments undertaking it. Examinations
Ultrasound techniques for central venous access
277
and interventions performed by staff not specifically nor adequately trained in ultrasound scanning and interpretation may be dangerous to patients.
Extending the provision of ultrasound services in the United Kingdom Bates J, Deane C, Lindsell D. Available from the British Medical Ultrasound Society website: www.bmus.org, September 2003
Guidance for training in ultrasound and medical non-radiologists ISBN 1872599 30 3 code BFCR (97)1. Available from the Royal College of Radiologists website: www.rcr.ac.uk, 1997 (this document dates from 1997 but is currently being updated). Additional information is also available from, for example, the European Federation of Societies for Ultrasound in Medicine and Biology (EFSUMB), Educational and Professional Standards Committee January 2003 website: www.efsumb.org. BACKGROUND. It is one thing for a single person or a small number of enthusiasts to take the trouble to research and learn the safe and effective use of ultrasound. It is an altogether more difficult task to get a whole department, with a large number of individuals, to attain and retain such skills. Such issues are being addressed by a number of specialties such as obstetrics (perinatal ultrasound), cardiology (echocardiography) and emergency physicians. INTERPRETATION. There is a very valuable discussion in these documents, but their true relevance to anaesthesia and intensive care can be debated. Many of the recommendations regarding the training time and experience required for each particular examination are probably impractical for many staff. Although most interest in anaesthesia and critical care is related to interventional procedures, i.e. getting the needle into the right place in the body, there is clearly an element of diagnosis required in order to identity reliably target structures and any abnormality, e.g. thrombosed or collateral veins in the case of venous access. Comment An anaesthetic department in a large hospital may have 100 consultant and trainee anaesthetists. Training requires a considerable time commitment from trainers and trainees with the risk of serious disruption to services. This is a major undertaking for vascular access alone, which may be considered relatively simple in comparison with ultrasound guidance for nerve blockade.
The year in anaesthesia and critical care
278
Nerve blockade There has been increasing interest in the use of ultrasound in visualizing nerves in recent years as ultrasonographers realized the potential for the use of high-resolution ultrasound for imaging peripheral nerves and more central nerve trunks. Ultrasound-assisted nerve blockade has been reported in the anaesthetic literature since 1978 |13|. Increasing interest in this field has been the result of advances in technology with higher frequency (and, thus, increased resolution) probes, together with improved imaging enhancement software. A recent review by Beekman and Visser gave a very good description of the ultrasound appearance of nerves and described the problems associated with imaging nerves using ultrasound |14|. Modern clinical ultrasound probes typically operate in the 2.5–15 MHz frequency range. The higher the ultrasound frequency, the better the spatial resolution, but this is at the expense of a reduced depth of penetration. Lower frequencies provide better depth penetration, but at a lower spatial resolution. There is thus a trade-off between depth penetration and image quality. Most nerve block applications require frequencies in the range of 10–14 MHz. Ultrasound assistance demands detailed knowledge of the regional anatomy in order to identify surrounding structures and interpret images when seeking out nerves. A measure of hand-to-eye coordination is necessary and the ability for processing the twodimensional images while thinking and working in a three-dimensional environment requires patience and practice. As a general rule, nerves do not show up particularly well with ultrasound, as their structure and composition is of similar acoustic impedance to surrounding soft tissues. Nevertheless, with high-frequency ultrasound they can be seen during transverse scanning as hypoechoic, oval or round areas surrounded by a relatively bright hyperechoic border. Good examples of this include the trunks of the brachial plexus. Peripherally, where the nerves are closer to the skin (e.g. the wrist), the use of very highresolution ultrasound will allow the individual fascicles making up the nerve to be appreciated both on cross-section and longitudinal imaging. If deeper nerve bundles cannot be seen directly with ultrasound, their position may be inferred by the presence of other structures, e.g. arteries and veins in the appropriate neurovascular bundle. Ultrasound guidance for nerve blockade could be divided into three techniques. 1. Where the nerves themselves are reliably visualized and the needle can be advanced directly towards the nerve. 2. Where the nerve itself cannot be reliably visualized, surrogate markers, such as the artery and vein, which run in the same neurovascular bundle, can be used for determining the locality of the target nerves. 3. Where other anatomical structures, such as bone, muscle or fascia, guide needle placement. Local anaesthetic solution is then injected to fill an anatomical space, e.g. ultrasound-guided psoas compartment block |15| or local anaesthetic injections along the long saphenous vein for laser ablation. Ilio-inguinal, rectus sheath and prostatic nerve blocks may be considered similarly.
Ultrasound techniques for central venous access
279
Ideally local anaesthetic should be delivered precisely adjacent to the target nerve, avoiding damage to the nerve itself or related structures. Blind blocks that rely solely on anatomical landmarks and fascial clicks (e.g. ilio-inguinal) are known to have the potential for producing serious complications |16, 17|. There is little doubt that ultrasound may help guide needle placement and potentially reduce damage to arteries, veins and pleura, but whether or not it will reduce the incidence of nerve trauma is yet to be established (see Marhofer et al. (2005) below). Ultrasound can be used for identifying the point at which a nerve divides. Local anaesthetic can then be administered proximal to this or directed at individual nerve branches. Ultrasound enables real-time visualization of the spread of local anaesthetic and the possibility of repositioning the needle in cases of maldistribution. Other potential advantages of ultrasound guidance for nerve blockade include a significant reduction of the dose of local anaesthetic, a faster sensory onset time, longer duration of blocks and improved block quality (see Marhofer et al. (2004) below and |10, 18, 19|). The use of ultrasound for nerve location may mean a nerve stimulator is not required. There is considerable interest and literature developing in the field of ultrasound for nerve blockade. There has been debate in the literature for and against techniques and it is currently best considered as a developmental and research tool. However, it is likely that the purchase of machines for vascular access and other uses will accelerate interest in this field.
Ultrasound guidance for infraclavicular brachial plexus anaesthesia in children Marhofer P, Sitzwohl C, Greher M, Kapral S. Anaesthesia 2004; 59: 642–6 BACKGROUND. This study extended work performed by the same authors in adults |20|. Forty patients aged 1–10 years and weighing 14–42 kg were randomized to have brachial plexus block performed under light sedation with or without ultrasound. The use of ultrasound reduced pain during the procedure and resulted in a shorter onset time for anaesthesia and a denser sensory and motor block. INTERPRETATION. By avoiding the use of a nerve stimulator in the ultrasound group, the pain associated with muscle contractions due to nerve stimulation was avoided. All anaesthetic procedures were uneventful with no clinical signs of pneumothorax, inadvertent puncture of major vessels, haematoma or infection. All surgical procedures were performed under brachial plexus anaesthesia with midazolam or propofol sedation. The authors reported no conversions to general anaesthesia. Comment In children regional anaesthesia is traditionally performed under general anaesthesia. There are likely to be differences in culture and attitude towards regional anaesthesia in an awake child in different countries. In the UK at the current time parents, children and hospital staff are likely to expect and demand a general anaesthetic.
The year in anaesthesia and critical care
280
In a child the nerve lies relatively closer to the skin and may be more easily located with ultrasound. Such findings may extend the role of ultrasound guidance for regional anaesthesia in the smaller child.
Ultrasound guidance speeds execution and improves the quality of supraclavicular block Williams SR, Chouinard P, Arcand G, et al.. Anesth Analg 2003; 97: 1518–23 BACKGROUND. Ultrasound-guided brachial plexus blocks have been well described in the literature |10, 21–23|. The nerves lie relatively superficially and can usually be visualized in relation to the surrounding artery and vein. During the supraclavicular and infraclavicular approaches the pleura, artery and vein may be identified and avoided. INTERPRETATION. In this study the authors randomized 80 patients into two groups. Group 1 had supraclavicular block using real-time ultrasound imaging. Group 2 had a supraclavicular block using the subclavian perivascular approach. Both groups used a nerve stimulator for confirming the final needle position. The authors concluded that ultrasound-guided, neurostimulator-confirmed supraclavicular block is significantly quicker to perform (P=0.0001). Ultrasound-guided blockade also provided a more complete block than that using anatomical landmarks and neurostimulation, but the difference was not statistically significant. Comment Ultrasound guidance for supraclavicular brachial plexus block is significantly quicker to perform than landmark-based techniques. The anatomy of the nerve plexus and its position in the neurovascular bundle is well known to those anaesthetists performing this block regularly and it may be the first of the regional techniques to be taught using ultrasound imaging. In the past the axillary approach to the brachial plexus has been the favoured technique for providing regional anaesthesia for the arm, as it avoids the complication of pneumothorax, which is often quoted to occur at approximately 1% of supraclavicular brachial plexus blocks. However, the block produced with an axillary approach is possibly inferior to that with the supraclavicular approach and perhaps there will be a resurgence in the use of the latter with ultrasound in order to ensure accurate needle placement while avoiding puncturing the pleura.
Ultrasound guided lumbar facet nerve block: a sonoanatomic study of a
Ultrasound techniques for central venous access
281
new methodologic approach Greher M, Scharbert G, Kamolz LP, et al. Anaesthesiology 2004; 100: 1242–8 BACKGROUND. Facet joint-mediated pain is commonly thought to be a major cause of lower back pain. A controlled series of injections selectively anaesthetizing the nerves innervating the facet joint is an accepted technique for diagnosing and treating this condition. This is currently performed under fluoroscopic or, less commonly, computed tomography guidance. INTERPRETATION. The authors explored the use of ultrasound-guided needle placement for lumbar facet joint blocks. The study was in three parts: (1) to guide needles into cadavers, (2) to image the lumbar region of 20 volunteers with ultrasound and (3) to assess the use of ultrasound guidance in a clinical situation. The needle positions were found to be correct in the cadaver group following dissection at three different levels. The majority of ultrasound landmarks in the volunteers were deemed to be good and acceptable for clinical practice. Twenty-five out of 28 guided needle placements were correct in the patients according to subsequent fluoroscopy. The remaining three were within 5 mm of the radiologically defined target points. In conclusion, the authors felt that this technique of using ultrasound for visualizing and performing lumbar facet joint injections warranted further investigation. Comment Fluoroscopic guidance is associated with issues of X-ray exposure, device availability and procedural cost. In certain circumstances, such as pregnancy, X-ray exposure may be undesirable. For ultrasound to have significant organizational benefits, lumbar facet joint injections would need to be grouped together on specific lists to free up X-ray equipment and staff for use elsewhere. However, this paper did show that bony landmarks can be used for facilitating nerve blocks in the absence of other defining anatomical landmarks. Such techniques have the potential to be used elsewhere, for example around the cervical plexus, pelvis, intercostal and paravertebral areas.
Ultrasound imaging for popliteal sciatic nerve block Sinha A, Chan VW. Reg Anesth Pain Meet 2004; 29:130–4 BACKGROUND. Popliteal nerve blocks have been shown to relieve pain after foot and ankle surgery. The authors suggested that the use of ultrasound may be able to increase the success rate, which is quoted at 21% for single-injection popliteal blocks |24|. INTERPRETATION. In this paper the authors explored the use of ultrasound for imaging the sciatic nerve in the popliteal fossa in ten patients. The point at which the nerve divided and the spatial relationship between the peroneal and tibial nerves distally was recorded. This was found to be highly variable. The authors suggested that this might be one cause of difficulty when using a nerve stimulator for locating the sciatic nerve or
The year in anaesthesia and critical care
282
its terminal branches at this site. Ultrasound visualization of the sciatic nerve and its terminal branches at the time of nerve blockade identified the optimal site of needle placement, thereby avoiding multiple blind attempts and potential nerve or vascular injury. Confirmation of nerve localization and needle placement was obtained using a nerve stimulator. Circumferential spread of the local anaesthetic within the fascial sheath correlated with a rapid onset and completeness of sciatic nerve blockade. The authors suggested that, by confirming the extent of the proximal spread of local anaesthetic, it might not be necessary to block the nerves individually. Comment There has been imaging work performed on the sciatic nerve in the thigh |25|. Pos-terior and lateral approaches to the sciatic nerve in the popliteal fossa have been described |26, 27|. Other nerve blocks in the leg that have been studied using ultra-sound include the femoral three-in-one nerve blocks |18, 19|. Sciatic nerve blocks are notoriously awkward to perform, often relying on subtle anatomical landmarks and requiring the patient to adopt various positions (prone, lateral or supine with the leg raised), which maybe particularly difficult in the patient with trauma. McCartney et al. described an ultrasoundguided lateral approach to the sciatic nerve in the popliteal fossa that would appear to avoid this problem |28|. Ultrasound guidance by the lateral approach allows the patient to remain supine and the needle tip and shaft maybe visualized throughout the procedure (see Figs. 12.4 and 12.5).
Fig. 12.4 Position of the patient and orientation of the ultrasound probe when performing a lateral popliteal sciatic nerve block |3|. Source: McCartney et al. (2004) |28|.
Ultrasound techniques for central venous access
283
Fig. 12.5 Ultrasound image of the popliteal fossa. The arrows indicate the needle shaft. A, popliteal artery; TN, tibial nerve; PN, common peroneal nerve. Source: McCartney et al. (2004) |28|.
Ultrasound imaging improves learning curves in obstetric epidural anesthesia: a preliminary study Grau T, Bartusseck E, Conradi R, Martin E, Motsch J. Can J Anesth 2003; 50: 1047–50 BACKGROUND. Epidural anaesthesia may be difficult in pregnancy. Teaching this core skill and assisting trainees to perform their first few epidurals is also not an easy task. INTERPRETATION. In this study two groups of anaesthetic residents performed their first 60 obstetric epidurals under supervision. The first group proceeded in the conventional way using the landmark and loss of resistance technique. The second group proceeded in the same way, but were supported by pre-puncture ultrasound imaging, giving them information about the optimal puncture point, depth of epidural space and predicted needle trajectory. The authors showed that the success rates were significantly higher in the group using ultrasound (control 60±16% after the first ten attempts versus ultrasound 86±15%) (P<0.001). The authors concluded that ultrasound imaging might have a role to play in the teaching and training of obstetric regional anaesthesia.
The year in anaesthesia and critical care
284
Comment Many trainee anaesthetists struggle with their first few epidurals, intimidated by the fear of performing a dural tap or damaging nerves. Pre-puncture knowledge may reduce this fear and help instil confidence. Despite the results of this study we would advise caution. Whilst possibly of benefit in children, ultrasound imaging of the axial skeleton becomes considerably more difficult with age. Difficulty with epidurals is mostly as a result of obesity: the combination of a deep epidural space and increased adipose tissue makes imaging unhelpful. We view it as unlikely that ultrasound will be routinely used in the delivery suite to aid epidural anaesthesia. However, ultra-sound may have a place in training junior anaesthetists to perform their first few epidurals.
Ultrasound guidance in regional anaesthesia Marhofer P, Greher M, Kapral S. Br J Anaesth 2005; 94:7–17 BACKGROUND. The technology and clinical understanding of ultrasound has evolved greatly over the past decade. In the Department of Anaesthesia and Intensive Care Medicine at the Medical University of Vienna, it has become routine for regional anaesthetic nerve blockades to be performed using ultrasound to guide needle placement and to monitor the distribution of local anaesthetic. INTERPRETATION. Nerves are blocked by local anaesthetic and not by the needle which delivers it. The success of regional anaesthetic blocks thus depends on the distribution of
Fig 12 6 Transverse view of the infraclavicular
Ultrasound techniques for central venous access
285
part of the brachial plexus. The arrows indicate the pleura. SA, subclavian artery; SV, subclavian vein. Source: Marhofer et al. (2005).
Fig. 12.7 Transverse view of the infraclavicular part of the brachial plexus near the coracoid process. The arrows in the bottom left corner indicate the pleura. CP, coracoid process; CV, cephalic vein; PMM, pectoralis major muscle; SA, subclavian artery. Source: Marhofer et al. (2005). local anaesthetic which may be optimally distributed under sonographic visualization. In the authors’ experience, it seems appropriate to start practising ultrasound-guided blocks on peripheral nerves under supervision before moving on to more central blocks (see Figs. 12.6, 12.7 and 12.8). In addition to their now almost 100% success rate, significant improvements
The year in anaesthesia and critical care
286
Fig. 12.8 This is a transverse view of the axillary part of the brachial plexus. AA, axillary artery; BV, basilic vein. The axillary approach to the plexus is commonly used and the anatomy is nicely demonstrated. Source: Marhofer et al. (2005). have been made in terms of sensory and motor block onset times. This high standard of perioperative analgesia has translated into improved patient satisfaction for both adults and children. Comment We recommend this review. The authors performed more than 4000 nerve blocks under ultrasound guidance over 10 years and published several landmark papers. The technological background is described, suitable equipment is recommended and a detailed account is given of which types of nerve block lend themselves to sonographic guidance and how they can be performed in a straightforward and safe manner. The authors used a large sophisticated ultrasound machine usually only available in radiology departments. Such higher specification machines are likely to be beyond the budget of most anaesthetic departments. These are very different machines from those commonly available for vascular access. The priority when investing in ultrasound machines for anaesthesia is likely to be centred on vascular access, which requires a lower specification machine. It is these machines and the images they generate that will likely be used for aiding regional nerve blockade in the next few years. Advances in technology should widen the scope for such procedures in the future. Whilst we agree with the authors that ultrasound will be the guidance technique of the future, we would be inclined to agree with the opinions expressed in recent editorials by
Ultrasound techniques for central venous access
287
Greher et al. |20| and Peterson et al. |29| who felt that the transition from the conventional technique of nerve stimulation will take another 10 years or even longer to complete. Conclusion The development of ultrasound for nerve blocks should be encouraged. The poten-tial advantages over conventional landmark-based techniques are significant. Such benefits may extend to both medical and non-medical practitioners. The cost-effectiveness of ultrasound has been assessed in the context of both vascular access procedures and nerve blockade |5, 30|. The potential cost savings in terms of reduced procedure time, the reduction in the cost of disposables, e.g. nerve stimulator needles and the lower risk of complications and potential litigation may more than balance the cost of ultrasound machines. Saved time is important: American papers quote the cost of operating room time as being calculated at $8 per minute. These potential savings must be balanced with the cost of training staff in the use of these devices. At present ultrasound guidance for nerve blockade may be considered an adjunct to the traditional nerve stimulator. With experience, reversal of this relationship may be expected, with ultrasound eventually surpassing the nerve stimulator in usefulness for routine nerve blockade. In the UK we are a long way off delivering on the NICE guidelines for ultrasound guidance for the placement of CVCs. The purchase of equipment and training for this will almost certainly take priority over the development of ultrasound imaging for regional anaesthesia. Increased availability of ultrasound machines for vascular access should bring about an upsurge in its use in regional anaesthesia. References 1. National Institute for Clinical Excellence. Guidance on the Use of Ultrasound Locating Devices for Central Venous Catheters (NICE technology appraisal no. 49.) London: NICE, 2002. (http://www.nice.org.uk/) 2. Hind D, Calvert N, McWilliams R, Davidson A, Paisley S, Beverley C, Thomas S. Ultrasonic locating devices for central venous cannulation: meta-analysis. BMJ 2003; 327(7411): 361. 3. Scott DH. Editorial II: the king of the blind extends his frontiers. Br J Anaesth 2004; 93(2): 175– 7. 4. Scott DH. It’s NICE to see in the dark. Br J Anaesth 2003; 90(3): 269–72. 5. Calvert N, Hind D, McWilliams R, Davidson A, Beverley CA, Thomas SM. Ultrasound for central venous cannulation: economic evaluation of cost-effectiveness. Anaesthesia 2004; 59(11): 1116–20. 6. Chalmers N. Ultrasound guided central venous access. NICE has taken sledgehammer to crack nut. BMJ 2003; 326(7391): 712. 7. Maury E, Guglielminotti J, Alzieu M, Guidet B, Offenstadt G. Ultrasonic examination: an alternative to chest radiography after central venous catheter insertion? Am J Respirat Crit Care Med 2001; 164(3): 403–5. 8. Galloway S, Bodenham A. Ultrasound imaging of the axillary vein—anatomical basis for central venous access. Br J Anaesth 2003; 90(5): 589–95.
The year in anaesthesia and critical care
288
9. Nickalls RW. A new percutaneous infraclavicular approach to the axillary vein. Anaesthesia 1987; 42(2): 151–4. 10. Sandhu NS, Capan LM. Ultrasound-guided infraclavicular brachial plexus block. Br J Anaesth 2002; 89(2): 254–9. 11. Stenzel JP, Green TP, Fuhrman BP, Carlson PE, Marchessault RP. Percutaneous central venous catheterization in a pediatric intensive care unit: a survival analysis of complica-tions. Crit Care Med 1989; 17(10): 984–8. 12. Alderson PJ, Burrows FA, Stemp LI, Holtby HM. Use of ultrasound to evaluate internal jugular vein anatomy and to facilitate central venous cannulation in paediatric patients. Br J Anaesth 1993; 70(2): 145–8. 13. La Grange P, Foster PA, Pretorius LK. Application of the Doppler ultrasound bloodflow detector in supraclavicular brachial plexus block. Br J Anaesth 1978; 50(9): 965–7. 14. Beekman R, Visser LH. High-resolution sonography of the peripheral nervous system—a review of the literature. Eur J Neurol 2004; 11(5): 305–14. 15. Kirchmair L, Entner T, Kapral S, Mitterschiffthaler G. Ultrasound guidance for the psoas compartment block: an imaging study. Anesth Analg 2002; 94(3): 706–10. 16. Rosario DJ, Jacob S, Luntley J, Skinner PP, Raftery AT. Mechanism of femoral nerve palsy complicating percutaneous ilioinguinal field block. Br J Anaesth 1997; 78(3): 314–16. 17. Johr M, Sossai R. Colonic puncture during ilioinguinal nerve block in a child. Anesth Analg 1999; 88(5): 1051–2. 18. Marhofer P, Schrogendorfer K, Koinig H, Kapral S, Weinstabl C, Mayer N. Ultrasono-graphic guidance improves sensory block and onset time of three-in-one blocks. Anesth Analg 1997; 85(4): 854–7. 19. Marhofer P, Schrogendorfer K, Wallner T, Koinig H, Mayer N, Kapral S. Ultrasonographic guidance reduces the amount of local anesthetic for 3-in-l blocks. Reg Anesth Pain Med 1998; 23(6): 584–8. 20. Greher M, Retzl G, Niel P, Kamolz L, Marhofer P, Kapral S. Ultrasonographic assessment of topographic anatomy in volunteers suggests a modification of the infraclavicular vertical brachial plexus block. Br J Anaesth 2002; 88(5): 632–6. 21. Perlas A, Chan VW, Simons M. Brachial plexus examination and localization using ultra-sound and electrical stimulation: a volunteer study. Anesthesiology 2003; 99(2): 429–35. 22. Ting PL, Sivagnanaratnam V. Ultrasonographic study of the spread of local anaesthetic during axillary brachial plexus block. Br J Anaesth 1989; 63(3): 326–9. 23. Kapral S, Krafft P, Eibenberger K, Fitzgerald R, Gosch M, Weinstabl C. Ultrasound-guided supraclavicular approach for regional anesthesia of the brachial plexus. Anesth Analg 1994; 78(3): 507–13. 24. Provenzano DA, Viscusi ER, Adams Jr SB, Kerner MB, Torjman MC, Abidi NA. Safety and efficacy of the popliteal fossa nerve block when utilized for foot and ankle surgery. Foot Ankle Int 2002; 23(5): 394–9. 25. Grechenig W, Clement HG, Peicha G, Klein A, Weiglein A. Ultrasound anatomy of the sciatic nerve of the thigh. Biomed Tech (Berl) 2000; 45(11): 298–303. 26. Di Benedetto P, Casati A, Bertini L, Fanelli G, Chelly JE. Postoperative analgesia with continuous sciatic nerve block after foot surgery: a prospective, randomized comparison between the popliteal and subgluteal approaches. Anesth Analg 2002; 94(4): 996–1000. 27. Hadzic A, Vloka JD. A comparison of the posterior versus lateral approaches to the block of the sciatic nerve in the popliteal fossa. Anesthesiology 1998; 88(6): 1480–6. 28. McCartney CJ, Brauner I, Chan VW. Ultrasound guidance for a lateral approach to the sciatic nerve in the popliteal fossa. Anaesthesia 2004; 59(10): 1023–5. 29. Peterson MK, Millar FA, Sheppard DG. Ultrasound-guided nerve blocks. Br J Anaesth 2002; 88(5): 621–4. 30. Sandhu NS, Sidhu DS, Capan LM. The cost comparison of infraclavicular brachial plexus block by nerve stimulator and ultrasound guidance. Anesth Analg 2004; 98(1): 267–8.
Part IV Critical care
The Critical Care Outreach conundrum: does the end justify the means? JANE HARPER, JENNIFER HUNTER Introduction Critical Care Outreach |1| and Medical Emergency Teams (METs) attempt to extend the skills available in critical care units to the whole hospital—to create ‘critical care without walls’ (pp. 9, 15). The concept of Critical Care Outreach has been enthusiastic-ally embraced in England, as have METs in Australia, but conclusive evidence sup-porting either initiative is still outstanding |2–6|. The clinical conundrum is whether critical care services and specialists, wherever they practice, should support this major use of limited resources. Why are Critical Care Outreach/Medical Emergency Teams needed? In England, there are 0.6 intensive care beds per 10 000 population, compared to 4.4 beds per 10000 population in the USA |7|. Patients who require a higher level of care are often admitted to general hospital wards because of a lack of critical care beds |8|. Simchen et al. |9| reported on experience in Israel, where more than 5% of hos-pitalized patients met the criteria for intensive care admission, but only 1.4% were actually admitted. The patients admitted to intensive care had better survival in the first 3 days, suggesting that critical care skills and facilities must be provided in a timely way wherever sick patients are in hospital. Up to 85% of inpatients who suffer a cardiorespiratory arrest have documented deterioration in their physiological observations in the hours before the arrest |10, 11| and patients who suffer in-hospital cardiac arrest have a poor outcome. Goldhill and Sumner |12| reported a profile of admissions to 24 intensive care units (ICUs) in the UK and concluded that the mortality for patients admitted from the wards was high, and that intervention may be necessary before ICU admission to have any impact on mortality. In addition, 27% of patients died after discharge from intensive care © Atlas Medical Publishing Ltd
and 9.8% of patients were readmitted to the ICU. This indicates that a significant proportion of patients discharged from intensive care have a continuing need for high intensity medical care. Goldhill et al. went on to study ICU admissions in their own hospital. Thirty-four per cent of patients from the wards (out of 79 patients) were admitted post-cardiac arrest, and
63% had an oxygen saturation of <90% in the 6 h prior to admission |13|. These seriously ill patients might have benefited from earlier intervention. McGloin et al. used expert assessors for auditing ICU admissions in an English teaching hospital |14|. Thirty-one out of 86 patients were considered to have received suboptimal care prior to admission, either due to a failure to recognize the severity of the problem or to inappropriate treatment. Thirteen of the 20 unexpected cardiac arrests were considered ‘potentially avoidable’. MacQuillan et al. assessed care before ICU admission |15| in two hospitals in England and confirmed these findings. Ward care was assessed as suboptimal in at least 54 out of 100 admissions, with late referral to critical care in 39 cases and increased mortality in the group where management was judged suboptimal. They suggested that ‘better care before admission may reduce intensive care bed days’ (p. 1856). Critical Care Outreach and METs were introduced on the premise that early intervention would prevent deterioration and improve outcomes. Evidence of benefit from early intervention comes from, for example, Rivers et al.’s trial of early goal-directed therapy for patients presenting with sepsis |16|: timely intervention is an integral part of the ‘Surviving Sepsis’ campaign |17|. Medical Emergency Teams The concept of sharing critical care skills outside the ICU originated with Professor K. Hillman and his colleagues in Australia in the early 1990s. They introduced a MET into a 375-bed teaching hospital in New South Wales |18, 19|. The cardiac arrest team was replaced by a team of critical care doctors and nurses who were charged with assessing and treating patients at risk of deterioration. Patients at risk were identified by a set of ‘calling criteria’, where deterioration in their physiological observations or concern of staff about a patient would trigger a call to the MET (Table IV.1). When called the MET would assess the patients and institute appropriate investigations and treatment. However, the composition of the MET was not prescribed, nor was calling the MET compulsory. Effectiveness of Medical Emergency Teams There are several papers that have demonstrated a positive effect of introducing METs, but most are non-randomized, cohort observational studies. Most have his-torical controls or offer comparisons between dissimilar institutions and, thus, offer at best level 2 evidence |21|. Bristow et al. made a prospective cohort comparison of three hospitals, one with a MET and two without |20|. After adjustment for case mix, they reported that the rate of unplanned admissions to intensive care was less in the MET hospital and the rate of ‘do not resuscitate’ orders was higher, at 77% in the MET hospital and 60% and 70% in the comparator hospitals, respectively.
Table IV.1 Calling criteria for a MET Cardiorespiratory arrest Threatened airway Respiratory rate <6 breaths/min Respiratory rate >35 breaths/min Heart rate <41 beats/min Heart rate >140 beats/min Systolic blood pressure <91 mm Hg Repeated or prolonged seizures Fall in Glasgow Coma Score >2 points Concern about patient status not detailed above Source: Bristow et al. (2000) |20|.
Buist et al. conducted a single-centre (300-bed hospital) study comparing un-expected cardiac arrest rates before and after the introduction of a MET |22|. Unex-pected cardiac arrests decreased from 3.7 to 2.05 per 1000 admissions (after case mix adjustment odds ratio [OR] 0.5 and confidence interval [CI] 0.35–0.73). However, the findings were questioned because, after the MET was introduced, more patients had ‘do not resuscitate’ orders agreed. In addition, the death rate was already falling before the introduction of the MET. Bellomo et al. reported another before-after observational study involving 20 000 patients admitted to hospital that compared the number of cardiac arrests, the number of deaths post-cardiac arrest and the number of intensive care and hospital bed days occupied by survivors of cardiac arrest |23|. The relative risk ratio (RRR) of cardiac arrest was 0.65 after the introduction of the MET (P<0.001) and the risk of dying post-arrest was reduced (RRR 0.56; P=0.005). In addition, the length of stay in intensive care for survivors of cardiorespiratory arrest decreased from 163 to 33 days (RRR 0.80; P<0.001), as did the hospital length of stay (from 1353 to 159 days) (RRR 0.89; P<0.001). However, the improvement was apparent during the education phase prior to the introduction of the MET and was therefore not directly related to its introduction |24|. In 2004, Bellomo et al. reported a prospective, controlled before-after study of a MET in post-operative patients |25|. Serious adverse outcomes (e.g. respiratory failure, severe sepsis, stroke and acute renal failure) decreased from 301 to 127 per 1000 patients (RRR 57.8%; P<0.001) after the introduction of a MET, and the number of deaths also decreased (RRR 36%; P 0.0178). North American and European experience with Critical Care Outreach and METs is limited, although DeVita et al. reported a 6-year experience with a MET in Pitts-burgh, USA, demonstrating a 17.4% decrease in the incidence of in-hospital cardiac arrest in a retrospective analysis of the data |26|. The Simpson Centre for Health Services Research in the University of New South Wales and the Australian and New Zealand Intensive Care Society Clinical Trials group have recently completed the Medical Early Response Intervention Trial (MERIT) study, a
cluster-randomized, controlled trial of METs conducted in 23 hospitals in Australia. Summary results of the study have been presented at scientific meetings, but have not yet been published (Simon Finfer, personal communication). Cluster randomization means that the research subjects are not sampled independently, but in a group allocated to the same intervention, for example by hospital |27| and is undertaken because randomization at the patient level is impractical. In the MERIT study, hospitals were randomized to introduce a MET system or to maintain current practices. The primary outcome for the study was the aggregate unexpected ward deaths, unexpected cardiac arrests and unplanned ICU admissions. The study found no difference between hospitals randomized to have METs or not to have METs. What is Critical Care Outreach? In England the pivotal documents Critical to Success (Audit Commission, 1999) |28| and Comprehensive Critical Care (Department of Health, 2000) |1| recommended that every acute hospital in England develop Critical Care Outreach services, which are based on the MET model. Critical Care Outreach aspires to be ‘an organizational approach to ensure equity of care for all critically ill patients regardless of location’ (p. 6) |29|. Consisting mainly of senior nurses with critical care skills, Critical Care Outreach teams clinically assess sick patients and patients at risk of deterioration on the wards, follow up patients discharged from intensive care and participate in the education of ward staff. The specific goals of Critical Care Outreach are as follows |29, 30|. 1. To avert admissions to critical care. 2. To facilitate timely admission to and discharge from critical care units. 3. To prevent re-admissions. 4. To educate ward staff about the management of critically ill patients and those at risk of deterioration. 5. To promote continuity of care and improve services for patients who have been critically ill. The composition of Critical Care Outreach teams varies according to the needs of and resources available to each hospital. Six to eight nurses are required for 24 h a day, 7 days a week cover |29|. At present, intensive care-trained doctors and allied health professionals have variable input and there is no agreement about what constitutes the ‘best’ team structure. By 2002, when the National Outreach survey |31| was carried out, 100 (60%) of the 167 acute trusts in England that responded had Critical Care Outreach teams during the working week (08.00–22.00 h) and 41 (24%) provided 24 h a day, 7 days a week followup of patients discharged from intensive care. Only 34% of responding trusts provided 24 h a day, 7 days a week clinical support to wards. The configuration of these services was variable, with different grades of nursing staff involved and with different levels of medical and allied health professional support from hospital to hospital.
What does Critical Care Outreach do? Critical Care Outreach aims to provide a similar service to METs, although the remit of the team is broadened to include the education of ward staff and longer-term follow-up and psychosocial support of patients discharged from critical care. The first role of Critical Care Outreach is to find and treat patients at risk of deterioration. The introduction of ‘track and trigger’ systems for identifying physiological abnormalities so that Critical Care Outreach teams are contacted in a timely way reflects the use of ‘calling criteria’ in Australia. In England, the ‘modified early warning score’ (MEWS) (Table IV.2) |32, 33|) is generally used. Subbe et al. showed that high MEWS are associated with increased risk of death and an increased rate of admission to critical care in medical patients |34|. Goldhill demonstrated that 30-day hospital mortality increased with the number of physiological abnormalities documented |7|. There have been many versions of early warning scores described to date, with no clear advantage shown for any one score |29|. The action pathway at the Royal Liverpool University Hospital is as follows. 1. Score of 0–1: continue routine observations. 2. Score of 2: increase the frequency of observations to 2–4 hourly, monitor trends and inform sister. 3. Score of 3–4: repeat observations in 30 min and inform doctor. 4. Score of 5–7: confirm with sister and fast bleep doctor. 5. Score of 8 or more: bleep registrar and consider ICU referral. Critical Care Outreach teams provide education for ward staff to enable the provision of skills such as tracheostomy care or continuous positive airway pressure (CPAP) on general wards. The teams not only provide education at the bedside, but are frequently involved in multidisciplinary courses (e.g. Acute Life-threatening Early Recognition and Treatment |35|) in order to raise awareness and skills about critical illness for all hospital staff. Critical Care Outreach teams follow patients discharged from intensive care and most often extend this follow-up beyond hospital discharge in specific critical care follow-up clinics. These clinics help patients come to terms with the physical and psychological sequelae of critical illness |36|. Table IV.3 summarizes some of the activities of the Critical Care Outreach team at the Royal Liverpool University Hospital from 1 December 2004 to 6 January 2005. During this time the team was called to 127 patients and made a total of 332 patient visits. The team operates during daytime hours (08.00–17.00 h) 7 days a week. It is made up of three nurses with postgraduate qualifications and a physiotherapist, supported by consultant intensivists during weekday morning sessions.
Table IV.2 The modified early warning score used at the Royal Liverpool University Hospital SCORE
3
2
1
0
1
Respiratory rate/min
–
≤8
–
9–14
15–20 26–30 ≥31
Pulse/min
–
≤40
41–50
51–100 101– 110
111– 130
≥131
Systolic blood ≤70 pressure mmHg
70–80
81–100
101– 199
–
≥200
–
Glasgow Coma Score
≤8
9–13
14
15
–
–
–
AVPU
–
–
New agitation or confusion
Alert
Voice
Pain
Unresponsive
Urine
≤10ml/hr ≤30ml/hr – for 2 hours for 2 hours
– ≥30 ml/h or RRT
–
–
Temperature (°C)
–
36.1– 38
≥38.5 –
≤35.1
35.2–35.9
38.1– 38.4
2
3
AVPU, alert, voice, pain, unresponsive; RRT, renal replacement therapy. Source: Guidelines for the Introduction of Outreach Services |30|.
Table IV.3 Sample activities of a Critical Care Outreach team over 5 weeks at the Royal Liverpool University Hospital, 1 December 2004–6 January 2005 The of number of patients seen was 127 and the number of patient visits was 332 (mean 2.6 visits per patient) Interventions Nine patients: transferred to level 3 care (ICU) |37|, with three patients post-cardiac arrest Five patients: do not attempt resuscitation orders agreed Thirty-eight patients: CPAP initiated on the ward/high dependency unit (HDU) Fourteen patients: high flow oxygen therapy initiated Twenty-six patients: oxygen therapy initiated Four patients: mini-tracheostomy inserted One patient: tracheostomy tube downsized Three patients: arterial line inserted (HDU)
Two patients: referred for bilevel positive airway pressure ventilation (BiPAP) Fourteen visits: oxygen therapy adjusted Fifteen visits: fluid therapy adjusted Six visits: analgesia review Three visits: microbiology review Six visits: central venous pressure (CVP) line removed Investigations Cardiac monitoring instituted (13 patients) and twelve-lead electrocardiograph (8 patients) Chest X-ray (8 patients) and computed tomography scan (3 patients) Follow-up clinic One afternoon per week for two patients: by request or ICU admission >4 days
Effectiveness of Critical Care Outreach Several non-randomized observational before-after comparative studies have been published on the effectiveness of ward-based Critical Care Outreach in England. At present there are no randomized trials evaluating its effectiveness. Goldhill et al. reported data on a ‘patient at risk’ team. He found that 3.6% of patients admitted to an ICU had a cardiac arrest before admission if seen by the team compared to 30% for those not seen by the patient at risk team. The overall ICU mortality was improved for patients reviewed by the team (25% in patients seen by the patient at risk team versus 45% amongst patients not seen) |38|. In another study using historical controls Ball et al. reported improved survival to hospital discharge (RRR 1.08; CI 1.0–1.18) and reduced re-admission rates (RRR 0.48; CI 0.26–0.87) after the introduction of Critical Care Outreach |39|. In contrast, Leary and Ridley found that the numbers of re-admissions within 48 h of ICU discharge and the reasons for re-admission remained the same after the introduction of Critical Care Outreach |40|. Garcea et al. published a retrospective observational study of patients readmitted to critical care |41|. There was no difference in the re-admission rate after the introduction of Critical Care Outreach, although the critical care unit mortality of readmitted patients decreased from 36.7 to 22.8% (CI=2.4–30.3%). As with the other studies using historical controls, it was difficult to demonstrate that this was wholly or partly due to Critical Care Outreach. In a complex ward, cluster-randomized before-after trial published by Priestley et al. the in-hospital mortality was reduced (two-level OR 0.5; CI 0.32–0.85) and the length of stay possibly increased after the introduction of Critical Care Outreach teams |42|. Pittard found decreases in the rates of unplanned admissions, readmissions, mortality and length of stay in another before-after analysis |43|. However, these studies were confounded by the nature of their design: because of the rapidity of the introduction of Critical Care Outreach teams it has not been possible to conduct a prospective randomized trial. The introduction of a Critical Care Outreach nurse for reviewing high-risk postoperative patients in Melbourne, sAustralia, led to a non-significant decrease in the
incidence of serious adverse events overall but, interestingly, a possible increase in the incidence of myocardial infarctions in patients having colorectal surgery. The number of calls to the MET increased |44|. In Scotland a report published by the Scottish Executive in 2000 concluded that Critical Care Outreach teams would not affect either the patient care or workload for ICUs |45|. Cuthbertson |2| and others |6| have questioned the introduction of a costly service with so few supportive data. The Critical Care Outreach conundrum Why is it so hard to establish that Critical Care Outreach or METs, concepts so intuitively sensible and seductive, actually work? There are several possible reasons. One is that restructuring processes within institutions take time to ‘bed in’: it may be that Critical Care Outreach teams are too new to have established their precise role within our institutions |3–5|. The Australian experience suggests that not all patients who fulfil ‘calling criteria’ are actually referred to METs: in MET hospitals in the MERIT study the MET was called to less than one-third of patients who had documented MET calling criteria prior to unplanned admission (Simon Finfer, personal communication). Perhaps a more discriminating score for assessing patients at risk is needed, or a radical change in the way we monitor patients and structure patient care is required, in order to enable staff to ask for assistance earlier. It may be difficult to compare the effect of teams with such a heterogeneous makeup. Critical Care Outreach teams may have senior nurses with postgraduate qualifications covering shifts up to 24 h a day, 7 days a week, whereas other hospitals may have a single more junior grade nurse working normal working hours and not available on call. Medical support is variable. MET teams, on the other hand, are generally medically based, with nursing input. The treatment received by patients will differ according to the configuration of the team. It is easier to assess binary outcomes such as mortality than it is to assess the quality of care delivered to patients. Although we perceive a higher quality of care delivered to our patients, it is difficult to measure this objectively. Care improves as awareness of and education by METs/Critical Care Outreach is introduced or discussed: there is a significant ‘Hawthorne’ effect described in many of the studies cited above. It may be that the only way to demonstrate the utility of METs or Critical Care Outreach is through qualitative research, which does not sit comfortably with our evidence-based paradigm. Ultimately, it is unlikely that level 1 evidence will be produced for Critical Care Outreach |2, 4, 5|. Interestingly, at the end of the cluster-randomized controlled trial in Australia, those hospitals randomized to have MET teams did not disband them even knowing the results of the trial (Simon Finfer, personal communication). Perhaps, as Williams et al. suggested, the introduction of Critical Care Outreach ‘has nothing to do with touchyfeely new age medicine but is a real cry for help…for skills and knowledge in dealing with an increasingly sick population…’ (p. 699) |46|. It maybe, as Kerridge and Saul suggested, unethical not to introduce and support Critical Care Outreach, given the clinical needs |4|.
Acknowledgements We are grateful to members of the Royal Liverpool University Hospital Outreach team (K Francis, J French, E Arthan and L McCrossan) for information and advice, and to Dr Simon Finfer for commenting on the manuscript. References 1. Department of Health (England). Comprehensive Critical Care. London, England: Department of Health, 2000 (http://www.dh.gov.uk/assetRoot/04/08/28/72/04082872.pdf). 2. Cuthbertson BH. Editorial II: Outreach Critical Care—cash for no questions? Br J Anaesth 2003; 90(1): 4–6. 3. Kerridge RK. The medical emergency team: no evidence to justify not implementing change. Med J Aust 2000; 173:228–9. 4. Kerridge RK, Saul WP. The medical emergency team, evidence based medicine and ethics. Med J Aust 2003; 179(6): 313–15. 5. Parr M. Critical Care Outreach: some answers, more questions (editorial). Int Care Med 2004; 30:1261–2. 6. Smith GB, Nolan J. Results may have been due to education of ward staff (letter). BMJ 2002; 324:1215. 7. Goldhill DR. Preventing surgical deaths: critical care and intensive care Outreach services in the post-operative period. Br J Anaesth 2005; 95: in press (July). 8. Bion J. Rationing intensive care. BMJ 1995; 310:682–3. 9. Simchen E, Sprung CL, Galai N, Zitser-Gurevich Y, Bar-Lavi Y, Gurman G, Klein M, Lev A, Levi L, Zveibil F, Mandel M, Mnatzaganian G. Survival of critically ill patients hospitalised in and out of intensive care units under paucity of intensive care unit beds. Crit Care Med 2004; 32:1654–61. 10. Franklin C, Matthew J. Developing strategies to prevent in-hospital cardiac arrest: analysing responses of physicians and nurses in the hour before the event. Crit Care Med 1994; 22:244–7. 11. Kause J, Smith G, Prytherch D, Parr M, Flabouris A, Hillman K. A comparison of antecedents to cardiac arrests, deaths and emergency intensive care admissions in Australia and New Zealand and the United Kingdom—the ACADEMIA Study. Resuscitation 2004; 62:275–82. 12. Goldhill DR, Sumner A. Outcome of intensive care patients in a group of British intensive care units. Crit Care Med 1998; 26:1137–45. 13. Goldhill DR, White SA, Sumner A. Physiological values and procedures in the 24 hours before admission from the ward. Anaesthesia 1999; 54:529–34. 14. McGloin H, Adam SK, Singer M. Unexpected deaths and referrals to intensive care of patients from wards: are some cases potentially avoidable? J Coll Phys Lond 1999; 33(3): 255–9. 15. MacQuillan P, Pilkington S, Allan A, Taylor B, Short A, Morgan G, Nielson M, Barrett D, Smith G. Confidential enquiry into quality of care before admission to intensive care. BMJ 1998; 316:1853–8. 16. Rivers E, Nguyen H, Havstad S, Ressler J, Muzzin A, Knoblich B, Peterson E, Tomlanovich M. Early Goal-Directed Therapy Collaborative Group. Early goal-directed therapy in the treatment of severe sepsis and septic shock. N Engl J Med 2001; 345:1368–77. 17. Rhodes A, Bennett ED. Early goal directed therapy: an evidence-based review. Crit Care Med 2004; 32: S448–50. 18. Lee A, Bishop G, Hillman KM, Daffurn K. The medical emergency team. Anaesth Intensive Care 1995; 23:183–6.
19. Parr MJ, Hadfield JH, Flabouris A, Bishop G, Hillman K. The medical emergency team: 12month analysis of reasons for activation, immediate outcome and not-for-resuscitation orders. Resuscitation 2001; 50(1): 39–44. 20. Bristow PJ, Hillman K, Chey T, Daffurn K, Jacques TC, Norman SL, Bishop GF, Simmons EG. Rates of in-hospital arrests, deaths and intensive care admissions: the effect of a medical emergency team. Med J Aust 2000; 173:236–40. 21. Guyatt G, Hayward R, Richardson WS, Green L, Wilson M, Sinclair J, Cook D, Glasziou P, Detsky A, Bass E. Grades of recommendation in moving from evidence to action. In: Guyatt G, Rennie D (eds). User’s Guide to the Medical Literature. Chicago: AMA Press, 2002; pp 190–2. 22. Buist MD, Moore GE, Bernard SA, Waxman BP, Anderson JN, Nguyen TV. Effects of a medical emergency team on reduction of incidence of and mortality from unexpected cardiac arrest in hospital: preliminary study. BMJ 2002:324;387–90. 23. Bellomo R, Goldsmith D, Uchino S, Buckmaster J, Hart GK, Opdam H, Silvester W, Doolan L, Gutteridge G. A prospective before-and-after trial of a medical emergency team. Med J Aust 2003; 179(6): 283–7. 24. Bellomo R. Author reply: a prospective before-and-after trial of a medical team. Med J Aust 2004:180:309–10. 25. Bellomo R, Goldsmith D, Uchino S, Buckmaster J, Hart G, Opdam H, Silvester W, Doolan L, Gutteridge G. Prospective controlled trial of effect of medical emergency team on postoperative morbidity and mortality rates. Crit Care Med 2004; 32:916–21. 26. DeVita MA, Braithwaite RS, Mahidhara R, Stuart S, Foraida M, Simmons RL. Use of medical emergency team responses to reduce hospital cardiopulmonary arrests. Qual Saf Health Care 2004; 13(4): 251–4. 27. Bland M. Cluster Randomised Trials in the Medical Literature. 2003; http://wwwusers.%20york.ac.uk/~mb55/talks/clusmlhtm. 28. Audit Commission (England). Critical to Success—The Place of Efficient and Effective Critical Care Services Within the Acute Hospital. London: Audit Commission, 1999; http://%20www.auditcommission.gov.uk/Products/NATIONAL-REPORT/40B50F26-ED9F4317-A056–042B31AEA454/CriticalToSuccess.pdf. 29. Modernisation Agency (Department of Health, England). Critical Care Outreach 2003: Progress in Developing Services. London, England: Department of Health; http://www.%20modern.nhs.uk/criticalcare/5021/7117/78001-DoH-CareOutreach.pdf. 30. Intensive Care Society. Guidelines for the Introduction of Outreach Services: Standards and Guidelines. London, England: Intensive Care Society, 2002; http://www.ics.ac.uk/downloads/icsstandards-outreach.pdf. 31. Morgan R, Mugglestone M. National Critical Care Outreach Survey. London, Engand: Modernisation Agency, 2003; http://www.modern.nhs.uk/criticalcare/5032/5197/%202002%%2020outreach%20survey%20re sults.doc. 32. Morgan RJM, Williams F, Wright MM. An early warning scoring system for detecting developing critical illness. Clin Intensive Care 1997; 8:100. 33. Subbe CP, Kruger M, Rutherford P, Gemmell L. Validation of a modified early warning score in medical admissions. QJM 2001; 94:521–6. 34. Subbe CP, Davies RG, Williams E, Rutherford P, Gemell L. Effect of introducing the modified early warning score on clinical outcomes, cardio-pulmonary arrests and intensive care utilisation in acute medical admissions. Anaesthesia 2003; 58:797–802. 35. Smith G. Multiprofessional Training in Critical Care: The ALERT (Acute Life threatening Events—Recognition and Treatment) Course. http://www.suht.soton.ac.uk/Courses/%20alert.htm.
36. Jones C, Skirrow P, Griffiths R, Humphris G, Dawson S, Eddleston J, Waldman C, Gager M. Rehabilitation after critical illness: a randomised controlled trial. Crit Care Med 2003; 31:2456– 61. 37. Intensive Care Society. Levels of Care for Adult Patients. London, England: Intensive Care Society, 2002; http://www.ics.ac.uk/downloads/icsstandards-levelsofca.pdf. 38. Goldhill DR, Worthington L, Mulcahy A, Tarling M, Sumner A. The patient-at-risk team: identifying and managing seriously ill ward patients. Anaesthesia 1999; 54:853–60. 39. Ball C, Kirkby M, Williams S. Effect of the Critical Care Outreach team on patient survival to discharge from hospital and readmission to critical care: a non randomised population-based study. BMJ 2003; 327:1014–18. 40. Leary T, Ridley S. Impact of an Outreach team on re-admissions to a critical care unit. Anaesthesia 2003; 58:328–32. 41. Garcea G, Thomasset S, McClelland L, Leslie A, Berry DP. Impact of a Critical Care Outreach team on critical care readmissions and mortality. Acta Anaesthesiol Scand 2004; 48(9): 1096– 100. 42. Priestley G, Watson W, Rashidian A, Mozley C, Russell D, Wilson J, Cope J, Hart D, Kay D, Cowley K, Peteraki J. Introducing Critical Care Outreach: a ward-randomised trial of phased introduction in a general hospital. Intensive Care Med 2004; 30:1398–404. 43. Pittard AJ. Out of our reach? Assessing the impact of introducing a Critical Care Outreach service. Anaesthesia 2003; 58:882–5. 44. Story DA, Shelton AC, Poustie SJ, Colin-Thome NJ, McNicol PL. The effect of Critical Care Outreach on post-operative serious adverse events. Anaesthesia 2004; 59:762–6. 45. Scottish Executive, Health Department. Better Critical Care: Report of a Short Life Working Group on ICU and HDU Issues. Edinburgh: Scottish Executive, 2000; http://www.show.%20scot.nhs.uk/sehd/publications/report.pdf. 46. Williams E, Subbe C, Gemmell L. Outreach Critical Care (letter). Br J Anaesth 2003; 90: 699.
13 Choice of resuscitation fluid in the critically ill SIMON FINFER, JAMES COOPER, DEBORAH COOK Introduction The administration of intravenous fluids to restore or maintain intravascular volume remains a fundamental treatment for critically ill patients. Fluid is the cornerstone of early resuscitation for hypovolaemic, vasodilatory and cardiogenic shock. For the former two conditions in particular, intravenous fluid administration treats life-threatening intravascular volume depletion due to haemorrhage and cutaneous or other ‘third space’ losses in patients with pancreatitis or burns and increased vascular permeability in patients with the systematic inflammatory response from trauma or sepsis. Despite the pivotal role that intravenous fluid plays in the intensive care unit (ICU), many controversies remain unresolved. Key amongst these is the long-running uncertainty about the choice of resuscitation fluid, which has found its expression in the crystalloid-colloid debate—including controversy over the role of albumincontaining fluids compared to both synthetic colloids and crystalloid solutions—and whether hypertonic solutions, notably hypertonic saline, offer benefits over other fluids. In the absence of large, high-quality, randomized controlled trials focused on appropriate patient-centred outcomes, perspectives vary about the relative merits and demerits of different fluids and the rigour and relevance of meta-analyses summarizing the existing trials |1–4|. Complicating the matter further is ambiguity about what importance to attach to studies examining the short-term effects of different fluid regimens on physiological variables and other surrogate outcomes, which may not correlate with major morbidity or mortality in the ICU setting (Quinlan et al., 2004). Major advances in the clinical investigation of fluid resuscitation were witnessed in 2004. In this chapter we outline the state of knowledge as of early 2004 and review publications that have added significantly to the evidence base. In addition, we review important publications that seek to advise clinicians in their daily practice and that illustrate important unanswered questions that future clinical research should address. © Atlas Medical Publishing Ltd
The crystalloid-colloid debate: edging towards resolution? Advocates of crystalloid solutions point out that they are less expensive and carry a minimal risk of anaphylaxis or the transmission of infectious agents: those favouring colloids counter that resuscitation with colloids requires less volume and less time and
The year in anaesthesia and critical care
302
may be less likely to cause pulmonary and peripheral oedema. Both groups of enthusiasts lack large, randomized controlled trials to support their position or to refute the arguments of the other side. When important clinical questions have not been investigated by high-quality, randomized controlled trials, clinicians can still be informed (but can also be misinformed) by systematic reviews and meta-analyses of existing randomized controlled trials. Whilst narrative review articles can provide some insight, they tend to contain selective and potentially biased interpretations of published data. Conversely, a high-quality systematic review uses a comprehensive literature search, explicit selection criteria and a critical appraisal of original studies. Furthermore, a metaanalysis adds a quantitative summary of the results of the original studies to provide a more reliable assessment of treatment effects. In 1998 and 1999 two meta-analyses from industry-independent research groups addressed the crystalloid versus colloid controversy |1, 21|. Schierhout and Roberts reviewed 37 randomized controlled trials: 26 compared colloids with isotonic crystalloid, ten compared colloids in hypertonic crystalloid with isotonic crystalloid and one compared colloid with hypertonic crystalloid. They concluded that colloid resuscitation was associated with a 4% (24 versus 20%) absolute increase in mortality when compared to crystalloid, with no evidence of differing effects in populations with different types of injury. This headline conclusion was drawn from 19 randomized controlled trials enrolling 1315 patients comparing colloids with isotonic crystalloid and for which mortality data were available. The longer version of the paper, which is only published electronically on the BMJ website |5|, also reported on nine trials enrolling 1297 patients that compared colloids in hypertonic crystalloid with isotonic crystalloid and for which mortality data were available. These nine trials favoured treatment with colloid in hypertonic crystalloid versus crystalloid with a relative risk (RR) of death of 0.84 (95% confidence interval [CI] 0.70–1.01). The difference represented a decrease in the absolute risk of death of 5% (95% CI 0–9%). Thus, although the authors concluded that colloid use increased the risk of death, the inferences from these randomized controlled trials are weak. Choi et al. |2| limited their meta-analysis to trials that compared isotonic crystalloids and colloids. They reviewed 17 randomized controlled trials including 814 patients and found no difference in morbidity or mortality. In five trials including 302 trauma patients, crystalloid resuscitation was associated with decreased mortality (RR 0.39; 95% CI 0.17–0.89). The contrasting conclusions of the two meta-analyses served to fuel rather than resolve the controversy. Safety and efficacy of albumin Albumin-containing fluids have long been administered for resuscitating critically ill patients in the ICU. Since albumin is a naturally occurring protein with many important physiological properties and since observational studies have suggested a direct correlation between the degree of hypoalbuminaemia and the risk of death, albumin has also been used for treating hypoalbuminaemia. Whilst the safety of albumin was taken for granted, its higher acquisition cost caused many to question whether its administration could be justified, given the absence of data demonstrating that it improved important patient-centred outcomes. In 1998 the Cochrane Injuries Group Albumin Reviewers not
Choice of resuscitation fluid in the critically ill
303
only challenged assumptions about the efficacy of albumin, but also the assumptions about its safety |3|. Their meta-analysis of 30 randomized controlled trials included 1419 patients, 156 of whom died, a mortality of 11%. The authors concluded that, compared to crystalloid or no fluid, the administration of albumin-containing fluids to patients with hypovolaemia, hypoalbuminaemia and/or burns resulted in a 6% absolute increase in mortality (95% CI 3–9%). This publication generated great academic and media interest and was followed by a 40% reduction in the use of albumin in the UK |6|. Possibly the only area of agreement between the reviewers, critiques of the review and editorialists was that higher quality, randomized controlled trial evidence was needed in order to resolve the controversy |3, 7|. In contrast, in 2001 a second meta-analysis examined the safety of albumin administration and reported that ‘overall, no effect of albumin on mortality was detected’ (p. 149) |4|. There are important methodological differences between the meta-analysis and the Cochrane Injuries Group Albumin Reviewers’ research. The second analysis compared albumin therapy with crystalloid therapy, no albumin or lower doses of albumin, the trial selection was not limited to studies in critically ill or ICU patients and there were no limits on the indications for administering albumin. These broader inclusion criteria resulted in the inclusion of 55 trials involving 3504 patients and 525 deaths and a mortality of 15%. The accompanying editorial pointed to several weaknesses in the original randomized controlled trials and meta-analysis and called for fresh evidence in this area |8|. The Annals of Internal Medicine published a ‘Summary for Patients’ stating that ‘whether albumin therapy improves or worsens survival of critically ill patients is not known’ |9|. Comparison of one colloid with another The colloid solutions for fluid resuscitation include albumin and less purified human protein solutions, dextrans, gelatins and hydroxyethyl starch. The solutes in colloid solutions vary in molecular weight and other properties and remain in the circulation for varying lengths of time. It is inappropriate to assume that all colloid solutions are equally safe or equally efficacious. The evidence favouring one colloid versus another was examined in a meta-analysis of 57 randomized controlled trials with a total of 3659 critically ill or surgical patients |10|. The authors found no evidence that one colloid was any more safe or efficacious than another, but added the caveat that the confidence intervals for the comparisons were wide and the available evidence was inadequate for excluding clinically important differences. Thus, at the beginning of 2004, after at least two decades of basic science and clinical research, the effect of resuscitation with albumin and other colloid solutions on major health outcomes in critically ill patients remained unknown.
A comparison of albumin and saline for fluid resuscitation in the intensive care unit
The year in anaesthesia and critical care
304
Finfer S, Bellomo R, Boyce N, French J, Myburgh J, Norton R; SAFE Study Investigators. N Engl J Med 2004; 350:2247–56 BACKGROUND. It remains uncertain whether the choice of resuscitation fluid for patients in ICUs affects survival. INTERPRETATION. The use of either 4% albumin or normal saline for fluid resuscitation resulted in similar outcomes at 28 days in patients in the ICU. Comment The Saline versus Albumin Fluid Evaluation (SAFE) Study investigators randomized 6997 patients to receive either normal saline or 4% albumin for all fluid resuscitation in the ICU. Allocation concealment was assured by conducting randomization via a secure password-protected encrypted website and randomization was stratified by institution and by a diagnosis of trauma. The trial was designed to enrol 7000 patients: this provided 90% power for detecting a 3% difference in absolute mortality from an estimated baseline mortality rate of 15%. The treating clinicians and patients were prevented from knowing which fluid was administered by the use of specially designed and manufactured masking cartons and administration sets (Fig. 13.1a and b). Almost all hypovolaemic patients were eligible with only patients admitted to the ICU following cardiac surgery, liver transplantation or burns excluded. Tshe majority of patients were enrolled using a provision for delayed consent. The primary outcome was all-cause mortality within 28 days of randomization. The secondary outcomes were the survival time during the first 28 days, the proportion of patients with new organ failures, the duration of mechanical ventilation, the duration of renal replacement therapy and the length of stay in the ICU and hospital. The primary outcome was also reported in six predefined subgroups of patients with and without trauma, severe sepsis and acute respiratory distress syndrome. The groups were well matched at baseline. Study fluid was administered to 97.4% of the randomized patients: 90 in the albumin group and 107 in the saline group did not receive study fluid. Three hundred and nine patients (8.8%) in the albumin group and 375 patients (10.7%) in the saline group also received non-study resuscitation fluid. The administration of non-study resuscitation fluid was due to error in the majority of cases (189 [5.4%] in albumin group and 190 [5.4%] in saline group). Non-study resuscitation fluid was given because the treating clinicians believed that there was a specific clinical indication in 68 (2.0%) of the patients assigned albumin and 103 (2.9%) of the patients assigned saline. At the completion of the trial the vital status 28 days after randomization was available for 6930 (99.0%) patients. In most cases the vital status was only missing because consent to the use of personal health data was withheld or withdrawn
Choice of resuscitation fluid in the critically ill
305
Fig. 13.1 (a) SAFE Study fluid in a masking carton. Reproduced from http://bmj.com/cgi/content/full/326/73 89/559/DC1 with permission, (b) SAFE Study administration set. Reproduced from http://bmj.com/cgi/contenVfull/326/73 89/559/DC1 with permission. . The patients randomized to albumin received significantly less study fluid than those allocated to saline on each of the first 3 days of the study, resulting in a significantly greater positive fluid balance in the saline group on each day. The ratio of the volume of study albumin to study saline administered over the first 4 days was 1:1.4. The patients in the two groups received similar volumes of other fluids during the first 4 days except that the patients randomized to receive albumin were transfused with greater volumes of packed red blood cells on the first 2 days and on day 2 when patients in the saline group received a greater volume of non-study fluids. The two groups were resuscitated to similar clinical end-points, but the serum albumin concentration was higher in the albumin group throughout the study period.
The year in anaesthesia and critical care
306
The two groups had similar 28-day mortalities: albumin 20.9% and saline 21.1% (RR 0.99; 95% CI 0.91–1.09). There was no difference in survival time, the number of patients developing new single- or multiple-organ failures, ICU and hospital stay, days of mechanical ventilation or days of renal replacement therapy (Table 13.1). Subgroup analyses were carried out. The RR of 28-day mortality for albumin versus saline for patients with trauma was 1.36 compared with an RR for non-trauma patients of 0.96, (P=0.04 for the test for a common RR). This increased RR of death in patients with trauma was due to an excess of deaths in patients with trauma and brain injury (mortality for albumin=24.5%, mortality for saline=15.1%; RR 1.62; 95% CI 1.12–2.34) (P=0.009). There was no difference in the mortality in trauma patients without brain injury, this being 6.2% in both groups. (RR 1.00; 95% CI 0.56–1.79) (P=1.00). For trauma patients overall there were 81 (13.6%) deaths in the albumin group versus 59 (10.0%) in the saline group (RR 1.36; 95% CI0.99–1.86) (P=0.06). The 28-day mortality RR for albumin versus saline for patients with severe sepsis was 0.87 compared with 1.05 for patients without severe sepsis (P=0.06). The mortality rates for patients with severe sepsis were 30.7% for albumin and 35.3% for saline (RR 0.87; 95% CI 0.74–1.02) (P=0.09). The RR of death for patients with acute respiratory distress syndrome for albumin versus saline was 0.93 versus 1.00 for patients without acute respiratory distress syndrome (P=0.74). The RRs of death for the albumin and saline groups for the whole study population and for predefined subgroups are presented in Fig. 13.2. The SAFE Study was the first adequately powered trial to provide reliable estimates of the relative treatment effects of different resuscitation fluids. With the exception of the subgroup of patients with trauma and brain injury in whom caution is clearly warranted, the study provides powerful evidence that purified 4% albumin is as safe as saline in a heterogeneous population of adult ICU patients. The Cochrane Injuries Group Albumin Reviewers have updated their systematic review to include the results of the SAFE Study and this is discussed below. Whilst the SAFE Study can reassure clinicians about the safety of albumin, the study does not provide evidence that albumin is superior to saline. Although there is no doubt that the acquisition cost of albumin exceeds that of saline and other crystalloid solutions, the SAFE Study provided no data to suggest that the use of albumin results in other cost savings that offset its greater acquisition cost. The SAFE Study investigators did not conduct a formal economic analysis, but the gross measures of resource use such as ICU and hospital length of stay, duration of mechanical ventilation and renal replacement therapy were similar in the two groups.
Table 13.1 Primary and secondary outcomes from the SAFE study Outcome
Albumin group
Saline group
Relative risk (95% Cl)
726/3473 (20.9)
729/3460 (21.1)
0.99 (0.91– 1.09)
Absolute difference (95% Cl)
Pvalue
Status at 28 days— no./total no. (%) Dead
0.87
Choice of resuscitation fluid in the critically ill
307
Alive in ICU
111/3473 (3.2)
87/3460 (2.5)
1.27 (0.96– 1.68)
0.09
Alive in hospital‡
793/3473 (22.8)
843/3460 (24.5)
0.93 (0.86– 1.01)
0.10
Length of stay in ICU— days
6.50 ± 6.6
6.20 ± 6.2
0.24 (−0.06 to 0.54)
0.44
Length of stay in hospital—days
15.30±9.6
15.60±9.6
−0.24 (−0.70 to 0.21)
0.30
Duration of mechanical ventilation—days
4.50±6.1
4.30±5.7
0.19 (−0.08 to 0.47)
0.74
Duration of renal replacement therapy— days
0.48±2.28
0.39±2.0
0.09 (−0.00 to 0.19)
0.41
New organ failure—no. %)‡
0.85§
No failure
1397 (52.7)
1424 (53.3)
1 organ
795 (30.0)
796 (29.8)
2 organs
369 (13.9)
361 (13.5)
3 organs
68(2.6)
75 (2.8)
4 organs
18(0.7)
17 (0.6)
5 organs
2(0.1)
0
Patients with trauma
81/596 (13.6)
59/590 (10.0)
1.36 (0.99– 1.86)
0.06
Patients with severe sepsis
185/603 (30.7)
217/615 (35.3)
0.87 (0.74– 1.02)
0.09
0.93 (0.61– 1.41)
0.72
Death within 28 days according to subgroup— no./total no. %)
Patients with acute respiratory distress syndrome
24/61 (39.3) 28/66 (42.4)
Plus-minus values are means±standard deviations. †Data include the numbers of patients in the ICU or the length of stay in the ICU, ‡Data were available for 2649 patients in the albumin group and 2673 patients in the saline group. New organ failure was defined as a Sequential Organ Failure Assessment score of 0, 1 or 2 in any individual organ system at baseline, followed by an increase in the score to 3 or 4 in the same system. § P-value pertains to the comparison between the albumin and saline groups in the numbers of patients who had no new organ failure or new failure of one, two, three, four or five organs. Source: Finfer et al. (2004).
The year in anaesthesia and critical care
308
Fig. 13.2 The RR of death for all patients and predefined subgroups assigned to albumin versus saline in the SAFE Study. The square boxes indicate the point estimate for the RR of death, the size of the box is proportional to the number of deaths in the subgroup and the horizontal lines indicate the 95% confidence limits for the RR of death. Source: Finfer et al. (2004). The trend towards lower mortality in the subgroup of patients with severe sepsis assigned to albumin may be considered clinically significant, but may well have occurred by chance. Confirmatory findings from other randomized controlled trials are required before albumin can be recommended as the resuscitation fluid of choice in patients with severe sepsis. Taken in conjunction with the systematic reviews by Choi et al. and the Cochrane Injuries Group Albumin Reviewers |1, 2|, the SAFE Study can best be interpreted as strong evidence that crystalloid and colloid solutions are equally effective. Of the colloid solutions currently available to clinicians, albumin is the only one proven to be as safe as a crystalloid solution, but it is also one of the most expensive.
Human albumin solution for resuscitation and volume expansion in critically ill patients Alderson P, Bunn F, Lefebvre C, et al.; Albumin Reviewers Cochrane Database Syst Rev 2004; 4: CD001208
Choice of resuscitation fluid in the critically ill
309
BACKGROUND. Human albumin solutions are used in a range of medical and surgical problems. The licensed indications are the emergency treatment of shock and other conditions where restoration of the blood volume is urgent, burns and hypoproteinaemia. INTERPRETATION. There was no evidence that albumin reduced mortality for patients with hypovolaemia when compared with cheaper alternatives such as saline. There was no evidence that albumin reduced mortality in critically ill patients with burns and hypoalbuminaemia. The possibility that there may be highly selected populations of critically ill patients in whom albumin may be indicated remains open to question. Comment This updated Cochrane collaboration review of randomized controlled trials reported an assessment of the effect of human albumin solutions or plasma protein fraction (PPF) on the mortality of critically ill patients with hypovolaemia, burns and hypoproteinaemia. The trial selection criteria were randomized controlled trials comparing albumin/PPF with no albumin/PPF or with a crystalloid solution. The original 1998 review included 30 randomized controlled trials including 156 deaths among 1419 patients. The updated review included 32 randomized controlled trials with 1632 deaths in 8452 patients (the SAFE Study provided nearly all the additional patients and events). The main results were as follows: for hypovolaemia the RR for albumin/PPF was 1.01 (95% CI 0.92–1.10), for burns the RR for albumin/PPF was 2.40 (95% CI 1.11–5.19) and for hypoalbuminaemia the RR for albumin/PPF was 1.38 (95% CI 0.94–2.03). Combining the categories, the RR of death with albumin or PPF administration was 1.04 (95% CI 0.95–1.13). The 1998 Cochrane review concluded that ‘there is no evidence that albumin administration reduces mortality in critically ill patients with hypovolaemia, burns, or hypoalbuminaemia and a strong suggestion that it may increase mortality’ (p. 235). Since incorporating the SAFE Study, which contributed 83% of the data to this updated review, the reviewers have removed the suggestion that albumin administration may increase the risk of death. Instead they noted that albumin is twice as expensive as starch and over 30 times as expensive as crystalloid solutions and recommended against its use on the basis of cost. Although cost may not be the primary factor determining choice of resuscitation fluid, it is a factor that many clinicians will at least consider. The absence of formal cost-effectiveness comparisons between different fluids with widely differing physiological properties means that cost-based decisions currently have to be made using data limited to fluid purchase price and presumed relative benefits and harms. Future research should concentrate on examining the efficacy of albumin in more highly selected populations of critically ill patients and, whenever possible, randomized controlled trials should include formal cost-effectiveness analyses.
The year in anaesthesia and critical care
310
Colloids versus crystalloids for fluid resuscitation in critically ill patients Roberts I, Alderson P, Bunn F, Chinnock P, Ker K, Schierhout G. Cochrane Database Syst Rev 2004; 4: CD000567 BACKGROUND. Colloid solutions are widely used in the fluid resuscitation of critically ill patients. There are several choices of colloid and there is ongoing debate about the relative effectiveness of colloids compared to crystalloids. INTERPRETATION. There was no evidence that resuscitation with colloids reduced the risk of death compared to resuscitation with crystalloids. Comment The Cochrane systematic review of randomized controlled trials comparing colloid and crystalloid fluids in critically ill patients requiring volume resuscitation has also been updated following publication of the SAFE Study. Eligible randomized controlled trials were all randomized and quasi-randomized trials of colloids compared to crystalloids in patients requiring volume replacement. Crossover trials and trials in pregnant women, neonates and pre-operative elective surgical patients were excluded. We noted extensive revision following comments on the original publication. In particular, as a result of concerns that it may be inappropriate to combine effect estimates from studies of different colloids, the analyses were stratified by the type of colloid. The main results of colloids compared to crystalloids showed the following. 1. The RR of death for the albumin or PPF in 19 trials including 7576 patients was 1.02 (95% CI 0.93–1.11). 2. The RR of death for hydroxyethyl starch in ten trials including 374 patients was 1.16 (95% CI 0.68–1.96). 3. The RR of death for modified gelatin in seven trials including 346 patients was 0.54 (95% CI 0.16–1.85). 4. The RR of death for dextran in nine trials including 834 patients was 1.24 (95% CI 0.94–1.65). The main results of colloids in hypertonic crystalloid compared to isotonic crystalloid showed the following. 1. The RR of death for dextran in hypertonic crystalloid in eight trials including 1283 patients was 0.88 (95% CI 0.74–1.05). 2. The RR of death for albumin and hypertonic saline in one trial including 14 patients was 0.50 (95% CI 0.06–4.33). This updated review reported a total of 64 randomized controlled trials including 10 432 participants and 67% of these participants were provided by the SAFE Study. In comparison, the original 1998 review reported 37 trials with 3082 participants on the BMJ website and 26 trials with 1622 participants in the hard copy version in the BMJ.
Choice of resuscitation fluid in the critically ill
311
The updated review does not support the finding of the original hard copy publication that resuscitation with colloids was associated with a 4% increase in absolute mortality. By stratifying the review by type of colloid, this meta-analysis underscored how most colloid solutions are relatively untested in randomized controlled trials. There is an urgent need for large, blinded, randomized controlled trials evaluating the safety, efficacy and cost-effectiveness of synthetic colloids in heterogeneous populations of critically ill patients and the safety and clinical and economic consequences of albumin and synthetic colloids in more highly selected populations of critically ill patients.
Morbidity in hospitalized patients receiving human albumin: a metaanalysis of randomized, controlled trials Vincent J-L, Navickis RJ, Wilkes MM. Crit Care Med 2004; 32:2029–38 BACKGROUND. The aim of this study was to determine the effect of albumin administration on morbidity in acutely ill hospitalized patients. INTERPRETATION. Albumin reduced morbidity in acutely ill hospitalized patients. Comment This meta-analysis included all randomized controlled trials in which the administration of albumin was compared with the administration of crystalloid, lower dose albumin or no albumin in acutely ill hospitalized patients and in which morbidity data were available. The end-point examined was morbidity, which was defined as the incidence of complications including death. The authors identified 71 trials that included 3782 patients suffering 3287 complications including 515 deaths. The median number of randomized patients per included trial was 36. The reported morbidity was reduced in those assigned albumin with a risk ratio of 0.92 (95% CI 0.86–0.98). In 32 trials in which the control group received no albumin the risk ratio was 0.77 (95% CI 0.67–0.88). Although published after the SAFE Study, this meta-analysis did not include SAFE Study data. The SAFE Study demonstrated no difference between the two groups in the number of patients suffering new single- or multiple-organ failure, no difference in time in the ICU or hospital and no difference in the time treated with renal replacement therapy or mechanical ventilation (Table 13.1). Inclusion of the SAFE Study data would have increased the number of patients from 3782 to 10 779, 65% of whom would have been SAFE Study patients. The effect of including the SAFE Study patients in the updated Cochrane meta-analyses has been to negate the claim that the administration of colloids in general and albumin in particular increases mortality. It is highly likely that the claim that albumin therapy reduces morbidity would also have been negated if the SAFE Study patients had been included in this meta-analysis.
The year in anaesthesia and critical care
312
Evidence-based colloid use in the critically ill: American Thoracic Society Consensus Statement American Thoracic Society. Am J Respir Crit Care Med 2004; 170: 1247–59 BACKGROUND. Colloids are widely used in critical care and have theoretical advantages. Colloids also have the potential to cause adverse effects, carry a relatively high cost and there is no consensus regarding their use. INTERPRETATION. Colloids have rarely been studied in trials designed for determining clinical outcome benefits. The SAFE Study was a notable exception and supported a neutral influence of colloids on mortality in a general ICU population. Further investigation in the ICU setting is needed: in particular, outcome-centred trials are required in septic shock, renal failure, acute lung injury/acute respiratory distress syndrome and certain forms of brain injury. Comment This lengthy and comprehensive publication represented the consensus of a working group constituted under the auspices of the Critical Care Assembly of the American Thoracic Society. The goal of the group was to critically review the literature related to natural and synthetic colloids with a particular focus on patients in medical, surgical and cardiovascular ICUs and to make statements about the literature and therapeutic implications. Children and patients with burns were excluded. The working group was divided into a basic science group, a pre-clinical group and a clinical group in order to review the original colloid literature and associated systematic reviews. The main results showed the following. 1. Colloids have various non-oncotic properties that may influence vascular integrity, inflammation and pharmacokinetics. The clinical relevance of these properties is unknown. 2. All colloids affect coagulation. Dextran and hydroxyethyl starch have the most potent anti-thrombotic effects. 3. Hydroxyethyl starch may be deposited in the reticuloendothelial tissues for prolonged periods: the clinical significance of this is unknown. 4. Colloids restore intravascular volume and tissue perfusion more rapidly than crystalloids in all shock states, regardless of vascular permeability. 5. There is no evidence of a benefit of colloids in treating ischaemic brain injury or subarachnoid haemorrhage. Colloids may impact adversely on survival in traumatic brain injury. 6. Hydroxyethyl starch may increase the risk of acute renal failure in patients with sepsis. 7. Meta-analyses of critical care colloid use are conflicting because of entry trial heterogeneity and varied analytic techniques and a large prospective trial suggested a neutral influence of colloids on clinical outcomes. The therapeutic implications were as follows.
Choice of resuscitation fluid in the critically ill
313
1. Crystalloids should be administered first in non-haemorrhagic shock resuscitation. 2. Hydroxyethyl starch should be used with caution in cardiopulmonary bypass and in patients with sepsis. 3. Colloids should be avoided or used with caution in patients with traumatic brain injury. Of necessity, the process of achieving consensus on such a diverse body of evidence and agreeing on recommendations for clinical practice resulted in many of the conclusions being qualitative with the attendant risk of introducing bias. Nevertheless, this extensive basic science and clinical literature review provides a comprehensive summary of the literature on the theoretical advantages and disadvantages of colloids. It also emphasized the paucity of high-quality, randomized controlled trials in this area and identified areas where further research is urgently required.
Albumin influences total plasma antioxidant capacity favorably in patients with acute lung injury Quinlan GJ, Mumby S, Martin GS, Bernard GR, Gutteridge JMC, Evans TW. Crit Care Med 2004; 32:755–9 BACKGROUND. The aim of this study was to ascertain the influence of albumin on antioxidant status in patients with acute lung injury. INTERPRETATION. Albumin administration favourably influenced antioxidant status as well as levels of protein oxidative damage in patients with acute lung injury.
Comment In this substudy of a prospective, randomized, placebo-controlled trial, Quinlan et al. administered 25% albumin (nine doses of 25 g at intervals of 8 h) or placebo to 20 patients with acute lung injury and measured the plasma antioxidant activity. The main results showed that there was a positive correlation for all patients between albumin and plasma thiol concentrations (r=0.983; P <0.01) and albumin and antioxidant capacity (r=0.885; P<0.01). There was a strong correlation between thiols and antioxidant capacity in the albumin-treated group (r=0.876; P<0.01). No such correlation was apparent in the placebo group. Oxidative damage to plasma proteins and lipids occurs in patients with acute lung injury and acute respiratory distress syndrome. The severity of the damage may predict the clinical outcome. Albumin may have antioxidant functions and this study provided data in support of that hypothesis. In addition, this study suggested that albumin therapy may augment antioxidant capacity in patients with acute lung injury. Whilst these data may encourage those who advocate the use of albumin in patients with acute lung injury and acute respiratory distress syndrome, we await an adequately powered, randomized controlled trial focused on important patient-centred outcomes in this population.
The year in anaesthesia and critical care
314
Influence of different volume replacement strategies on inflammation and endothelial activation in the elderly undergoing major abdominal surgery Boldt J, Ducke M, Kumle B, Papsdorf M, Zurmeyer E-L. Intensive Care Med 2004; 30:416–22 BACKGROUND. The influence of different volume replacement regimens on inflammation and endothelial activation in elderly surgical patients was assessed. INTERPRETATION. Markers of inflammation and endothelial injury and activation were significantly higher in elderly patients after crystalloid- than after hydroxyethyl starch-based volume replacement regimens. Comment Boldt et al. compared the influence of normal saline, Ringer’s lactate and low molecular weight hydroxyethyl starch on markers of inflammation and endothelial activation in 66 elderly patients undergoing major abdominal surgery. The main results were that C-reactive protein, interleukin-6 and interleukin-8 concentrations increased significantly more in the crystalloid-treated patients, whereas endothelial leukocyte adhesion molecule-1 and intercellular adhesion molecule-1 concentrations were unchanged in the hydroxyethyl starch-treated patients, but increased significantly in the crystalloid-treated patients. The hydroxyethyl starch-treated patients received more norepinephrine. There were no differences in the use of blood products or survival between the groups. This small, unblinded study lends support to the hypothesis that different resuscitation fluids produce different physiological responses and so the choice of fluid might theoretically influence important clinical outcomes. The clinical relevance of these findings can only be tested in large, high-quality, randomized controlled trials.
Human albumin and starch administration in critically ill patients: a prospective randomized clinical trial Veneman TF, Nijhuis JO, Woittiez AJJ. Wien Klin Wochenschr 2004; 116: 305–9 BACKGROUND. The aim of this study was to determine whether human albumin or hydroxyethyl starch may lead to an increase in colloid osmotic pressure (COP) and to a better clinical outcome in critically ill hypoalbuminaemic patients compared to normal saline. INTERPRETATION. The administration of human albumin effectively increased the COP in critically ill patients, whereas hydroxyethyl starch and saline did not.
Choice of resuscitation fluid in the critically ill
315
Comment In a small, unblinded, randomized controlled trial of 61 patients designed to detect changes in COP, Veneman et al. found that albumin was more effective at increasing the COP than hydroxyethyl starch or crystalloid. The 30-day mortality was 53.3% in the albumin group, 60.0% in the hydroxyethyl starch group and 31.3% in the saline group. The size of this study precludes making any conclusions about the relative merits of the fluids studied beyond their effects on the COP. These results illustrate how small studies may generate statistically significant differences between groups that could be real but may be found by chance. The results of the blinded SAFE Study allowed the numerically impressive but statistically insignificant differences in 30-day mortality to be viewed in the appropriate light. Hypertonic saline for fluid resuscitation Hypertonic saline has been extensively investigated in animal and human studies as a resuscitation fluid for hypovolaemic shock |11|. Studies have suggested that hypertonic saline increases the blood pressure and decreases the intracranial pressure more effectively than isotonic crystalloids |12, 13|. The potential of hypertonic saline is greatest in the pre-hospital and military environments where reliance on small volumes and rapid resuscitation may be particularly important. Hypertonic saline also decreases the intracranial pressure in ICU patients with severe traumatic brain injury by shifting water away from non-injured brain areas |14|. In this setting, hypertonic saline may be used as an alternative to mannitol as it may maintain the intravascular volume and cerebral perfusion better. The use of hypertonic saline (both alone and combined with dextran to increase the duration of action) for pre-hospital resuscitation of hypotensive trauma patients has been tested in eight randomized controlled trials, including one of 422 patients |12|. Whilst an acceptable safety profile for hypertonic saline was suggested by these trials, none of them demonstrated an increase in hospital survival |12, 13, 15|. Post hoc subgroup analysis in two trials suggested that the use of hypertonic saline might increase hospital survival in patients with traumatic brain injury |13, 15|. Wade et al. reported that survival increased significantly from 27 to 38% with hypertonic saline/ dextran (adjusted P=0.048) in a cohort analysis of 223 traumatic brain injury patients taken from six randomized controlled trials including patients with traumatic hypotension |16|. Neurological function is a better measure of outcome than hospital survival in patients with traumatic brain injury as it addresses the concern that new therapies might increase the number of survivors with severe disability. As function improves up to at least 6 months after injury, the 6-month neurological scores are accepted as an optimal outcome measure in trials of new therapies for patients with traumatic brain injury. Despite the suggestion that hypertonic saline might improve outcomes in patients with traumatic brain injury, at the beginning of 2004 no randomized controlled trials had investigated either pre-hospital hypertonic saline or hypertonic saline/dextran in traumatic brain injury patients and no fluid resuscitation trials had used neurological function as the primary outcome measure. Despite the lack of any randomized controlled trial supporting the use of hypertonic saline in traumatic brain injury patients, hypertonic saline/colloid solutions
The year in anaesthesia and critical care
316
have been in clinical use in Europe since 1991 and hypertonic saline/dextran has regulatory approval for pre-hospital resuscitation in 14 European countries |17, 18|.
Pre-hospital hypertonic saline resuscitation of patients with hypotension and severe traumatic brain injury; a randomized controlled trial Cooper DJ, Myles PS, McDermott FT, et al., for the HTS Study Investigators. JAMA 2004; 291:1350–7 BACKGROUND. The aim of this study was to determine whether pre-hospital resuscitation with intravenous hypertonic saline improved the long-term neurological outcome in patients with severe traumatic brain injury compared with resuscitation with conventional fluids. INTERPRETATION. Patients with hypotension and severe traumatic brain injury who received pre-hospital resuscitation with hypertonic saline had almost identical neurological function 6 months after injury as patients who received conventional fluid. Comment In the first, double-blind, randomized controlled trial of pre-hospital hypertonic saline resuscitation for traumatic brain injury, Cooper et al. randomized 229 patients to receive an identical bag of 250 ml of either 7.5% saline or Ringer’s lactate solution in addition to standard resuscitation fluids. The patients had non-penetrating traumatic coma (blunt head injury and a Glasgow coma score of less than 9) and hypotension (systolic blood pressure <100 mmHg). Patients with multisystem trauma were included. Randomization was in blocks of four and was stratified by ambulance road car and thereby also by destination hospital. Treatment allocation was concealed and all pre-hospital personnel, hospital staff and outcome assessors were blinded. Paramedics infused the study fluid in addition to their usual protocols of Ringer’s lactate or a colloid solution. Each patient’s neurological outcome was measured 3 and 6 months after injury by an extended (8 grade) Glasgow Outcome Scale score obtained by a single trained assessor using a standardized structured questionnaire. Ethics committees waived the requirement for pre-hospital informed consent and instead delayed consent for continuation in the study was obtained later from the patient’s next of kin or other legal surrogate. The main results were that the two groups were equivalent at baseline with a median injury severity score of 38. Both groups received a median of 1250 ml of intravenous crystalloid and colloid resuscitation fluids in addition to their study fluids. Pre-hospital hypotension had been corrected equally in both groups on arrival at hospital. The patients receiving hypertonic saline had increased serum sodium at hospital admission and this increase was maintained for the first 12 hours of hospital admission.
Choice of resuscitation fluid in the critically ill
317
Fig. 13.3 Extended Glasgow Outcome Scale in patients with traumatic brain injury at 6 months after injury. The extended Glasgow Outcome Scale is an eight-point scale whereby 1 indicates dead, 2 indicates vegetative, 3 indicates lower severe disability, 4 indicates upper severe disability, 5 indicates lower moderate disability, 6 indicates upper moderate disability, 7 indicates lower good recovery and 8 indicates upper good recovery. In the hypertonic saline group 62 out of 113 patients survived to 6 months and in the Ringer’s lactate solution group (control) 53 out of 113 patients survived to 6 months. There were 40 survivors (35%) in the hypertonic saline group with a Glasgow Outcome Scale score of 5 (moderate disability) or better and there were 42 survivors (37%) in the Ringers lactate group with a Glasgow Outcome Scale score of 5 or better. Source: Cooper et al. (2004).
The year in anaesthesia and critical care
318
They required a shorter duration of inotropic support (P=0.03), and there was a trend to a lower intracranial pressure (P=0.08) and more adequate cerebral perfusion pressure (P=0.06). However, the primary study end-point, the median Glasgow Outcome Scale score at 6 months after injury, was the same in both groups (P=0.45) and the proportion of patients with favourable neurological outcomes at 6 months was also the same (P=0.96). An encouraging trend towards increased survival at 6 months in those treated with hypertonic saline (55 versus 47%) (P=0.23) was balanced by an equivalent and discouraging trend to an increased number of survivors with severe disability (Fig. 13.3). This carefully designed and executed trial found that pre-hospital resuscitation with hypertonic saline did not improve functional neurological outcomes in severely injured multitrauma patients with brain injury. The study had 80% power to identify a 20% (one grade) change in the extended Glasgow Outcome Scale score grade and found no advantage for hypertonic saline. Pre-hospital hypertonic saline resuscitation produced potential short-term benefits in relation to an improved cerebral perfusion pressure and a shorter duration of inotropic support. However, the trial also pointed to the potential for harm in a trend for an increased number of patients given hypertonic saline surviving with severe disability. Hypertonic saline did not improve functional outcomes and cannot currently be recommended for pre-hospital resuscitation of patients with traumatic brain injury. Future randomized controlled trials investigating resuscitation fluids and other therapies for traumatic brain injury should report the long-term functional outcome. Any apparent survival advantage should be reported with the functional status of survivors. Conclusion At the end of 2004 the inferences from currently available randomized controlled trials offered little to support the routine preferential use of either colloids or crystalloids for reducing mortality, morbidity or cost in the ICU setting. Synthetic colloids have not been sufficiently studied in the critically ill. Whether colloids (including albumin) have niche roles in specific ICU populations remains to be determined in future trials that should evaluate both clinical and economic consequences. The year 2004 also marked two milestones in fluid resuscitation research. One conceptual advance was moving beyond vital status to incorporating a clinically important end-point measuring functional status (Glasgow Outcome Scale score) in the hypertonic saline trial reviewed above (Cooper et al., 2004). Another landmark event was completion of the SAFE Study, which has demonstrated the feasibility of conducting large high-quality fluid resuscitation trials |19|. The future for fluid resuscitation research is promising as several national and international consortia now exist with the express goal of rigorously addressing fundamental clinical research questions in the ICU |20|.
Choice of resuscitation fluid in the critically ill
319
References 1. Schierhout G, Roberts I. Fluid resuscitation with colloid or crystalloid solutions in critically ill patients: a systematic review of randomised trials. BMJ 1998; 316:961–4. 2. Choi PT, Yip G, Quinonez LG, Cook DJ. Crystalloids vs. colloids in fluid resuscitation: a systematic review. Crit Care Med 1999; 27:200–10. 3. Cochrane Injuries Group Albumin Reviewers. Human albumin administration in critically ill patients: systematic review of randomised controlled trials. BMJ 1998; 317:235–40. 4. Wilkes MM, Navickis RJ. Patient survival after human albumin administration: a meta-analysis of randomized, controlled trials. Ann Intern Med 2001; 135:149–64. 5. Schierhout G, Roberts I. Fluid resuscitation with colloid or crystalloid solutions in critically ill patients: a systematic review of randomised trials, http://bmj.bmjjournals.com/%20content/vol316/issue7136/fulltext/supplemental/961/index.shtm l. 6. Roberts I, Edwards P, McLelland B, Fogarty BJ, Khan K. More on albumin. BMJ 1999; 318:1214. 7. Offringa M. Excess mortality after human albumin administration in critically ill patients. BMJ 1998; 317:223–4. 8. Cook D, Guyatt G. Colloid use for fluid resuscitation: evidence and spin. Ann Intern Med 2001; 135:205–8. 9. Annals of Internal Medicine. Summaries for patients: whether albumin therapy improves or worsens survival of critically ill patients is not known. Ann Intern Med 2001; 135: S25. 10. Bunn F, Alderson P, Hawkins V. Colloid solutions for fluid resuscitation. Cochrane Database Syst Rev 2003; 1: CD001319. 11. Dubick MA, Wade CE. A review of the efficacy and safety of 7.5% NaCl/6% dextran 70 in experimental animals and in humans. J Trauma 1994; 36:323–30. 12. Mattox KL, Maningas PA, Moore EE, Mateer JR, Marx JA, Aprahamian C, Burch JM, Pepe PE. Prehospital hypertonic saline/dextran infusion for post-traumatic hypotension. The USA Multicenter Trial. Ann Surg 1991; 213:482–91. 13. Vassar MJ, Fischer RP, O’Brien PE, Bachulis BL, Chambers JA, Hoyt DB, Holcroft JW. A multicenter trial for resuscitation of injured patients with 7.5% sodium chloride. The effect of added dextran 70. The Multicenter Group for the Study of Hypertonic Saline in Trauma Patients. Arch Surg 1993; 128:1003–11. 14. Worthley LI, Cooper DJ, Jones N. Treatment of resistant intracranial hypertension with hypertonic saline. Report of two cases. J Neurosurg 1988; 68:478–81. 15. Vassar MJ, Perry CA, Gannaway WL, Holcroft JW. 7.5% sodium chloride/dextran for resuscitation of trauma patients undergoing helicopter transport. Arch Surg 1991; 126: 1065–72. 16. Wade CE, Grady JJ, Kramer GC, Younes RN, Gehlsen K, Holcroft JW. Individual patient cohort analysis of the efficacy of hypertonic saline/dextran in patients with traumatic brain injury and hypotension. J Trauma 1997; 42: S61–5. 17. Svensen CH. Hypertonic solutions: an update. Trauma Care J 2002; 2:6–12. 18. Kramer GC. Hypertonic resuscitation: physiologic mechanisms and recommendations for trauma care. J Trauma 2003; 54: S89–99. 19. Cook D. Is albumin safe? New Engl J Med 2004; 350:2294–6. 20. Cook DJ, Brower R, Cooper DJ, Brochard L, Vincent JL. Multicenter clinical research in adult critical care. Crit Care Med 2002; 30:1636–43.
14 Glucose control in the critically ill JOHAN GROENEVELD, EMMY RIJNSBURGER, MARINA HONING Introduction It has recently become increasingly clear that hyperglycaemia in acute illness, either caused by or aggravated by acute illness-associated insulin resistance, worsens the outcome in a variety of conditions, including stroke, myocardial infarction and trauma. In a recent hallmark study |1|, intensive insulin treatment for maintaining blood glucose levels at between 4.4 and 6.1 mmol/l, thereby preventing even mild insulin resistanceinduced hyperglycaemia, was associated with improved survival in a large group of critically ill patients, mostly post-cardiac surgery patients. The mechanisms by which hyperglycaemia aggravates organ damage and how insulin partly prevents it are unknown, but are likely to be multifactorial |2|. Moreover, insulin may have beneficial effects through controlling glycaemia, but may also have intrinsic effects, modulating cell signalling, limiting inflammation and promoting cell repair. Van den Berghe recently summarized some of the potential mechanisms of the survival benefit of intensive insulin treatment |3|. We will review some recently published reports addressing these issues. We will not include a discussion on the merits and detriments of glucose-insulin-potassium infusion in myocardial infarction, angioplasty or coronary artery surgery, nor will we consider experimental animal work related to the subject. Hyperglycaemia and acute illness: assessment and significance During critical illness hyperglycaemia is often caused by insulin resistance evoked by the underlying acute illness (Table 14.1) |4, 5|. Hyperglycaemia can only be detected if it is measured and correctly appraised. We will briefly review some methods published in 2004 that help to estimate the glycaemic stress in critically ill patients. © Atlas Medical Publishing Ltd
Table 14.1 Causes of hyperglycaemia and insulin resistance in the critically ill Adrenergic stimulation of hepatic gluconeogenesis and glycogenolysis Release of cortisol, glucagon and growth hormone (GH) Pro-inflammatory factors, e.g. tumour necrosis factor-α (TNF-α) and interleukin-6 (IL-6), acting at the insulin receptor complex level Inhibition of glucose transporter 4
Glucose control in the critically ill
321
High insulin-like growth factor (IGF)-binding proteins in serum
Hyperglycaemic index as a tool to assess glucose control: a retrospective study Vogelzang M, Van der Horst ICC, Nijsten MWN. Crit Care 2004; 8: R122–7 BACKGROUND. In a retrospective study in Groningen, Vogelzang et al., collected blood glucose determinations in critically ill patients. The hyperglycaemic index was defined as the area under the curve above the upper limit of normal (<6.0 mmol/l) glucose levels divided by the total length of stay in the intensive care unit (ICU). INTERPRETATION. A total of 1779 patients and 65 428 glucose determinations were studied. The hyperglycaemic index was a mean of 0.9 mmol/l in survivors and 1.8 mmol/l in non-survivors (17% of all patients) at day 30 (P<0.001). The discriminatory value of the hyperglycaemic index was slightly greater than that of the morning, maximum, admission or mean (in the ICU) glucose levels. The hyperglycaemic index was shown to contribute to mortality independently from age, sex and Acute Physiology and Chronic Health Evaluation (APACHE) II scores using multiple regression. Comment Obviously, this model and its associated computer programme, as well as the potentially greater association with the outcome than for single or mean glucose levels, need prospective confirmation. Also of particular interest is the relation with glucosylation products such as glycohaemoglobin A1c (HbA1c) |6|.
Glycemic control in the ICU: a multicenter survey McMullin J, Brozek J, Jaeschke R, et al. Intensive Care Med 2004; 30: 798–803 BACKGROUND. In order to investigate current practice, 233 Canadian ICU nurses and 89 physicians in five university-affiliated hospitals completed a questionnaire on their beliefs about and attitudes to glycaemic control in the ICU. INTERPRETATION. The clinically important threshold for hyperglycaemia in diabetic as well as non-diabetic patients was reported to average 10 mmol/I. Nurses had higher thresholds (by 0.5–0.8 mmol/l) than physicians and ICU physicians expressed more concern about the accuracy of glucometer measurements. Avoidance of hyperglycaemia was judged to be important, particularly by physicians, for diabetics, patients with acute brain injury, after a recent seizure, in advanced liver disease or after a myocardial infarction.
The year in anaesthesia and critical care
322
Comment This study illustrated that intensive insulin therapy |1| is not yet commonplace in Canadian ICUs and this may apply to other institutions as well. Hyperglycaemic significance Next we will discuss the effect of hyperglycaemia on organ damage.
Persistent post-stroke hyperglycemia is independently associated with infarct expansion and worse clinical outcome Baird TA, Parsons MW, Phanh T, et al. Stroke 2003; 34:2208–14 BACKGROUND. Twenty-five patients were studied within 24 h after a stroke by continuous and intermittent glucose monitoring until 72 h after admission. Brain magnetic resonance imaging (MRI) and clinical assessments were performed at 15 h, 5 days and 85 days. INTERPRETATION. Continuous and intermittently measured glucose levels, particularly when >7 mmol/l, correlated significantly with the acute and chronic infarct volume (MRI) as well as the eventual functional outcome, which was worse with hyperglycaemia. The HbA1c level did not contribute further to this prediction. Comment The study suggested that even mild hyperglycaemia is an independent determinant of tissue damage and infarct expansion during a stroke. It is associated with worse functional outcomes.
Admission blood glucose level as risk indicator of death after myocardial infarction in patients with and without diabetes mellitus Stranders I, Diamant M, van Gelder RE, et al. Arch Intern Med 2004; 164: 982–8 BACKGROUND. Hyperglycaemia may increase the mortality after myocardial infarction, even in non-diabetics. The authors retrospectively evaluated 846
Glucose control in the critically ill
323
Fig. 14.1 Kaplan-Meier survival curves for patients without known diabetes mellitus and admission blood glucose levels less than 7.8 mmol/l (Group 1), 7.8–11.0 mmol/l (Group 2) and 11.1 mmol/l or higher (Group 3) and patients with previously known diabetes (Group 4). Source: Stranders et al. (2004). (109 diabetic) patient charts for admission blood glucose values. The follow-up was 0–93 months. INTERPRETATION. Almost twice as many diabetics as non-diabetics died during the follow-up. An increase of 1 mmol/l in the glucose level was associated with a 4% increase in the mortality risk in non-diabetics and a 5% increase in diabetic patients. One hundred and one non-diabetic subjects had admission values above 11.1 mmol/l and their mortality (43%) was similar to that of patients with known diabetes mellitus (Fig. 14.1). Comment This large study again nicely illustrated that mild hyperglycaemia, even in non-diabetics, is associated with an increased long-term mortality after myocardial infarction.
The year in anaesthesia and critical care
324
Diagnostic significance There was also some information published in 2004 on the prognostics of hyperglycaemia in intensive care patients.
Glucose control and mortality in critically ill patients Finney SJ, Zekveld C, Elia A, Evans TW. JAMA 2003; 290:2041–7 BACKGROUND. Of 531 patients (median age 64 years) admitted into an ICU, 523 underwent prospective analysis of their blood glucose levels in order to evaluate the relative importance of glycaemic control and insulin treatment on mortality. The target glucose values were 5.0–8.0 mmol/l for routine practice, but no strict insulin infusion protocol was used.
Fig. 14.2 The figure illustrates the studied strata of the glucose values. Patients, whether surviving or not, were classified according to the times
Glucose control in the critically ill
325
(normalized for the total admission duration) spent in a given stratum. Source: Finney et al. (2003). INTERPRETATION. The rates of mortality in the ICU and hospital were 5.2 and 5.7%, respectively. The authors found that insulin treatment was already positively and significantly associated with ICU mortality at a prevailing glucose stratum of 6.1–8.0 mmol/l and beyond in multivariate logistic regression analysis, suggesting that the survival benefits of insulin are attributable to glycaemic control rather than an increased administration of insulin. At every level of duration of the glucose targets (Fig. 14.2) the administration of insulin was associated with increased mortality. The threshold glucose level was estimated to amount to approximately 8.0 mmol/l. Comment This study suggested, according to the authors, that glucose control by insulin explains the survival benefit of intensive insulin treatment in the critically ill |1|. They could not exclude a detrimental role of exogenous insulin. However, the results can also be interpreted as non-survivors having higher insulin requirements than survivors because of a greater insulin resistance in the course of more severe disease. The statistical associations given by the authors did not reveal cause and effect relationships.
Hyperglycaemia and mortality in critically ill patients; a prospective study Christiansen C, Toft P, Jørgensen HS, Andersen SK, Tønnesen E. Intensive Care Med 2004; 30:1685–8 BACKGROUND. The authors prospectively studied maximum blood glucose levels and their association with morbidity and mortality in 135 patients staying in an ICU for more than 2 days after exclusion of diabetic patients. INTERPRETATION. There was a maximum blood glucose concentration-dependent increase in mortality in surgical but not in medical patients (Fig. 14.3) and hyperglycaemia correlated with the admission APACHE II and maximum sequential organ failure assessment scores and the occurrence of infections in the surgical patients. Comment This study suggested that, in a heterogeneous group of critically ill patients, particularly after recent surgery, hyperglycaemia (>12 mmol/l) was associated with increased mortality, although the mechanisms behind this association remained unclear. This particularly applied to the severity of disease as a potential confounder.
The year in anaesthesia and critical care
326
Fig. 14.3 The relation between (a) the maximum blood glucose level and mortality and (b) the severity of illness by the first day APACHE II score, respectively. Stratified on reason for admission. The boxes indicate the 25th, 50th and 75th percentiles and the whiskers indicate the 5th and 95th percentiles, while the 1st and 99th percentiles are indicated by crosses.**P<0.01 (test of trend). Source: Christiansen et al. (2004).
Admission hyperglycemia as a prognostic indicator in trauma Yendamuri S, Fulda GJ, Tinkoff GH. J Trauma 2003; 55:33–8
Glucose control in the critically ill
327
BACKGROUND. The charts of 738 patients >17 years of age presenting to a trauma centre were reviewed. Moderate hyperglycaemia was defined by admission glucose values >11.1 mmol/l and mild hyperglycaemia by admission glucose values >7.5 mmol/l. INTERPRETATION. Both mild and moderate hyperglycaemia were associated with prolonged ICU and hospital stays and increased mortality (15.5 and 34.1%, respectively, versus 3% in normoglycaemic patients), independent of age, gender, injury severity and revised trauma scores (logistic regression). Infectious complications were more common in patients with hyperglycaemia. Again, this appeared to be an independent predictor. Comment Although this study suggested an independent contribution of even mild hyper-glycaemia to morbidity and mortality, it cannot be completely ruled out that hyper-glycaemia was only a marker rather than a mediator of disease severity.
Relationship of early hyperglycemia to mortality in trauma patients Laird AM, Miller PR, Kilgo PD, Meredith JW, Chang MC. J Trauma 2004; 56: 1058–62 BACKGROUND. In this population of 516 trauma patients, the authors retrospectively studied the incidence and significance of blood glucose levels on days 1 and 2 after admission. Patients with diabetes mellitus were excluded. INTERPRETATION. The cut-offs for hyperglycaemia were 6.1, 8.3 and 11.1 mmol/l. Infections were related to the injury severity score and hyperglycaemia above 8.3 mmol/l (in 311 patients) and 11.1 mmol/l (in 90 patients), although this was not statistically significantly for hyperglycaemia (univariate logistic regression). Early hyperglycaemia above 11.1 mmol/l appeared of predictive value for mortality in these patients, independent of age, injury severity score, Glasgow coma scale score and base deficit in multivariate logistic regression. Comment This observational study did not rule out a marker rather than a mediator role of hyperglycaemia. Section summary Hyperglycaemia is significantly associated with increased morbidity and mortality in a variety of conditions, including myocardial infarction, stroke, etc. The studies cited again suggested that even mild hyperglycaemia may increase ischaemia/reperfusion tissue damage as well as mortality in the critically ill patient. Although attempting to control for the severity of underlying disease, the studies did not answer the question of whether
The year in anaesthesia and critical care
328
hyperglycaemia is a marker or mediator of morbidity and mortality. Nevertheless, the data argue in favour of avoiding hyperglycaemia in acute and critical illness. However, the optimal range of blood glucose levels and particularly the safe upper limit in critically ill medical and surgical patients needs to be further defined in both diabetics and nondiabetics. This is important in order to avoid too strict a glucose control with an increased risk for episodes of hypoglycaemia. It is possible that the safe upper limit of normal will be in the range of 6.0–8.0 mmol/l rather than the 6.1 mmol/l proposed by the Van den Berghe et al. study |1| and the possibility that this range differs between medical and surgical patients, independently from a history of diabetes mellitus, cannot be excluded. Hyperglycaemia and hyperinsulinaemia: host response interactions Hyperglycaemia may have a variety of unwanted side effects, particularly regarding the non-specific host defence against infection, including decreased neutrophil function |7|. After cardiac surgery, for instance, hyperglycaemia is known to increase the risk of severe (sternal) wound infection and, with burn wounds, hyperglycaemia may increase the risk for bacteraemia. New mechanistic data have emerged.
High glucose impairs superoxide production from isolated blood neutrophils Perner A, Nielsen SE, Rask-Madsen J. Intensive Care Med 2003; 29:642–5 BACKGROUND. Superoxide production in neutrophils is a mechanism for killing invading bacteria. The authors assessed whether the superoxide production by neutrophils of normal healthy volunteers, as evaluated by a reduction in cytochrome C, was affected by incubation with glucose, dehydroepiandrosterone (DHEA), a glucose-6-phosphate-dehydrogenase inhibitor or mannitol. INTERPRETATION. It appeared that glucose and DHEA but not mannitol suppressed superoxide production by neutrophils in a concentration-dependent fashion (Fig. 14.4). Comment These findings elicited an editorial on the differences between neutrophils and endothelial cells in respect of oxygen (O2) radical production, since hyperglycaemia stimulates rather than inhibits radical production in endothelial cells |8|. The author stressed the different metabolic pathways in endothelial cells and neutrophils, with hyperglycaemia enhancing the mitochondrial production of oxyradicals in endothelial cells and inhibiting the pentose phosphate pathway for nicotinamide adenine dinucleotide (NADH) and oxyradical production in neutrophils. The data also suggested that hyperglycaemia suppresses glucose-6-phosphate dehydrogenase in neutrophils and that the latter enzyme, acting in the first step of the pentose phosphate pathway, is involved in O2 radical production. Decreased oxyradical formation may decrease the host defence
Glucose control in the critically ill
329
against invading bacteria. Finally, neutrophil dysfunction partly explains the increased microbial infection rate during hyperglycaemia and poorly controlled diabetes mellitus.
Fig. 14.4 Effects of glucose, DHEA and mannitol on O2–production in cultures of isolated blood neutrophils from healthy subjects. O2–production was induced by formyl-Met-Leu-Phe (fMLP) and measured by superoxide dismutase-inhibitable cytochrome c reduction, (a) Effect of extracellular glucose (P<0.05 for 5 or 10 versus 25 mM). (b) Effect of extracellular DHEA in cells cultured with 5 mm glucose (P<0.05). (c) Effect of extracellular mannitol in cells cultured with 5 mM glucose (P>0.05). The x-axis gives the final concentration of glucose, DHEA and mannitol and the bars are means (±standard errors of the mean). Source: Perner et al. (2003).
The year in anaesthesia and critical care
330
Apart from impaired neutrophil function, hyperglycaemia may have other proinflammatory effects that can be overcome by glucose control using insulin |7|.
Intensive insulin therapy exerts anti-inflammatory effects in critically ill patients and counteracts the adverse effect of low mannose-binding lectin levels Hansen TK, Thiel S, Wouters PJ, Christiansen JS, Van den Berghe G. J Clin Endocrinol Metab 2003; 88:1082–8 BACKGROUND. C-reactive protein (CRP) and mannan-binding lectin (MBL), two acute phase proteins, were studied in relation to conventional or intensive insulin therapy in 451 critically ill patients with a duration of stay >5 days in the ICU from an original cohort of 1548 |1|. INTERPRETATION. It appeared that CRP declined more in the intensive insulintreated group than in the conventional group, while MBL increased less in the course of the disease
Fig. 14.5 Serial measurements of CRP concentrations in patients receiving intensive insulin treatment (open boxes) or conventional treatment (solid boxes). The P values refer to betweengroup comparisons with MannWhitney U-test. Source: Hansen et al. (2003).
Glucose control in the critically ill
331
(Fig. 14.5). Multiple regression confirmed that the effect on CRP could be responsible in part for the morbidity- and mortality-lowering effect of intensive insulin treatment. Comment The findings suggested an anti-inflammatory effect of intensive insulin treatment in maintaining glucose levels at between 4.4 and 6.1 mmol/l, since CRP is produced by the liver mainly upon stimulation by IL-6. In addition, MBL takes part in the lectin route of complement activation for combating invading microorganisms, in addition to the classical and alternative pathways.
Effect of hyperglycemia and hyperinsulinemia on the response of IL-6, TNF-α and FFA to low-dose endotoxemia in humans Krogh-Madsen R, Møller K, Dela F, Kronborg G, Jauffred S, Pedersen BK. Am J Physiol Endocrinol Metab 2004; 286:E766–72 BACKGROUND. The interaction between sepsis, glucose levels and insulin is still unclear as far as modulation of inflammation is concerned. The authors studied the effect of low-dose endotoxin in ten human volunteers and evaluated metabolic and inflammatory factors in the blood, before and during hyperglycaemic (15 mm) (trial B) and euglycaemic hyperinsulinaemic clamping (trial C) versus control periods (trial A).
The year in anaesthesia and critical care
332
Fig. 14.6 (a) Plasma free fatty acids after endotoxin. The concentrations of free fatty acids before clamping were similar in all trials. Endotoxin increased (P<0.001) the free fatty acids in the trial A clamp, but this was abolished in the trial B and C clamps, (b) IL-6. A difference (P<0.05) between the baseline trial A and the hyperglycaemia trial B was
Glucose control in the critically ill
333
found at t=240, 300 and 360 min. (c) TNF-α. No difference between trials, (d) Neutrophil counts. No difference, (e) Lymphocyte counts were reduced after both clamps (P<0.05), persisting after endotoxin injection, which reduced the lymphocyte counts in all trials. Means and 95% confidence intervals. Source: Krogh-Madsen et al. (2004). INTERPRETATION. Endotoxin increased free fatty acids and cytokines in the blood. There was no difference in the response of TNF-α during either clamp, but IL-6 increased during euglycaemic or hyperglycaemic clamping versus the control, while free fatty acids were suppressed. The endotoxin-induced fall in lymphocyte counts was accentuated by insulin (Fig. 14.6). Comment The authors interpreted their findings as an anti-inflammatory effect of insulin during human endotoxaemia, regardless of glucose levels, but IL-6 may have both pro- and antiinflammatory properties so that it is unclear what the net effect would be. Lowering of free fatty acids by insulin can be important in lessening the (endothelial) toxicity during endotoxaemia, but demonstration of a direct effect is lacking.
Hyperglycaemia at admission to the intensive care unit is associated with elevated serum concentrations of interleukin-6 and reduced ex vivo secretion of tumor necrosis factor-α Wasmuth HE, Kunz D, Graf J, et al. Crit Care Med 2004; 32:1109–14 BACKGROUND. This was a prospective study of the admission values of circulating IL-6, IL-8, TNF-α and IL-10 as well as in vitro neutrophil stimulation by TNF-α in relation to glucose values in 189 critically ill patients. INTERPRETATION. Seventy-five per cent of patients presented with hyperglycaemia, which was related to the risk of death in the ICU (P=0.009). Hyperglycaemia was associated with high IL-6/IL-8/IL-10, reduced ex vivo production of TNF-α and prior diabetes independently of disease severity scoring. Comment The results suggested that admission hyperglycaemia is related to distinct changes in humural and cellular inflammatory function and prompt further study into the immunological effects of insulin in critical illness.
The year in anaesthesia and critical care
334
Insulin treatment improves the systemic inflammatory reaction to severe trauma Jeschke MG, Klein D, Herndon DN. Ann Surg 2004; 239:553–60 BACKGROUND. The authors retrospectively evaluated a group receiving insulin for maintaining blood glucose levels at between 6.7 and 10 mmol/l with a group not having insulin but with similar glucose values in 28 burn-wounded children. The groups were comparable for age and burned surface area. Pro-inflammatory mediators were determined in the plasma. INTERPRETATION. Insulin attenuated pro-inflammatory factors (CRP, TNF-α and IL-1β) and the acute phase reaction and promoted anti-inflammatory (IL-10) factors and serum albumin levels, probably by enhancing hepatic synthesis, up to 30 days after trauma. Insulin decreased triglycerides and free fatty acids in the serum for up to 40 days after trauma. IGF-1 and IGF-binding protein 3 had increased at days 20 and 30. Comment The authors interpreted from their findings that insulin has anti-inflammatory actions, even though the data had been collected retrospectively. Patients needing insulin for preventing severe hyperglycaemia had more insulin resistance at a similar caloric intake than the control group and, thus, were expected to have a higher pro-inflammatory status. Nevertheless, the plasma levels suggest a decreased pro-inflammatory and increased antiinflammatory (IL-10) acute phase response with insulin treatment. How this relates to morbidity (superimposed bacterial infection) and mortality remains unknown. In any case, the suggested anti-inflammatory actions of insulin may confirm the predictions made by others |2, 7|, as well as experimental research by the same authors, showing that insulin exerted anti-inflammatory actions by decreasing the serum levels and hepatic upregulation of pro-inflammatory cytokines while increasing anti-inflammatory cytokines in endo toxin-challenged rats |9|. Section summary Hyperglycaemia apparently increases oxidative stress at the endothelial level and decreases it at the neutrophilic level. Increased oxidative stress in the endothelium may stimulate nuclear factor KB to produce pro-inflammatory cytokines, so that the plasma levels increase, particularly in patients with bacterial infection. Insulin, by controlling hyperglycaemia, may be able to suppress this effect. In contrast, hyper-insulinaemia in the presence of euglycaemia (clamping) may have a pro-inflammatory effect, since it increased the pro-inflammatory cytokine blood levels in an endotoxin model of sepsis in human volunteers according to some investigators |10|. This may relate in part to an
Glucose control in the critically ill
335
increased level of free IGF-1. Intensive insulin therapy might reverse the immune anergy known to occur in prolonged critical illness and thereby exert beneficial effects rather than inhibit pro-inflammatory responses, as documented in part by the paper by Hansen et al. The papers by Krogh-Madsen et al. and Wasmuth et al. may also argue in favour of an anti-inflammatory effect of insulin on endotoxaemia or other critical illness. It is unclear how these divergent albeit incomplete results can be explained, but undoubtedly the underlying severity and stage of the inflammatory process and resultant hyperglycaemia from insulin resistance are crucial factors. More work should be performed in order to substantiate the idea that insulin exerts anti-inflammatory actions in the critically ill. Insulin resistance and treatment: mechanisms Insulin resistance denotes a condition of decreased insulin sensitivity that is manifested by a decrease in the (non-oxidative) glucose (disposal) plasma concentration and an increase in the insulin plasma concentration or both, as measured, for instance, during an insulin clamp. Insulin resistance involves a post-receptor defect caused by inflammatory mediators and counter-regulating hormones that can be partly overcome by the administration of high insulin doses |4, 5, 7|. Most acutely ill patients have some degree of insulin resistance because of increased glycogenolysis and gluco-neogenesis and decreased disposal in skeletal muscle (Table 14.1). Hence, hyper-glycaemia is the hallmark of insulin resistance in the critically ill |4, 5, 7|. The first therapeutic option during insulin resistance or insensitivity is to remove or attenuate the underlying cause, e.g. treatment directed at sepsis or trauma. Paradoxically, exogenous insulin is used in the treatment of the sequelae of insulin resistance, including hyperglycaemia. However, a completely normal metabolism cannot be restored, in spite of its ameliorating effects on hyperglycaemia. Attenuation of hyper-glycaemia may be mainly caused by insulin-induced increased non-oxidative glucose disposal in skeletal muscle, rather than inhibition of increased (hepatic) gluco-neogenesis. However, others have reported that the insulin resistance maybe greater in skeletal muscle than in the liver (|4| see below). Intensive insulin treatment, as undertaken in the Van den Berghe et al. study |1| and found to improve survival, was aimed at glucose level control at between 4.4 and 6.1 mmol/l. However, the survival benefit may not only relate to the attenuation of hyperglycaemia and its potential adverse effects, but also to other properties and actions of insulin in the critically ill (Table 14.2). Again, some of these potential mechanisms were evaluated by Van den Berghe et al. in the same study cohort as in the original paper |1|. First, however, we will discuss some literature on insulin algorithms to reach target glucose levels.
The year in anaesthesia and critical care
336
Table 14.2 The plethora of potentially beneficial effects of insulin in the critically ill Metabolic Attenuation of hyperglycaemia Reduction in trigtycerides and free fatty acids Increase in high-density lipoproteins Reduction in asymmetric dimethyl arginine Anti-inflammatory During hypergycaemia: decreased oxidative stress and release of cytokines, TNF and IL-6 Blunted CRP and MBL Functional Improved endothelial function and vascular reactivity Improved cardiac pre-conditioning Structural Anti-apoptotic: improved cell survival via the insulin receptor substrates P13 kinase, p70S6 kinase, Akt and eNOS or via Erk 1/2 Somatotropic axis and anabolic actions Increased IGF activity and decreased binding proteins
Standardization of intravenous insulin therapy improves the efficiency and safety of blood glucose control in critically ill adults Kanji S, Singh A, Tierney M, Meggison H, McIntyre L, Hebert PC. Intensive Care Med 2004; 30:804–10 BACKGROUND. Strict glucose control seems to improve morbidity and mortality in the critically ill, but the best manner or algorithm for achieving strict control, i.e. insulin dosing based on target glucose levels, is still open for discussion. INTERPRETATION. Two cohorts of critically ill patients received either insulin guided by target blood glucose ranges at the discretion of the treating physician or insulin according to a standardized protocol aiming at glucose values of between 4.4 and 6.1 mmol/l. In the latter cohort,the target glucose levels were reached more rapidly and kept within the target range for a longer time interval than in the former cohort. Severe hypoglycaemia was also less common, while the mean glucose levels over the day were lower. Standardization of insulin dosing on the basis of target glucose levels thus improves the efficacy and safety of strict glucose control in critically ill adults.
Glucose control in the critically ill
337
Comment More blood samples per day were taken for glucose determination in the insulin protocol group than in the conventionally managed group. Hence, the results can be explained in part by more frequently available glucose levels in association with protocol education and training that takes caloric intake into account. Thus, keeping the glucose within the indicated (relatively low) target range requires more frequent blood sampling and insulin dose adjustments, particularly with respect to changes in the amount and route of feeding.
Implementing intensive insulin therapy: development and audit of the Bath insulin protocol Laver S, Preston S, Turner D, McKinstry C, Padkin A. Anaesth Intensive Care 2004; 32:311–16 BACKGROUND. The authors developed and used an intravenous insulin infusion protocol for critically ill patients and compared their glucose values with those in historical controls. INTERPRETATION. Initially, 27 patients were started on the protocol (target value 4–7 mmol/l) and the median glucose value was 6.2 mmol/l, compared to 9.2 mmol/l before implementation. Three of the blood glucose recordings were below 2.2 mmol/l. Improved control (median 6.6 mmol/l), as compared to historical values (8.6 mmol/l), was obtained with respect to the glucose values during the whole ICU stay. Comment The authors concluded that their protocol was safe and efficient. Their target window was somewhat broader than in the Van den Berghe et al. study |1|, that difference being unexplained.
Implementation of a safe and effective insulin infusion protocol in a medical intensive care unit Goldberg PA, Siegel MD, Sherwin RS, et al. Diabetes Care 2004; 27:461–7 BACKGROUND. This study reported on a nurse-implemented insulin infusion protocol suitable for strict glucose control in a medical ICU, based on changes in glucose. INTERPRETATION. Fifty-two patients were treated according to the protocol in 69 episodes for a median of 61 h. The median time for reaching the target glucose levels (5.5–7.7 mmol/l) was 9 h. After the glucose values had fallen below 7.8 mmol/l, 93% of the subsequent hourly measurements were in the presumed clinically acceptable range of
The year in anaesthesia and critical care
338
4.4–11.1 mmol/l. The median insulin infusion rate was 4 U/h. Only 0.3% of the measurements were below 3.3 mmol/l. Comment The authors stated that their protocol was safe and effective. However, the narrow target levels were maintained in only 52% of the patients, suggesting less than optimal glucose control. The authors defended their policy by a comparison with historical controls, suggesting improved control with the protocol. Otherwise, the target glucose values were higher than those in the Van den Berghe et al. study |1|. The protocol was as efficient in diabetics as in non-diabetics. Mechanisms The next question is how insulin exerts beneficial effects in the critically ill, thereby contributing to a diminished morbidity and mortality (Table 14.2).
Outcome benefit of intensive insulin therapy in the critically ill: insulin dose versus glycemic control van den Berghe G, Wouters PJ, Bouillon R, et al. Crit Care Med 2003; 31: 359–66 BACKGROUND. Some 1548 critically ill patients were randomized to conventional (10.0–11.1 mmol/l of glucose) or intensive (between 4.4 and 6.1 mmol/l of glucose) insulin treatment. The authors analysed the factors determining the insulin requirements and the impact of insulin dose versus blood glucose on the observed outcome benefit of intensive insulin treatment. INTERPRETATION. Stepwise linear regression identified the body mass index, diabetes, admission diagnosis, glycaemia, caloric intake and length of stay as determinants of insulin requirements. Normoglycaemia was reached in the intensive insulin group within 24 h, in spite of a rapidly increasing caloric intake. The insulin requirements were a mean of 77 U/day for the first week. After day 7, the requirements fell by 40% on a stable caloric intake. A lowered blood glucose rather than insulin requirements related to reduced mortality and less critical illness neuropathy, bacteraemia and inflammation in multiple logistic regression. As compared with normoglycaemia, even mild hyperglycaemia (6.1–8.3 mmol/l) was associated with higher morbidity and a worse outcome. Hypoglycaemia occurred in 5.2% of intensively treated and 0.8% of conventionally treated patients.
Glucose control in the critically ill
339
Comment In contrast to what was deduced above, in this paper Van den Berghe et al. |1| argued in favour of the low glucose target levels indicated, since metabolic control of glycaemia rather than infused insulin per se contributes to favourable effects.
Intensive insulin treatment in critically ill trauma patients normalizes glucose by reducing endogenous glucose production Thorell A, Rooyackers 0, Myrenfors P, Soop M, Nygren J, Ljungqvist OH. J Clin Endocrin Metab 2004; 89:5382–6 BACKGROUND. Six polytrauma patients (and six controls) were studied using 6, 6–2H2-glucose at baseline, during total parenteral nutrition and during nutrition plus insulin for changes in glucose handling in order to study the mechanism of insulin controlling hyperglycaemia in critical illness. INTERPRETATION. Insulin restored normoglycaemia by suppressing (hepatic) endogenous glucose production (gluconeogenesis) without altering the (peripheral) whole body glucose disposal. Compared to the controls, the patients had higher glucose and insulin levels at baseline and higher insulin requirements for maintaining normoglycaemia during total parenteral nutrition, indicating insulin resistance. Comment Although the numbers were small, this study indicated that, in contrast to earlier suggestions, hepatic insulin resistance can be overcome by sufficient exogenous insulin dosing. Conversely, insulin resistance is greater in peripheral tissues than in the liver.
Extremity hyperinsulinemia stimulates muscle protein synthesis in severely injured patients Gore DC, Wolf SE, Sanford AP, Herndon DM, Wolfe RR. Am J Physiol Endocrinol Metab 2004; 286: E529–34 BACKGROUND. While it is clear that insulin resistance involves the rate of peripheral glucose disposal for a given insulin plasma level, insulin resistance may also be related to the effects on lipid and protein metabolism. To illustrate this, Gore et al., examined the protein balance across muscle in 14 burned (>40%) patients using isotopic dilution of the essential amino acid phenylalanine and fluxes across the femoral vascular bed and the effects of locally infused insulin thereupon. INTERPRETATION. Extremity hyperinsulism had only minor systemic effects. It increased the blood flow in the leg and increased protein synthesis in leg muscle without altering breakdown, both being calculated fom arteriovenous fluxes as well as from muscle biopsies (Fig 14 7) The study nicely documented the anabolic effect of
The year in anaesthesia and critical care
340
exogenous insulin in the skeletal muscle of severe trauma patients. Comment The insulin levels required to increase protein synthesis are far greater during trauma than in healthy volunteers, suggesting insulin resistance as the cause of protein metabolism in the former group.
Fig. 14.7 Measures of the fractional synthesis rate for muscle protein. The values for the healthy controls were taken from another paper by the authors. *P<0.05. Source: Gore et al. (2004).
Regulation of the somatotropic axis by intensive insulin therapy during protracted critical illness Mesotten D, Wouters PJ, Peeters RP, et al. J Clin Endocrinol Metab 2004; 89: 3105–13 BACKGROUND. In this paper the authors evaluated GH, IGF-1 and plasma protein binding levels, reflecting the anabolic somatotropic axis in critical illness. The hypothesis was that intensive insulin therapy has anabolic effects via the somatotropic axis and thereby contributes to a reduced morbidity and mortality. INTERPRETATION. Surprisingly, intensive insulin treatment down regulated the axis rather than exerting anabolic effects through stimulation of IGF-1. For instance, the levels of GH increased and those of IGF-1 decreased, indicating GH insensitivity, as reflected by a fall in IGF-binding protein (−3, but not 1 as in prior studies). Thus, the beneficial effect of intensive insulin treatment in the Van den Berghe et al study |1|
Glucose control in the critically ill
341
could not be explained on the basis of an increased activity of the somatotropic axis. Comment A decrease in IGF-1, an unexpected effect of intensive insulin treatment, did not exclude an anabolic effect of insulin at the cellular level through altered intracellular signalling (Table 14.2). Apart from the unchanged urea:creatinine ratio as an indirect measure of the catabolic state and protein balance, which was not improved by insulin, the authors did not supply any direct measure of protein synthesis (see also the papers by Jeschke et al. and Gore et al.). Hence, the mechanism of the effect of intensive insulin treatment in the critically ill remains unresolved by this study.
Contribution of circulating lipids to the improved outcome of critical illness by glycemic control with intensive insulin therapy Mesotten D, Swinnen JV, Vanderhoydonc F, Wouters PJ, Van den Berghe G. J Clin Endocrinol Metab 2004; 89:219–26 BACKGROUND. The mechanisms through which intensive insulin treatment beneficially affects the survival of the critically ill remain unclear. In the current paper the response of circulating lipids to the intensive insulin regimen was studied in part of the cohort (n=1548) of the Van den Berghe et al. study |1|. It involved 363 patients requiring intensive care for more than 7 days randomized to either intensive insulin or conventional therapy. INTERPRETATION. During the intensive insulin versus conventional treatment, low-and high-density lipoproteins increased, while serum triglycerides fell. Multiple regression analysis suggested that the lipid alterations, even more than glucose control and correction for risk factors and inflammation, were responsible for the beneficial effects of insulin on morbidity and mortality. The mRNA levels of the glucose transporter −4 were increased in the skeletal muscle of non-surviving patients in the intensive insulin group. Comment The authors suggested that intensive insulin treatment for maintaining normo-glycaemia increased skeletal muscle glucose uptake, but that alterations in lipids could contribute to the survival-enhancing effect. Although the authors gave some possible mechanisms for the latter, including endotoxin binding by lipoproteins, there was no direct confirmation available.
The year in anaesthesia and critical care
342
Section summary The optimal algorithm or directive for strict glucose control will depend on local conditions and educational levels. Both the optimal target values as well as the mechanisms by which insulin exerts beneficial actions in critically ill patients with insulin resistance need further clarification. For instance, the ability of exogenous insulin for inhibiting hepatic gluconeogenesis and increasing skeletal muscle glucose disposal is likely to be determined by the type, stage and severity of the underlying disease and the level of counter-regulating hormones. Insulin alternatives Since the treatment of insulin resistance by insulin carries the risk of hyperinsulinism and hypoglycaemia, alternative strategies for better control of hyperglycaemia are being explored.
Intravenous glucagon-like peptide 1 normalizes blood glucose after major surgery in patients with type 2 diabetes Meier JJ, Weyhe D, Michaely M, et al. Crit Care Med 2004; 32:848–51 BACKGROUND. In a preliminary study of eight patients with type 2 diabetes mellitus glucagon-like peptide 1 was infused after elective surgery in order to evaluate its metabolic effects. Glucagon-like peptide is an insulinotropic and glucagonostatic gut hormone. INTERPRETATION. Continuous infusion led to increases in the plasma insulin and decreases in the glucagon levels, while the glucose concentrations were lower, from a mean of 10 to 6 mmol/l. There were no instances of hypoglycaemia. Comment This preliminary study is worth citing since it documented that control of hyperglycaemia is possible even without insulin. Obviously, more investigation is needed into the cost-benefit ratio of this therapy and how it compares to that with insulin, as it potentially has effects beyond glucose control. The paper illustrated the efforts being made to treat insulin resistance more causally, rather than with insulin itself. Alternatively, the thiazolidinedione drugs may increase insulin sensitivity and may have anti-inflammatory actions. Further studies of these drugs in the critically ill can be expected.
Glucose control in the critically ill
343
Conclusion Since the landmark paper by Van den Berghe et al. |1|, interest in glucose metabolism in the critically ill has greatly increased. The sequelae of hyperglycaemia as well as the effects of insulin are being intensively studied and the beneficial mechanisms explored. This will help to elucidate fully the role of tight glucose control by insulin in the critically ill. References 1. Van den Berghe G, Wouters P, Weekers F, Verwaest C, Bruyninckx F, Schetz M, Vlasselaers D, Ferdinande P, Lauers P, Bouillon R. Intensive insulin treatment in critically ill patients. N Engl J Med 2001; 345:1359–67. 2. Groeneveld ABJ, Beishuizen A, Visser FC. Insulin: a wonder drug in the critically ill? Crit Care 2002; 6:102–5. 3. Van den Berghe G. How does blood glucose control with insulin save lives in intensive care? J Clin Invest 2004; 114:1187–95. 4. Ljungqvist O, Nygren J, Thorell A. Insulin resistance and elective surgery. Surgery 2000; 128:757–60. 5. Carlson GL. Insulin resistance in human sepsis: implications for the nutritional and meta-bolic care of the critically ill surgical patient. Ann R Coll Surg Engl 2004; 86:75–81. 6. Cely CM, Arora P, Quartin AA, Kett DH, Schein RMH. Relationship of baseline glucose homeostasis to hyperglycemia during medical critical illness. Chest 2004; 126:879–87. 7. Marik PE, Raghavan M. Stress-hyperglycemia, insulin and immunomodulation in sepsis. Intensive Care Med 2004; 30:748–56. 8. Leverve X. Hyperglycemia and oxidative stress: complex relationships with attractive prospects. Intensive Care Med 2003; 29:511–14. 9. Jeschke MG, Klein D, Bolder U, Einspanier R. Insulin attenuates the systemic inflammatory response in endotoxemic rats. Endocrinology 2004; 145:4084–93. 10. Soop M, Duxbury H, Agwunobi AO, Gibson JM, Hopkins SJ, Childs C, Cooper RG, Maycock P, Little RA, Carlson GL. Euglycemic hyperinsulinemia augments the cytokines and endocrine responses to endotoxin in humans. Am J Physiol Endocrinol Metab 2002; 282: E1276–85.
15 Non-invasive ventilation JONATHON TRUWIT Introduction Non-invasive ventilation (NIV) in the acute setting has now become routine therapy for patients with chronic obstructive lung disease and acute pulmonary oedema |1–3|. Some of the benefits associated with NIV over invasive mechanical ventilation (IMV) include reductions in mortality, length of stay, nosocomial infection and tracheostomy rates. However, the clinician must exercise good judgement in patient selection and stringent monitoring is needed, particularly during the initial hours of application. NIV has been used for patients with acute respiratory failure upon admission, during post-operative recovery, following extubation and as an adjunct to weaning from mechanical ventilation in the acute care setting |4, 5|. Successful outcomes have been documented in patients with chronic obstructive pulmonary disease (COPD) and associated acute exacerbations, acute cardiopulmonary oedema, immunosuppressed patients with bilateral infiltrates and those recovering from lung surgery |1|. Studies in patients with other disease states have either been too small or with too varied results to allow the recommendation of NIV. Non-invasive ventilation can be administered as negative pressure ventilation (NPV) or positive pressure ventilation (PPV). The former can be applied by an iron lung, poncho vest or curass ventilator. The latter can be applied via a full facemask, nasal mask or helmet. The advantages of NPV over traditional NPPV include improved patient comfort, less dependence on the patient’s level of consciousness for compliance, reduced skin breakdown and the removal of impediments to secretion clearance. This review will focus on studies where NIV is applied to patients in the acute setting (intensive care, ward or emergency department). The studies selected include patients with: COPD and acute exacerbations, acute pulmonary oedema, immunosuppression and acute hypoxic respiratory failure, respiratory distress post-extubation and do-notresuscitate orders. The discussion will also incorporate NPV and PPV as well as the use of a helmet, which is an emerging new appliance for the delivery of PPV. Most articles in this review chose common criteria for NIV exclusion as well as common criteria for endotracheal intubation after the initiation of NIV. These criteria are summarized in Tables 15.1 and 15.2. © Atlas Medical Publishing Ltd
Non-invasive ventilation
345
Table 15.1 NIV studies: common exclusion criteria Patient refusal Facial deformity sufficient for inhibiting an adequate mask to face seal Severe encephalopathy unrelated to hypoxaemia and/or hypercapnia Overt gastrointestinal bleeding Upper airway obstruction Hypotensive shock Acute cardiac ischaemia Life-threatening cardiac arrhythmia Tracheostomy Need for urgent intubation Cardiac arrest Respiratory arrest Prolonged respiratory pauses Psychomotor agitation requiring sedation Inability to clear secretions
Table 15.2 Common criteria for endotracheal intubation after NIV application Cardiac arrest or life-threatening cardiovascular alterations Respiratory arrest Prolonged respiratory pauses Psychomotor agitation requiring sedation Inability to clear secretions Failure to improve arterial blood sample values within 2 h Worsening gas exchange (at least two of following) pH values <7.30 and <0.02 from pre-NIV value Partial pressure of oxygen in arterial blood/fraction of inspired oxygen (PaO2/FiO2) values <150 and <20 from the pre-NIV value Partial pressure of carbon dioxide in arterial blood (PaCO2) value >20% and/or worsening level of consciousness
The year in anaesthesia and critical care
346
Non-invasive ventilation for patients with chronic obstructive pulmonary disease and acute respiratory failure The greatest successes with NIV have been in patients with COPD and associated acute exacerbations |1, 6|. It has become first-line therapy, as outlined by an international consensus conference |7|. The following papers addressed important questions. Can NIV be applied in patients whom most physicians would bypass with NIV and directly intubate and initiate mechanical ventilation? And do those failing NIV have worse outcomes due to the delay in intubating? Can comorbidities be used for predicting NIV success and long-term outcomes?
Non-invasive versus invasive ventilation in COPD patients with severe acute respiratory failure deemed to require ventilatory assistance Squadrone E, Frigerio P, Fogliati C, et al. Intensive Care Med 2004; 30: 1303–10 BACKGROUND. This study examined patients with COPD with severe acute hypercapnic respiratory failure (AHRF) as a result of an acute exacerbation or community-acquired pneumonia. Severe AHRF was defined as patients who, despite maximal medical therapy, had PaCO2 values >70 mmHg and pH values ≤7.25, a respiratory rate >35 breaths/min, severe dyspnoea and the use of accessory respiratory muscles. The outcomes were compared to a case-control group of patients admitted to the same wards in the previous 2 years who were intubated within 6 h of intensive care unit (ICU) admission and were matched to four criteria: age ±5 years, a simplified acute physiology score II of ±6, a forced expiratory volume in 1 s (FEV1) of ±5% and pH values prior to mechanical ventilation of ±0.02. Sixty-four patients with severe COPD and well-matched controls comprised the study populations. Severe respiratory failure was evidenced prior to NIV initiation by mean values of 7.18 for pH, 102 mmHg for PaCO2 and 43.5 mmHg for PaO2. The failure rate of NIV was 62.5%. Forty of the 64 patients required IMV. Gas exchange failed to improve with NIV in 15 patients and 25 did not tolerate it. The NIV group had a lower incidence of serious complications when measured in sum (P=0.012). When NIV was successful the duration of mechanical ventilation was significantly shorter (P<0.001). INTERPRETATION. This study demonstrated that NIV could be effective even in the most severe exacerbations of COPD with PaCO2 values of 100 mmHg. However, these patients should be cared for in the ICU setting given the high failure rate and, thus, the need for intubation. The failures were primarily due to a lack of improvement in gas exchange or patient intolerance to facemask ventilation. Comment Conti et al. demonstrated that NIV could be effective in patients with COPD and severe acute respiratory failure (mean PaCO2 value of 86 mmHg) in a small, randomized controlled trial of NIV versus IMV |8|. As with this study, the failure rate of NIV was
Non-invasive ventilation
347
high at 52% and no negative impact was associated with a delay in intubating, but the study was insufficiently powered. However, in a smaller study by Wood et al., a trend towards a higher mortality in the NIV failures was noted, questioning the impact of a delay in intubating |9|. Esteban et al. completed a large multicentre cohort study and found no difference in ICU mortality between those failing NIV and those receiving IMV as the first modality |10|. In order to assess whether a delay in intubation (mean 7.5 h) affected the outcome in the patients failing to respond to or tolerate NIV, the authors compared this subpopulation to the 64 case-control population. It is not clear why the comparison was not directly to the 40 case-control population. The NIV group did not have worse outcomes compared to the entire case-control population and, contrary to the entire NIV population, they did not have less serious complications as compared to the case-control population. Squadrone et al. studied 64 patients treated with NIV and had 40 failures, providing a larger group for analysis of NIV failures versus IMV, albeit case controlled. This permitted them to address the impact of delayed IMV, which did not prove to be harmful. This study also confirmed prior reports that NIV in COPD patients with severe AHRF is associated with a high failure rate, yet did not negate the benefit associated with a closely observed trial of NIV.
Co-morbidity and acute decompensations of COPD requiring non-invasive positive-pressure ventilation Scala R, Bartolucci S, Naldi M, Rossi M, Elliott MW. Intensive Care Med 2004; 30:1747–54 BACKGROUND. This study examined the impact of acute and chronic comorbidities on NIV success and the 6-month outcomes in those surviving. Scala et al., studied 120 consecutive patients with severe COPD and AHRF who failed to respond to medical therapy and met four criteria: (1) a pH value <7.33 or a PaCO2 value >55 mmHg, (2) a PaO2/FiO2 value <250, (3) dyspnoea at rest with a respiratory rate >25 breaths/min and (4) the use of accessory respiratory muscles or paradoxical abdominal breathing. Acute and chronic non-respiratory comorbidities were recorded. The Charlson comorbidity score was used for chronic comorbidities and acute non-respiratory comorbidities were defined using criteria established in a study by Moretti et al. |11|. Acute and chronic non-respiratory comorbidities were common, at 41.7 and 20%, respectively. Cardiac comorbidity accounted for 68% of acute non-respiratory comorbidities (34 out of 50), while chronic non-respiratory comorbidities were predominantly non-cardiac (58%). Patients with chronic and acute non-respiratory comorbidities were more likely to fail NIV: acute nonrespiratory comorbidities 33.3 versus 14.6% (P=0.043) and chronic non-respiratory comorbidities 32 versus 8.6% (P=0.001). The failure rates rose with the number of acute non-respiratory comorbidities: 8.6% for no comorbidities, 22.9% for one comorbidity and 53.3% for more than one comorbidity (P<0.001). The univariate and multivariate analyses are shown in Table 15.3. Forty-nine of the 100 patients were re admitted 80 times for COPD exacerbations with a mortality rate of 17 5%
The year in anaesthesia and critical care
348
(14 out of 80). The 6-month mortality was 35.3% (42 out of 119) and 23% of those surviving enrolment had a respiratory failure event (23 out of 100). The 6-month mortality of all patients enrolled was greater in those with chronic non-respiratory comorbidities, more than one acute non-respiratory comorbiditiy, lesser ability to perform activities of daily living (ADL) and non-cardiovascular chronic nonrespiratory comorbidities (Table 15.3). INTERPRETATION. The authors concluded that both chronic and acute nonrespiratory comorbidities impact on the success of NIV application as well as the 6month mortality. Cardiovascular comorbidity appears to be the most significant acute non-respiratory
Table 15.3 Variables associated with NIV failure and 6-month mortality NIV failure
Six-month mortality
Univariate
Univariate Multivariate
P
OR
P
P
OR
P
ADL
<0.001
–
–
<0.001 0.06
<0.001
APACHE III score
<0.001
–
–
0.022
–
–
Acute non-respiratory
0.001
8.34
0.003
–
–
–
More than one acute
–
–
–
0.007
6.30
0.046
non-respiratory comorbidity
–
–
–
–
–
–
Kelly-Matthay score
0.001
–
–
0.033
–
–
FEV1% pred
0.004
0.88
0.003
0.008
–
–
pH at 2 h NIV
0.013
–
–
–
–
–
Pneumonia
0.016
–
–
–
–
–
AHRF 2° cardiopulmonary
0.029
–
–
–
–
–
(acute hypercapnic respiratory
–
–
–
–
–
–
Chronic non-respiratory
0.043
–
–
–
–
–
Non-cardiovascular chronic
–
–
–
0.005
11.09 0.002
non-respiratory comorbidity
–
–
–
–
–
–
PaO2/FiO2 at 2 h NIV
0.045
–
–
–
–
–
BMI
–
–
–
0.032
–
–
Non-invasive ventilation
349
BMI
–
–
–
0.032
–
–
Home ventilation
–
–
–
0.008
–
–
APACHE, Acute Physiology and Chronic Health Evaluation; BMI, body mass index; OR, odds ratio. Source: Scala et al. (2004).
comorbidity, while the opposite is true for chronic comorbidities. This paper was the first to look at the impact of acute and chronic non-respiratory comorbidities on NIV failure and mortality. Moretti et al., did look at the impact of acute non-respiratory comorbidities on late NIV failure, despite an initial favourable response and found that acute nonrespiratory comorbidities, a lower pH and ADL were predictive of NIV failure |11|. Comment Patients with severe COPD commonly have comorbidities, both acute and chronic. Little is known of the impact of these comorbidities on the success of NIV during acute respiratory failure and the long-term follow-up of those treated with NIV. This study provided useful information, as knowledge of the impact of acute comorbidities may assist in triage decisions with regard to which patients treated with NIV should be considered for the ward versus ICU. Furthermore, long-term mortality data can be useful prognostic information for patient and family conferences.
Re-admission rates and life-threatening events in COPD survivors treated with non-invasive ventilation for acute hypercapnic respiratory failure Chu CM, Chan VL, Lin AW, Wong IW, Leung WS, Lai CK. Thorax 2004; 59: 1020–5 BACKGROUND. This study examined the long-term outcomes of successfully discharged COPD patients treated with NIV for AHRF without the need for IMV. AHRF was defined by having values of <7.35 for pH, >45 mmHg for PaCO2 and <60 mmHg for PaO2 from an arterial blood sample while breathing room air. Overall, 110 survivors out of 118 patients enrolled were available for long-term follow-up over 16 months. Fifty-two per cent of the patients used home oxygen and 32% had a do-not-intubate (DNI) status declared prior to NIV administration. The median follow-up was 403 days, during which 85 patients had at least one re-admission for a respiratory diagnosis. The median time to the first re-admission was 57 days. The median number of re-admissions in the year following discharge from the index admission was 2 and the range was 0–13. Nearly 80% of patients had an admission within 1 year. Multivariate analysis revealed that the ADL score and being hospitalized >21 days in the year preceding index hospitalization correlated with the risk of re-admission (P=0.018 and 0.016, respectively). Seventy-nine patients were re-admitted for a life-threatening event. The statistically significant factors favouring such an outcome on multivariate analysis were prior home oxygen use a
The year in anaesthesia and critical care
350
higher APACHE II score and a lower BMI. Sixty patients died in the follow-up period, 90% from respiratory causes and the 1-year mortality was 9.1%. Multivariate analysis revealed that a high Medical Research Council (MRC) dyspnoea score was the only independent risk factor associated with death (P<0.001). INTERPRETATION. This study demonstrated that patients with COPD and AHRF surviving a hospitalization in which NIV was used had high 1-year mortality and readmission rates. Furthermore, the likelihood of requiring another episode of mechanical ventilation was quite high. Comment Connors et al. studied 1016 patients admitted with COPD and AHRF with PaCO2 values >50 mmHg and reported a 1-year mortality rate of 43% |12|. In addition, 446 out of the 1016 (43%) patients were re-admitted within 6 months of the index hospitalization. While the 1-year mortality and morbidity were better than that seen in patients with advanced lung cancer, the data are still of concern and could be useful in decision making with regard to the therapies used during AHRF, code status and disposition planning. Non-invasive ventilation for patients with acute cardiopulmonary oedema Studies employing NIV in patients with cardiopulmonary oedema have been mostly positive, albeit with continuous positive airway pressure (CPAP) and not the bi-level (set inspiratory and expiratory airway pressures; IPAP and EPAP) modality |1|. While initial studies were predominantly performed in the ICU setting, the next two studies that are reviewed here examined the use of NIV in the emergency department. Defining benefit through reducing intubation rates in this population is often difficult as medical therapy for pulmonary oedema is usually effective and efficient. The third study identified the risk factors for intubation in patients presenting with pulmonary oedema and, hence, might allow for a target population for NIV intervention.
Non-invasive ventilation in cardiogenic pulmonary oedema: a multicenter randomized trial Nava S, Carbone G, DiBattista N, et al. Am J Respir Crit Care Med 2003; 168: 1432–7 BACKGROUND. This randomized controlled trial enrolled patients with acute respiratory failure from cardiopulmonary oedema, provided they had a PaO2/FiO2 value <250 despite breathing oxygen at 10 l/min, dyspnoea, a respiratory rate >30 breaths/min and typical signs of pulmonary oedema. The patients were randomized to standard medical therapy with supplemental oxygen, with and without NIV. The randomization schema was also stratified for a baseline PaCO value >45 mmHg
Non-invasive ventilation
351
One hundred and thirty patients were enrolled and 65 were randomized to standard therapy with NIV and 65 were randomized to standard therapy without NIV. The patient characteristics and laboratory values were similar in the two treatment arms as well as within those stratified by the PaCO2 values. The primary outcomes of intubation and mortality when evaluating all patients were not different between the treatment arms. However, when examining those with baseline PaCO2 values >45 mmHg, the application of NIV resulted in significantly fewer intubations (29 versus 15%) (P=0.015). A trend was also noted towards a reduced mortality. There were no greater occurrences of new acute myocardial infarctions in the NIV group. INTERPRETATION. This study demonstrated that NIV reduced intubation rates as compared to standard therapy in patients with hypercapnic cardiopulmonary oedema. Comment A study by Masip et al. also demonstrated reduced intubation rates with NIV. However, that study was too small for analysing those with hypercapnoea separately |13|. Improvements with NIV in the entire study population and those with PaCO2 values >45 mmHg were seen in arterial blood samples, vital signs and dyspnoea, but the mortality rates did not differ. Although Mehta et al. reported an increased myocardial infarction rate with the bi-level modality |14|, Nava et al.’s study did not find this, nor did a small study by Levitt |15|. Sharon et al. also found not only increased myocardial infarction rates but also increased mortality with NIV |16|. In Nava et al.’s study there was a trend for reduced mortality. The difference may be related to lower IPAP and EPAP levels.
Randomised controlled comparison of continuous positive airways pressure, bi-level non-invasive ventilation and standard treatment in emergency department patients with acute cardiogenic pulmonary oedema Crane SD, Elliott MW, Gilligan P, Richards K, Gray AJ. Emerg Med J 2004; 21: 155–61 BACKGROUND. This randomized controlled trial enrolled patients with acute cardiopulmonary oedema and consisted of three arms: standard medical and oxygen therapy alone or with one of two NIV modalities (CPAP or bi-level). The patients enrolled had a respiratory rate >23 breaths/min, radiographic evidence of pulmonary oedema and a pH value <7.35. The bi-level and CPAP modalities were applied at a mean pressure of 10 cmH2O and pressure settings of 15/5 and 10, respectively. Sixty patients were studied over a study period of 16 months, with an enrolment rate of 25%. The three groups were similar with regard to their patient demographic and physiological characteristics as well as doses of medications administered as part of the standard therapy. The treatment success and failure rates were not different between the groups. However, the survival to hospital discharge favoured those receiving CPAP plus standard therapy (100%) versus standard therapy (70%) and bi-level plus standard therapy (75%) (P=0.029). CPAP plus standard therapy proved superior to standard therapy and/or bi level plus
The year in anaesthesia and critical care
352
standard therapy. No difference in the incidence of myocardial infarctions was noted (for standard therapy, CPAP plus standard therapy and bi-level plus standard therapy there were six, three and nine patients, respectively). However, a trend towards higher creatine kinase values was noted in the bi-level plus standard therapy group. Also of note was that the survivors had higher systolic blood pressures and received a higher median total nitrate dose than non-survivors and were more likely to have received sublingual nitrates on route to hospital from ambulance personnel, regardless of the treatment arm to which they were assigned. INTERPRETATION. While this study was small, it was the largest randomized trial to evaluate the three therapies of standard therapy alone or in conjunction with CPAP or bi-level therapy. It was the first study to demonstrate survival benefit with the addition of CPAP therapy to standard therapy and survival benefit over bi-level therapy. Comment Prior studies have demonstrated that NIV reduces intubation rates, but not mortality and, as noted above, one study found increased mortality rates |13–19|. While increased rates of myocardial infarction were not observed with bi-level therapy, as seen in the studies by Mehta et al. and Sharon et al. |14, 16|, it is worth noting that a trends toward higher creatine kinase values was seen with bi-level therapy. Interestingly, acute physiological responses favoured bi-level therapy, yet the mortality data favoured CPAP therapy use.
Risk factors for intubation as a guide for non-invasive ventilation in patients with severe acute cardiogenic pulmonary edema Masip J, Paez J, Merino M, et al. Intensive Care Med 2003; 29:1921–8 BACKGROUND. This 5-year observational study addressed which patients with acute cardiopulmonary oedema are at risk for endotracheal intubation and, thus, are candidates for NIV. Patients presenting with acute cardiopulmonary oedema (physical signs and radiographic evidence) and a PaO2/FiO2 value <300 mmHg were evaluated. Thirty of an identified 110 patients received NIV. These 30, as compared to the other 80 patients, had significantly lower PaO2/FiO2 values and higher systolic blood pressures. The group treated with NIV also had a lower intubation rate (6.6 versus 26.3%) (P=0.034). Of the 80 patients treated conservatively, 21 were intubated and more likely to have hypertension and severe left ventricular dysfunction. Forty-six per cent of patients presenting with an acute myocardial infarction and cardiopulmonary oedema were intubated. Patients requiring intubation were more likely to have lower systolic blood pressures and PaO2/FiO2 and pH values and higher PaCO2 values on admission. INTERPRETATION. Multivariate analysis demonstrated that patients presenting with a pH value <7.25, acute myocardial infarction, hypercapnoea, systolic blood pressure <140 mmHg and a history of a severely reduced ejection fraction were more likely to be intubated However those presenting with blood pressures >180 mmHg were
Non-invasive ventilation
353
less likely to be intubated. The authors created an algorithm from this analysis (Fig. 15.1). Comment The authors provided an algorithm for identifying those patients with cardiopulmonary oedema likely to require intubation and, thus, candidates for NIV intervention (Fig. 15.1). Hence, NIV need not be applied to all patients with cardiopulmonary oedema. This study goes beyond that of Nava et al. (reviewed earlier) as it also identified hypo- or normocapnic patients that may benefit from NIV. In their study, by their algorithm NIV would be applied to 40–50% of all patients presenting with acute cardiopulmonary oedema. NIV would be provided to 80–90% of those with a high likelihood of requiring intubation.
Fig. 15.1 The algorithm: intervening with NIV in patients with acute cardiopulmonary oedema. Source: Masip et al. (2003).
The year in anaesthesia and critical care
354
Non-invasive ventilation and nosocomial infections If NIV is to reduce mortality by reducing intubation rates, it will probably also reduce the nosocomial infection rate, specifically ventilator-associated pneumonia. Earlier studies have indeed suggested this in both immunocompromised and non-immuno-compromised patients |20–25|. However, do these benefits translate into daily practice where NIV has been incorporated into standard therapy?
Secular trends in nosocomial infections and mortality associated with noninvasive ventilation in patients with exacerbation of COPD and pulmonary edema Girou E, Brun-Buisson C, Taille S, Lemaire F, Brochard L JAMA 2003; 290: 2985–91 BACKGROUND. This retrospective study examined patients admitted to an ICU for acute exacerbations of COPD or severe cardiopulmonary oedema and stratified them by their initial method of mechanical ventilation (NIV versus IMV). Overall, 521 out of 984 patients admitted for COPD with an acute exacerbation or severe cardiopulmonary oedema received NIV, IMV or both over a period of 8 years. Forty-two were excluded (24 with shock, 13 with coma and five had do-notresuscitate orders). Significant trends in increased NIV usage and reduced nosocomial infections and mortality were seen over the 8 years (Fig. 15.2). The NIV usage for these patients rose from 40% in the first 2 years to 85% in the latter 2 years and the nosocomial infection rates fell from 25–30 to 15–20%. Sixty-five per cent of the patients were treated with NIV throughout the study period with an intubation rate of 11% (35 patients). The nosocomial pneumonia rates fell from 20 to 8% for the 479 patients receiving ventilatory support (NIV and IMV). Significant decrements in mortality were seen over the same time frame: ICU mortality decreased from 21 to 7% and in-hospital mortality decreased from 24 to 11%. The risk of death was threefold lower with the application of NIV, but was even greater for patients with nosocomial infection and an increased simplified acute physiology score II. To address concerns that sicker patients would initially receive IMV, the authors developed a propensity score for NIV and analysed the population by quintiles. The application of NIV reduced the mortality in this analysis as well. INTERPRETATION. This article further strengthened the direct benefits of NIV in patients with acute exacerbations of COPD or severe cardiopulmonary oedema, at least during the initial hospitalization. It appeared that reducing nosocomial infections is a primary reason for this. Comment A reduction in nosocomial pneumonia with NIV has been demonstrated previously |20– 25|. This study also demonstrated that the incorporation of NIV into the standard care of patients with COPD and acute exacerbations or severe cardiopulmonary oedema is both achievable beyond an investigative trial with the associated resources and resulted in a
Non-invasive ventilation
355
reduction in nosocomial pneumonia and mortality. This reduction paralleled the increments in NIV usage (Fig. 15.2).
Fig. 15.2 Trends: reductions in nosocomial pneumonia and mortality parallel increases in NIV usage. Source: Girou et al. (2003).
The year in anaesthesia and critical care
356
Non-invasive ventilation for respiratory failure after extubation Non-invasive ventilation for patients developing respiratory failure after extubation has been studied. In patients with COPD or Cardiogenic Pulmonary Oedema, the application of NIV in this venue appears to be useful in preventing reintubation |26|. However, Keenan et al. did not find adjunctive therapy with NIV in those developing respiratory failure post-extubation to be useful. The study was limited by the small number of patients |27|. This set the stage for the larger multicentre trial described below.
Non-invasive positive-pressure ventilation for respiratory failure after extubation Esteban A, Frutos-Vivar F, Ferguson ND, et al. N Engl J Med 2004; 350: 2452–60 BACKGROUND. The application of NIV for facilitating weaning has been demonstrated in patients with COPD and cardiopulmonary oedema. The two studies examining the use of NIV for reducing re-intubation rates prior to Esteban et al., had divergent findings. Hilbert et al. |26| demonstrated a reduction in reintubation rates in patients with COPD in a case-control study. In contrast, Keenan et al, did not find benefit with the application of NIV in a randomized control trial. However, less than 10% of the patients in that study had COPD |27|. The patients enrolled in this randomized control trial had developed respiratory failure within 48 h of extubation. All had received mechanical ventilation for >48 h and successfully completed a spontaneous breathing trial. Common criteria for discontinuation of IMV were used across ICUs. Respiratory failure was defined by meeting two of the following criteria: respiratory acidosis (pH value <7.35 with PaCO2 value >45 mmHg), impending respiratory muscle fatigue as indicated by clinical examination, respiratory rate >25 breaths/min for two consecutive hours and hypoxia (oxygen saturation <90% or a PaO2 value <80 mmHg while the breathing fraction of oxygen exceeded 0.5. The patients were randomized to standard medical therapy with supplementary oxygen or NIV with standard therapy. Criteria similar to those listed in Table 15.2 were used for re-intubation. The primary outcome was mortality from all causes and the secondary outcomes were the re-intubation rate and ICU and length of stay. Nine hundred and eighty patients ventilated for >48 h were extubated and 244 developed respiratory failure within 48 h (25%). Two hundred and twentyone patients were randomized and 23 were excluded (10%) because of a reduced level of consciousness, increased respiratory effort, shock, hypoxaemia and upper airway obstruction. Of the 221, 114 received NIV with standard therapy and 107 received standard medical therapy alone. The mortality rates between the two groups were different and worse outcomes were seen in those given NIV (25 versus 14%) (P=0.048). The mortality differences between the groups appeared to be in those who were re-intubated (38 versus 22%) (P=0.06). Overall, the re-intubation rates did not differ between the NIV with standard therapy and standard therapy alone groups (48 versus 51%) (non-significant). However, the median time between the development of post extubation respiratory distress and re intubation was
Non-invasive ventilation
357
longer in the NIV with standard medical therapy group (12 versus 2.5 h) (P=0.02). INTERPRETATION. The authors concluded that NIV was not beneficial in preventing re-intubation in patients with respiratory failure as applied to a generalized ICU population. In fact, NIV may be harmful given the increased mortality rate seen in this trial. It was unclear whether, when limited to patients with COPD and cardiopulmonary oedema, the application of NIV in patients developing respiratory failure post-extubation will be useful. Too few of such patients were included in the study. However, it was clear that if NIV is used, close monitoring and a low threshold for intubation are needed. Comment In this study the delay to intubation appeared to be associated with increased mortality. This was contrary to the findings in COPD patients presenting with respiratory failure prior to intubation, as noted by Conti et al. |9| and Squadrone et al. (reviewed above). However, Wood et al. did suggest that a delay in intubation is associated with increased mortality |8|. Given the results of this large, randomized controlled trial, clinicians should exercise considerable caution when applying NIV to patients, particularly postextubation. As was true in this trial, improvements in arterial blood samples are generally seen within 2 h of the application of NIV |4|. Thus, it would be reasonable to re-evaluate patients after a period of NIV as short as 2 h in order to assess whether there is improvement and, if not seen, strongly advise re-intubation. Non-invasive ventilation in the do-not-intubate population A DNI order does not necessarily mean do not provide mechanical ventilation. Many patients choose NIV, but remain with a DNI order, even when the discomfort associated with NIV is conveyed. But how do we as clinicians predict hospital outcomes for such patients and what of the subsequent year after surviving an episode of acute respiratory failure treated with NIV?
Outcomes of patients with do-not-intubate orders treated with noninvasive ventilation Levy M, Tanios MA, Nelson D, et al. Crit Care Med 2004; 32:2002–7 BACKGROUND. This study was performed at four hospitals (two affiliated teaching hospitals and two community hospitals) and included patients having a declared DNI status. All patients were treated with bi-level NIV. The patients were classified by their primary diagnoses for respiratory failure, COPD, pneumonia, cardiopulmonary oedema and disseminated malignancy. Prior to initiation of NIV, respiratory therapists assessed the patient’s cough strength ability to protect their
The year in anaesthesia and critical care
358
airway and awake and agitation status. Overall, 1211 patients were treated with NIV and 114 (9.4%) had DNI orders over 10 months. COPD, cardiopulmonary oedema and pneumonia were the frequent diagnoses. Forty-nine of the DNI patients survived hospitalization (43%) and were predominantly those with COPD and cardiopulmonary oedema. Seventeen of the 49 patients were discharged home, 32 to a chronic care facility and one to an inpatient hospice unit. The patients’ survival correlated with the pre-NIV PaCO2 values, with higher survival more likely in patients with higher PaCO2 levels. Those with cough (strong or weak) and those awake (arousable and able to follow verbal commands) were also more likely to survive. INTERPRETATION. This study demonstrates that it is reasonable to provide NIV to patients who have a declared DNI status. Comment The predictors of survival from respiratory failure after NIV and which patients it is initiated in are affected by simple variables, which are easily assessed. They include a diagnosis of COPD or congestive heart failure (CHF), mental status, ability to cough and PaCO2 level.
Non-invasive ventilation in patients with acute hypercapnic exacerbation of chronic obstructive pulmonary disease who refused endotracheal intubation Chu CM, Chan VL, Wong IW, Leung WS, Lin AW, Cheung KF. Crit Care Med 2004; 32:372–7 BACKGROUND. This prospective study evaluated the long-term outcomes in patients with COPD having a declared DNI status after being admitted with respiratory failure and treated with NIV. These patients were compared to a cohort of COPD patients with full code status admitted during the same time frame to the same unit. The cohort group was intubated for reasons similar to those outlined in Table 15.2 (the first six criteria listed). The primary outcome was death and the secondary outcome was combined death or AHRF requiring NIV or intubation in the subsequent years of follow-up. Eighty of the 100 patients admitted to the NIV unit had COPD and 37, who proved to be older, had a declared DNI status. The median scores for dyspnoea, ADL and chronic comorbidities were significantly higher in the DNI group. Patients were followed for a mean of 551 days. The median survival for the DNI patients was 179 days and this was not reached in the cohort group. The 1-year survival was 29.7% in the DNI patients and 65.1% in the cohort group (P<0.0001). The median event-free survivals for the DNI and cohort patients were 102 and 292 days, respectively (P=0.0004). Overall, 83.3% of the DNI patients died or experienced a life-threatening event at 1 year, as did 53.5% in the cohort group (P=0.0004).
Non-invasive ventilation
359
INTERPRETATION. This study complemented the previous study by Levy et al. (reviewed above). It addressed the long-term outcomes in patients with COPD and a DNI status who, as in the Levy et al., paper, were found to have a higher than expected hospital survival rate. Comment In another study by Chu et al., the long-term follow-up of patients with COPD (DNI and non-DNI) revealed high 1-year re-admission and mortality rates: 79.9 and 63.3%, respectively (reviewed earlier). All this information may be useful in discussing the merits of NIV in patients with COPD and a DNI status. Alternatives to facemask ventilation The application of NIV is often impeded by patient intolerance secondary to the facemask or positive pressure applied. Alternatively, NPV could be used in the acute setting or non-invasive positive pressure ventilation (NPPV) can be delivered through a helmet device that encloses the neck and head. The following four papers examined the potential role of NPV and NPPV via a helmet. NPV can be applied with an iron lung, curass shell or poncho enclosing the patient and connected to a negative pressure generator.
Non-invasive negative and positive pressure ventilation in the treatment of acute or chronic respiratory failure Gorini M, Ginanni R, Villella G, Tozzi D, Augustynen A, Corrado A. Intensive Care Med 2004; 30:875–81 BACKGROUND. Consecutive patients admitted to a respiratory ICU with acute respiratory failure over a period of 18 months were assessed for enrolment. NIV was initiated in patients failing medical therapy with a pH value <7.32 and a PaCO2 value >60 mmHg or a PaO2 value <60 mmHg while the breathing fraction of oxygen equalled 0.5 and one of the following: dyspnoea, accessory muscle use or a respiratory rate >30 breaths/min. NPV was preferentially chosen for those with a reduced level of consciousness, copious airway secretions, facial deformities or severe respiratory acidosis (pH value <7.25). NPPV was selected as the initial mode of NIV for patients with a preserved level of consciousness, sleep apnoea syndrome or obesity. If the patients failed to respond to one modality after 4 h of treatment the other was applied, provided no urgent indication for endotracheal intubation was present. The criteria for endotracheal intubation are described in Table 15.2 (the first six criteria listed) and also included a failure to maintain an oxygen saturation >90% or improve respiratory acidosis (pH value <7.30 and below baseline value). Two hundred and fifty eight patients were enrolled 70% with a diagnosis of COPD
The year in anaesthesia and critical care
360
Other chronic respiratory illnesses included tuberculosis sequelae (8%), sleep apnoea-hypopnoea or obesity-hypoventilation syndromes (10%) and chest wall deformity (5%). The most frequent aetiologies of respiratory failure included COPD exacerbation (40%), pneumonia (37%) and heart failure (19%). Overall, 235 of the 258 patients required mechanical ventilation after a median of 4 h of medical and oxygen therapy. NPV was the first-line modality in 137 (58%) patients and NPPV was the first-line modality in 79 (34%) patients. Nineteen patients required IMV as the first modality, ten of whom were intubated in the emergency department. Arterial blood samples drawn at 1 and 4 h of NIV demonstrated significant improvements in both the NPV and NPPV groups in the values of pH, PaCO2 and PaO2/FiO2. Furthermore, patients treated sequentially with both modalities demonstrated significant improvements in their arterial blood samples between hour 4 of the first-line modality and hour 4 of the alternative modality, regardless of whether changing from NPV to NPPV or vice versa. NPV failed in 32 out of 137 (23.4%) patients, nine of whom met the criteria for endotracheal intubation. The reasons for switching from NPV to NPPV in the remaining 23 patients included intolerance of the iron lung (five patients), inability to maintain an oxygen saturation >90% (eight patients) and failure to improve respiratory acidosis (ten patients). Sixteen per cent of patients experienced upper airway obstruction while on NPV. The patients’ arterial blood samples improved with NPPV in all but three patients and, thus, the failure rate for NIV fell to 8.8%. NPPV failed in 20 out of 79 patients (25.3%), four of whom met the criteria for endotracheal intubation. NPV was provided to the remaining 16, but failed due to intolerance (two patients), inability to maintain an oxygen saturation >90% (one patient) and failure to improve respiratory acidosis (13 patients). Nineteen per cent of patients experienced large air leaks at the facemask-patient interface. The patients’ arterial blood samples improved with NPV in all patients and, thus, the NIV failure rate fell to 5%. INTERPRETATION. This paper by Gorini et al. pointed out that the dual modalities of NIV, when used sequentially, can be useful in preventing IMV. Both modalities had similar failure rates, but when used sequentially the overall failure rate of NIV fell from 24.1 to 7.4%. Comment As iron lungs may no longer be readily accessible, this study should be repeated with NPV as applied with a poncho or curass. If sequential NIV with NPV with a poncho and NPPV proves successful, it would be more easily adopted into clinical practice. This could provide for better tolerance and reduced intubation rates. The latter would result in reduced nosocomial pneumonias and mortality, as noted by Girou et al. (reviewed earlier).
Non-invasive ventilation
361
Treatment of acute exacerbations of chronic respiratory failure; integrated use of negative pressure ventilation and non-invasive positive pressure ventilation Todisco T, Baglioni S, Eslami A, et al. Chest 2004; 125:2217–23 BACKGROUND. Patients admitted over a period of 2 years with acute respiratory failure (pH value <7.30 and PaCO2 value >70 mmHg) and given nonIMV were evaluated. The exclusion criteria were similar to those listed in Table 15.1. All patients enrolled were to receive NPV unless the iron lung was contraindicated and then NPPV was provided preferentially. The iron lung contraindications included severe left heart failure, arrhythmias causing haemodynamic instability, upper airway obstruction, pneumothorax, obesity, obstructive sleep apnoea or patient refusal. Arterial blood samples were assessed after 2 h and patients with declining values were changed from NPV to NPPV or IMV and those receiving NPPV were intubated and given IMV. The criteria for intubation included respiratory arrest, respiratory pauses with a reduced level of consciousness, gasping, heart rate <50 beats/min with a reduced level of consciousness and haemodynamic instability and hypoxia (PaO2<45 mmHg). One hundred and fifty-two patients were included in the study: 72.4% had COPD. Ninety-seven patients received NPV and 52 NPPV as the first-choice NIV. Six patients promptly failed NPV and were switched to NPPV. The arterial blood samples had improved at the 2-hour juncture with the PaO2 value increasing by 15 mmHg and the PaCO2 value falling by 11 mmHg. However, 16 patients failed NIV early into treatment and required IMV (three NPV and 13 NPPV) (P<0.05). We do not know what portion of these 13 failures included the six patients that were changed from NPV to NPPV within the first 2 h of treatment. Over the ensuing days, treatment failed in seven out of 88 patients receiving NPV and five out of 48 patients receiving NPPV, with three deaths and nine requiring IMV. Ten of the 25 patients receiving IMV died, making a total of 13 deaths. The success rate of NIV (combined NPV and NPPV) as measured in the combined endpoint of avoiding intubation or death was 81.6%. Thirty patients were discharged home on NIV, while 27 on NPPV were discharged home. Eleven of these patients were from the NPV group and 19 from the NPPV group (P<0.001). INTERPRETATION. This study, as with that of Gorini et al. (reviewed above), demonstrated that NIV could be delivered by different modalities, which can be complementary to each other. The study was similar to Gorini et al., but was not designed to have NPV given sequentially after NPPV should patients be intolerant to or fail with NPPV. The authors pointed out that NPV might be useful in patients with hypercapnoeainduced coma, as eleven out of 13 responded well and were discharged alive. These patients may not have tolerated NPPV.
The year in anaesthesia and critical care
362
Comment Again, as the study used the iron lung for NPV, instead of a poncho vest or curass, concern about broad applicability remains. Non-invasive ventilation with a helmet The potential advantages for improved patient tolerance in delivering NPPV through a helmet include improved environmental interaction as the vision is not impeded, a reduced risk of facial skin breakdown and reduced facemask leakage at the patient interface. Such leakage results from either facial deformities or loose straps. The next two papers evaluated a latex-free transparent helmet enveloping the head and neck and resting upon the upper torso for applying NIV. The helmet is secured to the torso by armpit straps, which reduces leakage around the neck-shoulder-helmet interface. A common ICU ventilator is connected to the helmet and patients are ventilated with the head of the bed elevated to 45°. The helmet is produced in three different sizes based upon patient neck circumference.
Non-invasive positive pressure ventilation using a helmet in patients with acute exacerbation of chronic obstructive pulmonary disease: a feasibility study Antonelli M, Pennisi MA, Pelosi P, et al. Anesthesiology 2004; 100:16–24 BACKGROUND. This case-control series examined the feasibility of using the helmet for dyspnoeic patients with COPD and acute respiratory failure while in the ICU setting. Two out of three criteria needed to be present: respiratory rate >30 breaths/min, PaO2 value <45 mmHg and pH value <7.35. The exclusion criteria were similar to those listed in Table 15.1 and included the presence of two or more acute organ failures. The case-control subjects were selected from 198 patients admitted during the previous 12 months and treated with facemask NIV. They were matched as follows in priority ranking: PaCO2 values within 8 mmHg, pH values within 0.04 points, age within 10 years, simplified acute physiology scores II within four points and PaO2/FiO2 values while the breathing fraction of oxygen equalled 0.35 within 10 points. The initial NIV settings for the facemask group were similar to those used with the helmet patients. The same criteria for NIV discontinuation and intubation were applied to both groups. Thirty-three patients were enrolled. The subjects and controls were well matched for specified criteria as well as comorbidities, conditions exacerbating COPD and FEV1 values. Both groups had significant decreases in their PaCO2 values at 1 h, but larger reductions were seen in the facemask group (P=0.01). Similarly, at discontinuation of NIV, the reduction in the PaCO2 value was significantly greater in the facemask group. Both groups had an improved PaO2/FiO2 value to a similar extent. Greater levels of pressure support were applied in the helmet group (20±5 versus 15±7 cmH2O) and without major air leaks The failure rates were similar but the helmet failure rates were primarily a
Non-invasive ventilation
363
result of in efficacy and the facemask failure rates a combination of inefficacy and intolerance. Skin necrosis, gastric distension, conjunctivitis or intolerance were not seen in the helmet group but were present in 36% of the facemask patients (P<0.001). INTERPRETATION. The authors concluded that NIV could be applied via a helmet with similar success rates to that of facemask ventilation in patients with COPD and acute respiratory failure. Comment NIV delivered with a helmet could also be applied for longer periods, as the tolerance was considerably better. However, the improvements in ventilation were less, perhaps due to carbon dioxide re-breathing or less reduction of patient inspiratory effort.
Non-invasive ventilation by helmet or facemask in immunocompromised patients: a case-control study Rocco M, Dell’Utri D, Morelli A, et al. Chest 2004; 126:1508–15 BACKGROUND. Immunosuppressed patients admitted to the ICU with bilateral infiltrates on a radiograph, a PaO2/FiO2 value <150 while on oxygen, severe dyspnoea, a respiratory rate >30 breaths/min and the use of accessory respiratory muscles were evaluated. An immunosuppressed patient was defined as one with neutropenia, receiving immunosuppressive therapy or with acquired immune deficiency syndrome. The study’s case-control subjects were selected from patients admitted in the previous 3 years. They were matched in rank priority to the cause of immunosuppression, a PaO2/FiO2 value within 15 points, age within 10 years and simplified acute physiology score II of ±4 points. Common exclusion criteria, as listed in Table 15.1, were used and bleeding diathesis was included but not facial deformity. The criteria for intubation were similar to those listed in Table 15.2. However, the PaO2/FiO2 value <150 and less than 20 from pre-NIV value criteria was modified to a PaO2/FiO2 value <100. Nineteen patients were enrolled with 17 controls matching all four criteria and two matching three of four criteria. The baseline characteristics were similar. Both groups had similar percentages of patients with significant improvements in their PaO2/FiO2 values and these were of similar magnitude. A sustained improvement in the PaO2/FiO2 values was seen in 74% of the patients in the helmet group but only 37% of the patients in the facemask group (P=0.02). The PaCO2 values did not differ. Seven of the 19 patients in the helmet group and nine of the 19 patients in the facemask group required intubation (not significant). Patients treated with the helmet required less stoppages in the initial 24 h (1.2±1 versus 2.9±1) (P<0.001). Nine patients in the facemask group had nasal abrasions and two in the helmet groups had axillary abrasions (P=0.01). The trends towards a reduction in ICU and hospital mortality favoured the helmet group.
The year in anaesthesia and critical care
364
INTERPRETATION. This case-control study of helmet versus facemask NIV included only immunocompromised patients, another population where NIV appears to have substantial advantages over IMV. The authors concluded that NIV as delivered through a helmet was at least as effective as NIV delivered through a facemask. Comment Improved patient tolerance and less skin necrosis were reported in this study and, unlike the study by Antonelli et al. (reviewed above), the PaCO2 values did not differ with the mode of NIV. The application of NIV by means of a helmet should be considered in this population. Conclusion Non-IMV should now be standard initial therapy for the hypercapnic patient with COPD or cardiopulmonary oedema. While it does not require being initiated in the intensive care setting, the data support close monitoring capabilities and a low threshold for intubation. Defined end-points in respect to ventilation and oxygenation should not only be established for the failure of NIV, but also for its success. Improvements in gas exchange are generally seen within the first 2 h of NIV and certainly by 4 h. The studies reviewed here have expanded our knowledge of choosing which patients are appropriate for such therapy as well as enhancing our ability for providing prognostic information. Furthermore, we are given insights into use of mixed modalities of NIV and a relatively new patient-ventilator interface, the helmet. At this juncture I would recommend NIV for patients with COPD and AHRF, cardiopulmonary oedema, immunosuppression with bilateral infiltrates and after thoracic surgery (Fig. 15.3) |4|. With the exception of cardiopulmonary oedema, bi-level ventilation or NPV would be appropriate modalities. In cardiopulmonary oedema, CPAP should still be the preferred modality, as the data with bi-level ventilation are still confusing with regard to increased mortality and myocardial infarction. Furthermore, CPAP has a good track record. Medical centres applying NIV for their patients are best advised to do so in a controlled environment with close monitoring and experienced practitioners
Non-invasive ventilation
365
Fig. 15.3 Schema: an approach to applying NIV. Source: Truwit and Bernard (2004) |4|. .
The year in anaesthesia and critical care
366
References 1. Liesching T, Kwok H, Hill NS. Acute applications of non-invasive positive pressure ventilation. Chest 2003; 124:699–713. 2. British Thoracic Society Standards of Care Committee. Guidelines on Non-invasive ventilation in acute respiratory failure. Thorax 2002; 57:192–211. 3. Hess DR. The evidence for non-invasive positive-pressure ventilation in the care of patients in acute respiratory failure: a systematic review of the literature. Respirat Care 2004; 49:810–29. 4. Truwit JD, Bernard GR. Non-invasive ventilation—don’t push too hard. N Engl J Med 2004; 350:2512–15. 5. Sinuff T, Keenan SP. Clinical practice guideline for the use of non-invasive positive pressure ventilation in COPD patients with acute respiratory failure. J Crit Care 2004; 19: 82–91. 6. Ram FS, Lightowler JV, Wedzicha JA. Non-invasive positive pressure ventilation for treatment of respiratory failure due to exacerbations of chronic obstructive pulmonary disease. Cochrane Database Syst Rev 2003; (1): CD004104. 7. Organized jointly by the American Thoracic Society, the European Respiratory Society, the European Society of Intensive Care Medicine, and the Société de Réanimation de Langue Française, and approved by ATS Board of Directors, December 2000. International Con-sensus Conferences in Intensive Care Medicine: non-invasive positive pressure ventilation in acute respiratory failure. Am J Respir Crit Care Med 2001; 163:283–91. 8. Wood KA, Lewis L, Von Harz B, Kollef MH. The use of non-invasive positive pressure ventilation in the emergency department: results of a randomized clinical trial. Chest 1998; 113:1339–46. 9. Conti G, Antonelli M, Navalesi P, Rocco M, Bufi M, Spadetta G, Meduri GU. Non-invasive versus conventional mechanical ventilation in patients with chronic obstructive pulmonary disease after failure of medical treatment in the ward: a randomized trial. Intensive Care Med 2002; 28:1701–7. 10. Esteban A, Anzueto A, Frutos F, Alia I, Brochard L, Stewart TE, Benito S, Epstein SK, Apezteguia C, Nightingale P, Arroliga AC, Tobin MJ. Characteristics and outcomes in adult patients receiving mechanical ventilation: a 28-day international study. JAMA 2002; 287:345– 55. 11. Moretti M, Cilione C, Tampieri A, Fracchia C, Marchioni A, Nava S. Incidence and causes of non-invasive mechanical ventilation failure after initial success. Thorax 2000; 55: 819–25. 12. Connors AF, Dawson NV, Thomas C, Harrell FE, Desbiens N, Fulkerson WJ, Kussin P, Bellamy P, Goldman L, Knaus WA. Outcomes following acute exacerbation of severe chronic obstructive pulmonary disease. Am J Respirat Crit Care Med 1996; 154:959–67. 13. Masip J, Betbese AJ, Paez J, Vecilla F, Canizares R, Padro J, Paz MA, De Otero J, Ballus J. Non-invasive pressure support ventilation versus conventional oxygen therapy in acute cardiogenic pulmonary oedema: a randomised trial. Lancet 2000; 356:2126–32. 14. Mehta S, Jay GD, Woolard RH, Hipona RA, Connolly EM, Cimini DM, Drinkwine JH, Hill NS. Randomized, prospective trial of bilevel versus continuous positive airway pressure in acute pulmonary edema. Crit Care Med 1997; 25:620–8. 15. Levitt MA. A prospective, randomized trial of BiPAP in severe acute congestive heart failure. J Emerg Med 2001; 21:363–9. 16. Sharon A, Shpirer I, Kaluski E, Moshkovitz Y, Milovanov O, Polak R, Blatt A, Simovitz A, Shaham O, Faigenberg Z, Metzger M, Stav D, Yogev R, Golik A, Krakover R, Vered Z, Cotter G. High-dose intravenous isosorbide-dinitrate is safer and better than Bi-PAP ventilation combined with conventional treatment for severe pulmonary edema. J Am Coll Cardiol 2000; 36:832–7.
Non-invasive ventilation
367
17. Lin M, Chiang HT. The efficacy of early continuous positive airway pressure therapy in patients with acute cardiogenic pulmonary edema. J Formos Med Assoc 1991; 90:736–43. 18. Bersten AD, Holt AW, Vedig AE, Skowronski GA, Baggoley CJ. Treatment of severe cardiogenic pulmonary edema with continuous positive airway pressure delivered by face mask. N Engl J Med 1991; 325:1825–30. 19. Lin M, Yang YF, Chiang HT, Chang MS, Chiang BN, Cheitlin MD. Reappraisal of continuous positive airway pressure therapy in acute cardiogenic pulmonary edema: short-term results and long-term follow up. Chest 1995; 107:1379–86. 20. Girou E, Schortgen F, Delclaux C, Brun-Buisson C, Blot F, Lefort Y, Lemaire F, Brochard L. Association of non-invasive ventilation with nosocomial infections and survival in critically ill patients. JAMA 2000; 284:2361–7. 21. Guerin C, Girard R, Chemorin C, De Varax R, Fournier G. Facial mask non-invasive mechanical ventilation reduces the incidence of nosocomial pneumonia; a prospective epidemiological survey from a single ICU. Intensive Care Med 1997; 23:1024–32. 22. Antonelli M, Conti G, Rocco M, Bufi M, De Blasi RA, Vivino G, Gasparetto A, Meduri GU. A comparison of non-invasive positive-pressure ventilation and conventional mechanical ventilation in patients with respiratory failure. N Engl J Med 1998; 339: 429–35. 23. Nourdine K, Combes P, Carton M-J, Beuret P, Cannamela A, Ducreux JC. Does non-invasive ventilation reduce the ICU nosocomial infection risk? A prospective clinical survey. Intensive Care Med 1999; 25:567–73. 24. Carlucci A, Richard JC, Wysocki M, Lepage E, Brochard L; SRLF Collaborative Group on Mechanical Ventilation. Non-invasive versus conventional mechanical ventilation: an epidemiologic study. Am J Respirat Crit Care Med 2001; 63:874–80. 25. Hilbert G, Gruson D, Vargas F, Valentino R, Gbikpi-Benissan G, Dupon M, Reiffers J, Cardinaud JP. Non-invasive ventilation in immunosuppressed patients with pulmonary infiltrates, fever and acute respiratory failure. N Engl J Med 2001; 344:481–7. 26. Hilbert G, Gruson D, Portel L, Gbikpi-Benissan G, Cardinaud JP. Non-invasive pressure support ventilation in COPD patients with post-extubation hypercapnic respiratory insufficiency. Eur Respirat J 1998; 11:1349–53. 27. Keenan SP, Powers C, McCormack DG, Block G. Non-invasive positive-pressure ventilation for post-extubation respiratory distress: a randomized controlled trial. JAMA 2002; 287:3238– 44.
16 Brain death GILBERT PARK, BALACHANDRA MAIYA Introduction Death is often regarded as an event, but this is wrong: it is a process. Different parts of the body die at different rates when deprived of oxygen. The brain is the most sensitive and loss of consciousness from cerebral hypoxia and cardiac arrest occurs in seconds. Other organs may take hours to stop functioning, while connective tissue can take days to die. The sensitivity of the brain to hypoxia means that it can die while all of the other organs can survive. The ability to support organs and their functions, for example by ventilating a patient with a brain injury to see whether cerebral recovery is possible, has meant that modern medicine now interferes with the process of death. The brain is considered to be the person, something that physicians, healthcare workers, members of the clergy and lay people throughout the world have accepted and when the brain is dead the person is dead. However, if the vital functions are supported then the body is not cyanosed and cold: it remains warm, the chest moves with ventilation and the heart beats. The body may look alive, but the brain and the person are dead. Treatment is futile and this adds to the distress caused to the patient’s relatives and uses expensive intensive care unit (ICU) resources to no advantage. Brain death has been estimated to account for 1.2–4.3% of hospital deaths in Europe and the USA and 13% of all deaths in ICUs. In 1959 Mollaret and Goulon introduced the term irreversible coma (‘coma dépasse’) in 23 comatose patients with flat electroencephalographs (EEGs) |1|. This was followed in 1968 by a report of an Ad Hoc Committee of the Harvard Medical School who went on to define irreversible coma and death as unresponsiveness and the absence of movement, breathing and brainstem reflexes |2|. This committee made the important point that the cause of the coma needs to be identified. Three years later Mohandas and Chou |3| introduced the concept that damage to the brainstem was a critical component of severe brain damage. A further 5 years later the UK conference of the Academy of Medical Royal Colleges and their faculties published a statement on the diagnosis of brainstem death and this has been regularly updated |4|. It now forms the basis for the diagnosis of brainstem death in the UK. The diagnosis of death is one of the primary duties expected of the medical practitioner. However, there is no legal definition of death in the UK: the law accepts the © Atlas Medical Publishing Ltd
opinion of the suitably experienced, registered medical practitioner to say when a person is dead. ‘Conventional’ death is a skill usually delegated to junior doctors in hospital
Brain death
369
wards. Formal guidelines on how to diagnose this form of death are rare. Many medical undergraduate textbooks ignore this important role |5|. Patients with brainstem death are found in the ICU where more experienced doctors can usually be found, but they have needed guidance about brainstem death because of the complexities of modern day intensive care treatment. It is crucially important that there is no confusion in a doctor’s mind between a patient with a severe brain injury who has no hope of useful recovery and a patient whose brain is dead. The code of practice of the Academy of Medical Royal Colleges aims to clarify this. Furthermore, the diagnosis of brainstem death must be seen as part of the doctor’s responsibility to the patient in life. It is not a procedure undertaken merely to facilitate organ donation, but when brain death occurs then it is proper to consider whether any of the dead patient’s organs are suitable for transplantation into another patient. It also needs to be remembered that, if ventilation is continued despite the presence of brain death, then cardiac arrest eventually supervenes. In the UK and elsewhere death of the brainstem is taken as indicating loss of function of the whole brain. Diagnosis of brainstem death The UK code of practice for the diagnosis of brainstem death is based on clinical examination |4|. No confirmatory tests such as EEG recordings or angiography are performed, as in some other countries. The clinical diagnosis of brainstem death needs a series of careful neurological tests. However, because of the complexity of the condition and the treatment used for supporting the patient in life, the tests need to be performed by clinicians experienced in this area of medicine. Some might argue that only neurologists should make the decision. While they can perform the tests, it was felt that other disciplines are also suitably experienced, thus intensive care practitioners and neurosurgeons amongst others may also perform them. Confirmatory clinical testing is needed and a second suitably experienced doctor will usually perform this. One of the doctors must be of consultant status and the other one must have been registered for at least 5 years. No doctor performing the tests must have an immediate connection with transplantation. This would represent a conflict of interest. Preconditions There are certain preconditions that must be fulfilled before testing can take place. The cause of the patient’s coma must be known. This condition must be known to be irreversible and associated with brainstem death. If the cause is unknown then the tests cannot be performed. Similarly, if the cause is not associated with brainstem death then the patient should not be considered for testing. Common causes are shown in Table 16.1.
The year in anaesthesia and critical care
370
Table 16.1 Cause of death of patients for whom death was diagnosed following brainstem tests from the national Potential Donor Audit, 1 April 2003– 31 March 2004 (courtesy of United Kingdom Transplant) Cause of death Intracranial haemorrhage
n
%
826
59.60
Intracranial thrombosis
35
2.53
Brain tumour
33
2.38
Hypoxic brain damage: all causes
87
6.28
Intracranial: type unclassified
69
4.98
Trauma through road traffic accident: car
37
2.67
Trauma through road traffic accident: motorbike
12
0.87
Trauma through road traffic accident: pushbike
11
0.79
Trauma through road traffic accident: pedestrian
27
1.95
Trauma through road traffic accident: unknown type
4
0.29
Other trauma: suicide
3
0.22
Other trauma: accident
35
2.53
Other trauma: unknown cause
11
0.79
Cardiac arrest
18
1.30
Aneurysm
2
0.14
Congestive cardiac failure
1
0.07
Pulmonary embolism
1
0.07
Pneumonia
3
0.22
Respiratory failure
3
0.22
Carbon monoxide poisoning
1
0.07
Cancer other than brain tumour
4
0.29
Meningitis
29
2.09
Septicaemia
3
0.22
Renal failure
1
0.07
Multi-organ failure
2
0.14
Paracetamol overdose
1
0.07
Other drug overdose
10
0.72
Brain death
371
Not reported Other Unknown Total
1
0.07
114
8.23
2
0.14
1386
100.00
Source: Barber et al. (2005) (reviewed p 390).
A computed tomography (CT) scan is usually available. This may show brainstem herniation, multiple lesions with oedema or oedema alone. It needs to be remembered that these findings do not negate the need for a careful search for confounding factors. In addition, after some hypoxic injuries, meningitis or encephalitis a CT scan may be normal. The patient needs to be deeply unconscious with their breathing supported by mechanical ventilation. Clinicians must be certain that depressant drugs are not causing this. Opioids, hypnotics (such as midazolam) and tranquillizers may have a prolonged action, particularly if the patient is hypothermic or if renal or hepatic failure is present. Benzodiazepines are widely used in these patients for their anticonvulsant activity and also for facilitating synchronization with mechanical ventilation. They are also very cumulative. Part of this is because the enzyme that metabolizes many benzodiazepines (cytochrome P-450 3A4) is reduced in critical illness and is affected by many commonly used drugs. The metabolite of benzodiazepines may be active and accumulate in renal failure |6|. The drug chart must be carefully scrutinized and any possibility of intoxication being the cause or contributing to the patient’s comatose state should prevent a diagnosis of brainstem death being made. While antagonists exist for some drugs (opioids and benzodiazepines) there are none for drugs such as barbiturates. Plasma concentrations may be unreliable. Waiting is one answer, but it prolongs the process of dying and, if organ transplantation is being considered, may risk losing this valuable resource. Hypothermia will delay the elimination of drugs and is also a cause of unconsciousness that must have been excluded. Reversible endocrine, metabolic and other disturbances that may have caused or contributed to the coma must also be excluded. It is recognized that disturbances of this type are a likely accompaniment of brainstem death (for example hypernatremia associated with diabetes insipidus). However, these are the effects rather than the cause of that condition and do not preclude the diagnosis being made. Brainstem reflexes are absent 1. The pupils are dilated and do not respond to shining a bright light in the eyes. 2. Corneal stimulation produces no reflex. Care is needed when doing this test in order to exclude damage to the corneas if they are to be used for transplantation. 3. No eye movements are seen during or after the injection of at least 50 ml of ice-cold water over 1 min into each external auditory meatus. This shows that the vestibuloocular reflexes are absent. Clear access to the tympanic membrane must be established by direct inspection and the head should be flexed at 30°. If there is injury or disease
The year in anaesthesia and critical care
372
in one or both sides this test may not be possible. However, this does not affect other diagnosis of brainstem death. 4. There should be no motor responses within the cranial nerve distribution when a painful stimulus is applied to the hands or legs. It is important to note that movement of the arms and legs is sometimes seen when this test is performed. This is a simple spinal reflex and does not mean that the brainstem is not dead. Similarly, there should be no limb movement when supra-orbital pressure is applied. 5. There must be no gag reflex or other response when a catheter is passed down the trachea. 6. Apnoea is an essential part of the test. The patient should be disconnected from the ventilator and arterial blood gases measured before and after disconnection. No respiratory movement should be seen at a partial pressure of carbon dioxide oxygen in the arterial blood (PaCO2) of 6.65 kPa. Oxygenation may be maintained during this test by placing a suction catheter in the trachea through which oxygen is blown. Some patients have pre-existing chronic respiratory disease and may need supra-normal levels of carbon dioxide and depend on their hypoxic drive. Their management should be done in consultation with an expert in respiratory diseases. Children The British Paediatric Association and the Council of the Royal College of Physicians have suggested that the brainstem death criteria in children over 2 months old are the same as in adults. Below that age the diagnosis of brainstem death is particularly difficult. Repetition of testing Two sets of tests should always be performed. The two practitioners may carry these out separately or together. The interval between the tests is a matter of clinical judgement, but should be adequate for the reassurance of those directly concerned. The time of death is when the first set of tests are completed. Hypothermia This causes loss of brainstem reflexes and pupillary dilatation. The response to light is lost at core temperatures of between 28 and 32°C and brainstem reflexes disappear when the core temperature is below 28°C. All these are reversible on re-warming. Confirmatory tests In many countries confirmatory testing such as an EEG or a test of cerebral blood flow is needed. These are not needed in the UK. An EEG is widely used as a confirmatory test. Recordings should be obtained for at least 30 min. A sensitive machine is needed and unfortunately this is often interfered with in the ICU by the multiple electronic devices in use.
Brain death
373
Transcranial ultrasonography is a relatively sensitive test that may show the absence of flow in the middle cerebral and vertebral arteries. A bone window made during a craniotomy may interfere with the reliability of this test. Some tests need the patient to be taken out of the ICU. Nuclear imaging may show a lack of uptake of the tracer substance. Cerebral angiography may show that the intracranial arteries at the entrance to the skull do not fill because the intracranial pressure is too high to allow blood flow into the brain.
Brain death worldwide: accepted fact but no global consensus and diagnostic criteria Wijdics EFM. Neurology 2002; 58:20–5 BACKGROUND. The diagnosis of brainstem death and brain death varies around the world. This paper determined how this condition was diagnosed in 80 countries. A simple questionnaire was sent to clinicians in these countries who were identified through a review of the literature and personal contacts.
Table 16.2 Requirements for brain(stem) death testing in 80 nations Continent/Country Law Guideline Apnoea No. of Observation Confirmatory test physicians time (h) test North America USA
P
P
PaCO2
2*
6
Optional
Canada
P
P
PaCO2
1
6
Optional
Barbados
A
A
A
A
A
Optional
Cuba
A
P
PaCO2
2
6 (24)†
Optional
Jamaica
A
P
A
2
12
Mandatory
Trinidad and Tobago
A
P
PaCO2
2
A
Optional
Argentina
P
P
DVO
1
6
Mandatory
Brazil
P
P
DVO
1
6
Optional
Chile
P
P
DVO
2
A
Mandatory
Columbia
P
P
A
2(N)
A
Optional
Costa Rica
P
P
PaCO2
1
24
Optional
Caribbean
Central and South America
The year in anaesthesia and critical care
374
Ecuador
A
A
A
A
A
Optional
El Salvador
A
P
A
1
6(24)†
Mandatory
Guatemala
A
A
A
A
A
Not known
Honduras
A
A
A
A
A
Not known
Mexico
P
P
A
A
24
Mandatory
Paraguay
A
P
PaCO2
1
24
Optional
Uruguay
P
P
PaCO2
1
A
Optional
Venezuela
P
P
PaCO2
2(N)
12
Optional
Austria
P
P
DVO
1
12
Optional
Belgium
P
P
DVO
3
A
Optional
Cyprus
P
P
DVO
2
A
Optional
Czech Republic
A
P
PaCO2
2
A
Mandatory
Denmark
P
P
DVO
2
2(24)†
Optional
Estonia
P
P
PaCO2
1
12
Optional
Finland
P
P
DVO
1
A
Optional
France
P
P
PaCO2
2
A
Mandatory
Germany
P
P
PaCO2
2
12
Optional
Greece
P
P
DVO
3
6
Optional
Hungary
P
P
PaCO2
1
12(72)†
Mandatory
Ireland
P
P
PaCO2
2
A
Optional
Italy
P
P
PaCO2
1
6(24)†
Mandatory
Luxembourg
P
P
PaCO2
1
A
Mandatory
Malta
P
P
PaCO2
1
6
Optional
The Netherlands
P
P
PaCO2
1
A
Mandatory
Norway
P
P
DVO
A
24
Mandatory
Poland
P
P
DVO
1
3
Optional
Portugal
P
P
PaCO2
1
2–24
Optional
Romania
P
P
PaCO2
2
6
Mandatory
Russia
P
P
PaCO2
2
6(24)†
Optional
Slovak Republic
P
P
A
3
A
Mandatory
Spain
P
P
PaCO2
1
6(24)†
Optional
Europe
Brain death
375
Sweden
P
P
PaCO2
1
A
Mandatory
Switzerland
P
P
PaCO2
2
6(48)†
Optional
Turkey
P
P
PaCO2
4
A
Mandatory
Ukraine
A
P
DVO
A
A
Optional
UK
P
P
PaCO2
2
6
Optional
Yugoslavia
A
P
DVO
3
A
Mandatory
Egypt
A
A
A
A
A
Not known
Ghana
A
A
A
A
A
Not known
South Africa
P
P
PaCO2
2
A
Optional
Tanzania
A
P
A
1
A
Mandatory
Tunisia
P
P
DVO
1
A
Optional
Iran
A
P
A
3
12.24.36
Mandatory
Israel
P
P
PaCO2
1
6(24)†
Mandatory
Jordan
A
P
A
1
A
Mandatory
Lebanon
P
P
A
2
6
Mandatory
Oman
P
P
PaCO2
1
6
Optional
Qatar
P
P
PaCO2
1
A
Mandatory
Saudi Arabia
P
P
PaCO2
2
24†
Mandatory
Syria
A
A
A
A
A
Not known
United Arab Emirates
P
P
PaCO2
1
3
Optional
Armenia
A
A
A
A
A
Optional
Bangladesh
P
P
DVO
3
A
Optional
China
A
A
A
A
A
Optional
Georgia
P
P
DVO
3
24
Mandatory
Hong Kong†
P
P
PaCO2
1
A
Optional
India
P
P
DVO
4
A
Mandatory
Indonesia
A
P
PaCO2
3
24
Optional
Japan
P
P
PaCO2
1
A
Mandatory
Korea (South)
P
P
PaCO2
1
6
Optional
Africa
Middle East
Asia
The year in anaesthesia and critical care
376
Malaysia
P
P
PaCO2
2
12
Mandatory
Pakistan
A
A
A
A
A
Not known
Philippines
A
P
DVO
1
24
Optional
Singapore
P
P
PaCO2
2
A
Optional
Taiwan
A
P
PaCO2
1
6
Optional
Thailand
P
P
DVO
3
6
Optional
Vietnam
A
A
DVO
A
A
Optional
Australia
P
P
PaCO2
2
2
Optional
New Zealand
A
P
PaCO2
2
2
Optional
Oceania
* Eight US states only; the time within parentheses indicates the observation time required in conditions due to anoxia. † The observation time can be shortened or eliminated if one confirmatory test is positive for brain death. ‡ China resumed control of Hong Kong in 1997. PaCO2, target PaCO2 defined (50 or 60 mmHg); A, absent criterion or guideline; DVO, disconnection from ventilator only; Law, legal standard of organ donation; N, neurologist; MD, medical doctor; P, present criterion or guideline. Source: Wijdics (2002).
INTERPRETATION. Legal standards at present were given in 69% of the countries. Guidelines for helping clinicians diagnose this condition were available in 88% of these countries. More than one and up to four clinicians were needed for making this diagnosis in half the practice guidelines. All guidelines excluded confounders: the need for a diagnosis of irreversible coma, absent motor responses and absent brainstem reflexes as well as apnoea testing. Apnoea testing was recommended in almost 60% of the countries. Considerable differences existed in the time of observation and required expertise by the examining physicians (Table 16.2). Confirmatory testing was needed in 40% of the guidelines. Comment The diagnosis of brainstem death testing was clearly not uniform. However, should we expect such a complex and difficult diagnosis to be uniform throughout the world? The diagnosis of death has widespread social, ethical, moral and legal connotations. None of these are uniform within the world. Why should the diagnosis of brainstem death testing be the first to reverse this trend? One answer to this question might be that death is the universal final end-point for human beings, but brainstem death is more complex than ordinary death. Its diagnosis will depend on all the facilities and resources available and the ethical approach to brain death. However, protagonists of these tests will argue that it introduces a degree of certainty that clinical testing cannot do. There is absolutely no doubt that if blood does not go into the brain it is dead. However, a flat EEG could be due
Brain death
377
to drugs or even be an artefact. However, is the test really needed or is clinical diagnosis sufficient? Unfortunately, there is only a little information in the literature. One study compared the EEG, brainstem acoustic evoked potentials, Doppler ultrasonography and arterial digital subtraction angiography after the clinical diagnosis of brain death had been made. Although there was a lot of agreement between the tests, the authors went on to say that the angiography is needed if death of the whole brain is needed for diagnosing brain death |7|.
Brain stem death testing in the UK—time for reappraisal? Bell MDD, Moss E, Murphy PG. Br J Anaesth 2004; 92:633–40 BACKGROUND. In this study the authors sent a questionnaire to consultants working in neurocritical care units in order to determine whether the UK code of practice for brainstem death was being applied according to the recommendations. The questionnaire consisted of five sections and included details of the clinicians, how it was started, the criteria for testing, conduct of the tests and the process of organ donation. INTERPRETATION. The study revealed evidence of failure to apply the existing guidelines accurately. There was also considerable variation in practice where the recommendations were vague. In particular, difficulties were identified with the administration of glucose for manipulating the serum sodium, the apnoea test, the glucose concentration and temperature (Tables 16.3 and 16.4). The authors felt that the current code of practice should make more specific recommendations. Despite the code of practice not being accurately followed, this was unlikely to invalidate the time of death. Comment The failure to apply the code of practice properly is disturbing, but unlikely to result in a misdiagnosis of death in patients. Clinicians need to be aware that the public is sensitive to these issues, particularly with the decline in public confidence in the medical profession after Bristol, Liverpool and Dr Shipman. This could further undermine the profession’s standing. Discussion on this issue is worthwhile and may prompt an improvement in the code of practice. At the moment there are several vague areas. Whether one doctor, two doctors once or two doctors twice perform the tests is also confusing and explains the discrepancy. It could be helped with improved wording. Acute manipulation of values such as the serum sodium with glucose 5% is clearly valueless. The use of antagonists for excluding the effect of opioids or benzo-diazepines is a useful addition to the therapeutic armamentarium for clinicians. It should also be included in the next set of guidelines.
The year in anaesthesia and critical care
378
Table 16.3 Answers to questions on the preconditions for brainstem death tests and their timing Preconditions
Value
Preconditions
Value
Maximum plasma glucose (mmol/l−1)
Minimum temperature (°C) 34
6 8
35
4
60 10–12
>35 but <36 36.5
29
41* 15
36
6* 20
34
25 −1
Minimum plasma sodium (mmol/l )
10 −1
Maximum plasma sodium (mmol/l )
120
13 145
4
125
23 150
10
130
65 155
53
135
10 160
32
140
2 170
4
No limit
10
Timing (%) Immediately after clinical diagnosis
34
Wait 6 h
4
Wait 24 h
17
Differential delay based on pathology
56
Suggested delay after hypoxic brain damage (range in hours) 6–72 * Values outside the recommendations within current guidelines. Source: Bell et al. (2004).
Table 16.4 Answers to questions about the conduct of brainstem death tests Tests
%
One set of tests only
12
Tests carried out with colleague
50
Tests carried out independently
27
Tests carried out is either, depending on circumstances
23
Brain death
379
Tests carried out with delay between tests (range 20 min to 24 h)
73
Discussion with family between tests
15
Discussion with family after both tests
85
Apnoea test; simultaneous
52
Apnoea test: independently
35
Apnoea test: either
12
Maximum PaCO2 (kPa) (range 4.4–13) 4.5
15*
5–6
20*
6–6.5
23*
6.65
12
>7
29
Caloric testing Minimum volume of cold water (range 5–100 ml) <20 ml
40*
20–40 ml
20*
50 ml
40
Duration of infusion (range 10 s to 4 min) <20 s
50*
1 min
15
Gag reflex Tube manipulation
32*
Formal laryngoscopy
23
Wooden spatula/suction end
45
Peripheral pain stimulation Yes
65
No
35*
* Values outside the recommendations within current guidelines. Source: Bell et al. (2004).
The elimination of sedative drugs in critically ill patients is variable between patients and even within the same patient over time. Unlike general anaesthesia, with administration of a single dose or short infusion of a drug the serum concentration can be misleading when a drug is infused for days. Indeed, a report of two patients who became unconscious after an overdose of a benzodiazepine showed the plasma concentration was
The year in anaesthesia and critical care
380
greater than those associated with profound sedation when they were discharged awake 48 h later |8|. Other parts of this survey also show uncertainty in the minds of doctors during the tests. An intracranial pressure greater or equal to the mean arterial pressure must eventually result in brain death. There are some parts of the current code of practice that are unclear and despite this knowledge they have not been addressed. As an example, two doctors have to perform the tests, but there is a lack of clarity about whether these tests need to be performed once or twice by each doctor. Some clinicians argue that it is unnecessary to repeat tests you have performed once, particularly when confirmed by another doctor. Conversely, some prefer to confirm their own findings. If the code of practice is rewritten perhaps it should also recognize that there is variability in patients and clinical tests may benefit from help. The use of reversal agents such as flumazenil and naloxone as well as investigations such as cerebral angiography may be needed in some but not all patients. Whatever the outcome, it should be flexible so that the right tests can be performed in each patient.
Usefulness of venous oxygen saturation in the jugular bulb for the diagnosis of brain death: report of 118 patients Diaz-Reganon G, Minambres E, Holanda M, Gonzalez-Herrara S, Lopez-Espadas F, Garrido-Diaz C. Intensive Care Med 2002; 28:1724–8 BACKGROUND. Jugular bulb oximetry provides continuously available bedside information on the adequacy of cerebral perfusion. A jugular venous oxygen saturation approaching arterial oxygen saturation strongly suggests arrest of the cerebral blood flow. The investigators performed a prospective observational study in order to assess the usefulness of jugular bulb venous oxygen saturation as a complementary test for the diagnosis of brain death. The study was conducted in an acute care teaching hospital in Spain and included 118 patients who met the brainstem death criteria. The ratio (Cvj02) between the central venous saturation (SvO2) and jugular venous saturation (SjO2) was calculated. INTERPRETATION. The ratio was below unity at the time of brain death diagnosis in 114 patients (97%), whereas only four patients (3.4%) had a ratio above unity. However, amongst the 152 survivors, only one patient had a CvjO2 less than unity. The CvjO2 in a brain dead state has a sensitivity of 96.6% and a specificity of 99.3%. The study showed that jugular venous saturation is a useful complementary test for determining the arrest of cerebral flow in countries where it is mandatory to diagnose brain death. Comment This paper describes a technique widely used in neurosurgical ICUs and its value in predicting brainstem death. The theory is that the technique should be good at diagnosing this type of death. Unfortunately, the technique has a variety of pitfalls. It may not be
Brain death
381
reliable in patients with a haemangioma. More worrying is the fluctuation sometimes seen in the results produced. The authors did concede these points and indicated that this test should only been seen as a possible confirmatory test and not a test that can stand alone.
Detection of brain death onset using the bispectral index in severely comatose patients Vivien B, Paqueron X, Le Cosquer P, Langeron O, Coriat P, Riou B. Intensive Care Med 2002; 28:419–25 BACKGROUND. Bispectral index monitoring, which is derived from the EEG, has been used as a depth of anaesthesia and sedation monitor. Good correlation has been reported between the bispectral index and neurological status in unsedated comatose patients. In this study the accuracy of the bispectral index in the diagnosis of brain death was studied in 56 severely comatose patients (Glasgow coma score <5). The bispectral index was monitored throughout their stay in an ICU from the time of admission to time of death or discharge. Where necessary, brain death was confirmed by an EEG or cerebral angiography. INTERPRETATION. Twelve patients who were clinically brain dead on admission had bispectral index values of zero. An EEG or cerebral angiography was used for confirming brain death in these patients. The patients who were not clinically brain dead on admission (n=44) had bispectral index values of between 20 and 79. Twenty-seven of these patients eventually became brain dead and their bispectral index values fell to zero a few hours or days after admission. Seventeen patients who never became brain dead had values greater than 35. Two important limitations were noticed: (1) the bispectral index transiently decreased to zero in one patient who had an episode of acute intracranial hypertension and (2) in some clinically brain dead patients the bispectral index values increased from zero to 90 because of interference by muscle activity. Despite these limitations, the bispectral index has the potential to be a useful test for deciding on the timing of further confirmatory tests. Comment At first sight the bispectral index seems to be the method of solving all the problems associated with brainstem death. All patients with a bispectral index of zero were brainstem dead when two EEGs or a cerebral angiography were subsequently performed. However, closer reading of the paper shows some difficulties. First we are assured that no sedation remained, but there are no details of what was used and when or how it was excluded. Then there are the transient changes seen in the bispectral index: one of the zero values turned out to be wrong. High values (90 and above) are also seen in those already known to be brain dead. This was usually caused by electromyographic artefacts and could sometimes be stopped with a muscle relaxant. The authors postulated that excessive electromyographic activity is seen with brain death.
The year in anaesthesia and critical care
382
The bispectral index may well have a future in the diagnosis of brain death, but at present the most it can be used for is as a screening test for alerting clinicians to the development of brain death. This will possibly change in the future. It must be remembered that the bispectral index was not designed or intended for this purpose, but for the titration of anaesthesia and sedation. The algorithms it works from have been developed for these purposes. Perhaps if the algorithms can be adjusted for the diagnosis of brainstem death the artefacts that prevent its acceptance can be overcome. One fascinating aspect of this paper was the referral of brain dead potential organ donors to a specialist centre for retrieval of the organs. Although this may have some advantages over the system we operate where the retrieval team travel to the donor hospital, it may have potential disadvantages such as deterioration of the donor during travel. It would make an interesting area of study in the future.
A survey of patients admitted in severe coma: implications for brain death identification and organ donation Senouci K, Guerrini P, Diene E, et al. Intensive Care Med 2004; 30:38–44 BACKGROUND. This study looked at the incidence of brain death in Parisian patients with a Glasgow coma score <8. All patients admitted to 79 of the 108 ICUs in Paris who took part in this study and had a Glasgow coma score <8 were included. INTERPRETATION. In total 792 patients were identified over the 2 months of the study. One hundred and twenty of these patients became clinically brain dead (15.1%). These accounted for almost 12% of the deaths occurring in the ICUs and 3.3% of the deaths in the hospital during this period. The presence of the donor coordination team and the number of transplant coordinators were positively associated with the number of brain dead patients. Similarly, the number of patients transferred using a mobile emergency team was also associated with organ donation. The authors concluded that increasing the number of hospital coordinators in collaboration with the emergency teams would lead to the identification of more brain dead patients amongst comatose patients in Paris. Comment Large studies collecting detailed data present large organizational difficulties: this short study handled the data well by collecting the information prospectively rather than retrospectively. It validated the data by comparing them with other records to ensure that cases had not been added or removed. The study showed that those hospitals with a neurosurgical and a transplant unit had the highest rate of donation. This probably reflects the case mix of the hospital and the interest in this area. Interestingly, they did have a large number of patients who were considered medically unsuitable. These boundaries are constantly changing and now the only contraindication to the donation of an organ or tissue is the potential risk of transmission of a rapidly fatal disease that is incurable.
Brain death
383
The transfer of potential donors to a specialized ICU for maintenance until donation is a technique that is not practised in the UK, but also appears to be carried out in Greece (see above). It is worthy of further exploration in the UK and elsewhere to see if it offers any advantages over our current system.
Brain stem death; an organ donation 11 years on Park GR, Wilkins IN, Higgins T. Br J Anaesth 2003; 91:577–9 BACKGROUND. The authors studied patterns of organ donation in a major teaching hospital where there is a neurosurgical and transplant unit. The study was repeated 11 years later to see how patterns had changed. INTERPRETATION. All deaths in a 1-year period 11 years after the first study were investigated in order to identify potential heart-beating and non-heart-beating donors in the hospital’s ICUs. The actual organ donor’s reasons for not donating were identified. Overall, there was a significant reduction in the number of potential organ donors in the 11-year period. This was accompanied by an increase in the refusal rates by relatives by 10 to 29% and a decrease in the refusal rates by the coroner by 28 to 11% (Fig. 16.1). Comment In this hospital the number of potential and actual organ donors was reduced during the 11-year period for several reasons. There was a decreasing number of patients becoming brain dead. This may have been because treatment in the neurosurgical ICU in particular had improved dramatically over this time |9|. Improvements in road safety, road behaviour legislation and the design of motor vehicles as well as the wearing of protective headgear by cyclists |10| all decreased the number of people becoming brain injured. Disturbingly, there was an increase in the relative refusal rate and this has been shown in other studies throughout the UK. This reduction in the donation rate has been partially offset by the more liberal attitude of the coroner and an increase in the number of non-heart-beating donors.
The UK national potential donor audit Barber KM, Hussey JC, Bond ZC, Falvey SJ, Collett D, Rudge CJ. Transplant Proc 2005; in press BACKGROUND. This was the first national audit of all deaths performed in the UK since the one performed over 10 years ago by Gore et al. |11|. This one differed in that it was a review of the notes after the patient had died. The aim was to find out why so few patients who became brain dead did not become organ donors. This paper reported the results from the first year of the ongoing audit. INTERPRETATION The results from the first full year of the audit indicated that
The year in anaesthesia and critical care
384
the overall relative refusal rate for heart-beating solid organ donation was 41.5%. The age and gender of the potential heart-beating donor had little impact on the relative refusal rate, but relatives of ethnic minority groups were more than twice as likely to deny consent than those of white potential heart-beating donors. Of the 22 668 patient deaths, brainstem death was diagnosed in 1387 (6%) patients. Only seven patients had absolute medical
Fig. 16.1 Organ donation for 1 year in a teaching hospital. The numbers in italics in the box ‘Potential organ donors’ are the numbers from 11 years ago. NHBD non-heart-beating donor; BSD brainstem
Brain death
385
dead; TC, transplant coordinator. Source: Park et al. (2003). contraindications to heart-beating solid organ donation. Of the remaining 1380 potential heart-beating solid organ donors, there was either no record of discussion with the donors’ relatives or the relatives were known as not having been approached for consent in 198 (14%) of the potential donors. Heart-beating donation was discussed with the relatives of 1182 potential donors and consent for donation was given in 691 (58.5%) cases. Of these, 622 (90%) became heart-beating solid organ donors. Comment Despite the methodological differences between this and other studies, this large study confirmed that one of the biggest problems in organ donation and retrieval is a refusal by relatives. There may be many reasons for this. There may be a genuine desire not to donate, perhaps for fear of disfigurement of the body and these wishes, particularly if expressed by the patient in life, must be respected. There may have been complacency in thinking about end-of-life decisions and neither the organ donor card nor registration on the organ donor register may have been completed. In such circumstances the current practice is to ask the relatives. The UK has a system of ‘opting in’ for organ donation (unless specific consent is obtained, it is assumed that donation is not wanted), while other countries, such as Belgium, have an ‘opt out’ system (if you do not register your wish not to donate, it is assumed organ donation is your wish) |12|. Whether changing the law in order to adopt an opting out policy would work in the UK is unknown.
Analysis of terminal events in 109 successive deaths in a Belgian intensive care unit Gajewska K, Schroede M, De Francoise M, Vincent JL Intensive Care Med 2004; 30:1224–7 BACKGROUND. The authors conducted a study for determining the incidence of end-of-life decisions in intensive care patients. Data were prospectively collected from 109 successive deaths over a period of 3 months in a combined 31-bed medicosurgical ICU. The ICU staff were invited to complete a questionnaire indicating the mode of death, whether information was given to the family and whether they were present at the time of death. If a decision to withdraw was made, data were then collected on the type of life support modalities withdrawn, the use of sedatives and their thoughts on the timing of the decision. INTERPRETATION. A decision to forego life support treatment was made in 50 patients: withdrawal in 43 patients and withholding in seven patients. Questionnaires were completed by the attendants in 34 of the 50 patients in whom a decision to forego
The year in anaesthesia and critical care
386
life support was made. The patients for whom the decision to withdraw/withhold support was made were older (65±19 versus 58±20 years) (P=0.05), but the differences in the length of ICU stay were not statistically different. According to the staff questionnaire, the decision to withdraw/withhold was timely in 28 (82%) patients, too late in five (15%) and too early in one (3%) patient (where the family did not have the time to reach the hospital). The study showed that therapeutic limitations are frequent in the ICU with withdrawing more common than withholding support. The staff in this survey felt that appropriate end-of-life decisions were made most of the times. Comment This study looked at end-of-life decisions in one large Belgium ICU. Brain death was one of the end-of-life decisions that were made. Unfortunately, we are not given much detail about these particular cases. In the UK the relevance of this paper is the importance of every member of the team, as well as the relatives, being involved in the decision-making process. Despite this involvement, there still has to be one person who makes a decision and in this context the decisions are medical, putting the responsibility with the doctors. We were therefore somewhat surprised to see a significant number of decisions to withdraw being made by the nurses. However, if the number of donors is to increase, then the whole team needs to be aware of the potential for and benefits of donation. If they are not, then the approach to the relatives will not be made with conviction and commitment, thereby resulting in an increased risk of refusal.
Identifying the potential organ donor: an audit of hospital deaths Opdam HI, Silvester W. Intensive Care Med 2004; 30:1390–7 BACKGROUND. Australia has a lower organ donation rate (9.8 per million population) compared to Spain (32.1 per million population), the USA (24 per million population), France (16.6 per million population) and the UK (13.5 per million population). The authors conducted a prospective medical record audit in order to quantify the potential for organ donation and to identify missed opportunities for organ donation in the Victorian state of Australia. Data on deaths, total potential donors, organ donors and outcomes of requests for donation were collected over a period of 12 months by a medical record review. Patients in whom brain death occurred or was likely to occur and in whom donation was not requested (unrealized potential donors) were classified by an independent panel consisting of five ICU specialists who were blinded to the identity of the patients, treating clinicians and hospitals. INTERPRETATION. The independent panel reviewed the data and classified each case into the following categories: (1) confirmed brain death, (2) likely to progress to brain death within 24 h, (3) likely to progress to brain death in >24 but <72 h and (4) not likely to progress to brain death or medically unsuitable for donation. Categories (1)-(3) were considered to be unrealized potential donors The audit findings are shown in Fig
Brain death
387
16.2. The rates of organ donation and unrealized donors were determined as a proportion of the total potential donors and hospital deaths and the maximum potential donor rate was estimated. The maximum potential donor rate was estimated to be 30 per million population. The authors concluded that an increase in the organ donation rate may be possible through increasing consent and the identification and support of potential donors. Comment The object of this study was to find out why the donation rate was so low. Although the authors talked about ways of improving rates, one of the unanswered questions
Fig. 16.2 Organ donation in 1 year in Australian ICUs. Source: Opdam and Silvester (2004). (for us anyway) was the influence of the use of confirmatory testing. Thirteen out of 24 donors could have been tested for brain death but were not. Was this due to not wanting to wait, not wanting to do the tests or not wanting to move the potential donor to the imaging facility?
The year in anaesthesia and critical care
388
The criteria for donation are also stricter than in many other countries. The age range for donation is between 1 and 75 years, which is different from many countries that do not define an age range at all. The final interesting feature of this study was a consent rate of 59%, comparable to many other studies around the world. This raised the question of why do 40–50% of relatives refuse? Although in the UK we blame medical scandals, are all countries suffering them? Alternatively, are doctors and other healthcare workers just not as trusted as they used to be? Maybe the increasingly knowledgeable and sophisticated populations of these countries have a proportion of people that just do not want to donate organs?
Clinical course and determination of brainstem death in a children’s hospital Goh AY-T, Mok Q. Acta Paediatr 2004; 93:47–52 BACKGROUND. The authors hypothesized that, despite a widespread dissemination of guidelines about brain death by the Royal College of Paediatrics and Child Health, there are variations in the current practices that are used for declaring brainstem death. A retrospective chart review of all patients admitted with brainstem death between January 1995 and December 1998 was carried out in a large paediatric ICU in London, UK. The data included the cause, clinical features and time intervals between triggering events, examination for death, pronouncement of death and organ procurement for transplantation. The authors also determined the extent of variability of and conformity to accepted brainstem death guidelines. The guidelines were applicable to children aged >2 months who were comatose, apnoeic and on a ventilator. INTERPRETATION. There were a total of 31 children, three of whom were under 2 months old so the brainstem death guidelines could not be applied. The average time between the insult and the first neurological examination consistent with brainstem death was 27.9±75.2 h and the time from that to confirmation of brainstem death was 25.5±20.3 h. Fifteen patients had two sets of brainstem death testing whilst in eight children there was spontaneous cardiac arrest after the first set. Sixteen (52%) had ancillary tests, with an EEG in 14 patients, cerebral blood flow measurement in three and evoked potential recordings in one. The rate of organ procurement was less than 10%. Though the intensivists in this study followed accepted guidelines, there was wide use of ancillary tests for aiding the diagnosis. Comment Brain death in children is poorly studied, probably because of the relative rarity of this condition in this age group. In this retrospective study over 4 years the commonest cause was trauma, with death caused by a central nervous system infection being relatively rare. This is unlike emerging countries, where the reverse is true. Although in two-thirds of those declared brain dead treatment was withdrawn, in the remaining ones treatment was not withdrawn because of parental dissent for various
Brain death
389
reasons and asystole followed in approximately 2 days. This adds to the data, confirming that asystole follows soon after brain death is diagnosed |13|. In the very young the diagnosis of brainstem death can be both difficult and emotionally traumatic. The use of confirmatory tests in this situation can be helpful. As with other studies, the degree to which the codes and guidelines were followed was variable, particularly with respect to the number of doctors performing the tests. This may have been because there is no requirement to perform the tests if organ donation is not to follow. Indeed, only six families were approached about donation and, of those, only three agreed, similar to many other studies.
High prevalence of decreased cortisol reserve in brain-dead potential organ donors Dimopoulou I, Tsagarakis S, Anthi A, et al. Crit Care Med 2003; 31: 1113–17 BACKGROUND. Cortisol secretion normally increases after surgical stimulation. However, we do not know the ability of the brain dead patient to respond to a surgical stimulus. This was a prospective study for assessing adrenocortical function in brain dead patients. A total of 37 patients with severe brain injury admitted to this ICU were divided into two groups. Group A (n=20) consisted of brain-injured patients who did not deteriorate to brain death. Group B (n=17) consisted of patients who were either brain dead on admission or became brain dead during their ICU stay. Baseline cortisol was measured on admission: subsequently, after confirmation of brain death, cortisol was measured after 30 min of administration of a low dose of corticotrophin (1 µg). INTERPRETATION. After brain death the group B patients had a significantly lower baseline (8.5±6.2 versus 17±6.6 µg/dl) (P<0.001) and stimulated (16.9±6.3 versus 23.9±5.7 µg/dl) (P=0.001) plasma cortisol compared to group A patients. The baseline and stimulated cortisol levels were related in group B patients. This study highlighted the state of relative adrenocortical deficiency in a substantial portion of brain dead patients. Though previous studies did not show a difference in the inotropic requirements with steroid replacement in brain dead patients, it may improve the viability of the organs. Comment There is considerable variability in the changes in cortisol secretion seen after brainstem death develops. The investigators only used clinical brainstem death testing for diagnosing death. This variability may be related to whether the whole brain or just the brainstem has died. Unfortunately, we are not given any information about anterior or posterior pituitary function (such as the development of diabetes insipidus, for example) in this study. The major importance of this study is the finding that, after brain death, not only does adrenocorticotrophic hormone stimulation change, but cortisol secretion by the adrenal gland also changes. The authors did not know why this occurs, but there are several possibilities. It may be artefact. The patients who died were older and must have been
The year in anaesthesia and critical care
390
much sicker than those who survived. In such situations, blood concentrations of proteins may change. Cortisol is bound to several proteins and the concentrations of these may change |14|. The total (protein bound and free) cortisol may change, but the free (active) portion remains unchanged. Other possibilities include infarcts in the adrenal gland caused by the haemodynamic changes associated with brain death, some change in metabolic regulation (such as has recently been described in animal studies of cerebral inflammation |15|) or the lack of another hormone. Does the lack of cortisol matter? Unfortunately, the paper does not really help with the answer to this question. There was no difference between the use of inotropes between those with a poorly responsive adrenal gland and those with a normal response. In sepsis, a trial specifically designed to look at this point measured free cortisol and showed that giving steroids to patients with a low cortisol concentration (<25 µg/dl) did result in a reduction in inotrope use |16|. Future studies may help elucidate which patients benefit and which do not. One very small point about this paper was the admission criteria. Some of the patients were admitted with a diagnosis of clinical brainstem death already. This differs significantly from UK practice and is more in line with French practice. We wonder if they were admitted for donor maintenance before donation? Conclusion The diagnosis of brain(stem) death is important, firstly, to avoid further distress to the relatives and the nursing and medical attendants and, secondly, it will avoid futile and wasteful use of resources that in most countries can be used for benefiting others. Finally, the diagnosis allows organ donation. No current test is perfect. Clinical examination remains the cornerstone of diagnosis in the UK. Elsewhere in the world other tests are used for confirming clinical examination. Ideally, a small black box connected noninvasively to the patient that is 100% reliable and accurate is what is needed. For the moment that is science fiction, but we are fast approaching the time when it is real. The jugular bulb SvO2 may go some way towards that but is not yet perfect. Similarly, the bispectral index almost but not quite approaches what is needed. Brain (stem) death is a difficult area in medicine, often resulting in emotional discussion particularly when organ transplantation is involved. Some of the papers published recently will fuel these concerns. Are we performing the tests properly and recording that they have been performed? The evidence says we are not doing so and have not been doing so for some time |17|. The public expects that of us and we should accede. Organ donation rates have decreased over the years. Universally, relative refusal has been the most important cause. In the UK, the damage done by various medical scandals is not helping. United Kingdom Transplant has of late reversed the previous decline in donation rates, mostly by increasing non-heart-beating donation. Whether other initiatives such as presumed consent, which is successful elsewhere, would work in the UK is uncertain. What must not happen is that there is wastage of this precious resource if donation is at all possible.
Brain death
391
There also seems to be a trend in some countries for admitting potential organ donors to specialist ICUs for donor maintenance before donation. While this risks loss or deterioration of the donor during transfer, it does have some advantages. Familiarity with the care of an organ donor can only be obtained in centres that do it frequently. Most ICUs in this country only manage a few donors a year and may not be so used to the specific problems they pose. This might allow unsuitable donors to be converted into suitable ones |18|. Since the specialized donor ICU will probably be in the same hospital as the recipient, the donor and recipient operations can be timed for minimizing the ischaemic time for the various organs. There is no doubt that organ transplantation has the potential to help restore ill patients to health. It is at the cutting edge of our medical knowledge today. One thought always makes us smile: 100 years from now our successors will look back on what we are doing today and probably chuckle. To them, organ transplantation will be like the application of leeches to cure all ills for us. But then leeches have just come back into use for plastic surgery! Commonality of Interest statement Gilbert R Park is a non-executive director of the United Kingdom Transplant Special Health Authority. His salary from this post is paid to Addenbrooke’s Hospital. References 1. Mollaret P, Goulon M. Le coma dépassé (mémoire préliminaire). Rev Neurol 1959; 101: 3–5. 2. Harvard Medical School. A definition of irreversible coma; report of the Ad Hoc Committee of the Harvard Medical School to examine the definition of brain death. JAMA 1968; 205:337–40. 3. Mohandas A, Chou SN. Brain death: a clinical and pathological study. J Neurosurg 1971; 35:211–18. 4. Academy of Medical Royal Colleges. A Code of Practice for the Diagnosis of Brain Stem Death. 1998 see http://www.dh.gov.uk/PublicationsAndStatistics/Publications/PublicationsPolicyAndGuidance/ PublicationsPolicyAndGuidanceArticle/fs/en?CONTENT_ID= 4009696&chk=cWPWv9 5. Park GR. Death and its diagnosis by doctors. Br J Anaesth 2004; 92:625–8. 6. Park GR. Molecular mechanisms of drug metabolism in the critically ill. Br J Anaesth 1996; 77:32–49. 7. Nau R, Prange HW, Klingelhofer J, Kukowski B, Sander D, Tchorsch R, Rittmeyer K. Results of four technical investigations in fifty clinically brain dead patients. Intensive Care Med 1992; 18:82–8. 8. Greenblatt DJ, Woo E, Allen MD, Orsulak PJ, Schader RL Rapid recovery from massive diazepam overdose. JAMA 1977; 240:1872–4. 9. Patel HC, Menon DK, Tebbs S, Hawker R, Hutchinson PJ, Kirkpatrick PJ. Specialist neurocritical care and outcome from head injury. Intensive Care Med 2002; 28:547–53. 10. Cook A, Sheikh A. Cycle injury trends: helmets are most likely explanation. BMJ 2001; 322:1427. 11. Gore S, Cable DJ, Holland AJ. Organ donation from intensive care units in England and Wales: two year confidential audit of deaths in intensive care. BMJ 1992; 304:349–55.
The year in anaesthesia and critical care
392
12. Michielsen P. Presumed consent to organ donation: 10 years’ experience in Belgium. J R Soc Med 1996; 89:663–6. 13. Harding JW, Chatterton BE. Outcomes of patients referred for confirmation of brain death by 99m Tc-exametazime scintigraphy. Intensive Care Med 2003; 29:539–43. 14. Hamrahian AH, Oseni TS, Arafah BM. Measurements of serum free cortisol in critically ill patients. N Engl J Med 2004; 350:1629–38. 15. Renton KW, Nicholson TE. Hepatic and central nervous system cytochrome P450 are downregulated during lipopolysaccharide-evoked localized inflammation in brain. J Pharmacol Exp Ther 2000; 294:524–30. 16. Marik PE, Zaloga GP. Adrenal insufficiency during septic shock. Crit Care Med 2003; 31: 141–5. 17. Kafrawy U, Stewart D. An evaluation of brainstem death documentation: the importance of full documentation. Paediatr Anaesth 2004; 14:584–8. 18. Wheeldon DR, Potter CD, Oduro A, Wallwork J, Large SR. Transforming the ‘unacceptable’ donor: outcomes from the adoption of a standardized donor management technique. J Heart Lung Transplant 1995; 14:734–42.
List of abbreviations ACC
American College of Cardiology
ACE
angiotension-converting enzyme
ADL
activities of daily living
AEP
auditory evoked potential
AHA
American Heart Association
AHRF
acute hypercapnic respiratory failure
AHVR
acute hypoxic ventilatory response
AMD
airway management device
AMI
acute myocardial infarction
ANH
acute normovolaemic haemodilution
APACHE
Acute Physiology and Chronic Health Evaluation
ARX
autoregressive exogenous input
ASA
American Society of Anesthiologists
ATP
adenosine triphosphate
BiPAP
bilevel positive airway pressure ventilation
BMI
body mass index
C25
concentration causing a 25% increase in current
2+
CA
voltage-dependent calcium
CABG
coronary artery bypass graft/grafting
CAD
coronary artery disease
CcO2
capillary oxygen content
CePROP
effect site concentrations of propofol
CHF
congestive heart failure
CI
confidence interval
cLMA
classic laryngeal mask airway
Cmax
maximum concentration
CNS
central nervous system
CO2
carbon dioxide
COP
colloid osmotic pressure
COPD
chronic obstructive pulmonary disease
COX
cyclooxygenase
CPAP
continuous positive airway pressure
List of abbreviations
395
CPB
cardiopulmonary bypass
CRP
C-reactive protein
CS
cell salvage
CT
computed tomography
CVC
central venous catheter
CVD
cardiovascular disease
CvjO2
ratio between the central venous saturation and jugular saturation
CvO2
venous oxygen content
CVP
central venous pressure
DH
depth of hypnosis
DHEA
dehydroepiandrosterone
DHel-ind
electronic index of the depth of hypnosis
DHobs
observed depth of hypnosis
DHreal
real depth of hypnosis
DNI
do not intubate
ECG
electrocardiogram
ED50/95
dose associated with 50% and 95% effect
EEG
electroencephalograph
Emax
maximal effect
EMG
electromyograph/ electromyography
EPAP
expiratory positive airway pressure
FDA
Food and Drug Administration
FFP
fresh frozen plasma
FiO2
fraction of inspired oxygen
fMLP
formyl-Met-Leu-Phe
GABA
γ-aminobutyric acid
GH
growth hormone
Hb
haemoglobin
HbAlc
glycohaemoglobin Alc
© Atlas Medical Publishing Ltd HbcAb
relative B core antibody
HCOA
hypnotic component of anaesthesia
HIV
human immunodeficiency virus
HTLV1
human T cell lymnphoma virus 1
ICNARC
Intensive Care National Audit and Research Centre
List of abbreviations
396
ICU
intensive care unit
ID
internal diameter
IDD
indocyanine green dye dilution
IGF
insulin-like growth factor
IL
interleukin
IMV
invasive mechanical ventilation
IPAP
inspiratory positive airway pressure
KATP
ATP-sensitive potassium
ke0
rate constant for the plasma effect site equilibration
LDL
low-density lipoprotein
LMA
laryngeal mask airway
LOC
loss of consciousness
LORL
loss of response to laryngoscopy
LORnoxious
loss of response to a noxious stimulus
LORverbal
loss of response to a verbal command
ITS
Laryngeal Tube Sonda
M6G
morphine-6-glucoronide
MB
myocardial band
MBL
mannan-binding lectin
MERIT
Medical Early Response Intervention Trial
MET
Medical Emergency Team
MEWS
modified early warning score
MI
myocardial infarction
MLAEP
mid-latency auditory evoked potential
MODS
multiple organ dysfunction score
MRC
Medical Research Council
MRI
magnetic resonanceimaging
mRNA
messenger ribonuclear acid
NADH
nicotinamide adenine dinucleotide
NICE
National Institute for Clinical Excellence
NICO2
non-invasive carbon dioxide rebreathing technique
NIV
non-invasive ventilation
NMDA
N-methyl-D-aspartate
NPPV
non-invasive positive pressure ventilation
NPV
negative pressure ventilation
NSAID
non-steroidal anti-inflammatory drug
List of abbreviations
397
O2
oxygen
ODM
oesophageal Doppler monitor
OPCAB
off-pump coronary artery bypass
OR
odds ratio
OTFC
oral transmucosal fentanyl citrate
PAC
pulmonary artery catheter
PaCO2
arterial carbon dioxide pressure
PAD
pre-operative autologous blood donation
PaO2
arterial oxygen pressure
PAOP
pulmonary artery occlusion pressure
PCR
polymerase chain reaction
PCTA
percutaneous coronary transluminal angioplasty
PETCO2
primary outcome end-tidal carbon dioxide concentration
pHi
intramucosal pH
PK
prediction probability
PMN
polymorphonuclear neutrophil
PO2
oxygen tension
PPF
plasma protein fraction
PPV
positive pressure ventilation
PVC
polyvinyl chloride
RR
relative risk
RRR
relative risk ratio
SAFE
Saline versus Albumin Fluid Evaluation (study)
SARS
severe acute respiratory syndrome
SD
standard deviation
SE
standard error
SEM
standard error of the man
SjO2
jugular oxygen saturation
SNP
single-nucleotide polymorphism
SOFA
sequential organ failure assessment
SpO2
arterial haemoglobin oxygen saturation
SvO2
venous oxygen saturation
SWMA
segmental wall motion abnormality
T50
temperature inducing a pain intensity of 50
TDCO
thermodilution cardiac output
Tmax
time to reach the maximum concentration
List of abbreviations
398
TNF
tumour necrosis factor
TOE
transoesophageal echocardiography
TRICC
Transfusion Requirements in Critical Care
VAP
ventilator-associated pneumonia
VAS
visual analogue scale
vCJD
variant Creutzfeld-Jakob disease
Vi45
ventilation at a fixed end-tidal pressure of carbondioxide of 45 mmHg
VIGOR
VIOXX Gastrointestinal Outcomes Research
VTI
velocity-time integral
WBEB
whole-body electrical bioimpedance
Index of papers reviewed Akamatsu S, Oda A, Terazawa E, Yamamoto T, Ohata H, Michino T, Dohi S. Automated cardiac output measurement by transesophageal color Doppler echocardiography. Anesth Analg 2004; 98: 1232–8. 218 Akca O, Wadhwa A, Sengupta P, Durrani J, Hanni K, Wenke M, Yucel Y, Lenhardt R, Doufas AG, Sessler DI. The new perilaryngeal airway (CobraPLA) is as efficient as the laryngeal mask airway (LMA) but provides better airway sealing pressures. Anesth Analg 2004; 99: 272–8. 271 Albertin A, Casati A, Bergonzi P, Fano G, Torri G. Effects of two target-controlled concentrations (1 and 3 ng/ml) of remifentanil on MAC-BAR of sevoflurane. Anesthesiology 2004; 100:255–9. 178 Alderson P, Bunn F, Lefebvre C, Li WP, Li L, Roberts I, Schierhout G; Albumin Reviewers. Human albumin solution for resuscitation and volume expansion in critically ill patients. Cochrane Database Syst Rev 2004; 4: CD001208. 316 American Thoracic Society. Evidence-based colloid use in the critically ill: American Thoracic Society Consensus Statement. Am J Respir Crit Care Med 2004; 170:1247–59. 320 Antonelli M, Pennisi MA, Pelosi P, Gregoretti C, Squadrone V, Rocco M, Cecchini L, Chiumello D, Severgnini P, Proietti R, Navalesi P, Conti G. Non-invasive positive pressure ventilation using a helmet in patients with acute exacerbation of chronic obstructive pulmonary disease: a feasibility study. Anesthesiology 2004; 100:16–24. 371 Arkilic CF, Taguchi A, Sharma N, Ratnaraj J, Sessler DI, Read TE, Fleshman JW, Kurz A. Supplemental perioperative fluid administration increases tissue oxygen pressure. Surgery 2003; 133:49–55. 34 Asai T, Marfin AG, Thompson J, Popat M, Shingu K. Ease of insertion of the Laryngeal Tube during manual in-line neck stabilization. Anaesthesia 2004; 59: 1163–6. 261 Baird TA, Parsons MW, Phanh T, Butcher KS, Desmond PM, Tress BM, Colman PG, Chambers BR, Davis SM. Persistent post-stroke hyperglycemia is independently associated with infarct expansion and worse clinical outcome. Stroke 2003; 34: 2208–14. 331 Barber KM, Hussey JC, Bond ZC, Falvey SJ, Collett D, Rudge CJ. The UK national potential donor audit. Transplant Proc 2005; in press. 390 Bates J, Deane C, Lindsell D. Extending the provision of ultrasound services in the United Kingdom. Available from the British Medical Ultrasound Society website: www.bmus.org, September 2003. 282 Bein B, Worthmann F, Tonner PH, Paris A, Steinfath M, Hedderich J, Scholz J. Comparison of esophageal Doppler, pulse contour analysis, and real-time pulmonary artery thermodilution for the continuous measurement of cardiac output. J Cardiothorac Vasc Anesth 2004; 18: 185–9. 222 Bell MDD, Moss E, Murphy PG. Brain stem death testing in the UK—time for reappraisal? Br J Anaesth 2004; 92: 633–40. 384 © Atlas Medical Publishing Ltd
Index of papers reviewed
401
Boldt J, Ducke M, Kumle B, Papsdorf M, Zurmeyer EL. Influence of different volume replacement strategies on inflammation and endothelial activation in the elderly undergoing major abdominal surgery. Intensive Care Med 2004; 30: 416–22. 28, 322 Böttiger BW, Motsch J, Teschendorf P, Rehmert GC, Gust R, Zorn M, Schweizer M, Layug EL, Snyder-Ramos SA, Mangano DT, Martin E. Post-operative twelve-lead ECG predicts perioperative myocardial ischaemia associated with myocardial cell damage. Anaesthesia 2004; 59: 1083–90. 102 Bouillon T, Bruhn J, Radu-Radulescu L, Andresen C, Cohane C, Shafer SL. Mixed effects modelling of the intrinsic ventilatory depressant potency of propofol in the non-steady state. Anesthesiology 2004; 100: 240–50. 174 Bouillon T, Bruhn J, Radulescu L, Andresen C, Shafer TJ, Cohane C, Shafer SL. Pharmacodynamic interaction between propofol and remifentanil regarding hypnosis, tolerance of laryngoscopy, bispectral index, and electroencephalographic approximate entropy. Anesthesiology 2004; 100: 1353–72. 176, 243 Brandstrup B, Tønnesen H, Beier-Holgersen R, and the Danish Study Group on Perioperative Fluid Therapy. Effects of intravenous fluid restriction on post-operative complications: comparison of two perioperative fluid regimens: a randomized assessor-blinded multicenter trial. Ann Surg 2003; 238: 641–8. 33 Brimacombe J, Keller C, Dana J. Gum elastic bougie-guided insertion of the ProSeal™ laryngeal mask airway is superior to the digital and introducer tool techniques. Anesthesiology 2004; 100: 25–9. 255 Bucerius J, Gummert JF, Borger MA, Walther T, Doll N, Falk V, Schmitt DV, Mohr FW. Predictors of delirium after cardiac surgery delirium: effect of beating-heart (off-pump) surgery. J Thorac Cardiovasc Surg 2004; 127(1): 57–64. 65 Bunn F, Roberts I, Tasker R, Akpa E. Hypertonic versus near isotonic crystalloid for fluid resuscitation in critically ill patients. Cochrane Database Syst Rev 2004; (3): CD 002045. 32 Carless P, Moxey A, O’Connell D, Henry D. Autologous transfusion techniques: a systematic review of their efficacy. Transfus Med 2004; 14: 123–44. 55 Chen LC, Elliott RA, Ashcroft DM. Systematic review of the analgesic efficacy and tolerability of COX-2 inhibitors in post-operative pain control. J Clin Pharm Ther 2004; 29: 215–29. 138 Chittock DR, Dhingra VK, Ronco JJ, Russell JA, Forrest DM, Tweeddale M, Fenwick JC. Severity of illness and risk of death associated with pulmonary artery catheter use. Crit Care Med 2004; 32: 911–15. 226 Choi HK, Seeger JD, Kuntz KM. Effects of rofecoxib and naproxen on life expectancy among patients with rheumatoid arthritis: a decision analysis. Am J Med 2004; 116: 621–9. 134 Christiansen C, Toft P, Jørgensen HS, Andersen SK, Tønnesen E. Hyperglycaemia and mortality in critically ill patients; a prospective study. Intensive Care Med 2004; 30: 1685–8. 334 Chu CM, Chan VL, Lin AW, Wong IW, Leung WS, Lai CK. Re-admission rates and lifethreatening events in COPD survivors treated with non-invasive ventilation for acute hypercapnic respiratory failure. Thorax 2004; 59: 1020–5. 358 Chu CM, Chan VL, Wong IW, Leung WS, Lin AW, Cheung KF. Non-invasive ventilation in patients with acute hypercapnic exacerbation of chronic obstructive pulmonary disease who refused endotracheal intubation. Crit Care Med 2004; 32: 372–7. 367 Conzen PF, Fischer S, Detter C, Peter K. Sevoflurane provides greater protection of the myocardium than propofol in patients undergoing off-pump coronary artery bypass surgery. Anesthesiology 2003; 99(4): 826–33. 67 Cook TM, McKinstry C, Hardy R, Twigg S. Randomized crossover comparison of the ProSeal™ laryngeal mask airway with the Laryngeal Tube™ during anaesthesia with controlled ventilation. Br J Anaesth 2003; 91: 678–83. 259
Index of papers reviewed
402
Cook TM, Porter MV. Randomised comparison of the classic Laryngeal Mask Airway™ with the Airway Management Device™ during anaesthesia. Br J Anaesth 2003; 91: 672–7. 265 Cooper DJ, Myles PS, McDermott FT, Murray LJ, Laidlaw J, Cooper G, Tremayne AB, Bernard SS, Ponsford J; HTS Study Investigators. Pre-hospital hypertonic saline resuscitation of patients with hypotension and severe traumatic brain injury; a randomized controlled trial. JAMA 2004; 291: 1350–7. 324 Cortinez LI, Hsu YW, Sum-Ping ST, Young C, Keifer JC, Macleod D, Robertson KM, Wright DR, Moretti EW, Somma J. Dexmedetomidine pharmacodynamics: Part II, crossover comparison of the analgesic effect of dexmedetomidine and remifentanil in healthy volunteers. Anesthesiology 2004; 101: 1077–83. 146 Cotter G, Moshkovitz Y, Kaluski E, Cohen AJ, Miller H, Goor D, Vered Z. Accurate, noninvasive continuous monitoring of cardiac output by whole-body electrical bioimpedance. Chest 2004; 125:1431–40. 219 Crane SD, Elliott MW, Gilligan P, Richards K, Gray AJ. Randomised controlled comparison of continuous positive airways pressure, bi-level non-invasive ventilation and standard treatment in emergency department patients with acute cardiogenic pulmonary oedema. Emerg Med J 2004; 21:155–61. 360 Dahaba AA, Mattweber M, Fuchs A, Zenz W, Rehak PH, List WF, Metzler H. The effect of different stages of neuromuscular block on the bispectral index and the bispectral index-XP under remifentanil/propofol anesthesia. Anesth Analg 2004, 99: 781–7. 246 Deusch E, Thaler U, Kozek-Langenecker SA. The effects of high molecular weight hydroxyethyl starch solutions on platelets. Anesth Analg 2004; 99: 665–8. 24 Diaz-Reganon G, Minambres E, Holanda M, Gonzalez-Herrara S, Lopez-Espadas F, Garrido-Diaz C. Usefulness of venous oxygen saturation in the jugular bulb for the diagnosis of brain death: report of 118 patients. Intensive Care Med 2002; 28: 1724–8. 387 Dimitriou V, Voyagis GS, latrou C, Brimacombe J. A comparison of the PAxpress™ and facemask plus Guedel airway by inexperienced personnel after mannequin-only training. Anesth Analg 2003; 96: 1214–17. 268 Dimitriou V, Voyagis GS, latrou C, Brimacombe J. The PAxpress™ is an effective ventilatory device but has an 18% failure rate for flexible lightwand-guided tracheal intubation in anesthetized paralyzed patients. Can J Anesth 2003; 50: 495–500. 270 Dimopoulou I, Tsagarakis S, Anthi A, Milou E, Ilias I, Stavrakaki K, Charalambidis C, Tzanela M, Orfanos S, Mandragos K, Thalassinos N, Roussos C. High prevalence of decreased cortisol reserve in brain-dead potential organ donors. Crit Care Med 2003; 31: 1113–17. 396 Doufas AG, Bakhshandeh M, Bjorksten AR, Shafer SL, Sessler DI. Induction speed is not a determinant of propofol pharmacodynamics. Anesthesiology 2004; 101: 112–21. 179 Dressler O, Schneider G, Stockmans G, Kochs EF. Awareness and EEG power spectrum: analysis of frequencies. Br J Anaesth 2004; 93(6): 806–9. 235 Drover DR, Litalien C, Wellis V, Shafer SL, Hammer GB. Determination of the pharmacodynamic interaction of propofol and remifentanil during esophago-gastro-duodenoscopy in children. Anesthesiology 2004; 100: 1382–6. 187 Durazzo AES, Machado FS, Ikeoka DT, De Bernoche C, Monachini MC, Puech-Leao P, Caramelli B. Reduction in cardiovascular events after vascular surgery with atorvastatin: a randomized trial. J Vasc Surg 2004; 39: 967–76. 100 Esteban A, Frutos-Vivar F, Ferguson ND, Arabi Y, Apezteguia C, Gonzalez M, Epstein SK, Hill NS, Nava S, Soares MA, D’Empaire G, Alia I, Anzueto A. Non-invasive positive-pressure ventilation for respiratory failure after extubation. N Engl J Med 2004; 350: 2452–60. 365
Index of papers reviewed
403
Falcone RA, Nass C, Jermyn R, Hale CM, Stierer T, Jones CE, Walters GK, Fleisher LA. The value of pre-operative pharmacologic stress testing before vascular surgery using ACC/AHA guidelines: a prospective, randomized trial. J Cardiothorac Vasc Anesth 2003; 17: 694–8. 92 Fechner J, Ihmsen H, Hatterscheid D, Jeleazcov C, Schiessl C, Vornov JJ, Schwilden H, Schuttler J. Comparative pharmacokinetics and pharmacodynamics of the new propofol prodrug GPI 15715 and propofol emulsion. Anesthesiology 2004; 101 (3): 626–39. 129 Finfer S, Bellomo R, Boyce N, French J, Myburgh J, Norton R; SAFE Study Investigators. A comparison of albumin and saline for fluid resuscitation in the intensive care unit. N Engl J Med 2004; 350: 2247–56. 30, 312 Finney S J, Zekveld C, Elia A, Evans TW. Glucose control and mortality in critically ill patients. JAMA 2003; 290: 2041–7. 333 Fishbain DA, Fishbain D, Lewis J, Cutler RB, Cole B, Rosomoff HL, Rosomoff RS. Genetic testing for enzymes of drug metabolism: does it have clinical utility for pain medicine at the present time? A structured review. Pain Med 2004; 5:81–93. 163 Fleisher LA, Corbett W, Berry C, Poldermans D. Cost-effectiveness of differing perioperative β blockade strategies in vascular surgery patients. J Cardiothorac Vasc Anesth 2004; 18:7–13. 95 Gaitini LA, Vaida SJ, Somri M, Yanovski B, Ben-David B, Hagberg CA. A randomized controlled trial comparing the ProSeal™ laryngeal mask airway with the Laryngeal Tube suction in mechanically ventilated patients. Anesthesiology 2004; 101: 316–20. 262 Gajewska K, Schroede M, De Francoise M, Vincent JL. Analysis of terminal events in 109 successive deaths in a Belgian intensive care unit. Intensive Care Med 2004; 30:1224–7. 392 Gajic O, Rana R, Mendez JL, Rickman OB, Lymp JF, Hubmayr RD, Moore SB. Acute lung injury after blood transfusion in mechanically ventilated patients. Transfusion 2004; 44: 1468–74. 49 Girard T, Kindler CH. Pharmacogenetics and anaesthesiology. Curr Pharmacogenom 2004; 2:119–35. 159 Girou E, Brun-Buisson C, Taille S, Lemaire F, Brochard L. Secular trends in nosocomial infections and mortality associated with non-invasive ventilation in patients with exacerbation of COPD and pulmonary edema. JAMA 2003; 290: 2985–91. 363 Goh AY-T, Mok Q. Clinical course and determination of brainstem death in a children’s hospital. Acta Paediatr 2004; 93: 47–52. 395 Goldberg PA, Siegel MD, Sherwin RS, Halickman JI, Lee M, Bailey VA, Lee SL, Dziura JD, Inzucchi SE. Implementation of a safe and effective insulin infusion protocol in a medical intensive care unit. Diabetes Care 2004; 27: 461–7. 345 Gore DC, Wolf SE, Sanford AP, Herndon DN, Wolfe RR. Extremity hyperinsulinemia stimulates muscle protein synthesis in severely injured patients. Am J Physiol Endocrinol Metab 2004; 286: E529–34. 347 Gorini M, Ginanni R, Villella G, Tozzi D, Augustynen A, Corrado A. Non-invasive negative and positive pressure ventilation in the treatment of acute or chronic respiratory failure. Intensive Care Med 2004; 30: 875–81. 368 Grau T, Bartusseck E, Conradi R, Martin E, Motsch J. Ultrasound imaging improves learning curves in obstetric epidural anesthesia: a preliminary study. Can J Anesth 2003; 50: 1047–50. 289 Grebenik CR, Boyce A, Sinclair ME, Evans RD, Mason DG, Martin B. NICE guidelines for central venous catheterisation in children. Is the evidence base sufficient? Br J Anaesth 2004; 92(6): 827–30. 277 Greher M, Scharbert G, Kamolz LP, Beck H, Gustorff B, Kirchmair L, Kapral S. Ultrasoundguided lumbar facet nerve block: a sonoanatomic study of a new methodologic approach. Anaesthesiology 2004; 100: 1242–8. 286 Grow MP, Singh A, Fleming NW, Young N, Watnik M. Cardiac output monitoring during offpump coronary artery bypass grafting. J Cardiothorac Vasc Anesth 2004; 18: 43–6. 72, 217
Index of papers reviewed
404
Hansen TK, Thiel S, Wouters PJ, Christiansen JS, Van den Berghe G. Intensive insulin therapy exerts anti-inflammatory effects in critically ill patients and counteracts the adverse effect of low mannose-binding lectin levels. J Clin Endocrinol Metab 2003; 88: 1082–8. 338 Hofer CK, Buhlmann S, Klaghofer R, Genoni M, Zollinger A. Pulsed dye densitometry with two different sensor types for cardiac output measurement after cardiac surgery: a comparison with the thermodilution technique. Acta Anaesthesiol Scand 2004; 48: 653–7. 225 Hsu YW, Cortinez LI, Robertson KM, Keifer JC, Sum-Ping ST, Moretti EW, Young CC, Wright DR, Macleod DB, Somma J. Dexmedetomidine pharmacodynamics: Part I, crossover comparison of the respiratory effects of dexmedetomidine and remifentanil in healthy volunteers. Anesthesiology 2004; 101: 1066–76. 145 Hullett B, Gibbs N, Weightman W, Thackray M, Newman M. A comparison of CardioQ and thermodilution cardiac output during off-pump coronary artery surgery. J Cardiothorac Vasc Anesth 2003; 17(6): 728–32. 74, 215 Jensen EW, Litvan H, Struys M, Martinez Vazquez P. Pitfalls and challenges when assessing the depth of hypnosis during general anaesthesia by clinical signs and electronic indices. Acta Anaesthesiol Scand 2004; 48: 1260–7. 237 Jeschke MG, Klein D, Herndon DN. Insulin treatment improves the systemic inflammatory reaction to severe trauma. Ann Surg 2004; 239:553–60. 341 Johnson KB, Egan TD, Kern SE, McJames SW, Cluff ML, Pace NL. Influence of hemorrhagic shock followed by crystalloid resuscitation on propofol; a pharmacokinetic and pharmacodynamic analysis. Anesthesiology 2004; 101: 647–59. 188 Jungheinrich C, Sauermann W, Bepperling F, Vogt NH. Volume efficacy and reduced influence on measures of coagulation using hydroxyethyl starch 130/0.4 (6%) with an optimised in vivo molecular weight in orthopaedic surgery: a randomised, double-blind study. Drugs R D 2004; 5:1– 9. 25 Jüttner B, Kuse ER, Elsner HA, Heine J, Jaeger K, Piepenbrock S, Scheinichen D. Differential platelet receptor expression following hydroxyethyl starch infusion in thrombocytopaenic orthotopic liver transplantation recipients. Eur J Anaesthesiol 2004; 21: 309–13. 25 Kanji S, Singh A, Tierney M, Meggison H, McIntyre L, Hebert PC. Standardization of intravenous insulin therapy improves the efficiency and safety of blood glucose control in critically ill adults. Intensive Care Med 2004; 30: 804–10. 344 Keller C, Brimacombe J, Bittersohl P, Lirk P, Von Goedecke A. Aspiration and the laryngeal mask airway: three cases and a review of the literature. Br J Anaesth 2004; 93: 579–82. 253 Kern SE, Xie G, White JL, Egan TD. Opioid-hypnotic synergy: a response surface analysis of propofol-remifentanil pharmacodynamic interaction in volunteers. Anesthesiology 2004; 100: 1373–81. 177 Klepstad P, Rakvåg TT, Kaasa S, Holthe M, Dale O, Borchgrevink PC, Baar C, Vikan T, Krokan HE, Skorpen F. The 118 A→G polymorphism in the human µ-opioid receptor gene may increase morphine requirements in patients with pain caused by malignant disease. Acta Anaesthiol Scand 2004; 48: 1232–9. 167 Krogh-Madsen R, Møller K, Dela F, Kronborg G, Jauffred S, Pedersen BK. Effect of hyperglycemia and hyperinsulinemia on the response of IL-6, TNF-α and FFA to low-dose endotoxemia in humans. Am J Physiol Endocrinol Metab 2004; 286: E766–72. 339 Kuitunen AH, Hynynen MJ, Vahtera E, Salmenpera MT. Hydroxyethyl starch as a priming solution for cardiopulmonary bypass impairs hemostasis after cardiac surgery. Anesth Analg 2004; 98: 291–7. 26
Index of papers reviewed
405
Laird AM, Miller PR, Kilgo PD, Meredith JW, Chang MC. Relationship of early hyperglycemia to mortality in trauma patients. J Trauma 2004; 56: 1058–62. 336 Landesberg G, Mosseri M, Wolf YG, Bocher M, Basevitch A, Rudis E, Izhar U, Anner H, Weissman C, Berlatzky Y. Pre-operative thallium scanning, selective coronary revascularization and long-term survival after major vascular surgery. Circulation 2003; 108: 177–83. 90 Landesberg G, Shatz V, Akopnik I, Wolf YG, Mayer M, Berlatzky Y, Weissman C, Mosseri M. Association of cardiac troponin, CK-MB and post-operative myocardial ischemia with longterm survival after major vascular surgery. J Am Coll Cardiol 2003; 42 :1547–54. 88 Laver S, Preston S, Turner D, McKinstry C, Padkin A. Implementing intensive insulin therapy: development and audit of the Bath insulin protocol. Anaesth Intensive Care 2004; 32: 311–16. 344 Leather HA, Vuylsteke A, Bert C, M’Fam W, Segers P, Sergeant P, Vandermeersch E, Wouters PF. Evaluation of a new continuous cardiac output monitor in off-pump coronary artery surgery. Anaesthesia 2004; 59: 385–9. 221 L’E Orme RM, Pigott DW, Mihm FG. Measurement of cardiac output by transpulmonary arterial thermodilution using a long radial artery catheter: a comparison with intermittent pulmonary artery thermodilution. Anaesthesia 2004; 59: 590–4. 210 Levy M, Tanios MA, Nelson D, Short K, Senechia A, Vespia J, Hill NS. Outcomes of patients with do-not-intubate orders treated with non-invasive ventilation. Crit Care Med 2004; 32: 2002–7. 366 Leykin Y, Pellis T, Lucca M, Lomangino G, Marzano B, Gullo A. The pharmacodynamic effects of rocuronium when dosed according to real body weight or ideal body weight in morbidly obese patients. Anesth Analg 2004; 99: 1086–9. 184 Liem EB, Lin C-M, Suleman M-I, Doufas AG, Gregg RG, Veauthier JM, Loyd G, Sessler DI. Anesthetic requirement is increased in redheads. Anesthesiology 2004; 101: 279–83. 161 Mack MJ, Pfister A, Bachand D, Emery R, Magee MJ, Connolly M, Subramanian V. Comparison of coronary artery bypass surgery with and without cardiopulmonary bypass in patients with multivessel disease. J Thorac Cardiovasc Surg 2004; 127(1): 167–73. 63 Maltby JR, Beriault MT, Watson NC, Liepert DJ, Pick GH. LMA-Classic™ and LMAProSeal™ are effective alternatives to endotracheal intubation for gynecologic laparoscopy. Can J Anaesth 2003; 50: 71–7. 252 Marhofer P, Greher M, Kapral S. Ultrasound guidance in regional anaesthesia. Br J Anaesth 2005; 94: 7–17. 289 Marhofer P, Sitzwohl C, Greher M, Kapral S. Ultrasound guidance for infraclavicular brachial plexus anaesthesia in children. Anaesthesia 2004; 59: 642–6. 285 Marx G, Schuerholz T, Sumpelmann R, Simon T, Leuwer M. Comparison of cardiac output measurements by arterial transcardio-pulmonary as well as pulmonary arterial thermodilution with direct Pick. Eur J Anaesthesiol 2005; 22: 129–34. 212 Masip J, Paez J, Merino M, Parejo S, Vecilla F, Riera C, Rios A, Sabater J, Ballus J, Padro J. Risk factors for intubation as a guide for non-invasive ventilation in patients with severe acute cardiogenic pulmonary edema. Intensive Care Med 2003; 29: 1921–8. 361 McIntyre L, Hebert PC, Wells G, G, Fergusson D, Marshall J, Yetisir E, Blajchman MJ; Canadian Critical Care Trials Group. Is a restrictive transfusion strategy safe for resuscitated and critically ill trauma patients? J Trauma 2004; 57: 563–8. 43 McMullin J, Brozek J, Jaeschke R, Hamielec C, Dhingra V, Rocker G, Freitag A, Gibson J, Cook D. Glycemic control in the ICU: a multicenter survey. Intensive Care Med 2004; 30: 798– 803. 330 Mehta Y, Kulkarni V, Juneja R, Sharma KK, Mishra Y, Raizada A, Trehan N. Spinal (subarachnoid) morphine for off-pump coronary artery bypass surgery. Heart Surg Forum 2004; 7(3): E205–10. 69
Index of papers reviewed
406
Meier JJ, Weyhe D, Michaely M, Senkal M, Zumtobel V, Nauck MA, Holst JJ, Schmidt WE, Gallwitz B. Intravenous glucagon-like peptide 1 normalizes blood glucose after major surgery in patients with type 2 diabetes. Crit Care Med 2004; 32: 848–51. 349 Mesotten D, Swinnen JV, Vanderhoydonc F, Wouters PJ, Van den Berghe G. Contribution of circulating lipids to the improved outcome of critical illness by glycemic control with intensive insulin therapy. J Clin Endocrinol Metab 2004; 89: 219–26. 348 Mesotten D, Wouters PJ, Peeters RP, Hardman KV, Holly JM, Baxter RC, Van den Berghe G. Regulation of the somatotropic axis by intensive insulin therapy during protracted critical illness. J Clin Endocrinol Metab 2004; 89: 3105–13. 348 Miller A, Sleigh JW, Barnard J, Steyn-Ross DA. Does bispectral analysis of the electroencephalogram add anything but complexity? Br J Anaesth 2004; 92: 8–13. 234 Moretti EW, Robertson KM, El-Moalem H, Gan TJ. Intra-operative colloid administration reduces post-operative nausea and vomiting and improves post-operative outcomes compared with crystalloid administration. Anesth Analg 2003; 96: 611–17. 28 Myles PS, Leslie K, McNeil J, Forbes A, Chan MT. Bispectral index monitoring to prevent awareness during anaesthesia: the B-Aware randomised controlled trial. Lancet 2004; 363: 1757– 63. 238 Nava S, Carbone G, DiBattista N, Bellone A, Baiardi P, Cosentini R, Marenco M, Giostra F, Borasi G, Groff P. Non-invasive ventilation in cardiogenic pulmonary oedema: a multicenter randomized trial. Am J Respir Crit Care Med 2003; 168: 1432–7. 359 Noguchi H, Kitazumi K, Mori M, Shiba T. Electroencephalographic properties of zaleplon, a non-benzodiazepine sedative/ hypnotic, in rats. J Pharmacol Sci 2004; 94(3): 246–51. 130 Opdam HI, Silvester W. Identifying the potential organ donor: an audit of hospital deaths. Intensive Care Med 2004; 30: 1390–7. 393 Park GR, Wilkins N, Higgins T. Brain stem death; an organ donation 11 years on. Br J Anaesth 2003; 91: 577–9. 390 Paul M, Dueck M, Kampe S, Fruendt H, Kasper SM. Pharmacological characteristics and side effects of a new galenic formulation of propofol without soya bean oil. Anaesthesia 2003; 58(11): 1056–62. 127 Perner A, Nielsen SE, Rask-Madsen J. High glucose impairs superoxide production from isolated blood neutrophils. Intensive Care Med 2003; 29: 642–5. 337 Quinlan GJ, Mumby S, Martin GS, Bernard GR, Gutteridge JMC, Evans TW. Albumin influences total plasma antioxidant capacity favorably in patients with acute lung injury. Crit Care Med 2004; 32: 755–9. 321 Rao SV, Jollis JG, Harrington RA, Granger CB, Newby LK, Armstrong PW, Moliterno DJ, Lindblad L, Pieper K, Topol EJ, Stamler JS, Califf RM. Relationship of blood transfusion and clinical outcomes in patients with acute coronary syndromes. JAMA 2004; 292: 1555–62. 41 Roberts I, Alderson P, Bunn F, Chinnock P, Ker K, Schierhout G. Colloids versus crystalloids for fluid resuscitation in critically ill patients. Cochrane Database Syst Rev 2004; 4: CD000567. 318 Rocco M, Dell’Utri D, Morelli A, Spadetta G, Conti G, Antonelli M, Pietropaoli P. Noninvasive ventilation by helmet or facemask in immunocompromised patients: a case-control study. Chest 2004; 126: 1508–15. 372
Index of papers reviewed
407
Rocco M, Spadetta G, Morelli A, Dell’Utri D, Porzi P, Conti G, Pietropaoli P. A comparative evaluation of thermodilution and partial CO2 rebreathing techniques for cardiac output assessment in critically ill patients during assisted ventilation. Intensive Care Med 2004; 30: 82–7. 224 Romberg R, Olofsen E, Sarton E, Den HJ, Taschner PE, Dahan A. Pharmacokineticpharmacodynamic modeling of morphine-6-glucuronide-induced analgesia in healthy volunteers: absence of sex differences. Anesthesiology 2004; 100: 120–33. 144 Romberg R, Olofsen E, Sarton E, Teppema L, Dahan A. Pharmacodynamic effect of morphine6-glucuronide versus morphine on hypoxic and hypercapnic breathing in healthy volunteers. Anesthesiology 2003; 99: 788–98. 143 Royal College of Radiologists. Guidance for training in ultrasound and medical non-radiologists. ISBN 1872599 30 3 code BFCR (97)1. 282 Salvi L, Sisillo E, Brambillasca C, Juliano G, Salis S, Marino MR. High thoracic epidural anesthesia for off-pump coronary artery bypass surgery. J Cardiothorac Vasc Anesth 2004; 18(3): 256–62. 71 Sandhu NPS, Sidhu DS. Mid-arm approach to basilica and cephalic vein cannulation using ultrasound guidance. Br J Anaesth 2004; 93(2): 292–4. 279 Scala R, Bartolucci S, Naldi M, Rossi M, Elliott MW. Co-morbidity and acute decompensations of COPD requiring non-invasive positive-pressure ventilation. Intensive Care Med 2004; 30: 1747– 54. 356 Scharbert G, Deusch E, Kress HG, Greher M, Gustorff B, Kozek-Langenecker SA. Inhibition of platelet function by hydroxyethyl starch solutions in chronic pain patients undergoing peridural anesthesia. Anesth Analg 2004; 99: 823–7. 23 Schultz A, Grouven U, Zander I, Beger FA, Siedenberg M, Schultz B. Age-related effects in the EEG during propofol anaesthesia. Acta Anaesthesiol Scand 2004; 48: 27–34. 233 Sedrakyan A, Gondek K, Paltiel D, Elefteriades JA. Volume expansion with albumin decreases mortality after coronary artery bypass graft surgery. Chest 2003; 123: 1853–7. 30 Segal JB, Blasco-Colmenares E, Norris EJ, Guallar E. Pre-operative acute normovolemic hemodilution: a meta-analysis. Transfusion 2004; 44: 632–44. 54 Senouci K, Guerrini P, Diene E, Atinault A, Claquin J, Bonnet F, Tuppin P. A survey of patients admitted in severe coma: implications for brain death identification and organ donation. Intensive Care Med 2004; 30: 38–44. 389 Sharma A, Bodenham AR, Mallick A. Ultrasound-guided infraclavicular axillary vein cannulation for central venous access. Br J Anaesth 2004; 93: 188–92. 276 Shibutani K, Inchiosa MA, Sawada K, Bairamian M. Accuracy of pharmacokinetic models for predicting plasma fentanyl concentrations in lean and obese surgical patients; derivation of dosing weight (‘pharmacokinetic mass’). Anesthesiology 2004; 101: 603–13. 181 Sinha A, Chan VW. Ultrasound imaging for popliteal sciatic nerve block. Reg Anesth Pain Med 2004; 29: 130–4. 287 Sivasankar R, Bahlmann UB, Stacey MR, Sehgal A, Hughes RC, Hall JE. An evaluation of the modified Airway Management Device. Anaesthesia 2003; 58: 558–61. 264 Solomon DH, Schneeweiss S, Glynn RJ, Kiyota Y, Levin R, Mogun H, Avorn J. Relationship between selective cyclooxygenase-2 inhibitors and acute myocardial infarction in older adults. Circulation 2004; 109: 2068–73. 135 Song D, Hamza MA, White PF, Byerly SI, Jones SB, Macaluso AD. Comparison of a lowerlipid propofol emulsion with the standard emulsion for sedation during monitored anesthesia care. Anesthesiology 2004; 100(5): 1072–5. 126 Song D, Hamza MA, White PF, Klein K, Recart A, Khodaparast O. The pharmacodynamic effects of a lower-lipid emulsion of propofol: a comparison with the standard propofol emulsion. Anesth Analg 2004; 98(3): 687–91. 125
Index of papers reviewed
408
Squadrone E, Frigerio P, Fogliati C, Gregoretti C, Conti G, Antonelli M, Costa R, Baiardi P, Navalesi P. Non-invasive versus invasive ventilation in COPD patients with severe acute respiratory failure deemed to require ventilatory assistance. Intensive Care Med 2004; 30: 1303–10. 355 Stamer UM, Lehnen K, Hothker F, Bayerer B, Wolf S, Hoeft A, Stuber F. Impact of CYP2D6 genotype on post-operative tramadol analgesia. Pain 2003; 105: 231–8. 164 Stevens RD, Burri H, Tramér MR. Pharmacologic myocardial protection in patients undergoing non-cardiac surgery: a quantitative systematic review. Anesth Analg 2003; 97: 623–33. 97 Stranders I, Diamant M, van Gelder RE, Spruijt HJ, Twisk JW, Heine RJ, Visser FC. Admission blood glucose level as risk indicator of death after myocardial infarction in patients with and without diabetes mellitus. Arch Intern Med 2004; 164: 982–8. 331 Suttner S, Piper SN, Kumle B, Lang K, Rohm KD, Isgro F, Boldt J. The influence of allogeneic red blood cell transfusion compared with 100% oxygen ventilation on systemic muscle oxygen tension after cardiac surgery. Anesth Analg 2004; 99: 2–11. 47 Sweeney BP. Watson and Crick 50 years on; from double helix to pharmacogenomics. Anaesthesia 2004; 59: 150–65. 160 Thorell A, Rooyackers O, Myrenfors P, Soop M, Nygren J, Ljungqvist OH. Intensive insulin treatment in critically ill trauma patients normalizes glucose by reducing endogenous glucose production. J Clin Endocrin Metab 2004; 89: 5382–6. 346 Tian J, Lin X, Guan R, Xu JG. The effects of hydroxyethyl starch on lung capillary permeability in endotoxic rats and possible mechanisms. Anesth Analg 2004; 98: 768–74. 27 Todisco T, Baglioni S, Eslami A, Scoscia E, Todisco C, Bruni L, Dottorini M. Treatment of acute exacerbations of chronic respiratory failure; integrated use of negative pressure ventilation and non-invasive positive pressure ventilation. Chest 2004; 125: 2217–23. 369 Trapani A, Laquintana V, Lopedota A, Franco M, Latrofa A, Talani G, Sanna E, Trapani G, Liso G. Evaluation of new propofol aqueous solutions for intravenous anesthesia. Int J Pharm 2004; 278(1): 91–8. 124 Vaida SJ, Gaitini D, Ben-David B, Somri M, Hagberg CA, Gaitini LA. A new supraglottic airway, the Elisha Airway Device: a preliminary study. Anesth Analg 2004; 99: 124–7. 267 Vamvakis EC. White blood cell containing allogeneic blood transfusion, post-operative infection and mortality: a meta-analysis of observational before and after studies. Vox Sanguinis 2004; 86: 111–19. 52 van den Berghe G, Wouters PJ, Bouillon R, Weekers F, Verwaest C, Schetz M, Vlasselaers D, Ferdinande P, Lauwers P. Outcome benefit of intensive insulin therapy in the critically ill: insulin dose versus glycemic control. Crit Care Med 2003; 31: 359–66. 345 Van der Linden PJ, De Hert SG, Daper A, Trenchant A, Schmartz D, Defrance P, Kimbimbi P. 3.5% urea-linked gelatin is as effective as 6% HES 200/0.5 for volume management in cardiac surgery patients. Can J Anaesth 2004; 51: 236–41. 21 van Hilten JA, van der Watering LMG, van Bockel JH, van de Velde CJ, Kievit J, Brand R, van den Hout WB, Geelkerken RH, Roumen RM, Wesselink RM, Koopman-van Gemert AW, Koning J, Brand A. Effects of transfusion with red cells filtered to remove leucocytes: randomised controlled trial in patients undergoing major surgery. BMJ 2004; 328: 1281–9. 51 Vanluchene A, Struys M, Heyse B, Mortier E. Spectral entropy measurement of patient responsiveness during propofol and remifentanil: a comparison with the bispectral index. Br J Anaesth 2004; 93(5): 645–54. 242 Vanluchene A, Vereecke H, Thas O, Mortier E, Shafer S, Struys M. Spectral entropy as an electroencephalographic measure of anaesthetic drug effect: a comparison with bispectral index and processed midlatency auditory evoked response. Anesthesiology 2004; 101: 34–42. 239
Index of papers reviewed
409
van Zundert AA, Fonck K, Al-Shaikh B, Mortier E. Comparison of the LMA-Classic™ with the new disposable Soft Seal laryngeal mask in spontaneously breathing adult patients. Anesthesiology 2003; 99: 1066–71. 257 Veneman TF, Nijhuis JO, Woittiez AJJ. Human albumin and starch administration in critically ill patients: a prospective randomized clinical trial. Wien Klin Wochenschr 2004; 116: 305–9. 323 Vincent J-L, Navickis RJ, Wilkes MM. Morbidity in hospitalized patients receiving human albumin: a meta-analysis of randomized, controlled trials. Crit Care Med 2004; 32: 2029–38. 319 Vincent JL, Wilkes MM, Navickis RJ. Safety of human albumin—serious adverse events reported worldwide in 1998–2000. Br J Anaesth 2003; 91: 625–30. 29 Vivien B, Paqueron X, Le Cosquer P, Langeron O, Coriat P, Riou B. Detection of brain death onset using the bispectral index in severely comatose patients. Intensive Care Med 2002; 28: 419– 25. 388 Vogelzang M, Van der Horst ICC, Nijsten MWN. Hyperglycaemic index as a tool to assess glucose control: a retrospective study. Crit Care 2004; 8: R122–7. 330 Walsh TS, McArdle F, McLellan SA, Maciver C, Maginnis M, Prescott RJ, McClelland DB. Does the storage time of transfused red blood cells influence regional or global indices of tissue oxygenation in anemic critically ill patients? Crit Care Med 2004; 32: 364–71. 45 Wang J, Filipovic M, Rudzitis A, Michaux I, Skarvan K, Buser P, Todorov A, Bernet F, Seeberger MD. Transesophageal echocardiography for monitoring segmental wall motion during off-pump coronary artery bypass surgery. Anesth Analg 2004; 99(4): 965–73. 76 Wasmuth HE, Kunz D, Graf J, Stanzel S, Purucker EA, Koch A, Gartung C, Heintz B, Gressner AM, Matern S, Lammert F. Hyperglycaemia at admission to the intensive care unit is associated with elevated serum concentrations of interleukin-6 and reduced ex vivo secretion of tumor necrosis factor-α. Crit Care Med 2004; 32: 1109–14. 341 Wheeler M, Birmingham PK, Lugo RA, Heffner CL, Cote CJ. The pharmacokinetics of the intravenous formulation of fentanyl citrate administered orally in children undergoing general anaesthesia. Anesth Analg 2004; 99: 1347–51. 185 Wijdics EFM. Brain death worldwide: accepted fact but no global consensus and diagnostic criteria. Neurology 2002; 58: 20–5. 381 Williams SR, Chouinard P, Arcand G, Harris P, Ruel M, Boudreault D, Girard F. Ultrasound guidance speeds execution and improves the quality of supraclavicular block. Anesth Analg 2003; 97: 1518–23. 285 Wilson SH, Fasseas P, Orford JL, Lennon RJ, Horlocker T, Charnoff NE, Melby S, Berger PB. Clinical outcome of patients undergoing non-cardiac surgery in the 2 months following coronary stenting. J Am Coll Cardiol 2003; 42: 234–40. 93 Yendamuri S, Fulda GJ, Tinkoff GH. Admission hyperglycemia as a prognostic indicator in trauma. J Trauma 2003; 55: 33–8. 335 Zaugg M, Lucchinetti E, Garcia C, Pasch T, Spahn DR, Schaub MC. Anaesthetics and cardiac preconditioning. Part II: clinical implications. Br J Anaesth 2003; 91: 566–76. 107 Zaugg M, Lucchinetti E, Uecker M, Pasch T, Schaub MC. Anaesthetics and cardiac preconditioning. Part I: signalling and cytoprotective mechanisms. Br J Anaesth 2003; 91: 551–65. 103
General index A A1 18G polymorphism 166–8 A-line autoregressive model with exogenous input (ARX) index 239–41 accuracy measures 197 accurate burst suppressive detection 240–1 acetaminophen 136 acute cardiopulmonary oedema, non-invasive ventilation 359–62 acute coronary syndromes blood transfusion and clinical outcomes, relationship 41–3 transfusion trigger 5 acute hypercapnic respiratory failure (AHRF) 355–6, 358 acute hypoxic ventilatory response (AHVR) 143 acute lung apoptosis, hydroxyethyl starch solution and Ringer’s lactate 3 acute lung injury albumin influences, total plasma antioxidant capacity 321–2 blood transfusion, mechanically ventilated patients 49–50 acute myocardial infarction, older adults, selective COX-2 inhibitors, relationship between 135–7 acute normovolaemic haemodilution (ANH) 53, 56 pre-operative 54–5 acute pain treatment 133 Acute Physiology and Chronic Health Evaluation (APACHE) II Scores 43–4, 51, 227(fig.), 358 acute pulmonary oedema, non-invasive ventilation 353, 359–62 acute trauma (hypovolaemia), awareness, risk factor 238 Adenomatous Polyp Prevention on Vioxx Study 134 adenosine triphosphate (ATP) 6 adrenocortical function, assessment, brain(stem) death 396–7 adult respiratory distress syndrome (ARDS), pulmonary aspiration 253–5 ‘aggressive’ fluid management 34 airway equipment new 251–73 single use 201 Airway Management Device (AMD) 263–6 LMA Classic during anaesthesia, randomised comparison with 265–6 modified, evaluation 264 airway management, nurses, no previous experience 268–9 airway rescue, supraglottic airway 203 airway (supraglottic), development, progress 200–1 albumin 19 antioxidant function, acute lung injury 321–2 containing fluids 309, 311 efficacy 311 haemostasis, effects on 23
General index
412
hypovolaemia, treatment 29–31 hypoproteinaemia, critically ill intensive care patients 29 mortality associated 29 immunoactivating activity 3 platelet dysfunction 4 safety 311 therapy, survival of critically ill patients, Annals of Internal Medicine 311 volume expansion, CABG surgery, mortality, effect on 30 albumin (4%) 21 versus normal saline, fluid resuscitation in the ICU 30–1 allergic-anaphylactic reactions, blood transfusion therapy 5 allergy, β-2-adrenoceptor gene, polymorphism 157–8 allogeneic blood transfusion 39, 53 donor leucocytes 50–3 white blood cell containing, postoperative infection and mortality 52–3 allogeneic red cell transfusion, versus 100% oxygen ventilation, influence on systemic muscle oxygen tension 47–9 α-2 agonists 87, 109 pharmacologic myocardial protection, non-cardiac surgery 97–100 © Atlas Medical Publishing Ltd α-melanocyte-stimulating hormone 162 Am-149 123, 127–9 American College of Cardiology, pre-operative cardiac assessment 9 American College of Cardiology/American Heart Association (ACC/AHA) Guidelines for Perioperative Evaluation in Non-cardiac Surgery 91 pre-operative pharmacologic stress testing before surgery 92 American Heart Association, pre-operative cardiac assessment 9 Ampofol 123 pharmacodynamics 125 (table), 126 sedation during monitored anaesthesia care 125 (table), 126 anaemia, beneficial 5 anaesthesia and pharmacogenetics 165–9 anaesthetic pharmacology, what is new 117–21 anaesthetic practice, synergistic interactions 119 anaesthetic requirements, variables 161–2 anaesthetics preconditioning 10–11 cardiac, signalling and cytoprotective mechanisms 103–7 analgesia, multi-modal approach 133 analgesics, new 133–48 angioplasty 108–9 coronary artery stenting 93–4 thrombosis 94 glucose-insulin-potassium infusion 329 prior to surgery 109 surgery to be delayed at least 6 weeks 109 angiotensin-converting enzyme (ACE) 159 anion gap 3 Annals of Internal Medicine, albumin therapy, survival in critically ill patients 311 anti-fibrinolytics 57
General index
413
anti-platelet drugs coronary stenting 9 thrombosis, reduction 94 anticoagulation neuraxial haematoma formation 69 spinal haematoma formation 70 Aortech 223 apnoea, brain(stem) death, test 380–1 aprotonin 54 Aquavan 118 arterial catheter, peripherally placed, cardiac output determination by arterial thermodilution 210– 12 arterial thermodilution with pulse contour analysis 196 arterial transcardiopulmonary thermodilution, pulmonary arterial thermodilution with direct Pick, comparison with 212–15 aspirin 11, 136, 137(fig.) acute myocardial infarction, older patients 135–7 asthma, β-2-adrenoceptor gene, polymorphism 157–8 atenolol 86 atorvastatin, cardiovascular events reduction after vascular surgery 100–2 auditory evoked potentials (AEP) 237 autologous blood transfusion 39, 53 techniques 55–7 autologous red cell transfusion, effectiveness compared to stored red cells 45–7 automated cardiac output measurement 218–19 awareness 238–9 avoidance 231–2 detection 231 isolated forearm technique 236 during anaesthesia, prevention 198–9 EEG power spectrum analysis of frequencies 235–6 axillary artery, ultrasound imaging 290 (fig.) B B-aware Trial 232, 238–9, 248 back pain 286–7 Bard Site-Rite3 ultrasound probe 279 basilic vein, ultrasound imaging 290 (fig.) basilic/cephalic intravenous access, ultrasound guidance 279–80 benzodiazepines 379–80 β-1-adrenergic receptor 156 β-2-adrenoceptor agonists 158 β-2-adrenoceptor polymorphism 157 (table) pressor response to laryngoscopy 158 β-blocking agents 91, 109 perioperative 86–7 cardiac protection 9–11 different protocols, vascular surgery patients, cost-effectiveness 95–6 pharmacologic myocardial protection, non-cardiac surgery 97–100 β-endorphin 166 bias 197
General index
414
bioimpedance 220 BIS-XP 246–7 bisoprolol 87 bispectral index (BIS) 126, 146, 234–5 brain death onset detection, comatose patients 388–9 electroencephalogram 234–5 monitoring to prevent awareness during anaesthesia, B-aware trial 238–9 neuromuscular block, different stages effect on 246–7 spectral entropy measurement, comparison with 242–3 under remifentanil/propofol anaesthesia 243–7 pharmacodynamic interaction between 176–7 bleeding cardiac surgery patients, HES as a priming solution for CABG 26 crystalloids, replacing by 4–5 dextrans, side effects 19 life-threatening, blood transfusions 57 plasma substitutes, side effects, perioperative hypovolaemia 22–6 blood conservation, perioperative 53–7 blood loss see bleeding blood ‘storage lesion’ 6 blood transfusion therapy acute coronary syndromes, clinical outcomes, relationship 41–3 benefits 5 immunomodulation 5–6 leukoreduction 5–6 morbidity 5 mortality 5 old blood 6 perioperative 5–6, 39–58 blood conservation 53–7 British Medical Journal, editorials 39–40 complications 49–50 costs 40 effectiveness 40–9 leucodepletion of red cells, importance of 51–3 microbiological safety 40 (table) risks 39 risks 5 storage lesion 6 transfusion trigger 5 body homeostasis 117 body mass index 181, 183–5 brachial plexus trunks 284 ultrasound guidance axillary approach 286 axillary part, transverse view 290 (fig.) infraclavicular part, children 285 infraclavicular part, transverse view 287–8 (figs.) pneumothorax, risk 286 supraclavicular block 285–6 bradykinin 104 (fig.), 108
General index
415
brain, sensitivity to hypoxia 377 brain(stem) death 377–97 cortisol secretion, response to surgical stimulation 396–7 diagnosis 378–97 Academy of Medical Royal Colleges 377–8 ITU patients 378 preconditioning 379–81 electromyographic activity 388 identification, patients admitted in severe coma, organ donation, implications 389 intracranial pressure 386 onset, detection using bispectral index, comatose patients 387–8 organ donation 378, 389–97 public sensitive to this issue 385 testing 380–7 requirements in 80 nations 382–3 (tables) UK, time for reappraisal 384–7 venous oxygen saturation, jugular bulb 387 worldwide 381–4 brainstem reflexes 380–1 hypothermia 381 British Medical Journal, editorial relating to blood transfusion 39–40 brotizolam 130 burns 309 hyperglycaemia 337 butyrylcholinesterase 160 C C-reactive protein (CRP) 67, 338–9 Caesarian section, awareness, risk factor 238 calcium channel blocking agents, pharmacologic myocardial protection, non-cardiac surgery 97– 100 cancer patients, post-operative infection rates, transfusion-associated immunomodulation 50–1 capillary leakage 33 capillary to venous oxygen content (CcO2-CvO2) difference 225 carbon dioxide narcosis 175 cardiac arrest, cerebral hypoxia 377 cardiac assessment (pre-operative), American College of Cardiology 9 cardiac disease (pre-existing), patient undergoing anaesthesia 85 cardiac output monitoring 196–7 accuracy measures 197 less invasive 209–28 measurement (2004) 195 cardiac protection, perioperative 8–11 coronary revascularization, pre-operative 8–9 pharmacological protection 9–11 cardiac stress testing 8 dobutamine stress 92–3 cardiac surgery awareness, risk factor 238 hyperglycaemia after 337
General index
416
cardiac troponin, postoperative, myocardial ischaemia, major vascular surgery, long-term survival 88–90 cardiac wall stabilization devices 7–8 CardioQ 74–6 off-pump CAB surgery, thermo dilution output, comparison with 215–16 cardiogenic pulmonary oedema non-invasive ventilation 359–61 risk factors for intubation as a guide, algorithm 361–2 cardiopulmonary resuscitation Laryngeal Tube 261 supraglottic airway 203 celecoxib, acute myocardial infarction, older patients 135–7 Celecoxib Long-term Arthritis Safety study 136 cell salvage, perioperative 53–4, 56 central venous catheters insertion, ultrasound guidance, NICE recommendations 275–82 cerebral angiography 381, 387–8 cerebral blood flow, arrest 387 cerebral hypoxia 377 chemotherapy, peripherally inserted central catheter 280 children, brain(stem) death criteria criteria 381 determination, clinical course 395–6 chloride solutions, fluid resuscitation, large volumes 3 cholecystectomy, LMA management 253, 254(table) chronic neuropathic pain 133 chronic obstructive pulmonary disease (COPD) non-invasive positive pressure ventilation, co-morbidity and acute decompensations 356–7 non-invasive ventilation 353–8 acute hypercapnic respiratory failure 358 non-invasive vs invasive ventilation, patients with severe acute respiratory failure 355–6 chronic pain syndromes, after acute post-operative pain 140 CK-MB, postoperative, myocardial ischaemia, major vascular surgery, long-term survival 88–90 clonidine 87 pharmacologic myocardial protection, non-cardiac surgery 97–100 closed-loop system 120, 191, 232 clotting proteins 22–3 coagulation abnormalities 22–6 Cobra PLA airway 271–2 Cochrane Injuries Group Albumin Reviewer’s (1998) research 311, 314, 317 Cochrane Injury Group 29–30 cognitive dysfunction, following CABG 6–7, 65–6 colloids coagulation, effect on 320 comparison of one with another 311–22 evidence-based critically ill patients, American Thoracic Society Consensus Statement 320–1 hyperoncotic 21 hypo-oncotic 21 ischaemic brain injury 320 iso-oncotic 21 non-oncotic properties 320 non-protein 30 haemostasis, impaired 4
General index
417
post-operative nausea and vomiting, vs crystalloids, intraoperative administration 28–9 synthetic 19 tissue oxygenation, effects on 34 traumatic brain injury, impact of 320–1 water-binding capacity 21 Combitube 202 common peroneal nerve, ultrasound imaging 291 (fig.) compensatory neurohumoral mechanisms, activation 19 Comprehensive Critical Care (DOH 2000) 300 computed tomography (CT), brain(stem) death 379 conscious sedation 148 ‘conservative’ fluid management 34 coronary angiography 9, 11, 85 coronary artery bypass graft (CABG) surgery 6–9 anaesthetic management 66–7 with cardiopulmonary bypass vs off-pump coronary artery bypass (OPCAB), patients with multivessel disease 63 cognitive dysfunction 6–7 extubation time 67 graft patency 7 heat-conserving/active-warming techniques 67 HES, priming solution, homeostasis after 26 inotropic support 7–8 preconditioning by volatile agents 108 stroke, associated with 6–7 volume expanders, mortality, effect on 30 xenon, neurocognitive deficits 119 coronary artery disease 85 correctable 8–9 coronary plaque characteristics 11 perioperative stabilization 11 statins, role 10 rupture 85–6 coronary revascularization, pre-operative 8–9 coronary stenting 9 anti-platelet treatment 9 myocardial infarction, perioperative 9, 11 patients undergoing non-cardiac surgery two months following 93–4 cortisol secretion, brain dead patient, response to surgical stimulus 396–7 creatine kinase 67 myocardial 67 Creutzfeld-Jacob disease (vCJD) 39–40, 57–8 airway equipment, cross-infection 201 test for, in blood donors 20(table) Critical Care Assembly of the American Thoracic Society 320–1 Critical Care Outreach 297–305 conundrum 304–5 effectiveness 303–4 roles 301–2 Scottish Executive (200) 304 specific goals 300–1
General index
418
teams 300–4 Critical to Success (Audit Commission, 1999) 300 crystalloids blood loss, replacement (3:1 ratio) 4–5 hypercoagulability 34 hypertonic vs near isotonic, fluid resuscitation incritically ill patients 32 isotonic, immunological effects 3–4 metabolic acidosis 34–5 non-haemorrhagic shock resuscitation 321 post-operative nausea and vomiting, vs colloid intraoperative administration 28–9 shifting, after infusion 20–1 tissue oxygenation, effects on 34 crystalloids vs colloids debate 3–5, 19–21, 309 critically ill patients, fluid resuscitation 318–19 edging towards resolution? 309–10 cutaneous hyperalgesia 143 cyclodextrin 119, 124, 131 cyclooxygenase-2 inhibitors (COX-2) 118 coronary artery disease, risk for 139 ischaemic preconditioning, myocardium 136 low-density lipoprotein, effect on 136 post-operative pain control, analgesic efficacy, systemic review 138–9 rheumatoid arthritis 135 rofecoxib withdrawal, role after 133–4 selective, acute myocardial infarction in older patients, relationship between 135–7 CYP2D genotype 164–5 cystic fibrosis, β-2-adrenoceptor gene, polymorphism 157–8 cytochrome P-450 3A4 164 cytochrome P-450 genetics 162–5 genetic testing for enzyme of drug metabolism clinical utility for pain medicine, 163–4 post-operative tramadol analgesia, impact of CYP2D6 genotype 164–5 D DDG 2001 analyser 225–6 deep sedation becomes unconsciousness 198 delirium, after cardiac surgery, predictors 65–6 densitometry, peripheral dye estimation 196 depth of anaesthesia, monitoring 198–200, 231–48 awareness 238–9 fundamental research 232–7 pharmacological approach 239–47 depth of hypnosis, monitoring 232–7 desflurane 123 redheads, requirements 161–2 dexmedetomidine 133, 145–8 lignocaine, added to intra/post-operative analgesia 145 pharmacodynamisc vs remifentanil, crossover comparison of the alangesic effect 146–8 vs remifentanil, crossover comparison of the respiratory effect 145–7 shivering 145 dextran (10%) 21
General index
419
dextrans 19 anti-thrombotic effects 320 bleeding tendency 23 side effects 19 diabetes insipidus 380 diabetes mellitus poorly controlled, microbial infection rate 337–8 preconditioning 11 tissue perfusion 34 type-2, glucagon-like peptide-1 349–50 diazoxide 136 dibucaine inhibition 160 dibucaine number 157 diltiazem, pharmacologic myocardial protection, non-cardiac surgery 97–100 2,3-diphosphoglycerol (2,3 DPG) 6, 45 Diprifusor 174, 246 Diprivan see propofol dipyrone 164 disodium edetate 118 do-not-intubate population, non-invasive ventilation 366–8 dobutamine stress echocardiography 92–3 drug administration, non-invasive routes 181 delivery, optimization 119–20 equipment 121 ‘dry vs wet’ philosophy 32 dural tap 289 E Ecto-5’-nucleotidase 108 EEG approximate entropy, under remifentanil/propofol anaesthesia, pharmacodynamic interaction between 176–7 EEG-based monitors 199–200 hypnotic component of anaesthesia 200 egg lecithin 124–5 elderly surgical patients, major abdominal surgery, different volume replacement strategies 322 elective surgical procedures fluid administration 4–5 prophylactic coronary revascularization before 9 electroencephalography (EEG) 231 bispectral index 234–5 brain(stem) death, confirmatory tests 378, 381, 384 during propofol anaesthesia, age-related effects 223–4 irreversible coma 377 monitors, implementation 232 power spectrum and awareness, analysis of frequencies 235–6 electromyography, interpretation 231 Elisha Airway Device 267–8 eltanolone 123 end tidal carbon dioxide pressure (PETCO2) 175 endothelial nitric oxide synthase 159
General index
420
endotoxaemia, lung capillary permeability, HES effect on 27 endotracheal (ET) extubation, respiratory failure after, non-invasive ventilation 364–6 endotracheal (ET) intubation, alternatives 251–3 enkephalin 166 epidural anaesthesia chronic pain patients, platelet function inhibition by HES 23–4 high thoracic, off-pump coronary artery bypass surgery 71–2 obstetric, learning curves, ultrasound imaging 289 erythrocyte(s) deformability, storage lesion 6 tissue oxygenation, impairment 6 etomidate, induction agent in patients with shock, after fluid resuscitation 189–91 etoricoxib 136, 139 extremity hyperinsulinaemia, muscle protein synthesis stimulation 347 F facemask ventilation, alternatives 368–70 facet joint-mediated pain 286–7 factor VIII 24 febrile transfusion reactions 50 fentanyl 69, 166 intravenous formulation administered orally, children undergoing general anaesthesia 185–6 pharmacokinetic models for prediction of plasma concentration in lean/obese patients 181–3 Fentanyl Oralet (oral transmucosal fentanyl citrate) 185–7 Pick principle 209, 212–15, 224 FiO2 224–5 flow cytometry 24–5 fluid resuscitation in critically ill, choice 309–26 albumin influences total plasma antioxidant capacity 321–2 chloride solution, large volumes, metabolic effects 3 colloids vs crystalloids 318–19 different volume replacement strategies 322 evidence-based colloid use, American Thoracic Society Consensus statement 320–1 human albumin morbidity in hospitalized patients 319–20 and starch administration 323 volume expansion 316–17 hypertonic saline solutions 323–6 inadequate, trauma deaths 5 saline vs albumin 312–16 flumazenil 387 fluoride inhibition 160 fluoroscopic guidance, facet joint pain 286–7 free fatty acids, hyperglycaemia and hyperinsulinaemia, effects on 339–41 fresh frozen plasma 50 acute lung injury 50 G G-protein coupled receptors 104 (fig.), 107–8, 159 GABAA/B receptors 140 gabapentin 133, 139–42
General index
421
gag reflex 380 gastric tonometry 47 critically ill patients 6 gastromucosal tonometry (pH) 27 gelatine (3.5%) 21–2 gelatins 19 haemostasis, effects on 23 platelet dysfunction 4 genetic alterations 155 genetic factors, pre-operative assessment 169 genetic screening 159–60 genotyping, pharmacogenetics 157 Glasgow Outcome Scale score 325–6 global oxygen consumption 47–9 glucagon-like peptide-1 349–50 glucose (blood) levels admission hyperglycaemia, prognostic indicator in trauma 335–6 control and mortality in critically ill patients 333–4 maintenance, insulin treatment 329 myocardial infarction, indicator of death, patients with/without diabetes mellitus 331–2 sepsis 339–41 glucose transport-4, inhibition 330(table) glycaemic control 88 ICU patients 329–50 glycerol 124–5 glycoprotein IIb-IIIa receptor 24–5 ‘goal-directed’ fluid administration 31, 33 ‘goal-directed’ therapy 4–5 GPI-15715 124 propofol, comparative pharmacokinetics and pharmacodynamics 129–30 Guedel airway 268–9 gum elastic bougie, guided LMA-ProSeal insertion 255–7 H haeme oxygenase-1 10 haemodynamic restoration to ‘normal’ 19, 20 (fig.) haemoglobin ischaemic heart disease, when to transfuse 41–3 transfusion trigger 5 trauma patients not requiring intensive care, when to transfuse 44 haemoglobin dissociation curve 225 haemophilia A 155 helmet, non-invasive ventilation 370–2 heparin 69–71 hepatitis B tests 40 (table) hepatitis C antibody test 40 (table) hepatitis C PCR antibody test 40 (table) Hetastarch 23–4 Hextend 23 anti-platelet effect, in vitro study 24
General index
422
HIV antibody test 40 (table) HIV PCR test 40 (table) Holter monitoring 102–3 human albumin 19 critically ill patients, volume expansion and resuscitation 316–17 and HES administration in critically ill patients 323 morbidity in hospitalized patients 319–20 safety, serious adverse events reported worldwide (1998–2000) 29–30 human albumin (20%) 21 human albumin (4%), priming solution for CABG, homeostasis after 26 Human Genome Project 118–19, 155 hydroxyethyl starch (10% HES) solution 21 hydroxyethyl starch (6% HES) solution 21–3 thrombocytopaenic orthotopic liver transplantation, differential platelet receptor expression 25 hydroxyethyl starch (HES) solution 19 acute lung apoptosis 3 acute renal failure, risk for 321 bleeding tendency, increased 23 cardiopulmonary bypass 321 priming solution for, homeostasis after 26 coagulation, effect on 4 different in vivo molecular weights, volume efficacy, orthopaedic surgery 25–6 high-molecular weight, platelets, effect on 24 immunoactivating activity 3 inflammatory response, effect on 27 low-molecular weight, elderly patients 322 lung capillary permeability, effects/mechanisms in endotoxic rats 27 newer generation 23 physico-chemical characteristics 21 (fig.) platelet function inhibition, chronic pain patients undergoing peridural anaesthesia 23–4 sepsis 321 hypercapnic arousal phenomena 146 hyperglycaemia acute illness 329–37 admission hyperglycaemia, prognostic indicator in trauma 335–6 causes 330(table) early hyperglycaemia, mortality in trauma patients 336 mortality in critically ill patients 334–5 hyperinsulinaemia, host response interaction 337–2 post-stroke, associated with infarct expansion 331 hyperglycaemic index, tool to assess glycose control 330 hyperinsulinaemia 337–42 in the presence of euglycaemia (clamping) 342 hypernatraemia 380 hypertonic saline 309 fluid resuscitation 323–6 pre-hospital resuscitation, patients with hypotension and severe traumatic brain injury 324–6 traumatic brain injury 324–6 hypertonic solutions 19 hypertonic volume replacement strategy 31–2 sodium concentration range 31 hypnosis
General index
423
during general anaesthesia, depth assessment, clinical sciences and electronic indices, pitfalls and challenges 237 under remifentanil/propofol anaesthesia, pharmacodynamic interaction between 176–7 hypnotic component of anaesthesia (HCOA) 231, 233–5 ideal monitor 248 monitoring and clinical practice 231, 248 quantifying, clinical/electronic methods, problems 237 hypnotic-anaesthetic drugs 118 hypnotics, new 123–31 hypoalbuminaemia, treatment 311 hypothermia 380 brainstem reflexes 381 hypovolaemia 19 compensatory neurohumoral mechanisms, activation 19 related changes 20 (fig.) shock, treatment with hypertonic or hypertonic-hyperoncotic solutions 31–2 I ibuprofen 139 ideal volume replacement regimen 19–21 ilio-inguinal block 284 immune function, intravenous fluids 3–4 immunomodulation 5–6 blood transfusion, associated 49–51 in-line neck stabilization, Laryngeal Tube insertion 261 indocyanine green dye dilution (IDD) 196 technique 225–6 inflammatory mediators 19 infusion pumps 120 inhaled anaesthetics, new closed-circuit delivery systems 120–1 inotropic support 7–8 insulin 88 free fatty acids, lowering 339–41 intensive therapy 343–9 anti-inflammatory effects, critically ill patients 338–9, 342 Bath insulin protocol, development and audit 344–5 benefit/outcome in critically ill patients, insulin vs glycaemic control 345–6 circulating lipids contribution, improved outcome of critical illness 348–9 endogenous glucose production, critically ill trauma patients 346 intravenous therapy standardization, critically ill patients 344 safe and effective infusion protocols, medical ICU 345 somatotropin axis regulation during protracted critical illness 348 sepsis 339–41 treatment alternatives 349–50 blood glucose levels, maintenance 329 glycaemia control, beneficial effects 329 systemic inflammatory reaction to severe trauma, improvement 341–2 insulin resistance critically ill patients 329, 330(table) treatment, mechanisms 343–9
General index
424
insulin-like growth factor 330 (table) Intensive Care National Audit and Research Centre (ICNARC) 196 interleukin-6 68, 330 (table) hyperglycaemia and hyperinsulinaemia, effects on 339–41 intermittent pulmonary artery termodilution long radial artery catheter 210–12 intra-operative dreaming 198–9 intravenous drug abusers, intravenous access, ultrasound guidance 279–80 intravenous fluid therapy overload 32 perioperative 3–5, 19–35 albumin 29–31 fluid administration, volume and timing 4–5 type of 3–4 hypertonic volume replacement 31–2 ideal substance 20–2 is less/more better? 32–4 plasma substitutes 22–9 restriction, effects on post-operative complications 33–4 IntraVent Orthofix 202 invasive mechanical ventilation 353 iron lungs 369 irreversible coma (coma dépassé) 377 ischaemic cardiac events, perioperative, aetiology 11 ischaemic cardiac preconditioning, in vivo/in vitro 103 ischaemic heart disease, blood transfusion 41–3 isoflurane 136 during CABG 108 isolated forearm technique, detector of awareness 236 J jugular bulb oximetry 387 K κ-opioids 162 keeping the patient ‘dry’ 19 ketamine 118 L labetalol 86 laparoscopic cholecystectomy, airway management 253–5 laparoscopic gastric banding 184 laryngeal mask airway (LMA) during adenotonsillectomy 251 reusable 201–2 single-use 251, 257 safety 201–2 Laryngeal Tube 259–63 cardiopulmonary resuscitation 261 efficacy 259–60 manual in-line, neck stabilization 261
General index
425
restricted neck movement, ease of insertion 260 suction 261–3 mechanically ventilated patients 262–3 Laryngeal Tube Sonda (LTS) 200–1,261 laryngoscopy tolerance 243–6 under remifentanil/propofol anaesthesia, pharmacodynamic interaction between 176–7 lecocytes (allogeneic) transfusion, immunomodulation 5–6 leucodepletion, red blood cells 50–3 leukoreduction 5–6 lithotomy position 254–5 LMA intubating see LMA-Fastrach LMA Soft Seal 257 spontaneously breathing adult patient 257–8 LMA Unique 202, 251, 257 LMA-Classic (cLMA) 200–2, 251–3 Airway Management Device during anaesthesia, randomized comparison with 265–6 endotracheal intubation, alternative for gynaecological laparoscopy 252–3 insertion, success rate 255 pulmonary aspiration, risk of 202, 253–5 spontaneously breathing adult patients 257–8 LMA-Fastrack 251 pulmonary aspiration, risk of 253–4 LMA-Flexible 251 LMA-ProSeal 202, 251 efficacy 251–2 endotracheal intubation, alternative for gynaecological laparoscopy 252–3 gum elastic bougie-guided insertion 255–7 obese patients, undergoing laparoscopic surgery 253 pulmonary aspiration, risk of 253–5 local anaesthetics, nerve blockade 284 locoregional anaesthetic techniques 119 loss of consciousness 198 low output states 19 low-dose endotoxemia in humans, before and during hyperglycaemia 339–41 low-flow, fresh gas flow 121 LTS II 202 lumbar facet nerve block, ultrasound guidance 286–7 lumbar puncture, heparinisation 70 lumiracoxib 139 lung capillary permeability, animal model (endotoxic rats), HES effect on 27 M Magill position, patient’s head and neck 261 malignant hyperthermia 156, 158–9 mannan-binding lectin (MBL) 338–9 mechanically ventilated patients, PreSeal LMA, Laryngeal Tube suction, comparison 262–3 Medical Early Response Intervention Trial (MERIT) study 300, 304 Medical Emergency Team 295–305 calling criteria 299(table) effectiveness 298–300 melanocortin 1-receptor gene 161–2
General index
426
meloxicam 136 meperidine 145 metabolic acidosis, resuscitation, chloride solution, large volumes 3 metoprolol 86 microbial infection rate, hyperglycaemia 337–8 microcirculation 32–3 dysfunction 19, 20(fig.) mid-latency auditory evoked potentials (MLAEPs) 231, 248 monitors 239–41 implementation 232 midazolam 69, 123, 131, 379 cardioprotection 67 minute ventilation 175 mitral anulus, automated cardiac output measurement 218–19 mivacurium 160 atypical response to succinylcholine 157 mivazerol 87 mixed venous oxygen saturation (SvO2) 217–18 µ-agonists 144 µ-opioid receptor polymorphism 165–8 A1 18G polymorphism, morphine requirements in patients with cancer pain 167–8 modified early warning score (MEWS), Critical Care Outreach Liverpool University Hospital 301– 3 molecular genetics for anaesthetist 160–1 morphine 166 off-pump coronary artery bypass surgery 166 spinal (subarachnoid) injection 69–70 morphine-6-glycuronide (M6G) 118, 133, 142–5 induces analgesia in healthy volunteers, pharmacokinetic/pharmacodynamic modelling 144–5 vs morphine, pharmacodynamic effect on hypoxic/hypercapnic breathing in healthy volunteers 143 multi-compartmental pharmacokinetic models 120 multi-organ failure microcirculation impairment 33 microcirculatory dysfunction 19, 20(fig.) myocardial infarction blood glucose level, indicator of death, patients with/without diabetes mellitus 331–2 glucose-insulin-potassium infusion 329 hyperglycaemia 329 perioperative coronary stenting 9 glycaemic control 88 homeostasis, alterations 11 mechanism of 85–6 neuroaxial anaesthesia, reduced incidence 87–8 myocardial ischaemia emergence from anaesthesia 86 perioperative, myocardial cell damage postoperatively, 12-lead ECG 102–3 post-operative, cardiac troponin and CK-MB, major vascular surgery 88–90 protracted 86–7 myocardial oxygen supply 11 myocardial protection (pharmacologic), patients undergoing non-cardiac surgery 97–100
General index
427
N N-methyl-D-aspartate (NMDA) receptor-acting drugs 118 naloxone 387 naproxen 134, 139 effects on life expectancy among patients with rheumatoid arthritis 134–5 Narcotrend 223 National Institute for Clinical Excellence (NICE) ultrasound guidance 203–4 elective central venous catheterization 275–9 National Outreach survey (2002) 300–1 negative pressure ventilation 353 nerve blockade, ultrasound techniques 283 nerve stimulator 284 nerves, visualizing 283–4 neuraxial blockade 87–8, 109 neurological injury, following cardiac surgery 65–6 neuromuscular blocking agents 119 neuropathic cancer pain 139 neutrophils, superoxide production 337–8 NICO2 (partial non-invasive carbon dioxide rebreathing technique) device 224–5 nitrazepam 130 nitric oxide 104(fig.), 107–8 nitroglycerin, pharmacologic myocardial protection, non-cardiac surgery 97–100 non-cardiac surgery, protecting the heart 85–109 β-blockade 86–7 glycaemic control 88 infarction, mechanisms 85–109 preventive strategies 86 sympatholytic therapies 87–8 volatile anaesthetics and ischaemic preconditioning 86–108 non-invasive negative pressure ventilation, acute/chronic respiratory failure 368–9 non-invasive positive pressure ventilation acute/chronic respiratory failure, treatment 368–9 chronic respiratory failure, acute exacerbation, treatment 369–70 delivered through a helmet device 368 respiratory failure after extubation 365–6 non-invasive ventilation 353–4, 353–74 endotracheal intubation, common criteria after non-invasive ventilation 354(table) in: acute cardiopulmonary oedema 359–62 COPD and acute respiratory failure 354–8 do-not-intubate population 366–8 face mask ventilation, alternatives 368–70 with a helmet 370–2 nosocomial infections 362–4 respiratory failure, after extubation 364–6 schema: an approach to applying 373(fig.) studies, common exclusion criteria 354(table) non-steady-state blood flow 174–5
General index
428
non-steroidal anti-inflammatory drugs (NSAIDs) 135–7 NONMEM 241 normal saline 19, 322 elderly surgical patients, undergoing major abdominal surgery 28 versus albumin (4%), fluid resuscitation in the ICU 30–1 normovolaemic ‘well hydrated patient’ 19 nosocomial infection, non-invasive ventilation 362–4 Nunn isoshunt plots 224–5 O obesity 181, 184 epidural anaesthesia, difficulties 289 laparoscopic gastric banding 184 laparoscopic surgery, LMA-ProSeal 253 peripheral intravenous access, ultrasound guidance 279–80 pulmonary aspiration 254–5 rocuronium pharmacodynamic effects when dosed according to real/ideal body weight 184–5 surface landmarks 275 obstetric epidural anaesthesia, learning curves, ultrasound imaging 289 oesophageal Doppler 196, 209, 222–3 monitor (CardioQ) 74–6 off-pump coronary artery bypass (OPCAB) surgery 6–11, 61–78 anaesthetic management 66–72 cardiac output monitoring, during 72–4, 217–18 CardioQ and thermodilution cardiac output, comparison 74–6, 215–16 historical note 61–2 management, intraoperative 7–8 monitoring 72–7 new continuous cardiac output monitor, evaluation 221–2 who benefits? 6–7 why? 62–6 old blood organ failure 44 storage time, red blood cells 45 oncotic pressure, colloid 21 open-loop systems 120 opioid-hypnotic synergy, propofol-remifentanil pharmacodynamic interaction 177–8 opioids 119 acute-chronic pain 165–6 side effects 133 oral hypoglycaemic drugs 11 ORG21465 117, 123 ORG25435 117, 123 Org25969 119 organ donation, brain(stem) death 378, 389–97 end-off-life decisions, Belgian ICU 392–3 rates, decrease, potential organ donor 390, 392, 397 decreased cortisol reserve 396–7
General index
429
identification, an audit of hospital deaths 393–5 relative refusal rate, increase 390 UK national potential donor audit 390–2 organ dysfunction, microcirculatory dysfunction 19, 20 (fig.) organ failure, hypovolaemia 19 organs, vital/non-vital 19 orthotopic liver transplantation, thrombocytopaenia, HES 6% 25 oxprenolol 86 oxycodone/paracetamol 137–8 oxygen consumption index (VO2I) 48(fig.) oxygen delivery index (DO2I) 48(fig.) oxygen free radicals 104(fig.), 107–8 oxygen radical production, hyperglycaemia 337–8 oxygen-carrying capacity, red blood cell transfusion 5 P P-450 family of enzymes 159 pain therapy children 187 future 118 multi-modal approach 148 pancreatitis 309 paracetamol 136, 137(fig.) partial carbon dioxide rebreathing technique, cardiac output assessment, assisted ventilation, critically ill patients 224–5 percutaneous coronary interventions 8–9 percutaneous coronary transluminal angioplasty (PCTA) 90, 91(fig.) perilaryngeal airway (Cobra PLA) 271–2 perioperative care 3–11 blood transfusion therapy 5–6 cardiac protection 8–11 intravenous fluid therapy 3–5 off-pump coronary artery bypass surgery 6–8 perioperative genomics 159 perioperative volume deficits 19 peripheral intravenous access, ultrasound guidance 279–80 peripheral perfusion 19 phantom limb pain 139 pharmacodynamics, clinical applications 173–91 ‘pharmacogenetic disease’ 156, 158–9 pharmacogenetics 118–19 ‘anaesthesia diseases’ 158–9 pharmacogenetics and anaesthesia 165–9 pharmacogenomics 118–19, 155 from double helix to Watson and Crick, 50-years on 160–1 pharmacokinetic mass 181, 183(fig.) pharmacokinetic simulation modules 173 pharmacokinetics clinical applications 173–91 pharmacodynamics relationship, clinical applications, special patient groups 180–1 Pharyngeal Airway xpress (PAxpress) 200, 268–71
General index
430
phenylketonuria 155 Physiometrix 235–6 PiCCO 210–12 piritramide 164 plasma, immunoactivating activity 3 plasma cholinesterase activity 156, 158–9 plasma protein fraction (PPF) 317 Plasma Protein Therapeutics Association 29–30 platelet function colloids, non-protein 4 HES 6%, adverse effect on 25 inhibition, HES, chronic pain patients undergoing peridural anaesthesia 23–4 pneumonia nosocomial 369 ventilator-associated 362 pneumothorax, brachial plexus block 285–6 popliteal artery, ultrasound imaging 291 (fig.) popliteal fossa, ultrasound imaging 291 (fig.) popliteal sciatic nerve block, ultrasound imaging 287–8, 290(fig.) Portex device 202 positive pressure ventilation 353 possible awareness during anaesthesia 198 post-herpetic neuralgia 139, 142 post-traumatic cell injury 3–4 post-operative nausea and vomiting, intraoperative colloid administration 28–9 pre-operative autologous blood donation (PAD) 53, 55–6 pregabalin 133, 139–42 premedication 181 prion protein transmission, airway equipment 201 prion removal filtration 40 (table) procalcitonin 68 propofol 69, 124 administration, target-controlled infusion devices 173–4 albumin, bound to 189 best controllable intravenous hypnotic agent 123 cardioprotection, off-pump coronary artery bypass surgery 67–9 cardiovascular depressive effect 123 EEG, age-related effects 223–4 effect site concentration (Ceprop) 240–1 elderly patients, standard induction dose, level of hypnosis 233–4 GPI 15715, comparative pharmacokinetics and pharmacodynamics 129–30 hypoalbuminaemia 189 induction agent in patients with shock, after fluid resuscitation 189–91 induction of anaesthesia 174 influence of haemorrhagic shock followed by crystalloid resuscitation 188–91 lipid emulsion 123 bacterial growth 118 infection risk 123 lipid intake, ICU sedation 123 lower lipid emulsion compared with standard propofol emulsion pharmacodynamics 125–6 sedation during monitored anaesthesia care 126–7
General index
431
manually controlled infusions 120 new formulation 118, 124 non-steady state, intrinsic ventilatory depressant potency 174–5 pain on injection 118, 123 pharmacodynamics 125 (table), 126 induction speed is not a determinant 179–80 and remifentanil pharmacodynamic interaction 176–8 pharmacodynamic interaction during oesophago-gastro-duodenoscopy in children 187–8 pharmacodynamic interaction regarding hypnosis and EEG approximate entropy 243–6 responsiveness, spectral entropy measurement 242–3 sedation during monitored anaesthesia care 125(table), 126 target-controlled infusion 67 without soya bean oil, pharmacological characteristics and side effects 127–9 prostaglandin I2 136 protein kinase C 104(fig.), 107–8 pseudo-cholinesterase deficiency 156–7 testing 156–7 pseudocholinesterase deficiency 160 public confidence, decline 385 pulmonary arterial thermodilution with direct FICK, arterial transcardiopulmonary thermodilution, comparison with 212–15 pulmonary artery flow 218–19 pressure, resuscitation, chloride solution, large volumes 3 pulmonary artery catheter (PAC) 195–6, 209 complications 195–6 National Heart, Lung and Blood Institute consensus 209 patients: for elective major surgery 209 with sepsis 209 repeatability coefficient 195 risk-benefit ratio 209 severity of illness and risk of death, associated with 226–7 Pulmonary Artery Catheter in Patient Management (PAC-man) study 196 pulmonary aspiration ARDS 253–5 LMA, review of literature 253–5 risk, predisposing factors 254 (table) supraglottic airways 202–3 pulmonary carbon dioxide elimination 196 pulmonary oedema, blood transfusion, associated 5 pulse contour analysis 209–10, 222–3 pulse oximeter 224–5 pulsed dye densitometry 225–6 Q Q wave/non-Q wave myocardial infarction 94 quality control, anaesthesia 117
General index
432
R r2 value 197 real-time pulmonary artery thermodilution 222–3 red blood cells leucodepletion, importance of 50–3 storage time, tissue oxygenation in anaemic critically ill patients, influence of 45–7 transfusion 5 filtered to remove leucocytes, patients undergoing major surgery 51–2 red heads, anaesthetic requirements 161–2 regional anaesthesia, ultrasound guidance 289–92 regurgitation see also pulmonary aspiration 202–3 remifentanil 117, 123 administration, target-controlled infusion devices 173–4 cardioprotection 67 and propofol pharmacodynamic interaction 176–8 pharmacodynamic interaction during oesophago-gastro-duodenoscopy in children 187–8 pharmacodynamic interaction regarding hypnosis and EEG approximate entropy 243–6 responsiveness, spectral entropy measurement 242–3 two target-controlled concentrations, effects on MAC-BAR of sevoflurane 178–9 versus dexmedetomidine, crossover comparison of the respiratory/analgesic effects 145–7 renin-angiotensin system 19 repeatability coefficient 197 respiratory acidosis 3 response entropy 239–43 restricted neck movement, LMA insertion 260 restrictive transfusion strategy, safety for resuscitated and critically ill trauma patients 43–4 resuscitation, organ failure after, circulatory derangements 19, 20 (fig.) rigid bronchoscopy, awareness, risk factor 238 Ringer’s lactate 19, 23, 189, 322 acute lung apoptosis 3 elderly surgical patients, undergoing major abdominal surgery 28 immunoactivating activity 3 post-operative nausea and vomiting 28–9 traumatic brain injury 324–5 RO48–4791 117 RO48–6791 123 rocuronium 119, 252 morbidly obese patients, dosed according to real/ideal bodyweight 184–5 rofecoxib acute myocardial infarction, older patients 135–7 effects on life expectancy among patients with rheumatoid arthritis 134–5 myocardial infarction, risk of 134 thromboembolic adverse events 134 withdrawal 133 ryanadine receptor, genetic variants 160 S SAFE (Saline versus Albumin Fluid Evaluation) study 4–5, 31, 312–16, 326 SARS (severe acute respiratory syndrome) 40 (table)
General index
433
sciatic nerve, ultrasound guidance 287–8 sedative drugs, elimination in critically ill patients 386–7 sepsis, cardiac output ‘optimising’ 4–5 sevoflurane cardioprotection, off-pump coronary artery bypass surgery 67–9 MAC-BAR, two target-controlled remifentanil concentration, effect on 178–9 shock, anaesthetic drugs in 189–91 sickle cell disease 155 single-nucleotide polymorphisms (SNP) 155–8, 160 responses during anaesthesia 157–8 ‘small volume resuscitation’ 31–2 sore throat, after LMA and ET 253 soya bean oil 124–5 spinal cord injury 139 Spongiform Encephalopathy Advisory Committee 201 state entropy 239–43 state of wakefulness 199 statins 11, 91, 109 cardioprotection, perioperative 9–11 cardiovascular events reduction after vascular surgery 100–2 ‘pleiotropic’ effects 10 stents eluting, anti-proliferative drugs 9 thrombosis 9 stroke following CABG 6–7, 65 hyperglycaemia 329, 331 stroke volume, automated cardiac output measurement 218–19 subclavian vein (inflaclavicular axillary vein) cannulation, ultrasound guided 276–7 succinylcholine abnormal responses to 157, 160 mivacurium, delayed recovery 156–7 succinylcholine see also suxamethonium sufentanil, administration, target-controlled infusion devices 173–4 supraclavicular brachial plexus block, ultrasound guidance 285–6 supraglottic airways 251 airway rescue 203 cardiopulmonary resuscitation 203 contraindications for use 202 pulmonary aspiration 202–3 surface landmarks 275 surgical genomics 159 suxamethonium 252 sympathetic nervous system, activation 19 SynchFastSlow 234–5 T T wave inversion 103 target-controlled drug delivery 120
General index
434
infusion systems 120,180 targeted inhaled agent concentration 121 technology (new), anaesthesia and critical care 195–205 termodilution techniques 209–10 cardiac output 73–6 right ventricular intra-extra cardiac shunts 210 valvular regurgitation 210 technical limitations 210 thalium scanning, major vascular surgery, long-term survival 90–1 Therapeutic Arthritis Research and Gastrointestinal Event Trial 139 thiazolidinedione drugs 350 ‘third space’ losses 5, 309 thoracic bioimpendance 209 thromboelastogram 26 thromboxane A2 136 THRX-918661 117, 131 tibial nerve, ultrasound imaging 291 (fig.) tissue oxygen pressure, supplemental perioperative fluid administration 34 tissue oxygen tension 47–9 tissue oxygenation blood transfusion, physiological effect on 47 and perfusion 34 tramadol, post-operative analgesia, impact of CYP296 genotype 164–5 tranexamic acid 54 transcranial ultrasonography 381 transfusion triggers, resuscitated trauma patients, TRICC Trial 43–4 transoesophageal echocardiography aortic flow velocity 217 aortic flow velocity, measurement 196 with colour Doppler 196 colour Doppler, automated cardiac output measurement 218–19 regional wall motion, abnormalities 218 segmental wall motion, monitoring 76–7 transpulmonary arterial thermodilution, long radial artery catheter, intermittent pulmonary artery thermodilution 210–12 trauma deaths, fluid resuscitation, inadequate 5 traumatic brain injury colloids 320–1 Glasgow Outcome Scale Score 325 (fig.) hypertonic saline solutions 324–6 Trendelenburg position 8 triazolam 130 TRICC (Transfusion Requirements in Critical Care) trial 41, 43–4, 58 trigeminal neuralgia 139 troponin I 67, 68 (fig.) TruCath/TruCCOMS monitor 221–2 tumour necrosis factor α 330 (table) hyperglycaemia and hyperinsulinaemia, effects on 339–41 two target-controlled concentration, remifentanil effects on MAC-BAR of sevoflurane 178–9 U
General index
435
ultrasound interventional procedures 275 NICE guidelines 203–4 probes 283–4 services, training and accreditation 282–3 techniques 204 CV access 203–4, 275–83 regional anaesthesia 203–4, 283–92 V valdecoxib 136, 138–9 vaporiser, out-of-circle 121 vascular access ultrasound guidance 275–82 limitations 276 vecuronium 69 verapamil, pharmacologic myocardial protection, non-cardiac surgery 97–100 vestibulo-occular reflex 380 Vioxx 133 VIOXX Gastrointestinal Outcomes Research (VIGOR) Trial 133–5 volatile anaesthetic agents cardioprotective effect 10–11, 67, 88 preconditioning 88 signalling pathways 106 (fig.) volume replacement therapy, adequate 10–20 Von Willebrand factor 26 Von-Willebrand’s (type I)-like syndrome 24 W West Nile virus 40 (table) whole-body electrical bioimpedance (WBEB) 196, 219–21 X xenon 119 neurocognitive deficit, CABG 119 Z zaleplon 117–18, 124 electroencephalographic properties, non-benzodiazepine sedative/hypnotic in rats 130–1 Zeus (anaesthesia machine from Dräger) 191 zopiclone 130
