Prim Care Clin Office Pract 33 (2006) xi–xii
Preface
Robert L. Rogers, MD Joseph P. Martinez, MD Guest Editors
Primary Care physicians (PCPs) are on the front lines of patient care. In addition to treating patients with chronic illnesses, they also evaluate and treat patients with acute complaints. This includes patients with complaints that may be potentially life-threatening, such as headache, chest pain, and abdominal pain. Most of these patients will not have a life-threatening cause of their symptoms. The astute PCP must always be vigilant for that patient who presents with serious disease masquerading as a benign illness. This issue of Primary Care: Clinics in Office Practice focuses on disease entities that absolutely must be recognized by PCPs. While providing a general review of each topic, the guest editors’ goal is to focus on atypical presentations and high-risk patient populations. Most outpatient practice facilities do not have the capability to perform extensive testing to rule out all of these life-threatening conditions. The aim of this issue is to increase awareness of some of these ‘‘masqueraders’’ and to encourage PCPs to always consider the worst case first. In some cases, the worst case may be ruled out simply by considering it and discarding it as a possibility, at which point more benign diagnoses may be considered. An example is the patient who presents with headache. A focused history and physical may be all that is required to rule out subarachnoid hemorrhage, meningitis, temporal arteritis, acute glaucoma, and stroke. At this point, more benign causes such as tension headache or migraine (which are much more common etiologies of headache) may be entertained. Conversely, if a patient’s headache is labeled ‘‘tension’’ or ‘‘migraine,’’ and subarachnoid hemorrhage is never even thought of, you may not get a second chance to make the diagnosis!
0095-4543/06/$ - see front matter Ó 2006 Elsevier Inc. All rights reserved. doi:10.1016/j.pop.2006.06.011 primarycare.theclinics.com
xii
PREFACE
We have all been taught that diagnoses such as pulmonary embolus present very dramatically. This is unequivocally false. As this issue highlights, many serious, life-threatening diseases present quite subtly. Knowledge of risk factors for various diseases and an idea of their atypical presentations will serve the astute clinician well. The job of a PCP is undeniably difficult. As noted above, most patients who present with symptoms of potentially life-threatening disease will actually have a more benign entity. It is simply not feasible or prudent to refer every patient with abdominal or chest pain to the emergency department or a subspecialist. A busy PCP may go days, weeks, or even months without seeing a true emergency. The trick is to avoid becoming complacent to the point that when a patient does have an emergent condition, it is missed. The guest editors sincerely hope that this issue of Primary Care: Clinics in Office Practice will be useful in your day-to-day practice in detecting those emergencies and intervening in a timely fashion. Robert L. Rogers, MD Department of Emergency Medicine The University of Maryland School of Medicine 110 South Paca Street Sixth Floor, Suite 200 Baltimore, MD 21201 E-mail address:
[email protected] Joseph P. Martinez, MD Department of Emergency Medicine The University of Maryland School of Medicine 110 South Paca Street Sixth Floor, Suite 200 Baltimore, MD 21201 E-mail address:
[email protected]
Prim Care Clin Office Pract 33 (2006) 613–623
Hypertensive Urgencies and Emergencies David L. Stewart, MD, MPH, Sharon E. Feinstein, MD, Richard Colgan, MD* Department of Family Medicine, University of Maryland School of Medicine, 29 South Paca Street, Baltimore, MD 21201, USA
Hypertension afflicts 50 million people in the United States and approximately one billion people worldwide. Estimates are that 30% of the population is unaware they have hypertension, while only 34% of those who are being treated for hypertension are at a goal blood pressure measurement of !140/90 mmHg [1]. Although it is common for patients to be seen in the office with poorly controlled hypertension, hypertensive urgencies, and emergencies are less common. It is important for the primary care physician to recognize those patients with hypertensive urgencies and emergencies to avoid missing a life-threatening medical condition as well as not to overtreat elevations in blood pressure that may precipitate unnecessary patient morbidity.
Definitions According to JNC VII guidelines, patients with elevated blood pressures may be characterized into one of three stages: prehypertension, Stage I, and Stage II hypertension (Table 1). The degree to which a patient’s blood pressure is higher than recommended guidelines compels the clinician to further define the blood pressure elevation, with the implication being that more aggressive treatment may be necessary. Terminology for severely elevated blood pressure measurements is confusing. Although clinicians may feel that they know a hypertensive urgency or emergency when they see one, there is lack of uniformity in terms used to describe dangerously high blood pressures. The term ‘‘accelerated hypertension’’ has been used to describe individuals with chronic hypertension who had associated group 3 Keith-Wagener-Barker retinopathy, marked by retinal hemorrhages and exudates. Malignant hypertension has been used to * Corresponding author. E-mail address:
[email protected] (R. Colgan). 0095-4543/06/$ - see front matter Ó 2006 Elsevier Inc. All rights reserved. doi:10.1016/j.pop.2006.06.001 primarycare.theclinics.com
614
STEWART
et al
Table 1 Classification of blood pressure for adults BP classification
SBP mmHg
DBP mmHg
Normal Prehypertension Stage I Stage II
!120 120–139 140–159 Oor ¼ 160
and !80 or 80–89 or 90–99 orOor ¼ 100
describe those individuals with group 4 Keith-Wagener-Barker retinopathy, marked by papilledema. The term hypertensive crisis has been abandoned by some individuals because of confusion. Some interpret hypertensive crisis to mean a hypertensive emergency and others interpret it to mean hypertensive urgency. We prefer the terms normal blood pressure, prehypertension, stage 1 hypertension, and stage 2 hypertension. Stage 2 hypertension may be further delineated as hypertensive urgency and hypertensive emergency. There is no absolute blood pressure that indicates a hypertensive emergency. The typical individual presents with a severe elevation in blood pressure. Severe is defined by a blood pressure measurement O180/120 mmHg according to JNC VII. Key to the diagnosis of hypertensive emergency is evidence of target end-organ damage most commonly affecting the central nervous, cardiovascular, or renal systems. The conditions noted in Box 1 constitute evidence of end-organ damage, in the face of hypertensive emergency [2]. Hypertensive urgency is characterized by a severe elevation in blood pressure that may or may not be associated with symptoms such as severe headache, anxiety, or shortness of breath. The individual with hypertensive urgency has no physical findings that indicate impending target endorgan damage.
Box 1. Conditions constituting evidence of end-organ damage 1. Hypertensive encephalopathy 2. Intracerebral hemorrhage 3. Ischemic heart diseasedthe most common form of target organ damage associated with hypertension a. Acute myocardial infarction b. Acute left ventricle failure with pulmonary edema c. Unstable angina 4. Aortic dissection 5. Eclampsia 6. Stroke 7. Head trauma 8. Life-threatening arterial bleeding
HYPERTENSIVE URGENCIES AND EMERGENCIES
615
The most important consideration for the primary care provider is the clinical context for the patient’s elevated blood pressure. When a patient with a severe elevation in blood pressure is encountered the clinician must entertain whether the elevation is acute or chronic, and whether there is evidence of acute target end organ damage. Causes The most common cause of severe hypertension is an abrupt increase in blood pressure in patients with chronic hypertension. An important factor in hypertensive emergency and urgency is the rate at which the mean arterial pressure rises [3,4]. Noncompliance with medical therapy may therefore play a role in some patients who present with this severely elevated blood pressure [5,6]. Pathophysiology The pathophysiology of severely elevated blood pressure differs, depending on whether hypertensive urgency or emergency is present and in the case of hypertensive emergency the target end organ(s) affected. Normally with a rise in blood pressure, the blood vessels change in diameter to autoregulate the pressure flow, thus limiting damage. When the mean arterial pressure increases abruptly, the body’s ability to hemodynamically adjust to such a rapid change is thwarted, allowing for damage to the end organ [7]. Additional postulates of the mechanisms by which hypertensive emergencies create damage involve the rennin–angiotensin–aldosterone system [8]. Evaluation The role of a primary care doctor in the evaluation of severely elevated blood pressure in an outpatient setting is to distinguish as quickly as possible those patients who require aggressive blood pressure reduction from those who can be managed in the office. Rapid triage of the most concerning patients requires a concise but complete history and physical examination, supported by appropriate and available laboratory data. Although no perfect algorithm exists for the evaluation of a patient with a severely elevated blood pressure in the outpatient setting, initial steps involve the search for the presence of acute target end-organ damage, specifically heart failure, renal failure, encephalopathy, or papilledema, or the presence of any concurrent conditions that would necessitate aggressive blood pressure control such as aortic dissection, acute myocardial infarction, stroke, pheochromocytoma crisis, or preeclampsia [8]. If no hypertensive emergency is identified, there is still a determination to be made between hypertensive urgency that would require rapid but not immediate lowering of blood pressure and severe uncontrolled hypertension that would benefit from tighter long-term
616
STEWART
et al
control but does not require rapid blood pressure reduction. Figure 1 presents a flowchart for the triage of patients in the outpatient setting. Most of the hypertensive urgency and emergency review papers focus on evaluation and treatment in the emergency room setting. The approach in the outpatient setting is necessarily different, and will be limited by resources available in any given office. Because most laboratory data and imaging modalities are not readily accessible, the history and physical become the most critical tools in determining the severity of the situation. History Patients with severe elevations in blood pressure can present at any age, and may be the initial presentation for a previously undiagnosed hypertensive patient. The incidence of hypertensive urgencies and emergencies is Severe Hypertension in the Office (Blood Pressure >180/120)
Evidence of End Organ Compromise? Heart Failure Renal Failure Encephalopathy Papilledema
No
Yes
Concurrent condition which may mandate intensive blood pressure control ? Cardiovascular Aortic Dissection Acute MI
Cerebrovascular SAH/ IC hemorrhage Acute Cerebral Infarction
Other Acute renovascular hypertension Pheochromocytoma Severe Burns Severe Epistaxis Eclampsia
No
Yes
Conservative Management
Consider Transfer to ED
Oral Antihypertensive Therapy
For Rx and Invasive Monitoring
Fig. 1. Algorithm for triage of patients with severe hypertension in the office. (Modified from Gilmore RM, Miller SJ, Stead LG. Severe hypertension in the emergency department patient. Emerg Med Clin North Am 2005;23(4):1141–58; with permission.)
HYPERTENSIVE URGENCIES AND EMERGENCIES
617
greater in the elderly and African Americans. Most patients who present with severely elevated blood pressure are those with chronic essential hypertension who have been noncompliant or inadequately treated [1,5,6]. Therefore, it is critical to gain a thorough hypertension history. This includes current blood pressure medications, any recent changes, compliance with medications, and if the blood pressure had been previously controlled. The patient should be specifically asked if he or she abruptly discontinued a blood pressure medication or if they have had a previous ‘‘crisis’’ [9]. A full medical history, medication list including any nutritional supplements, and use of any illicit drugs will aid the evaluation. It is important to know if a patient currently uses a monoamine oxidase (MAO) inhibitor. Although rarely used today, when combined with foods or medications containing tyramine, MAO inhibitors can trigger a hypertensive crisis [10]. Baseline medical information is critical in assessing the acute nature of the presentation. Any previous renal insufficiency or failure, heart attacks, strokes, and any baseline deficits from a previous stroke should be noted. A patient’s outpatient chart may provide critical information in these areas as well as baseline laboratory data, especially in cases when a patient does not easily recall the information. Next, a patient’s current symptoms should be addressed with a focus on determining whether there is evidence for target end-organ damage. Chest pain, shortness of breath, a prominent apical pulse, or signs of congestive heart failure may indicate cardiovascular involvement. Visual changes, severe headache, dizziness, somnolence, or altered mental status may signify neurologic compromise. Renovascular damage may be evidenced by an acute onset of oliguria or anuria [11]. Severe headache or shortness of breath have been associated with hypertensive urgency as well, but in the presence of elevated blood pressure, these symptoms warrant further investigation to rule out myocardial infarction, congestive heart failure, or stroke [1]. Although a full discussion of the evaluation and treatment of elevated blood pressure in pregnancy is beyond the scope of this article, it is important to note that minor elevations (O140/90 mmHg) warrant further investigation. Hypertension in pregnancy can fall into several categories when it occurs after 20 weeks gestation. Preeclampsia is defined as new-onset elevated blood pressure with the presence of proteinuria (0.3 g or more of protein in a 24-hour urine collection, or R1 þ on a urine dipstick) [12,13]. Women with mild preeclampsia may be asymptomatic at the time of diagnosis [14]. Severe preeclampsia in which the blood pressure is R160/110 mmHg with significant proteinuria (R5 g of protein on 24-hour urine or 3 þ on a dipstick) can be accompanied by oliguria, blurred vision, scotomata, epigastric or right upper quadrant pain, pulmonary edema or cyanosis, impaired liver function, thrombocytopenia, and intrauterine growth restriction [12,13]. The definitive treatment for preeclampsia is delivery; however, the management of preeclampsia requires a careful consideration of the risks to the mother and the fetus [12]. The severity of the illness and the gestational age are
618
STEWART
et al
key components in decision making. If there is no proteinuria present in a patient with new hypertension after 20 weeks gestation, this is considered gestational hypertension (previously known as pregnancy-induced hypertension). Patients with chronic hypertension can develop superimposed preeclampsia before 20 weeks gestation and require special evaluation and monitoring throughout the pregnancy [13]. Physical examination First, the blood pressure measurement must be confirmed. The blood pressure should be rechecked, making sure that the patient is in the proper position and that the appropriate cuff size is used [15]. Blood pressures should be checked in both arms and some advocate measurements in the lower limbs [6]. The physical examination should focus on the organs commonly affected by elevated blood pressure seeking evidence of acute or chronic injury. Although many chronic hypertensives will have findings on fundoscopic examination, focus should be on the presence of acute changes. These include the presence of new hemorrhages (superficial or flame shaped, or deep and punctuate), retinal exudates (hard or ‘‘cotton wool’’), or papilledema [8]. A cardiovascular examination should include signs of heart failure such as pulmonary rales, elevated jugular venous pressure, and an S3 heart sound. Pulse discrepancy among limbs and a new aortic regurgitation murmur may indicate aortic dissection and requires further investigation [11]. Neurologic findings such as a change in mental status, abnormal visual fields, and focal neurologic signs point to possible intracranial hemorrhage, acute ischemic stroke, and hypertensive encephalopathy. A review of a complete history and physical is found in Box 2. Laboratory evaluation The laboratory evaluation in an outpatient setting will be limited by the resources available in any given office. An EKG should be performed in anyone with a severely elevated blood pressure or any signs or symptoms of heart involvement to identify evidence of myocardial ischemia or hypertrophy [16]. If possible, any EKG abnormalities should be compared with a baseline EKG. Some practitioners advocate using a urinalysis in the office to assess for end-organ damage. Hematuria or significant proteinuria may be indicative of acute or progressive renal failure. Gross hematuria is not usually present in a hypertensive emergency, and requires additional workup once the crisis has been managed [17]. A chest radiograph if immediately or quickly available may reveal evidence of pulmonary edema of heart failure. The presence of a widened mediastinum would be concerning for aortic dissection [18]. Further assessment with a head CT or laboratory data such as complete blood count, basic metabolic panel, urine toxicology if urgent,
HYPERTENSIVE URGENCIES AND EMERGENCIES
619
may be triaged to the emergency department. If an emergency does not exist, the above data can be collected in an outpatient setting to assist in longer term care. Treatment A large number of pharmacologic agents now exist to treat essential hypertension, making the need to treat hypertensive urgencies and emergencies less likely than several decades ago. However, the need has not totally disappeared. Hypertensive urgencies and emergencies may account for as many as 27.5% of all medical emergencies presenting to an emergency department and 3% of all emergency room visits. As many as 50% of individuals who present with a hypertensive urgency or emergency will already have a diagnosis of essential hypertension [19,20]. Both hypertensive urgencies and emergencies require immediate medical attention but use different treatment strategies. How the hypertension is treated depends on how the patient presents. The initial challenge for the clinician is to quickly and accurately determine whether the patient has target end-organ damage associated with the severely elevated blood pressure. Appropriate therapeutic decisions start with the triage process in the primary care physician’s office, and include such considerations as those listed in Box 3, which are indicative of office readiness. Hypertensive emergencies develop over hours or days, and require immediate blood pressure reduction, but not necessarily blood pressure normalization, to prevent or limit target end-organ damage. In the office, a patient with a hypertensive emergency may require the clinician to immediately evaluate the patient’s airway, breathing, or circulation because of a stroke, myocardial infarct, or other life-threatening condition. Patients with hypertensive emergency should be transported to the hospital using the emergency medical system. Such patients require continuous monitoring of blood pressure in a hospital setting where parenteral administration of antihypertensive agents are used to reduce mean arterial blood by no more than 25% within minutes to 1 hour of therapy initiation. If the patient is stable, the blood pressure is then titrated to 160/100 to 110 within the next 2 to 6 hours. If the patient remains stable, the blood pressure may be reduced to normal over the next 24 to 48 hours. Caution must be used to avoid excessive lowering of blood pressure, which may compromise renal, cerebral, or coronary blood flow. For this reason short-acting nifedipine is no longer considered acceptable in the initial treatment of severely elevated blood pressures. Caution should be used in patients with ischemic stroke in which there is no clear evidence from clinical trials to support the use of immediate antihypertensive treatment, patients with aortic dissection who should have the systolic blood pressure lowered to !100 mmHg if tolerated, and patients in whom blood pressure is lowered to enable the use of thrombolytic agents [1].
620
STEWART
et al
Box 2. Key aspects of the history and physical examination of the hypertensive patient Hypertension history Last known normal blood pressure, prior diagnosis and treatments, dietary and social factors Cardiac history Previous heart attacks, angina, arrhythmias Symptoms of dyspnea, chest pain, claudication, flank pain, back pain Neurologic history History of prior strokes, neurologic dysfunction Visual changes, blurriness, loss of visual fields, headache, nausea, and vomiting Renal history History of proteinuria, underlying renal disease Changes in urinary frequency Endocrine history Diabetes, thyroid dysfunction, Cushing’s Syndrome Family history Early hypertension in family members, cerebrovascular and cardiovascular disease, diabetes, pheochromocytoma Social history Smoking, alcohol, illicit drugs (especially cocaine, stimulants), noncompliance Medications Steroids, estrogens, sympathomimetics, nutritional supplements (eg, ephedra, mah huang), monoamine oxidase inhibitors (MAO-Is) Nutritional supplements (eg, ephedra, mah huang) Other comorbidities Organ transplant (especially cardiac or renal), current pregnancy (eclampsia or preeclampsia) Physical examination Vital Signs: blood pressure, pulse rate, weight, body habitus, buffalo hump, moon facies Cardiovascular: enlarged heart, presence of S3 heart sounds, asymmetric pulses, arrhythmias Neck: enlarged thyroid, carotid pulses, jugular venous distention Pulmonary: signs of left ventricular dysfunction (crackles, rhonchi)
HYPERTENSIVE URGENCIES AND EMERGENCIES
621
Renal: Presence of renal bruit or abdominal masses Neurologic: abnormal examination (evidence of stroke) Ophthalmologic: funduscopic examination (papilledema, hemorrhage), exophthalmia, lid lag Extremities: diminished or delayed pulses, edema, acromegaly
Adapted from Hsiao EC, Chung SA, Klag MJ. Hypertensive urgency and emergency. In: Cheng A, Zaus A. The Oster medical handbook. 1st edition. Johns Hopkins University; 2003.
Hypertensive urgency develops over days to weeks. In the office, a patient with hypertensive urgency will have a marked elevation in blood pressure with or without symptoms such as headache, shortness of breath, or anxiety. They will have no physical findings indicative of target end-organ damage. The elevated blood pressures in these individuals often make the clinician feel the need to act aggressively. There is no evidence to suggest that patients who present with severe blood pressure elevation and no indication of target end-organ damage have an increased short-term risk when their blood pressure is not aggressively lowered in the clinical setting [1]. The term ‘‘urgency’’ has often lead to aggressive dosing with intravenous or oral antihypertensives to rapidly lower the blood pressure in these individuals. Such an approach is not without risk. Oral or intravenous loading of antihypertensives in the office may lead to cumulative effects of blood pressure lowering that are not experienced by the patient until they have left the office exposing the patient to unnecessary morbidity.
Box 3. Office readiness for hypertensive urgency or emergency 1. An office process to identify and immediately assess blood pressure measurement for those patient with complaints which may indicate target end-organ damage from elevated blood pressure 2. Appropriate office equipment including accurately functioning syphgmomanometers with various blood pressure cuff sizes and electrocardiogram machine(s) 3. An office process to summon providers for emergent clinical need 4. An understanding by all clinicians of medication available for office administration 5. A clear process for activating the emergency system when transport to the emergency room is indicated
622
STEWART
et al
When presented with a patient with hypertensive urgency the clinician has to not only select the appropriate antihypertensive agent but also assess how rapidly the blood pressure should be lowered. If the patient has a current diagnosis of hypertension the clinician may choose to adjust the patient’s medication regimen, or if necessary, address issues of compliance with current medications. Such an approach will generally produce blood pressure lowering, which occurs over days to weeks. For some patients, the blood pressure elevation may warrant the reduction of blood pressure in hours to days. Such patients rarely require hospitalization. The approach is to treat with an oral antihypertensive agent, conduct serial blood pressure measurements in the office, and arrange reasonable short-term follow-up to assure long-term effectiveness of and adherence to treatment [17]. Oral hypertensive agents given in single or multiple dosages may promptly lower the blood pressure, but are associated with variable lag times before onset of action, and may be difficult to titrate because of long durations of action. The ideal pharmacologic agent selected to lower blood pressure with hypertensive urgency should have the following characteristics: no adverse effects on the patients underlying clinical profile, relative short onset of action, and relatively short duration of action. Agents most commonly used to lower blood pressure in the office setting have included clonidine, labetalol, captopril, and minoxidil. Although diuretics have little role in managing hypertensive urgency, they are known to potentiate the effect of other antihypertensive agents, and therefore may have an important role in the overall management of the patient with poorly controlled blood pressure. Once the hypertensive urgency has resolved and the blood pressure is stable, the clinician should investigate possible reasons for the patient’s elevation in blood pressure. Questions should explore factors that affect the patient’s ability to be compliant with prescribed therapy. The clinician should be satisfied that any medication prescribed will be filled and taken by the patient until seen in the office for follow-up. Be aware that financial difficulty may prohibit some patients from obtaining prescribed therapy. Once necessary medication adjustment(s) are made, close follow-up should be arranged for the patient with severely elevated blood pressure. References [1] Chobanian AV, Bakris GL, Henry R, et al. The seventh report of the Joint National Committee on prevention, detection, evaluation, and treatment of high blood pressure: the JNC 7 report. JAMA 2003;289(19):2560–72. [2] The sixth report of the Joint National Committee on prevention, detection, evaluation, and treatment of high blood pressure. Arch Intern Med 1997;157(21):2413–46. [3] Vaughan CJ, Delanty N. Hypertensive emergencies. Lancet 2000;356(9227):411–7. [4] Finnerty FA. Hypertensive encephalopathy. Am J Med 1972;52(5):672–81. [5] Lip GY, Beevers M, Potter JF, et al. Malignant hypertension in the elderly. QJM 1995;88(9): 641–7.
HYPERTENSIVE URGENCIES AND EMERGENCIES
623
[6] Varon J, Marik PE. The diagnosis and management of hypertensive crises. Chest 2000; 118(1):214–27. [7] Kitiyakara C, Guzman NJ. Malignant hypertension and hypertensive emergencies. J Am Soc Nephrol 1998;9(1):133–42. [8] Gilmore RM, Miller SJ, Stead LG. Severe hypertension in the emergency department patient. Emerg Med Clin North Am 2005;23(4):1141–58. [9] Calhoun DA, Oparil S. Treatment of hypertensive crisis. N Engl J Med 1990;323(17): 1177–83. [10] Chase S. Hypertensive crisis. RN 2000;63(6):62–8. [11] Kaplan NM. Hypertensive crisis. In: Zipes DP, Libby P, Bonow RO, Braunwald E, editors. Braunwald’s heart disease: A textbook of cardiovascular medicine. 7th edition. Philadelphia (PA): WB Saunders; 2005. p. 983–5. [12] ACOG Committee on Obstetric Practice. ACOG practice bulletin. Diagnosis and management of preeclampsia and eclampsia. No. 33, January 2002. American College of Obstetricians and Gynecologists. Obstet Gynecol 2002;99(1):159–67. [13] Report of the National High Blood Pressure Education Program Working Group on High Blood Pressure in Pregnancy. Am J Obstet Gynecol 2000;183(1):S1–22. [14] Wagner L. Diagnosis and management of preeclampsia. Am Fam Physician 2004;70(12): 2317–24. [15] Kaplan NM. Clinical hypertension. 7th edition. Baltimore (MD): Williams & Wilkins; 1998. [16] Shayne PH, Pitts SR. Severely increased blood pressure in the emergency department. Ann Emerg Med 2003;41(4):513–29. [17] Elliot WJ. Hypertensive emergencies. Acute Cardiac Care 2001;17(2):435–51. [18] Bales A. Hypertensive crisis; how to tell if it’s an emergency or an urgency. Postgrad Med 1999;105(5):119–30. [19] Cherney D, Straus S. Management of patients with hypertensive urgencies and emergencies: a systematic review of the literature. J Gen Intern Med 2002;17:937–45. [20] Cheng SL. Treating HTN crisis: How long? How fast? RN 2005;68(6):37–41.
Prim Care Clin Office Pract 33 (2006) 625–642
Identifying Chest Pain Emergencies in the Primary Care Setting Michael E. Winters, MDa,b,*, Scott M. Katzen, MDb a
Department of Emergency Medicine, University of Maryland School of Medicine, 110 South Paca Street, 6th Floor, Suite 200, Baltimore, MD 21201, USA b University of Maryland Medical Center, 22 South Greene Street, Baltimore, MD 21201, USA
Patients who have acute chest pain account for approximately 2% of visits to primary care physicians [1]. Thankfully, the majority of these patients have non-life–threatening etiologies. In fact, fewer than 40% of cases are from cardiac, pulmonary, or gastrointestinal causes [2]. The most common etiology of chest pain in the primary care setting is musculoskeletal conditions, accounting for 36% of cases [2]. Nevertheless, a life-threatening etiology must be considered when evaluating any patient who has acute chest pain. Potentially catastrophic conditions that must be excluded include acute coronary syndrome (ACS), aortic dissection (AD), pulmonary embolism (PE), esophageal rupture, and pericarditis. Timely diagnosis of these life-threatening conditions is essential, because any delays in diagnosis or therapy lead to substantially higher morbidity and mortality. Over the past decade, countless studies have been published regarding these catastrophic causes of chest pain. The following will serve to update the primary care physician on current evidence-based information pertaining to ACS, AD, PE, esophageal rupture, and pericarditis. A comprehensive review of each etiology is beyond the scope of this article; rather, it focuses on recognizing atypical clinical presentations, identifying risk factors for disease, defining the utility of the history and physical examination, and understanding the limitations of office-based diagnostic testing pertaining to these five conditions. Armed with this information, the primary care physician can more effectively identify patients who harbor potentially life-threatening etiologies of chest pain.
* Corresponding author. 1406 Chessie Court, Mount Airy, MD 21771. E-mail address:
[email protected] (M.E. Winters). 0095-4543/06/$ - see front matter Ó 2006 Elsevier Inc. All rights reserved. doi:10.1016/j.pop.2006.06.006 primarycare.theclinics.com
626
WINTERS & KATZEN
Acute coronary syndrome Coronary atherosclerotic disease (CAD) is the leading cause of cardiovascular mortality in the United States, accounting for 350,000 to 450,000 out-of-hospital cardiac arrests per year [3]. Currently, there are estimated 13 million Americans living with CAD [4]. In 2001, nearly 1 million individuals in the United States had a new or recurrent coronary event [4]. Currently, ACS accounts for up to 16% of chest pain visits to a primary care physician [2]. Although acute myocardial infarction makes up just 1.5% of cases [2], it is imperative that the primary care physician identify patients who have ACS. Mortality for patients sent home with ACS is twice as high as that of patients admitted to the hospital [5,6]. Clinical presentation Clearly, ACS would be considered in the patient presenting with chest tightness that is difficult to localize, radiates to the left arm, is associated with dyspnea, and is relieved with rest or nitroglycerin. Unfortunately, up to 25% of patients who have ACS do not have classic clinical presentations [7]. Patients who have atypical presentations of ACS are more likely to be elderly, female, or diabetic [8]. Many will have different descriptions of chest discomfort, such as burning, ‘‘heartburn,’’ a knot in the center of the chest, a lump in the throat, or a bandlike sensation across the chest [8]. In addition to atypical descriptions of pain, patients frequently have primary complaints other than chest pain. Dyspnea, diaphoresis, vomiting, near syncope, and fatigue are common presenting complaints in patients who have atypical presentations of ACS. In fact, fatigue is the most common presenting symptom of elderly patients who have an ACS [8]. According to the Global Registry of Acute Coronary Events (GRACE) trial, almost 10% of patients who have ACS do not complain of chest pain [9]. Thus the primary care physician must be suspicious of ACS across a myriad of patient populations and complaints. Radiation of chest pain must be interpreted with caution. Although clinicians commonly ask about radiation of discomfort to the left upper extremity, radiation to the right upper extremity is more predictive of ACS. Radiation of chest pain to the right shoulder increases the likelihood of ACS threefold, whereas radiation to both shoulders results in a sevenfold increase in the likelihood of ACS [7]. In a prospective study of emergency department patients who had chest pain, 48 out of 51 patients who had radiation of pain to the right arm were diagnosed with ACS [10]. Factors that aggravate or alleviate chest discomfort can be misleading. Chest discomfort in the setting of eating is typically associated with gastrointestinal etiologies. Postprandial pain, however, can be a marker for significant CAD, namely left main occlusion or severe three-vessel disease [11]. Similar to the onset of pain with eating, chest discomfort that is relieved with antacids is commonly ascribed to a gastrointestinal disorder.
IDENTIFYING CHEST PAIN EMERGENCIES
627
Unfortunately, relief of pain with antacids has been shown to be unreliable in differentiating cardiac and gastrointestinal causes of chest pain [8,12]. Risk factors Established risk factors for CAD include age over 65, male gender, cigarette use, hypertension, dyslipidemia, and diabetes mellitus. In addition to these, it is crucial that that clinician know emerging risk factors for ACS. Systemic lupus erythematosus (SLE), rheumatoid arthritis (RA), and HIV have been identified as independent risk factors for CAD. For patients who have SLE, the incidence of CAD is fifty times greater than in agematched controls [13]. It is hypothesized that the combination of vascular inflammation, immune-mediated vasculopathy, increased incidence of dyslipidemias, and hypercoagulability contribute to the development of CAD in patients who have SLE [13,14]. It is for these same reasons that patients who have RA are at increased risk for CAD [15]. Although HIV is not traditionally associated with CAD, postmortem studies have demonstrated significant atherosclerotic heart disease among HIV-positive patients [16]. Furthermore, the majority of patients in these studies lacked the traditional risk factors previously discussed. Reasons postulated for the increased incidence of CAD among patients who have HIV include endothelial cell dysfunction, chronic inflammation, increased incidence of hypertriglyceridemia, and the use of highly active antiretroviral therapy [9]. Physical examination There are no physical examination findings that reliably diagnose ACS. Nonetheless, a complete physical examination, including vital signs, is necessary in patients who have suspected ACS. The presence of hypotension, although predictive of a catastrophic cause of chest pain, is nonspecific. Cardiopulmonary findings that increase the likelihood of ACS include an S3 or S4 gallop, a new mitral insufficiency murmur, or signs of congestive heart failure such as pulmonary rales and elevated jugular venous pressures [8]. Chest pain that is reproducible upon palpation of the chest wall should not immediately be attributed to a musculoskeletal etiology. Up to 15% of patients who have an acute myocardial infarction will have reproducible chest pain upon palpation [17]. Electrocardiogram All patients who have suspected ACS should have an electrocardiogram (ECG) performed as soon as possible. Failure to perform an ECG is one of the most common reasons for physicians to lose malpractice litigation involving patients who have ACS [5]. Elevation of the ST segment is the most sensitive and specific marker for ACS, and is seen in 30% to 40% of patients [18]. In patients who have 1 mm or more of new ST segment
628
WINTERS & KATZEN
elevation, the prevalence of acute myocardial infarction is 80% [5]. Additional ECG signs of ACS include flat or downsloping ST segment depression or T-wave inversions. Up to 20% of patients who have new ST segment depression or T-wave inversion have an acute myocardial infarction [5]. Thus patients who have ST segment depression or segmental T-wave inversion must be considered to have ACS until proven otherwise. Finally, a normal ECG does not rule out ACS as an etiology of chest pain. Anywhere from 20% to 33% of patients who have ACS have a normal initial ECG [19]. Office-based treatment Once ACS is suspected, preparations should be made for immediate patient transport to the closest emergency department. While awaiting transfer, patients should be placed on a cardiac monitor and given supplemental oxygen. In addition, patients should receive a chewable aspirin as soon as possible. With the exception of hypotensive patients and those taking sildenafil, sublingual nitroglycerin should be administered for persistent pain. In patients who have no absolute contraindications, a beta-blocker can be given, because it decreases the progression of AMI by 13% [18].
Aortic dissection Aortic catastrophes result in 16,000 deaths per year in the United States [20]. For AD, the incidence ranges from 5 to 30 cases per million per year [21–24]. Unfortunately, the diagnosis is missed in up to 38% of patients on initial evaluation [21]. Reasons cited for missed AD include the failure to recognize atypical presentations, the inability to identify risk factors for AD, inattention to key physical examination features, and the failure to understand the limitations of diagnostic imaging. Mortality for untreated AD is significant. For acute ascending AD, mortality increases 1% to 3% per hour during the first 24 hours following symptom onset [21,25,26]. If the diagnosis is not made, mortality reaches 30% at 48 hours, 40% by day 7, and 50% at 1 month [27]. For patients who have an uncomplicated descending AD, mortality is lower, reaching 10% by day 30 [27]. Clinical presentation The classic description of AD is the instantaneous onset of chest pain that is maximal at its inception, is described as sharp or tearing, and radiates to the interscapular region. Regrettably, this classic presentation is often the exception rather than the rule. Anywhere from 16% to 26% of patients do not recall an instantaneous onset of pain [28]. According to data from the International Registry of Acute Aortic Dissection, only 50% of patients who had an AD described pain as tearing or ripping [22]. Furthermore, the location of pain is dependent upon the segment of aorta that is involved. Patients with
IDENTIFYING CHEST PAIN EMERGENCIES
629
a dissection of the ascending aorta commonly report anterior chest pain, whereas patients who have involvement of the descending aorta frequently complain of posterior chest, back, or abdominal pain [21]. In addition to varying locations and descriptions of pain, more than one third of patients present with symptoms attributable to secondary organ involvement [29]. The two most common organ systems involved in patients who have AD are the cardiovascular and neurologic systems. Cardiovascular involvement is reported to occur in 18% to 50% of cases of AD [21,30,31]. More often this occurs in dissections of the ascending aorta, resulting in acute congestive heart failure (CHF), cardiac tamponade, myocardial infarction, or cardiogenic shock. Neurologic symptoms are reported in 18% to 30% of patients with AD [21,32,33]. Although signs of cerebral ischemia are the most common, symptoms attributable to spinal cord ischemia can be seen in up to 10% of patients [21,32]. Patients who have AD and spinal cord ischemia may present with quadriplegia, paraplegia, or unilateral parasthesias. Additional clinical presentations of AD include syncope, dysphagia, and hoarseness. Risk factors Male gender, advanced age, and chronic hypertension are established risk factors for AD. In fact, the incidence of disease is highest among male patients between the ages of 50 and 70 years [21,25,34]. Depending on the study population, hypertension is present in almost 80% of patients who have AD [21,30]. Additional recognized risk factors for AD include smoking, hyperlipidemia, aortic instrumentation, connective tissue syndromes, bicuspid aortic valve, coarctation of the aorta, decelerating trauma, and inflammatory conditions such as giant cell and Takayasu’s arteritis [21]. Pregnancy and cocaine use have recently been identified as risk factors for AD. Aortic dissection in pregnancy occurs more often during the third trimester and the initial stages of labor. In fact, 50% of dissections in women less than 40 years of age occur during pregnancy [21]. Physical examination Classic physical examination findings associated with AD include hypertension, blood pressure differentials, pulse amplitude deficits, and a new diastolic murmur. Many of these findings are transient, and thus absent in a significant percentage of patients with AD. Up to 25% of patients who have AD present with a systolic blood pressure less than 100 mm Hg [21,25,35]. Similarly, pulse amplitude deficits are found in just 25% to 30% of patients [21,28,36]. Although bilateral blood pressure measurements are standard in the evaluation for AD, a difference of 20 mm Hg between extremities is seen in up to 20% of normal individuals [25]. Fewer than one third of patients who have AD have a diastolic murmur [28]. Thus results of the physical examination in patients with suspected AD must be
630
WINTERS & KATZEN
interpreted with caution. A diagnosis of AD should never be excluded based upon a normal physical examination. Diagnostic evaluation Imaging modalities used in the evaluation of AD consist of chest radiography (CXR), computed tomography (CT), magnetic resonance imaging (MRI), transesophageal echocardiography (TEE) and aortography. Classic CXR findings of AD include a widened mediastinum, widening of the aortic knob, an aortic wall shadow that extends beyond intimal calcification, a double aortic shadow, and a left pleural effusion. Unfortunately, CXR has limited utility in the evaluation of AD. The overall sensitivity and specificity for CXR are reported to be just 64% and 86%, respectively [37]. Furthermore, 10% to 40% of patients who have an AD have a normal initial CXR [27]. In most centers, helical CT scanning is the imaging modality of choice for AD. Diagnostic accuracy of CT for AD approaches 100% [37]. Advantages to CT include its accessibility, quick study times, and the ability to provide accurate measurements. Although the sensitivity and specificity for MRI ranges from 95% to 100%, technical limitations of long study time, poor patient access, restricted cardiac monitoring, and lack of immediate availability limit its use in the detection of AD [21]. In experienced hands, TEE is an acceptable imaging modality, because sensitivity is reported to be as high as 98% [21,38]. Limitations to TEE include the inability to visualize the aorta below the celiac axis and the strong dependence on operator experience. Aortography, although long considered the gold standard in diagnosing AD, is invasive and time-consuming. Sensitivity of aortography for AD ranges from 86% to 88%, whereas specificity ranges from 75% to 94% [21]. Office-based treatment As with ACS, once AD is suspected arrangements should be made for immediate transfer to the closest emergency department. Patients should be placed on a cardiac monitor and intravenous access obtained. Antihypertensive therapy should be started immediately, except in patients who are hypotensive. Goals of antihypertensive therapy are to reduce the force of left ventricular contractions, decrease the rise of aortic pulse wave, and decrease the systemic arterial pressure to as low as tolerable [21]. Standard antihypertensive therapy includes the combination of a beta-blocker and vasodilator, namely sodium nitroprusside. To prevent rebound tachycardia, the betablocker should be started before nitroprusside. Labetolol is recognized as an acceptable alternative to combination therapy [21]. Pulmonary embolism Pulmonary embolism is the third leading cause of cardiovascular mortality in the United States [39–41]. Although exact figures are unknown, it is
IDENTIFYING CHEST PAIN EMERGENCIES
631
estimated that the incidence of PE in the United States exceeds 1 per 1000 [39,42]. This translates into over 600,000 episodes each year, accounting for approximately 100,000 to 200,000 deaths [43,44]. Unfortunately, many cases of PE are not diagnosed. In fact, it is commonly cited that only 30% of emboli are diagnosed antemortem [45,46]. Mortality for undetected, and thus untreated, PE is significant, and ranges from 18% to 35% [47]. Although recent literature indicates mortality rates for ambulatory patients are lower [46], untreated PE remains a significant concern. It is imperative that primary care physicians be knowledgeable regarding current concepts and controversies in the evaluation of patients who have suspected PE. Clinical presentation Arguably, the most important step in diagnosing PE is to consider it as a diagnostic possibility. As such, the clinician must recognize that the clinical presentation of PE is variable and often nonspecific. Patients who have PE frequently present with such nonspecific symptoms as palpitations, cough, anxiety, lightheadedness, back pain, hiccoughs, or syncope [39,48]. In fact, syncope is reported in up to 13% of patients who have PE [39,49]. Moreover, PE may masquerade as an exacerbation of a chronic medical condition, namely CHF or chronic obstructive pulmonary disease (COPD) [50]. Because these illnesses predispose patients to venous thromboembolic (VTE) disease, PE must always be considered as an etiology of a CHF or COPD exacerbation. Although chest pain, dyspnea, and hemoptysis are classically associated presentations, they are unreliable and inconsistent findings among patients with PE. In a study of 500 patients, Miniati and colleagues [51] reported that 22% of patients who had PE did not complain of chest pain. Similarly, dyspnea can be absent in up to 8% of patients [39,52]. Fewer than 10% of patients who have PE have hemoptysis as a component of their clinical presentation [39,40,49]. Thus, the lack of these classic clinical presentations should not be used to exclude the diagnosis. The primary care physician must maintain suspicion for PE across a myriad of patient complaints. Risk factors Once PE is considered a diagnostic possibility, a directed history and physical examination must be performed. Central to the history in patients who have suspected PE is the identification of risk factors for VTE disease. Failure to assess and document risk factors is one of the most frequently cited reasons for missed PE [53]. Classic risk factors include malignancy, trauma, recent surgery, prolonged immobilization, pregnancy, oral contraceptive therapy, diseases that alter blood viscosity, inherited thrombophilias, and a prior history of VTE disease [39,40,42,43]. Of these, a prior history of VTE is the strongest risk factor for current disease [39,54]. Although textbooks emphasize these classic risk factors, it is important to
632
WINTERS & KATZEN
recognize that advanced age, obesity, hypertension, cigarette smoking, and medical conditions such as pneumonia, stroke, and CHF are risk factors for VTE disease [42]. In fact, obesity is the most common reversible risk factor for PE in our society [42]. Unfortunately, many patients who have suspected PE reveal no risk factors for disease, despite an exhaustive search. This should not dissuade the physician from pursuing a diagnostic evaluation, because up to 12% of patients who have PE lack any identifiable risk factors [39,55]. Physical examination Physical examination findings in patients who have suspected PE must be interpreted with caution. Many patients with PE, especially young patients and those who have normal cardiopulmonary reserve, can appear deceivingly well despite a large clot burden. Furthermore, classic vital sign abnormalities such as tachypnea and tachycardia are inconsistent findings among patients who have PE. In fact, tachypnea is absent in up to 13% of patients diagnosed with PE [39,46,54]. Similarly, tachycardia is absent in up to 30% of patients over 40 years of age [39]. In patients younger than 40 years of age who have PE, up to 70% have normal heart rates [39,56]. Additional examination findings that can be misleading include fever, wheezing, rales, or signs of consolidation. Stein and coworkers [57] reported that up to 14% of patients who have PE are febrile upon presentation. Nonetheless, no physical examination finding is considered sensitive or specific for PE. Clinical prediction rules Before pursuing a diagnostic evaluation, the clinician must classify the patient into low, intermediate, or high clinical probability of disease. Patients classified as low probability have a disease prevalence ranging from 5% to 10%, whereas those with an intermediate probability have a prevalence of 25% to 40% [43]. Patients felt to have a high clinical probability have a prevalence of PE ranging from 70% to 90% [43]. Classifying patients into one of these three groups has traditionally been based upon clinical gestalt. Within the past several years numerous scoring systems have been published in an attempt to standardize the clinical assessment of probability. The oldest and most frequently used scoring system is the Wells clinical prediction rule [58]. Although criticized for its subjectivity regarding the likelihood of alternative diagnoses, the Wells rule has been prospectively validated and can be used in the outpatient setting. Other scoring systems, although useful, require results from the electrocardiogram, chest radiography, or arterial blood gas measurement, making their application more challenging in the outpatient setting. Unfortunately, there is currently no consensus on the best clinical prediction rule for PE. For clinicians who chose not to use a standardized scoring system, clinical gestalt remains a reasonable alternative. In a recent review, clinical gestalt was found to have
IDENTIFYING CHEST PAIN EMERGENCIES
633
similar accuracy in discriminating among patients who had low, intermediate, or high probability of PE [59]. Electrocardiogram The primary role of the electrocardiogram in patients with suspected PE is to exclude alternative diagnoses, namely acute myocardial infarction. Unfortunately, the majority of patients who have PE have nonspecific abnormalities, thereby limiting the utility of the electrocardiogram. The most common abnormality seen in patients who have PE is T-wave inversions, most notably in leads V1 to V4 [39,60]. Additional nonspecific abnormalities include ST segment changes, left or right axis deviations, sinus tachycardia, and right bundle branch block [48]. The classic S1Q3T3 abnormality is neither sensitive nor specific for PE, and is found in only a small percentage of patients. Pulmonary embolism should never be excluded simply because patients lack an S1Q3T3 abnormality on their electrocardiogram. D-dimer D-dimer is a cross-linked degradation product formed by plasmin, and is a marker of endogenous fibrinolysis [46]. It is elevated in nearly all patients with pulmonary embolism [42,43]. Unfortunately, it is also elevated in a number of other conditions, such as pregnancy, trauma, postoperative states, cancer, inflammatory conditions, and advancing age [43]. Thus, D-dimer is a sensitive, but not specific, serum marker for PE. Over the past several years multiple D-dimer assays have been developed and evaluated. At present, the most sensitive D-dimer assay is the standard, or rapid, enzyme-linked immunosorbent assay (ELISA). Latex agglutination and whole-blood assays are less sensitive and should not be used in the evaluation of patients who have suspected PE. The greatest utility of D-dimer is in the outpatient evaluation of the patient who has low clinical probability of PE. Several studies have demonstrated that a normal ELISA d-dimer level, combined with a low clinical probability, can safely rule out PE. Dunn and colleagues [61] demonstrated a 99.6% negative predictive value for a normal ELISA D-dimer level in patients suspected of having PE. These results indicate that a normal D-dimer level is as predictive as a normal ventilation-perfusion scintigraphy or negative duplex ultrasound [46]. For patients who have intermediate or high clinical probability, there is currently not enough evidence to safely rule out the diagnosis of PE based upon a normal D-dimer level. Diagnostic imaging Available imaging modalities used in the evaluation of patients who have suspected PE include CXR, ventilation-perfusion scintigraphy, helical CT, venous ultrasound, and pulmonary angiography. Like that of the
634
WINTERS & KATZEN
electrocardiogram, the utility of chest radiography lies in its ability to exclude alternative diagnoses such as pneumothorax. Unfortunately, the chest radiograph is often abnormal, with the findings nearly always nonspecific. Common abnormalities include atelectasis, pulmonary infiltrates, elevation of the hemidiaphragm, pleural effusion, and cardiomegaly [48]. Perhaps most worrisome is that up to 12% of patients who have PE have an entirely normal chest radiograph [62]. Pulmonary embolism should be strongly suspected in patients presenting with cardiopulmonary complaints and a normal chest radiograph. Ventilation-perfusion scintigraphy has been used for over 3 decades in the evaluation of the patient who has suspected PE. When the results are definitive, normal or high probability, ventilation-perfusion scanning is useful. Unfortunately, the majority of patients suspected of having PE do not have definitive studies. In fact, anywhere from 50% to 70% of scans are indeterminate [61]. Thus, ventilation-perfusion scanning fails to provide a definitive answer in almost 70% of patients [46,61]. Furthermore, ventilation-perfusion scanning has been shown to have poor interobserver correlation [63]. It is for these reasons that many centers have replaced ventilation-perfusion scanning with helical CT in the evaluation of patients with suspected PE. Helical CT has become the preferred diagnostic imaging modality in patients who have suspected PE. Advantages to CT over ventilation-perfusion scanning include the ability to directly visualize thrombi, to detect abnormalities that may indirectly indicate thrombus, and to provide an alternative etiology for the presenting symptoms [43]. Initial studies using CT reported widely varying degrees of sensitivity and specificity, ranging from 57% to 100% and 78% to 100% respectively [43]. Nearly all of these studies used older-generation, single-slice conventional scanners in which false negative rates up to 30% have been reported [64]. Current-generation multidetector-row CT scanners have dramatically faster scanning times, less motion artifact, higher resolution, and better peripheral visualization. As a result of this improved technology, current CT scanners can image down to sixth order branches, significantly increasing the detection of subsegmental pulmonary emboli [64]. Thus far, animal studies demonstrate that CT is at least as accurate as pulmonary angiography in the detection of small peripheral emboli [64]. In addition to improved visualization, the interobserver agreement for CT is significantly higher than that of ventilation-perfusion scanning [63]. Outcome data are mounting on the safety of withholding anticoagulation in patients with a negative helical CT [65,66]. Based upon several retrospective and prospective studies, the negative predictive value of a normal CT scan approaches 98% [67]. In fact, the frequency of PE diagnosed after a normal CT scan is lower than for a normal or low-probability ventilation-perfusion scan [63,64]. Combining CT venography of the lower extremities, or lower extremity venous ultrasound, with chest CT for a more thorough evaluation for VTE disease is a common practice in many medical
IDENTIFYING CHEST PAIN EMERGENCIES
635
centers. The feasibility of this approach has been demonstrated in several studies [64,68]. The Prospective Investigation of Pulmonary Embolism Diagnosis II (PIOPED-II) study is a recently completed prospective multicenter study that will serve to solidify the role of CT in the evaluation of patients with suspected PE. The utility of lower extremity venous ultrasonographic screening in patients who have suspected PE, but without leg symptoms, is controversial. Venous ultrasound has been shown to be insensitive for diagnosing DVT in asymptomatic patients [64,69]. Furthermore, venous ultrasound may provide false positive results, or may detect residual abnormalities from a previous thrombosis. As such, the possibility of embolism cannot be reliably excluded based upon a negative lower extremity ultrasound. With the technological advances in CT scanning, the use of pulmonary angiography in the evaluation of patients who have suspected PE is declining. Long hailed as the gold standard diagnostic imaging modality for PE, pulmonary angiography is not without risks or limitations. Morbidity and mortality rates for the procedure range from 3.5% to 6% and 0.2% to 0.5%, respectively [64]. Major nonfatal complications include hematoma at the puncture site, respiratory failure, and contrast-induced nephropathy [43]. Furthermore, pulmonary angiography requires expertise in both performance and interpretation, which is often not available at many centers. Office-based treatment Once the primary care physician suspects PE as a diagnostic possibility, arrangements should be made for transport to the closest emergency department. Patients should be placed on a cardiac monitor and given supplemental oxygen. Intravenous access should be obtained and the patient should receive intravenous fluids. For patients in whom PE is considered the most likely diagnosis, heparin therapy should be initiated while pursuing definitive imaging. Although low-molecular–weight heparin preparations have advantages over unfractionated heparin, either is sufficient for the initial treatment of patients who have suspected PE.
Esophageal perforation Esophageal perforation is the most fatal perforation of the gastrointestinal tract [70]. Often, the diagnosis is delayed or missed entirely. Any delay in diagnosis leads to unacceptably high morbidity and mortality. When the diagnosis is delayed by just 24 hours, mortality exceeds 60% [71]. Reasons cited for delays in diagnosis include the failure to recognize atypical presentations, a failure to identify patient risk factors, and a failure to understand the limitations of traditional diagnostic imaging. Although esophageal perforation accounts for a small percentage of cases, the primary care physician must be able to identify this life-threatening etiology of chest pain.
636
WINTERS & KATZEN
Clinical presentation The clinical presentation of esophageal perforation is dependent upon the location of perforation. Patients who have a thoracic esophageal perforation present with the classic symptoms of chest pain, dyspnea, and odynophagia. Patients who have a cervical esophageal perforation, however, may only complain of neck pain, neck spasm, hoarseness, or dysphagia [71,72]. Those who have perforation of the intra-abdominal esophagus may only have midepigastric discomfort, thereby potentially leading to a misdiagnosis of pancreatitis, gastritis, peptic ulcer disease, or gastroesophageal reflux disease. With such a variable clinical presentation, it is easy to see how even the seasoned clinician can miss the diagnosis of esophageal perforation. Risk factors Identified risk factors for esophageal perforation include recent medical procedures, foreign body ingestion, caustic ingestion, trauma, esophageal malignancy, and immunocompromised states. Medical procedures are the greatest risk factor for esophageal perforation, accounting for up to 75% of cases [71,72]. Procedures that have a risk of perforation are pneumatic dilatation, sclerotherapy, upper endoscopy, surgical procedures of the head and neck, nasogastric tube insertion, and endotracheal intubation [71,72]. Patients who are profoundly immunocompromised, such as those who have advanced HIV, are at risk for esophageal perforation. These patients have a higher incidence of esophageal infection due to Candida, herpes simplex virus, and Mycobacterium tuberculosis, which have been shown to predispose patients to perforation [73]. Physical examination Classic physical examination findings in patients who have esophageal perforation are fever and subcutaneous emphysema. Unfortunately, both are inconsistent and unreliable findings in patients who have perforation. In an institutional review of cases of esophageal perforation, Eroglu [74] found that only 33% were febrile upon presentation. In patients who have a cervical esophageal perforation, only 60% have subcutaneous emphysema on examination [71]. For those who have a thoracic perforation, just 30% have evidence of subcutaneous emphysema [71]. Therefore, a normal physical examination does not rule out the possibility of esophageal perforation. Diagnostic evaluation Definitive diagnosis of esophageal perforation is challenging. Because the majority of patients report pain somewhere in the thoracic region, CXR is often obtained as the initial imaging modality. Chest radiograph abnormalities suggestive of perforation include a new pleural effusion, pneumothorax,
IDENTIFYING CHEST PAIN EMERGENCIES
637
or pneumomediastinum [70]. These CXR abnormalities, however, may take several hours to become evident on plain radiography. As a result, up to 12% of patients who have a perforation have a normal initial CXR [70]. To exclude an esophageal perforation, most primary care physicians rely on a water-soluble, contrast-enhanced esophagram. Because of the mechanical properties of water-soluble agents, false-negative results can occur. In fact, anywhere from 15% to 25% of esophagrams are initially negative in patients who have an esophageal perforation [70,71,75]. If clinical suspicion remains high, many radiologists recommend obtaining a CT scan. CT has been shown to compliment contrast-enhanced esophageal radiography, and is able to detect additional findings of perforation such as periaortic air, mediastinal fluid collections, and esophageal thickening [76]. Office-based treatment Once an esophageal perforation is suspected, rapid transport to the closest emergency department should be arranged. While awaiting transfer, patients should be placed on cardiac monitor, given supplemental oxygen, and intravenous access should be obtained. Definitive treatment for a perforation centers on emergent surgical repair. If available, broad spectrum antibiotics should be administered, but should not delay timely transport. Antibiotics should be given intravenously and cover gram-positive, gram-negative, and anaerobic organisms. Pericarditis Although generally not an immediate life-threatening condition, pericarditis must be considered in patients who have acute chest pain. Often, patients with pericarditis develop a pericardial effusion that, if severe, can result in cardiac tamponade and hemodynamic collapse. In addition to an effusion, patients can develop constrictive pericarditis, thereby mimicking tamponade physiology. The exact incidence of pericarditis is difficult to determine, because disease prevalence is dependent upon the study population. For patients presenting to an emergency department with acute chest pain, pericarditis accounts for approximately 5% of cases [77]. Clinical presentation Acute chest pain is the most common presenting symptom of pericarditis [78]. Pain is typically described as the sudden onset of sharp, anterior chest pain that is exacerbated by cough, deep inspiration, or the supine position. Although this classic presentation is helpful, it is not present in all patients who have pericarditis. Often, the pain caused by pericarditis is similar to the symptoms of an ACS. Furthermore, radiation of pain in pericarditis is common. In fact, pain that radiates to the trapezius ridge is frequently seen in pericarditis caused by inflammation of the phrenic nerve [77,79]. Patients
638
WINTERS & KATZEN
who have significant amounts of pericardial fluid may present with nonspecific complaints. More often these patients complain of generalized weakness, dyspnea with exertion, cough, or even dysphagia [80]. Risk factors Although the majority of cases are caused either by viral or idiopathic etiologies, a number of medical conditions place patients at risk for pericarditis. These include collagen vascular diseases such as systemic lupus erythematosus and rheumatoid arthritis, renal insufficiency, malignancy, and recent myocardial infarction [77]. In addition, a thorough review of medications is necessary, because several medications have been shown to incite pericarditis. Physical examination The textbook cardiac examination finding in patients who have pericarditis is the pericardial friction rub. It is classically described as a highpitched, scratchy sound that is heard best at the left sternal border during expiration [77]. A rub is reportedly 100% specific for the diagnosis of pericarditis [79]. Unfortunately, a pericardial rub often varies in intensity and intermittently disappears, thereby limiting its overall sensitivity. Thus the diagnosis of pericarditis should not be excluded based upon the absence of a friction rub. In addition to listening for a rub, the clinician should look for signs of cardiac tamponade. ‘‘Beck’s triad’’ of tamponade is hypotension, muffled heart sounds, and jugular venous distention. If tamponade is suspected, the patient should be examined for the presence of pulsus paradoxus. Although considered the hallmark of cardiac tamponade, the prevalence of pulsus paradoxus ranges from 12% to 75% of patients who have tamponade [79]. Electrocardiogram Electrocardiographic abnormalities in pericarditis are classically divided into four stages. Unfortunately, fewer than 50% of patients demonstrate all four stages on their ECG [81]. The most common abnormalities seen in pericarditis are diffuse ST segment elevation and PR segment depression. These findings are absent in up to 20% of patients who have acute pericarditis [77]. Depending upon the study, up to 16% of patients who have pericarditis have an entirely normal ECG [82]. As such, acute pericarditis should not be excluded based upon a normal ECG. Office-based treatment Unlike the previously discussed life-threatening etiologies of chest pain, patients who have acute pericarditis often do not require immediate hospitalization. For those that have viral or idiopathic etiologies, the course of
IDENTIFYING CHEST PAIN EMERGENCIES
639
pericarditis is generally benign. These patients can effectively be treated with nonsteroidal anti-inflammatory medications [83]. In recent years, a number of studies have attempted to risk stratify patients who have acute pericarditis. Patients at high risk of complications should be admitted for observation and further treatment. Prognostic signs that indicate patients at high risk include subacute onset, immunosuppressed states, trauma, oral-anticoagulation therapy, serologic evidence of myopericarditis, severe pericardial effusion, or evidence of tamponade [83]. To evaluate for the presence of a pericardial effusion, it is recommended that all patients with suspected pericarditis receive an echocardiogram [77].
Summary The majority of patients presenting to a primary care physician with acute chest pain will have non-life–threatening etiologies. Nevertheless, catastrophic cause of chest pain such as ACS, AD, PE, esophageal perforation, and pericarditis must be considered in the differential diagnosis. Often, these deadly conditions have atypical clinical presentations that must be recognized. Furthermore, the physical examination can be deceptively benign in patients harboring a catastrophic etiology of chest pain. By identifying these atypical presentations, recognizing the utility of the physical examination, and understanding of the limitations of traditional diagnostic imaging, primary care physicians can effectively diagnose patients who have life-threatening cause of acute chest pain. References [1] Woodwell DA. National ambulatory medical care survey: 1998 summary. Adv Data 2000; 19:1–26. [2] Klinkman MS, Stevens D, Gorenflo DW. Episodes of care for chest pain: a preliminary report from MIRNET. Michigan Research Network. J Fam Pract 1994;38:345–52. [3] Callans DJ. Out-of-hospital cardiac arrestdthe solution is shocking. N Engl J Med 2004; 351(7):632–4. [4] Clinical performance measures: chronic stable coronary artery disease. American College of Cardiology, American Heart Association, Physician Consortium for Performance Improvement. Available at: www.acc.org/clinical/measures/CAD/cadmeasures.pdf. Accessed: December 1, 2005. [5] Lee TH, Goldman L. Evaluation of the patient with acute chest pain. N Engl J Med 2000; 342(16):1187–95. [6] Lee TH, Rouan GW, Weisberg MC, et al. Clinical characteristics and natural history of patients with acute myocardial infarction sent home from the emergency department. Am J Cardiol 1987;60:219–24. [7] Panju AA, Hemmelgarn BR, Guyatt GH, et al. Is this patient having a myocardial infarction? JAMA 1998;280(14):1256–63. [8] Meisel JL. Diagnostic approach to chest pain in adults. Availabale at: http://www.utdol. com/utd/content/topic.do?topickey¼pri_card/2346&type¼A&selectedTitle¼1~88. Accessed December 2005.
640
WINTERS & KATZEN
[9] Brieger D, Eagle KA, Goodman SG, et al. Acute coronary syndromes without chest pain, an underdiagnosed and undertreated high-risk group. Insights from the Global Registry of Acute Coronary Events. Chest 2004;126:461–9. [10] Berger JP, Buclin T, Haller E, et al. Right arm involvement and pain extension can help to differentiate coronary diseases from chest pain of other origin: a prospective emergency ward study of 278 consecutive patients admitted for chest pain. J Intern Med 1990;227:165–72. [11] Berlinerblau R, Shani J. Postprandial angina pectoris: clinical and angiographic correlations. J Am Coll Cardiol 1994;23:627–9. [12] Castrina FP. Unexplained noncardiac chest pain. Ann Intern Med 1997;126:663–4. [13] Mattu A, Petrini J, Swencki S, et al. Premature atherosclerosis and acute coronary syndrome in systemic lupus erythematosus. Am J Emerg Med 2005;23:696–703. [14] Karrar A, Sequiera W, Block JA. Coronary artery disease in systemic lupus erythematosus: a review of the literature. Semin Arthritis Rheum 2001;30:436–43. [15] Chung CP, Oeser A, Raggi P, et al. Increased coronary-artery atherosclerosis in rheumatoid arthritis. Arthritis Rheum 2005;52(10):3045–53. [16] Passalaris JD, Sepkowitz KA, Glesby MJ. Coronary artery disease and human immunodeficiency virus infection. Clin Infect Dis 2000;31:787–97. [17] Aufderheide TP, Gibler WB. Acute ischemic coronary syndromes. In: Rosen P, Barkin R, editors. Emergency medicine concepts and clinical practice. 4th edition. St. Louis (MO): Mosby; 1998. p. 1655–716. [18] Masud SP, Mackenzie R. Acute coronary syndrome. J R Army Med Corps 2003;149(4): 303–10. [19] Norell M, Lythall D, Coghlan G, et al. Limited value of the resting electrocardiogram in assessing patients with recent onset chest pain: Lessons from a chest pain clinic. Br Heart J 1992;67(1):53–6. [20] National Center for Health Statistics. Vital statistics of the United States, 2000. Available at: http://wonder.cdc.gov/wonder/sci_data/mort/mcmort/type_txt/mcmort03/techap99.pdf. Accessed December 1, 2005. [21] Khan IA, Nair CK. Clinical, diagnostic, and management perspectives of aortic dissection. Chest 2002;122(1):311–28. [22] Hagan PG, Nienaber CA, Isselbacher EM, et al. The International Registry of Acute Aortic Dissection (IRAD): new insights into an old disease. JAMA 2000;283:897–903. [23] Eisenberg MJ, Rice SA, Paraschos A, et al. The clinical spectrum of patients with aneurysms of the ascending aorta. Am Heart J 1993;125:1380–5. [24] Bickerstaff LK, Pairolero PC, Hoiler LH, et al. Thoracic aortic aneurysms: a population based study. Surgery 1982;92:1103–8. [25] Rogers RL, McCormack R. Aortic disasters. Emerg Med Clin North Am 2004;22(4): 887–908. [26] Pitt MP, Bonser RS. The natural history of thoracic aortic aneurysm disease: an overview. J Card Surg 1997;12:270–8. [27] Nienaber CA, Eagle KA. Aortic dissection: new frontiers in diagnosis and management: Part I: From etiology to diagnostic strategies. Circulation 2003;108(5):628–35. [28] Haro LH, Krajicek M, Lobl JK. Challenges, controversies, and advances in aortic catastrophes. Emerg Med Clin N Am 2005;23:1159–77. [29] Khan IA. Clinical manifestations of aortic dissection. Journal of Clinical and Basic Cardiology 2001;4:265–7. [30] Hennessy TG, Smith D, McCann HA, et al. Thoracic aortic dissection or aneurysm: clinical presentation, diagnostic imaging and initial management in a tertiary referral center. Ir J Med Sci 1996;165:259–62. [31] Rahmatullah SI, Khan IA, Nair VM, et al. Painless limited dissection of the ascending aorta presenting with aortic valve regurgitation. Am J Emerg Med 1999;17:700–1. [32] Alverez Sabin J, Vazquez J, Sala A, et al. Neurologic manifestations of dissecting aneurysm of the aorta. Med Clin (Barc) 1989;92:447–9 [in Spanish].
IDENTIFYING CHEST PAIN EMERGENCIES
641
[33] Prendes JL. Neurovascular syndromes in aortic dissection. Am Fam Physician 1981;23: 175–9. [34] Auer J, Berent R, Eber B. Aortic dissection: incidence, natural history and impact of surgery. Journal of Clinical and Basic Cardiology 2000;3:151–4. [35] Garcia-Jiminez A, Paraza-Torres A, Martinez-Lopez G, et al. Cardiac tamponade by aortic dissection in a hospital without cardiothoracic surgery. Chest 1993;104:290–1. [36] von Kodolitsch Y, Schwartz AG, Nienaber CA. Clinical prediction of acute aortic dissection. Arch Intern Med 2000;160:2977–82. [37] Higgins CB. Modern imaging of the acute aortic syndrome. Am J Med 2004;116:134. [38] Vignon P, Gueret P, Vedrinne JM, et al. Role of transesophageal echocardiography in the diagnosis and management of traumatic aortic disruption. Circulation 1995;92:2959–68. [39] Laack TA, Goyal DG. Pulmonary embolism: an unsuspected killer. Emerg Med Clin North Am 2004;22:961–83. [40] Wells PS, Rodger M. Diagnosis of pulmonary embolism: when is imaging needed? Clin Chest Med 2003;24:13–28. [41] Gillum RF. Pulmonary embolism in the United States, 1970–1985. Am Heart J 1987;113: 1262–4. [42] Goldhaber SZ, Elliott CG. Acute pulmonary embolism: Part I, epidemiology, pathophysiology, and diagnosis. Circulation 2003;108 2726–29. [43] Fedullo PF, Tapson VF. The evaluation of suspected pulmonary embolism. N Engl J Med 2003;349:1247–56. [44] Dalen JE, Alpert JS. Natural history of pulmonary embolism. Prog Cardiovasc Dis 1975;17: 259–70. [45] Morgenthaler TI, Ryu JH. Clinical characteristics of fatal pulmonary embolism in a referral hospital. Mayo Clin Proc 1995;70:417–24. [46] Boie ET. Initial evaluation of chest pain. Emerg Med Clin North Am 2005;23:937–57. [47] Calder KK, Herbert M, Henderson SO. The mortality of untreated pulmonary embolism in emergency department patients. Ann Emerg Med 2005;45:302–10. [48] Tapson VF. Acute pulmonary embolism. Cardiol Clin 2004;22:353–65. [49] Brilakis ED, Tajik AJ. 82-year old man with recurrent syncope. Mayo Clin Proc 1999;74: 609–12. [50] Bosomworth J. Diagnosing pulmonary embolism in a rural setting. Can J Rural Med 2005; 10:100–6. [51] Miniati M, Prediletto R, Fromichi B, et al. Accuracy of clinical assessment in the diagnosis of pulmonary embolism. Am J Respir Crit Care Med 1999;159:864–71. [52] Susec O, Boudrow D, Kline JA. The clinical features of acute pulmonary embolism in ambulatory patients. Acad Emerg Med 1997;4:891–7. [53] Sullivan D. Pulmonary embolismdmedical error and risk reduction part I. Available at: www.thesullivangroup.com. Accessed December 5, 2005. [54] Colucciello SA. Protocols for deep venous thrombosis: a state of the art review: Part I: Risk factor assessment, physical examination, and current diagnostic modalities. Emerg Med Rep 1999:13–24. [55] Morgenthaler TI, Ryu JH. Clinical characteristics of fatal pulmonary embolism in a referral hospital. Mayo Clin Proc 1995;70:417–24. [56] Green RM, Meyer TJ, Dunn M, et al. Pulmonary embolism in younger adults. Chest 1992; 101:1507–11. [57] Stein PD, Afzal A, Henry JW, et al. Fever in acute pulmonary embolism. Chest 2000;117: 39–42. [58] Wolf SJ, McCubbin TR, Feldhaus KM, et al. Prospective validation of Wells Criteria in the evaluation of patients with suspected pulmonary embolism. Acad Emerg Med 2004;44(5): 503–10. [59] Chunilal SD, Eikelboom JW, Attia J, et al. Does this patient have pulmonary embolism? JAMA 2003;290(21):2849–58.
642
WINTERS & KATZEN
[60] Ferrari E, Imbert A, Chevalier T, et al. The ECG in pulmonary embolism: predictive value of negative T waves in the precordial leadsd80 case reports. Chest 1997;111:537–43. [61] Dunn KL, Wolf JP, Dorfman DM, et al. Normal d-dimer levels in emergency department patients suspected of acute pulmonary embolism. J Am Coll Cardiol 2002;40:1475–8. [62] Ramzi DW, Leeper KV. DVT and pulmonary embolism: Part I: Diagnosis. Am Fam Physician 2004;69(12):2829–36. [63] Blachere H, Latrabe V, Montaudon M, et al. Pulmonary embolism revealed on helical CT angiography: comparison with ventilation-perfusion radionuclide lung scanning. American Journal of Roentgenology 2000;174:1041–7. [64] Schoepf UJ, Goldhaber SZ, Costello P. Spiral computed tomography for acute pulmonary embolism. Circulation 2004;109:2160–7. [65] Tillie-Leblond I, Mastora I, Radenne F, et al. Risk of pulmonary embolism after a negative spiral CT angiogram in patients with pulmonary disease: a 1-yr follow up study. Radiology 2002;223:461–7. [66] Moores LK, Jackson WL, Shorr AF, et al. Meta-analysis: outcomes in patients with suspected pulmonary embolism managed with computed tomographic pulmonary angiography. Ann Intern Med 2004;141:866–74. [67] Ost D, Rozenshtein A, Saffran L, et al. The negative predictive value of spiral computed tomography for the diagnosis of pulmonary embolism in patients with nondiagnostic ventilation-perfusion scans. Am J Med 2001;110:16–21. [68] Cham MD, Yankelevitz DF, Shaham D, et al. Deep venous thrombosis: detection by using indirect CT venography: the Pulmonary Angiography-Indirect CT Venography Cooperative Group. Radiology 2000;216:744–51. [69] Mac Gillavry MR, Sanson BJ, Buller HR, et al. Compression ultrasonography of the leg veins in patients with clinically suspected pulmonary embolism: is a more extensive assessment of compressibility useful? Thromb Haemost 2000;84:973–6. [70] Singh H, Warshawsky ME, Herman S, et al. Spontaneous esophageal rupture: Boerhaave’s syndrome. Clinical Pulmonary Medicine 2003;10(3):177–82. [71] Duncan M, Wong RKH. Esophageal emergencies: things that will wake you from a sound sleep. Gastroenterol Clin North Am 2003;32(4):1035–52. [72] Williamson W, Ellis H. Esophageal perforation. In: Taylor MB, editor. Gastrointestinal emergencies. 2nd edition. Baltimore (MD): Lipincott, Williams & Wilkins; 1997. p. 31–49. [73] Grubbs BD, Baldwin DR, Trenkner SW, et al. Distal esophageal perforation caused by tuberculosis. J Thorac Cardiovasc Surg 2001;121:1003–4. [74] Eroglu A. Esophageal perforation: the importance of early diagnosis and primary repair. Dis Esophagus 2004;17:91–4. [75] Gimenez A, Franquet T, Erasmus J, et al. Thoracic complications of esophageal disorders. Radiographics 2002;22:S247–58. [76] Ghanem N, Altehoefer C, Springer O, et al. Radiological findings in Boerhaave’s syndrome. Emerg Radiol 2003;10:8–13. [77] Lange RA, Hills LD. Acute pericarditis. N Engl J Med 2004;351(21):2195–202. [78] Braunwald E, Zipes DP, Libby P. Heart disease: a textbook of cardiovascular medicine. 6th edition. Philadelphia: W.B. Saunders Company; 2001. p. 1823–76. [79] Spodick DH. Acute pericarditis: current concepts and practice. JAMA 2003;289(9):1150–3. [80] Spodick DH. Acute cardiac tamponade. N Engl J Med 2003;349(7):684–90. [81] Marinella MA. Electrocardiographic manifestations and differential diagnosis of acute pericarditis. Am Fam Physician 1998;57:699–710. [82] Bruce MA, Spodick DH. Atypical electrocardiogram in acute pericarditis: characteristics and prevalence. J Electrocardiol 1980;13:61–6. [83] Imazio M, Demichelis B, Parrini I, et al. Day-hospital treatment of acute pericarditis: a management program for outpatient therapy. J Am Coll Cardiol 2004;43(6):1042–6.
Prim Care Clin Office Pract 33 (2006) 643–657
Evaluation of the Dyspneic Patient in the Office Saiyad Sarkar, MD, Pamela J. Amelung, MD* Division of Pulmonary and Critical Care Medicine, Department of Medicine, University of Maryland School of Medicine, 10 N. Greene Street 3D-122, Baltimore, MD 21201, USA
Dyspnea literally means disordered breathing. It is a term generally applied to the sensations experienced by individuals who complain of unpleasant or uncomfortable respiratory sensations. Dyspnea occurs in healthy individuals (ie, with exercise or at high altitude), but respiratory patients experience dyspnea at lower levels of exercise or altitude. Dyspnea may be considered part of the warning system for humans to recognize when they are at risk of receiving inadequate ventilation. Dyspnea has been shown to be an independent predictor of mortality [1], and has been found to be related to quality of life more than pulmonary function tests [2]. A precise or widely accepted definition of dyspnea does not exist, because patients use an array of terms to describe their breathing sensation, and the term dyspnea represents a number of qualitatively distinct sensations. The specific descriptive words used by patients to describe their breathing may provide insight into the underlying pathophysiology of their disease [3,4]. The American Thoracic Society has defined dyspnea in the following way: ‘‘Dyspnea is a term used to characterize a subjective experience of breathing discomfort that is comprised of qualitatively distinct sensations that vary in intensity. The experience derives from interactions among multiple physiological, psychological, social and environmental factors, and may induce secondary physiological and behavioral responses’’ [5]. For the purpose of this article, the authors use the terms ‘‘dyspnea’’, ‘‘breathlessness,’’ and ‘‘shortness of breath’’ interchangeably.
* Corresponding author. E-mail address:
[email protected] (P.J. Amelung). 0095-4543/06/$ - see front matter Ó 2006 Elsevier Inc. All rights reserved. doi:10.1016/j.pop.2006.06.007 primarycare.theclinics.com
644
SARKAR & AMELUNG
Prevalence of dyspnea Breathing discomfort is a common symptom experienced by patients, and can be quite distressing. It has more than 30 attributed causes involving multiple organ systems [6]. There are no precise data on the prevalence of dyspnea. The actual scope of the problem varies among clinical settings and patient subgroups. Population-based surveys have estimated the prevalence to be between 17% and 38% [7–9]. In an ambulatory setting the prevalence of dyspnea was 3.7% [10]. Morbidity associated with dyspnea can range from minor to disabling. Dyspnea is often associated with cardiac and respiratory diseases, but it can also be caused by obesity and deconditioning. As these conditions increase, the population with the complaint of dyspnea will rise too.
Mechanisms of dyspnea The difficulty in determining the cause of shortness of breath results from the complex and poorly understood pathophysiology involved in the production of the sensation of breathlessness. The mechanism of dyspnea originates with the activation of sensory systems involved with respiration. One or more receptors can be individually or collectively stimulated to initiate an afferent signal to the central nervous system (CNS). This afferent impulse is then transmitted to the central nervous system. Here the message is processed and efferent impulse is directed to the respiratory system. The major receptor sites considered in the sensation of dyspnea include chemoreceptors, mechanoreceptors, and lung receptors. Chemoreceptors are located both centrally (medulla) and peripherally (carotid and aortic bodies), and are responsible for detection of changes in oxygen and carbon dioxide. Stimulation of these receptors causes changes that adjust breathing to maintain blood gas and acid-base homeostasis. An increase in carbon dioxide stimulates central receptors and results in an increase in ventilation. Hypoxia stimulates respiration through its effects on the peripheral chemoreceptors, which may cause breathlessness in patients who have underlying lung disease. It has also been observed that supplemental oxygen administration relieves dyspnea in some patients who have lung disease, even in the absence of any changes in ventilation. Various mechanoreceptors, including chest wall receptors, may also be important in the generation of the sensation of dyspnea [11]. Upper airway receptors can modify the sensation of dyspnea, based on clinical observations that patients sometimes report a decrease in the intensity of their shortness of breath when sitting by a fan or open window. Afferent information from pulmonary vagal receptors project to the CNS and are important in shaping the pattern of breathing and to the sensation of dyspnea [12,13]. These impulses, with afferent impulses generated from various receptors, are received and processed in the CNS. The motor cortex or brainstem respiratory neurons are thought
EVALUATION OF THE DYSPNEIC PATIENT IN THE OFFICE
645
to transmit a signal to the sensory cortex, which might contribute to a ‘‘sense of effort’’ to breathing [14,15]. This sensation increases whenever the central motor command is increased or whenever the respiratory muscles become weak or fatigued. Based on response to the afferent information, the CNS sends an efferent impulse via the phrenic nerve to the diaphragm and other respiratory muscles to increase respiration. It is not completely understood how this impulse affects different aspects of breathing. In 1963 the concept of ‘‘length-tension inappropriateness’’ was proposed by Campbell and Howell [16]. According to their theory, there was a disassociation between the force or tension generated by the respiratory muscles and the lung volume generated by that force. Currently, mismatch between afferent information to the CNS and outgoing motor command is thought to lead to the sensation of breathlessness. The afferent feedback from peripheral sensory receptors may allow the brain to assess the effectiveness of the motor commands issued to the ventilatory muscles; that is, the appropriateness of the response in terms of flow and volume for the command. When changes in respiratory pressure, airflow, or movement of the lungs and chest wall are not appropriate for the outgoing motor command, the intensity of dyspnea is heightened. In other words, dissociation between the motor command and the mechanical response of the respiratory system may produce a sensation of respiratory discomfort. Today the theory has been generalized to include not only information arising in the ventilatory muscles, but information emanating from receptors throughout the respiratory system, and has been termed ‘‘neuromechanical dissociation’’ [14,17]. Approach to patients who have dyspnea In the outpatient setting, dyspnea is a common symptom. Establishing a cause can at times be difficult, because dyspnea is a symptom of a myriad of disorders, ranging from the completely benign to the relatively serious; however, asthma, congestive heart failure (CHF), chronic obstructive pulmonary disease (COPD), pneumonia, cardiac ischemia, psychogenic, and interstitial lung disease have been found in some series to account for approximately 85% of all cases of shortness of breath [18–22]. For a dyspneic patient presenting to the office, the initial goal is to determine the severity of the dyspnea and the need for urgent intervention such as intubation. The clinical approach to the patient depends on the acuteness of the problem. Although most patients who are unstable usually present to local hospitals, primary care physicians must be prepared and equipped to triage, manage, and stabilize patients who have acute dyspnea. Evaluation of acute dyspnea For acute shortness of breath, an initial quick evaluation should consist of assessment of airway patency and auscultation of the lungs. The
646
SARKAR & AMELUNG
breathing pattern and rate should be determined and use of accessory muscles noted. Cardiac rhythm, vital signs, and pulse oximetry should be monitored. The mental status should be evaluated, and a brief history of cardiac and pulmonary disease obtained if not already known. Unstable patients typically have hypotension, hypoxemia, tracheal deviation, altered mental status, unstable arrhythmia, stridor, retractions, cyanosis, or absent breath sounds signaling the acuity of their problem. These patients should be administered oxygen, have intravenous (IV) access established, and be given initial treatment as appropriate (bronchodilators, diuretics). Consideration should be given to needle decompression if the initial diagnostic impression is pneumothorax, and to intubation if necessary. Unstable patients should be transported to the closest emergency department for further evaluation and treatment. Trained health care personnel should accompany the patient and continue management until supervision is transferred to the emergency department team. For acute problems, the differential diagnosis is relatively narrow; they are in general related to disease of the respiratory or cardiac systems. The differential diagnosis of these two systems will cover the most common diseases encountered with acute shortness of breath: COPD, asthma, pneumonia, pulmonary embolism, pneumothorax, heart failure, and myocardial infarction. There are many other physical causes of acute shortness of breath, however, including acute renal failure, diabetic ketoacidosis, septicemia, or other metabolic acidoses with respiratory compensation. Acute dyspnea may also be psychogenic in origin, although this should be a diagnosis of exclusion (Table 1). Once an emergent situation has been excluded, the assessment of the stable patient who complains of acute shortness of breath includes a medical history, physical examination, and appropriate laboratory testing. A comprehensive patient history is the starting point for evaluating dyspnea. It is imperative to characterize the dyspnea in terms of descriptive qualities, onset, frequency, intensity, duration, triggers (exposures), provoking activities (ambulation, eating, changing position), associated respiratory symptoms, and strategies or actions (medications, positions) that provide relief. Intermittent dyspnea may be caused by asthma or heart failure, whereas persistent or progressive dyspnea suggests other chronic conditions, such as COPD, interstitial fibrosis, or pulmonary hypertension. Nocturnal dyspnea is may be indicative of asthma, CHF, or gastroesophageal reflux. Dyspnea occurring independent of physical activity suggests possible psychological etiology, or possibly allergic or mechanical problems. Dyspnea occurring mainly after exercise suggests exercise-induced asthma. A complete history should emphasize any coexisting cardiac and pulmonary symptoms. For example, the presence of cough may imply asthma or pneumonia; cough combined with a change in character of sputum may be caused by exacerbation of COPD. The symptoms of fever, sore throat, and acute dyspnea may suggest epiglottitis. Chest pain during dyspnea
647
EVALUATION OF THE DYSPNEIC PATIENT IN THE OFFICE
Table 1 Acute dyspnea: etiologies and characteristics Disorder
History findings
Cardiac: CHF, ACS, arrhythmia, anemia pericarditis COPD exacerbation
Chest pain, orthopnea, PND, edema, palpitations
Asthma exacerbation
Pneumonia
Pulmonary embolism
Pneumothorax
Upper airway obstruction: laryngospasm, aspirated foreign body Psychogenic: hyperventilation, anxiety, panic attacks
Worsening dyspnea, increased sputum volume, increased sputum purulence History of asthma, allergy history, increased reliance on beta-agonists, chest tightness Fever, cough, purulent sputum
Pleuritic chest pain, lower extremity pain/swelling, predisposing risk factors Pleuritic chest pain
History of choking, gurgling respirations, persistent pneumonias Emotional upset, feeling impending doom, neurotic personality
Physical examination
Chest radiograph
Cyanosis, crackles, edema, JVD, murmurs, S3 or S4, HJR, hypertension Pursed-lip breathing, wheezing, barrel chest, decreased breath sounds, prolonged expiratory phase Wheezing, cough, tachycardia, prolonged expiratory phase
Cardiomegaly, pleural effusion, interstial edema
Fever, crackles, decreased breath sounds, increased fremitus Wheezing, friction rub, lower extremity swelling
Parenchymal infiltrate
Unilateral hyperresonance, absent breath sounds, tracheal shift Stridor, wheezing
Air in pleural space with collapsed lung, shift of mediastium
Sighing
Hyperinflated lungs
Hyperinflated lungs
Normal, atelectasis, effusion, wedge-shape density
Visualized foreign body, air trapping, hyperinflation
Normal
Abbreviations: ACS, acute coronary syndrome; CHF, congestive heart failure; HJR, hepatojugular reflux; JVD, jugular venous dyspnea; PND, paroxysmal nocturnal dyspnea.
648
SARKAR & AMELUNG
may be caused by coronary or pleural disease, depending on the quality and description of the pain. Pleuritic chest pain could be caused by pneumothorax, pulmonary embolism, pneumonia, or pleuritis. Palla and colleagues [23] found that dyspnea or tachypnea with pleuritic chest pain occurred in 97% of patients who had clinically significant pulmonary embolism. Anginal chest pain accompanied by shortness of breath may signify ischemia associated with left ventricular dysfunction. Lusiani and coworkers [24] found that parxosymal dyspnea or pulmonary edema may be the only clinical presentation in 10% of patients who have myocardial infarction. Based on this information, along with the physical examination, the physician should be able to categorize the cause of the acute dyspnea as either suspected pulmonary, suspected cardiac, or other. Although cardiopulmonary disease may be most likely, it is important that consideration of causes of acute dyspnea other than cardiopulmonary be kept in mind. Inquire about indigestion or dysphagia, which may indicate gastroespophageal reflux with aspiration. Anxiety symptoms may imply psychogenic causes of dyspnea, but organic etiologies always should be excluded. For example, Saisch and colleagues [25] found that the most common complaint in patients diagnosed with acute hyperventilation syndrome was dyspnea. Additional information about social history (cigarette smoking, occupation, current or previous inhalational exposures, hobbies, and so forth) is essential. Table 1 summarizes clues in the history that help in the diagnosis in dyspnea. The physical examination should include the neck, thorax, lungs, heart, and extremities. Selected abnormal findings on the physical examination and clinical relevance are shown in Table 1. The general appearance and vital signs can be used to determine the severity of dyspnea by observing respiratory effort, use of accessory muscles, mental status, and ability to speak. The neck area might reveal a shift of the trachea, jugular venous distention, an enlarged thyroid gland, or adenopathy. Auscultate for stridor, because it may be indicative of an upper airway obstruction. Inspection of thorax might show an increased anterior-posterior diameter or chest wall deformity. Palpate the chest for subcutaneous emphysema and crepitus, and percuss for dullness, an indication of consolidation or effusions. On the other hand, hyperresonance on percussion suggests pneumothorax or bullous emphysema. Auscultation of breath sounds should focus on intensity, timing of the respiratory phases, and any adventitial sounds. Absent breath sounds may be consistent with pneumothorax or pleural effusion. Wheezing indicates turbulent airflow, which can be caused by asthma, COPD, and left ventricular failure. Key features of the heart examination include point of maximal impulse, the presence of any heart murmur, and a possible gallop. Rapid or irregular pulse may signify a dysarrhythmia. An S3 gallop suggests a left ventricular dysfunction in congestive heart failure. A loud P2 may be heard in patients who have pulmonary hypertension and cor pulmonale. Check the lower extremities for cyanosis or clubbing. Digital clubbing can be notable for cancer or a chronic respiratory condition
EVALUATION OF THE DYSPNEIC PATIENT IN THE OFFICE
649
other than COPD. Edema of lower extremities suggests congestive failure if symmetric and thromboembolic disease if asymmetric. Using information obtained from the medical history and physical examination, the physician should be able to form a working hypothesis as to the most likely cause of the dyspnea. Developing a differential diagnosis is important to focus further diagnostic testing. This is preferable to ordering a battery of tests at the initial evaluation. The office work-up depends on the modalities available. The initial diagnostic work-up usually includes a chest radiograph, a resting electrocardiogram, and a complete blood count. On the other hand, if a diagnosis of asthma or COPD is initially suspected, then pulmonary function testing (PFT) and pulse oximetry are important first steps. The prevalence of positive chest radiographic findings in patients who present to acute care clinics with complaints of chest-related symptoms has been shown to be 34.8% for all ages and up to 47% for patients over the age of 40 years [26]. Although the chest radiograph is usually not diagnostic, it can provide useful information about cardiac size and configuration, lung parenchyma, pulmonary vasculature, the pleural space, mediastinum, and position of the diaphragm. The 12-lead resting electrocardiogram rarely establishes an initial diagnosis, but it does provide important data about ischemia, arrhythmias, and chamber size. The complete blood count screens for anemia, which can effect the oxygen carrying capacity and exert a physiologic stress that may contribute to dyspnea. On the other hand, polycythemia may be the only clue to hypoxia, and indicates a more chronic, rather than acute, process. Additional testing depends on whether a primary pulmonary or cardiac disease is considered as the most likely cause for breathlessness. Left heart failure owing to ischemia or cardiomyopathy, pericardial effusion, and valvular dysfunction can all lead to dyspnea. Although the history, physical examination, and basic laboratory tests may be helpful to evaluate these conditions, sending the patient for an echocardiogram will provide extremely useful information about chamber size, ventricular function, valvular function, possible pericardial effusion, and an estimate of pulmonary artery systolic pressure. Studies have shown several potential mechanisms for dyspnea in patients who have cardiac disease: an elevated pulmonary venous pressure, respiratory muscle weakness, bronchial hyperresponsiveness, and an augmented ventilatory response to exercise associated with an increased dead space to tidal volume ratio [27–30]. Concomitant obstructive airway or restrictive disease may also be present in patients who have cardiac disease. For example, Myers and Froelicher [29] looked at exercise hemodynamic determinants of exercise capacity in patients who have heart failure, and found that relative hyperventilation is commonly observed during exercise in patients who have heart failure, and is related to ventilationperfusion mismatching in the lung caused by a higher-than-normal fraction of physiologic dead space [29]. Also, Snashall and Chung [30] described how airway narrowing may be precipitated by acute elevation of pulmonary or bronchial vascular pressures. This appears to be mainly caused by reflex
650
SARKAR & AMELUNG
bronchoconstriction, and can occur in left ventricular failure, mitral stenosis, and pulmonary edema. Therefore selected pulmonary function tests may also be indicated to evaluate dyspnea in patients who have suspected cardiac disease. Establishing heart failure as a cause of dyspnea in patients presenting to primary care clinics is not always possible based on symptoms and physical findings, and patients may need to be sent for laboratory testing. With heart failure ventricular cells are recruited to secrete both atrial natriuretic peptide and brain natriuretic peptide (BNP). These markers are secreted from both the left and the right cardiac ventricle in response to ventricular volume expansion and pressure overload. Recent studies have suggested that these neurohormones are reliably elevated in the setting of congestive heart failure and may be very helpful in its diagnosis [31–33]. The use of rapid BNP testing in addition to clinical judgment increased the accuracy of the clinical evaluation. McCullough and colleagues [34] evaluated the value of rapid measurement of plasma BNP for distinguishing between heart failure and a pulmonary cause of dyspnea in approximately 1600 patients presenting with a major complaint of acute dyspnea. The final diagnosis was heart failure in 47%, no heart failure in 49%, and noncardiac dyspnea in patients who had a past history of left ventricular dysfunction in 5%. The study demonstrated that plasma concentrations of BNP were markedly higher in patients who had clinically diagnosed heart failure (both left and right) compared with those who did not have heart failure (mean of 675 versus 110 pg/ml). Intermediate values were found in the patients who had baseline left ventricular dysfunction but not acute exacerbation (346 pg/ml). A value greater than 100 pg/ml diagnosed heart failure with a sensitivity, specificity, and predictive accuracy of 90%, 76%, and 83%, respectively. The predictive accuracy of plasma BNP for heart failure was equivalent to or better than other parameters such as cardiomegaly on chest radiograph, a history of heart failure, or rales on physical examination [34]. Currently, with the rapid assay most dyspneic patients who have heart failure have values above 400 pg/ml, whereas values below 100 pg/ml have a very high negative predictive value for heart failure as a cause of dyspnea. A pulmonary function test is the starting point for evaluation of suspected respiratory disease. In the acute setting, simple office spirometry may be the only test easily available. The presence of acute airways obstruction will be suggested by the finding of reduced forced expiratory volume in one second (FEV1) and reduced FEV1 to forced vital capacity (FVC) ratio. This will typically indicate acute asthma or an exacerbation of COPD. As an alternative to spirometry, peak expiratory flow rates (PEFR) can be measured with a simple handheld device. These measurements are especially useful in acutely dyspneic patients who have known baseline measurements so that comparisons can be made. The flow-volume loop is most useful in the evaluation of upper-airway obstruction. Although an abnormal flow-volume loop may not be diagnostic, it can provide clues to specific diagnoses
EVALUATION OF THE DYSPNEIC PATIENT IN THE OFFICE
651
such as obstruction, restriction, or intrathoracic/extrathoracic airways obstruction. The utility of measuring D-dimer as a noninvasive method to diagnosis deep venous thromboembolism and pulmonary embolism has been studied extensively. D-dimer is a specific degradation product released into the circulation when cross-linked fibrin undergoes endogenous fibrinolysis. Perrier and colleagues [35] have shown that a D-dimer level measured by an enzyme-linked immunosorbent assay of less than 500 ng/ml has a negative predictive value of greater than 95% of excluding a pulmonary embolism in a patient who has low pretest clinical probability. A negative quantitative rapid enzyme-linked immunosorbent assay (ELISA) result is as diagnostically useful as a normal lung scan or negative duplex ultrasonograpy finding for excluding venous thromboembolism (VTE). The D-dimer is unidirectional in that a negative test is useful in ruling out pulmonary embolism, but a positive test does not have a sufficiently high specificity or positive likelihood ratio to be helpful in increasing the certainty of a diagnosis of pulmonary embolism. D-dimer results are unlikely to be helpful in patients who have had recent surgery (within 3 months) or who have malignancy, because these patients often have values above 500 ng/ml [36]. Evaluation of chronic dyspnea The patient who has chronic shortness of breath may be more difficult to diagnose because the dyspnea typically develops over weeks to months, patients alter their activities in response to the dyspnea so the severity may not apparent, and dyspnea is frequently out of proportion to any physiologic impairment that is found. In spite of this difficulty, however, the majority of patients who have chronic shortness of breath will have either asthma, COPD, interstitial lung disease, or cardiomyopathy [21]. The history is important in chronic dyspnea, but should be considered only part of the systematic evaluation that is necessary; the initial physician impression of the etiology of dyspnea based on the history alone was correct only 66% of the time in one study [21]. Patients may have difficulty describing the exact sensation of their dyspnea. They should be questioned about the onset, frequency and duration of breathlessness, as well as triggers and strategies that provide relief. Intermittent dyspnea is more likely caused by reversible events, such as bronchoconstriction, pleural effusion, CHF, or even chronic recurrent thromboemboli. Progressive dyspnea more likely stems from COPD, neuromuscular disorders, or interstitial lung disease. Associated symptoms should be sought, including cough, sputum, wheezing, orthopnea, chest pain, heartburn, and paroxysmal nocturnal dyspnea. In general, the sensation of increased work or effort of breathing is common to most dyspneic patients; however, if asked to choose from a list of descriptors, the sensation that most accurately describes their dyspnea, patients who have different diseases will choose different groups of descriptors [37,38].
652
SARKAR & AMELUNG
Physical examination should focus on the neck, chest, lungs, heart, and extremities. Obvious findings such as rales in CHF or distant breath sounds in COPD and pleural effusion should be noted, but also note if deep inspirations cause cough, which may indicate asthma or interstitial lung disease. Wheezing may only become apparent after the patient is asked to forcefully exhale. Careful evaluation of the heart sounds is needed to detect pulmonary hypertension, and clubbing should not be overlooked. Table 2 outlines the history and physical findings in chronic dyspnea. Testing in chronic dyspnea should be targeted in attempt to answer specific questions. For example, if asthma or COPD is suspected, spirometry should be ordered; a reduced FEV1 and FEV1/FVC ratio indicates obstructive airway disease. Bronchoprovocation can diagnose reactive airways Table 2 Chronic dyspnea: etiologies and characteristics Disorder
History findings
Right and left heart failure (CHF)
Chest pain, orthopnea, PND, edema
COPD
Tobacco use, chronic cough
Asthma
Childhood history, allergy history Gradual onset of dyspnea, occupational & environmental exposure
ILD
Malignancy
Psychogenic: hyperventilation, anxiety, panic attacks Anemia
Cough, hemoptysis, shortness of breath, fatigue, fevers, night sweats, weight loss Emotional upset, feeling impending doom, neurotic personality Fatigue, dyspnea with exertion
Abbreviation: ILD, interstitial lung disease.
Physical examination Cyanosis, crackles, edema, JVD, murmurs, S3 or S4, HJR, hypertension Pursed lip breathing, wheezing, barrel chest, decreased breath sounds Wheezing, cough
Chest radiograph Cardiomegaly, pleural effusion, interstial edema
Hyperinflated lungs
Decreased breath sounds, clubbing
Hyperinflated lungs Decreased lung volumes, increased interstial markings, fibrosis Mass, hilar adenopathy
Sighing
Normal
Tachycardia, pale conjunctiva
Normal
Fine inspiratory crackles
EVALUATION OF THE DYSPNEIC PATIENT IN THE OFFICE
653
disease caused by asthma if spirometric values are normal. Flow volume loops obtained during spirometry can suggest upper airway obstruction. Restrictive lung disease can be suggested by spirometric results (reduced FVC and normal or increased ratio of FEV1/FVC ratio), but the measurement of lung volumes is necessary to confirm restriction (reduced FVC and total lung capacity). A reduced diffusing capacity, (DLco) is usually a sensitive indicator of an abnormality in pulmonary gas exchange. It may be helpful to identify patients who have emphysema (ie, a reduced DLco in the presence of obstructive airway disease). In a patient who has normal spiromety and lung volumes but a reduced DLco, the differential diagnosis could include anemia, early interstitial lung disease and pulmonary vascular disease. The measurement of inspiratory (PImax) and expiratory (PEmax) mouth pressures is important in the evaluation of neuromuscular causes of dyspnea. Cardiac etiologies of dyspnea include left heart failure, ischemia, pericardial effusion, and valvular disease. As for acute dyspnea, an echocardiogram can be extremely useful in the evaluation of chronic dyspnea, and provides useful information about chamber size, valvular function, and cardiac function. An electrocardiogram (EKG) will reveal evidence of ischemia or arrhythmia. An EKG done in conjunction with stress or exercise may unmask a cardiac condition not evident at rest. Chest radiograph (CXR) may show an enlarged heart, pulmonary vascular congestion, or vascular pruning consistent with pulmonary hypertension. If directed testing is not possible because the cause of dyspnea is unclear from the history and physical examination, the basic work-up should include spirometry and diffusing capacity, a CXR, a resting 12-lead EKG, and oximetry as initial screening tests. If the results of these tests are normal, then anxiety/hyperventilation, deconditioning, and respiratory muscle weakness are the likely etiologies. As previously suggested, PImax and PEmax should be measured to evaluate respiratory muscle strength. Muscle weakness can be isolated to the respiratory system or can be part of a systemic process. For example, Flaherty and coworkers [39] described 28 patients referred for unexplained breathlessness who had normal spirometry, lung volumes, and gas exchange, but had reduced values for PImax and PEmax. Based on muscle biopsies, it was confirmed that these patients were diagnosed with mitochondrial myopathy, affecting their muscles of respiration. If the source of dyspnea cannot be determined despite a complete pulmonary and cardiac evaluation, cardiopulmonary exercise testing (CPET) can be performed. The test will hopefully simulate the patient’s experience of breathlessness and differentiate between pulmonary, cardiac, psychogenic, and deconditioning as the source of the patient’s dyspnea. It can also be useful if the severity of the patient’s dyspnea is greater than expected from the results of objective measurements, or if the patient likely has both cardiac and respiratory factors contributing to dyspnea. CPET stresses the oxygen transport system, a complex network in which pathology at any point could generate dyspnea. Measured responses to exercise are taken from various
654
SARKAR & AMELUNG
systems, so CPET can help identify whether the major abnormality lies in the pulmonary or cardiac system. It can point to deconditioning, or peripheral vascular or muscular disease as potential contributors as well. Other less commonly used tests in the work-up of chronic dyspnea include ventilationperfusion scanning (chronic thromboembolic disease), thyroid function tests (occult hyper- or hypothyroidism) or gallium scanning (inflammatory lung disease or infection).
Treatment of dyspnea Treatment of the underlying disease is the most effective method of alleviating dyspnea: bronchodilators for acute or chronic dyspnea related to asthma or COPD, diuretics for breathlessness caused by acute or CHF, oxygen and antibiotics for shortness of breath related to pneumonia. If the specific cause of the dyspnea is elusive, or if a specific treatment is not available, then treatment should be aimed at treating the symptom. Patients who have chronic dyspnea can be taught a variety of methods to help them alleviate or cope with their breathlessness. Pursed-lips breathing can increase the tidal volume, decrease the respiratory rate, and improve saturation in patients who have COPD, thereby relieving their dyspnea [40]. Diaphragmatic breathing is another strategy to help reduce dyspnea. Patients can be taught energy conservation techniques to both reduce their respiratory effort and improve their respiratory muscle function. Pulmonary rehabilitation, or physical training, has been shown to reduce dyspnea, based on ability to perform activities of daily living and perform exercise testing [41,42]. The physical training need not be extreme to afford benefit; even simple resistance training with weights has been shown to improve muscle strength and endurance [43,44]. Patients should understand that reaching goals may take months of training, and therefore motivation and commitment on their part is essential. Oxygen can be used to reduce respiratory drive, thereby reducing dyspnea. Oxygen can improve respiratory muscle function [45] and pulmonary artery pressure [46]. Oxygen is usually dosed to prevent desaturation, but higher levels may improve exercise performance [47]. Psychotropic medications, including anxiolytics and antidepressants, have not proven to be of benefit in relieving dyspnea [48]; however, these may be considered in patients who exhibit these symptoms in relation to their dyspnea. Although the role of opiates is well-accepted in relieving terminal dyspnea, the benefits have been inconsistent and side effects frequent when studied in long-term but nonterminal dyspneic patients [49].
Summary Dyspnea is a nonspecific symptom of any disease involving the respiratory system. Although diseases of the lungs, chest wall, pleura, diaphragm,
EVALUATION OF THE DYSPNEIC PATIENT IN THE OFFICE
655
upper airway, and heart are most common, diseases of many other organ systems (eg, neuromuscular, skeletal, renal, endocrine, rheumatologic, hematologic, and psychiatric) may involve the respiratory system and present with dyspnea. Dyspnea should be evaluated systematically, and a thorough history and physical examination and baseline tests of heart and lung function are necessary to establish a complete database. More sophisticated testing may be needed when the cause is not readily apparent from the initial work-up. Treatment is best and most effective when geared toward a specific etiology, but if this is not possible, nonspecific treatment of the symptom pf dyspnea may afford the patient some benefit.
References [1] O’Connor GT, Anderson KM, Kannel WB. Prevalence and prognosis of dyspnea in the Framingham Study [abstract]. Chest 1987;92(Suppl 2):90S. [2] Guyatt GH, Townsend M, Berman LB, et al. Quality of life in patients with chronic airflow limitation. Br J Dis Chest 1987;81:45–54. [3] Simon PMSRM, Weiss JW, Fenel V, et al. Distinguishable types of dyspnea in patients with shortness of breath. Am Rev Respir Dis 1990;142(5):1009–14. [4] Mahler DA, Harver A, Lentine T, et al. Descriptors of breathlessness in cardiorespiratory diseases. Am J Respir Crit Care Med 1999;159:321–40. [5] American Thoracic Society. Dyspnea-mechanisms, assessment and management. A consensus statement. Am J Respir Crit Care Med 1999;159:321–40. [6] Mulrow C, Lucey C, Farnett L. Discriminating causes of dyspnea through clinical examination. J Gen Intern Med 1993;8:383–92. [7] Renwick DS, Connoly MJ. Do respiratory symptoms predict chronic airflow obstruction and bronchial hyperresponsiveness in older adults? J Gerontol 1999;54:M136–9. [8] Dow LCD, Osmond C, Holgate ST. A population survey of respiratory symptoms in the elderly. Eur Respir J 1991;4:267–72. [9] Horsley JR, Sterling IJ, Waters WE, et al. Respiratory symptoms among elderly people in the New Forest area as assessed by postal questionnaire. Age Ageing 1991;20:325–31. [10] Kroenke K, Manglesdorff D. Common symptoms in ambulatory care: incidence, evaluation, therapy and outcome. Am J Med 1989;86:262–6. [11] Sibuya M, Yamada M, Kanamara A, et al. Effect of chest wall vibration on dyspnea in patients with chronic respiratory disease. Am J Respir Crit Care Med 1994;149:1235–40. [12] Banzett RB, Lansing RW, Brown R. High-level quadriplegics perceive lung volume change. J Appl Physiol 1987;62:567–73. [13] Manning HL, Shea SA, Schwartzstein RM, et al. Reduced tidal volume increases air hunger at fixed PC02 in ventilated quadriplegics. Respir Physiol 1992;90:19–30. [14] Schwartzstein RM, Manning HL, Weiss JW, et al. Dyspnea: a sensory experience. Lung 1990;168:185–99. [15] Manning HL, Mahler DA, Harver A. Dyspnea in the elderly. In: Mahler DA, editor. Pulmonary disease in the elderly patient. New York: Marcel Dekker; 1993. p. 81–112. [16] Campbell EJM, Howell JBL. The sensation of breathlessness. Br Med Bull 1963;19:36–40. [17] O’Donnell DE. Exertional breathlessness in chronic respiratory disease. In: Mahler DA, editor. Dyspnea. New York: Marcel Dekker; 1998. p. 97–148. [18] DePaso W, Winterbauer R, Lusk J, et al. Chronic dyspnea unexplained by history, physical examination, chest roentgenogram and spirometry. Chest 1991;100:1293–9. [19] Fedullo A, Sinburne A, McGuire-Dunn C. Complaints of breathlessness in the emergency department. N Y State J Med 1986;86:4–6.
656
SARKAR & AMELUNG
[20] Pearson S, Pearson E, Mitchell J. The diagnosis and management of patients admitted to the hospital with acute breathlessness. Postgrad Med J 1981;57:419–24. [21] Pratter M, Curley F, Dubois J, et al. Cause and evaluation of chronic dyspnea in a pulmonary disease clinic. Ann Intern Med 1989;149:2277–82. [22] Schmitt B, Kushner M, Wiener S. The diagnostic usefulness of the history of the patient with dyspnea. J Gen Intern Med 1986;1:386–93. [23] Palla A, Petruzzelli S, Donnamaria V, et al. The role of suspicion in the diagnosis of pulmonary embolism. Chest 1995;107(Suppl 1):21S–4S. [24] Lusiani L, Perrone A, Pesavento R, et al. Prevalence, clinical features and acute course of atypical myocardial infarction. Angiology 1994;45:49–51. [25] Saisch SG, Wessley S, Gardner WN. Patients with acute hyperventilation presenting to an inner-city emergency department. Chest 1996;110:952–7. [26] Benacerraf B, McLoud T, Rhea J, et al. An assessment of the contribution of chest radiography in outpatients with acute chest complaints: a prospective study. Radiology 1981;138: 293–9. [27] Hammond MD, Bauer KA, Sharp JT. Respiratory muscle strength in congestive heart failure. Chest 1990;98:1091–4. [28] McParland C, Krishnan B, Wang Y, et al. Inspiratory muscle weakness and dyspnea in chronic heart failure. Am Rev Respir Dis 1992;146:467–72. [29] Myers J, Froelicher VF. Hemodynamic determinants of exercise capacity in chronic heart failure. Ann Intern Med 1991;115:377–86. [30] Snashall PD, Chung KF. Airway obstruction and bronchial hyperresponsiveness in left ventricular failure and mitral stenosis. Am Rev Respir Dis 1991;144:945–56. [31] Maisel A. B-type natriuretic peptide levels: diagnostic and therapeutic potential. Cardiovasc Toxicol 2001;1:159–64. [32] Meuller C, Scholer A, Laule-Kilian K, et al. Use of B-type natriuretic peptide in the evaluation and management of acute dyspnea. N Engl J Med 2004;350(7):647–54. [33] Mueller T, Gegenhuber A, Poelz W, et al. Diagnostic accuracy of B type natriuretic peptide and amino terminal proBNP in the emergency diagnosis of heart failure. Heart 2005;91(5): 606–12. [34] McCullough PA, Nowak RM, McCord J, et al. B-type natriuretic peptide and clinical judgment in emergency diagnosis of heart failure: analysis from Breathing Not Properly (BNP) Multinational Study. Circulation 2002;106(4):416–22. [35] Perrier A, Roy PM, Aujesky D, et al. Diagnosing pulmonary embolism in outpatients with clinical assessment, D-dimer measurement, venous ultrasound, and helical computed tomography: a multicenter management study. Am J Med 2004;116(5):352–3. [36] Shutgens RE, Esseboom EU, Haas FJ, et al. Usefulness of a semiquantitative D-dimer test for the exclusion of deep venous thrombosis in outpatients. Ann Intern Med 2002;112(8): 617–21. [37] Elliott MW, Adams L, Cockcroft A, et al. The language of breathlessness. Use of verbal descriptors by patients with cardiopulmonary disease. Am Rev Respir Dis 1991;144(4): 826–32. [38] Simon PM, Schwartzstein RM, Weiss JW, et al. Distinguishable types of dyspnea in patients with shortness of breath. Am Rev Respir Dis 1990;142(5):1009–14. [39] Flaherty KR, Wald J, Weismann IM, et al. Unexplained exertional limitation: characterization of patients with a mitochondrial myopathy. Am J Respir Crit Care Med 2001;164(3): 425–32. [40] Teip BL, Burns M, Kao D, et al. Pursed lips breathing training using ear oximetry. Chest 1986;90(2):218–21. [41] ACCP/AACVPR Pulmonary Rehabilitation Guidelines Panel. Pulmonary rehabilitation: joint ACCP/AACVPR evidence-based guidelines. Chest 1997;112:1363–96. [42] American Thoracic Society. Pulmonary rehabilitationd1999. Am J Respir Crit Care Med 1999;159:1666–82.
EVALUATION OF THE DYSPNEIC PATIENT IN THE OFFICE
657
[43] Stewart K. Resistance training effects on strength and cardiovascular endurance in cardiac and coronary prone patients. Med Sci Sports Exerc 1989;21(6):678–82. [44] Simpson K, Killian K, McCartney N, et al. Randomised controlled trial of weightlifting exercise in patients with chronic airflow limitation. Thorax 1992;47(2):70–5. [45] Bye PT, Esau SA, Levy RD, et al. Ventilatory muscle function during exercise in air and oxygen in patients with chronic air-flow limitation. Am Rev Respir Dis 1985;132(2):236–40. [46] Dean NC, Brown JK, Himelman RB, et al. Oxygen may improve dyspnea and endurance in patients with chronic obstructive pulmonary disease and only mild hypoxemia. Am Rev Respir Dis 1992;146(4):941–5. [47] Dewan NA, Bell CW. Effect of low flow and high flow oxygen delivery on exercise tolerance and sensation of dyspnea. A study comparing the transtracheal catheter and nasal prongs. Chest 1994;105(4):1061–5. [48] Man GCW, Hsu K, Sproule BJ. Effect of alprazolam on exercise and dyspnea in patients with chronic obstructive pulmonary disease. Chest 1986;90(6):832–6. [49] Woodcock AA, Johnson MA, Geddes DM. Breathlessness, alcohol and opiates. N Engl J Med 1982;306:1363–4.
Prim Care Clin Office Pract 33 (2006) 659–684
Acute Abdominal Pain Mark H. Flasar, MDa,b,*, Raymond Cross, MDa,b, Eric Goldberg, MDa,b a
Division of Gastroenterology and Hepatology, Department of Medicine, University of Maryland Medical Center, 22 South Greene Street, Baltimore, MD 21201, USA b Veterans Affairs, Maryland Health Care System, 10 North Greene Street, Baltimore, MD 21201, USA
Abdominal pain is a complaint seen commonly in the outpatient setting. Many etiologies, both acute and chronic, can be evaluated on an outpatient basis. However, several causes of abdominal pain necessitate prompt, focused, and structured evaluation, given associated morbidity and mortality. The differential diagnosis of a patient presenting with acute abdominal pain is exhaustive, necessitating that the physician understand not only the underlying pathophysiology of the pain, but also the clinical presentation, course, and initial management of more harmful causes. A focused history, physical examination, and adjunctive testing strategy will allow for those patients with concerning presentations to be identified, initially managed, and appropriately referred for continued care. Abdominal pain is an extremely common complaint in all settings of medical practice. In primary care practices in 2002, abdominal pain was a complaint in more than 13.5 million patient visits [1]. Oftentimes, patients with severe abdominal pain will self-triage to an emergency department, hospital, or contact emergency medical services. However, some patients with potentially life-threatening abdominal catastrophes will initially present to the primary care physician. Ease of access to the primary care physician and preexisting appointments occurring shortly after the pain onset offer some explanation. Abdominal pain may often be a symptom of a disease process with a benign course, but it may also herald a severe, life-threatening condition that demands prompt recognition and management. The purpose of this review is to provide the practitioner with a framework for understanding
* Corresponding author. E-mail address:
[email protected] (M.H. Flasar). 0095-4543/06/$ - see front matter Ó 2006 Elsevier Inc. All rights reserved. doi:10.1016/j.pop.2006.06.004 primarycare.theclinics.com
660
FLASAR
et al
abdominal pain, so that they may determine those patients that need a more expedited evaluation. Furthermore, the pathophysiologic mechanisms underlying abdominal pain will be reviewed. A general approach to the patient with acute abdominal pain will be outlined, and several gastrointestinal etiologies of abdominal pain will be considered in detail, focusing on the most severe and commonly encountered. A general understanding of abdominal anatomy, physiology, and pathophysiology is vital when formulating a differential diagnosis for abdominal pain. In addition, it is important to understand how abdominal pain is generated and perceived by the patient. The abdominal viscera are innervated with nocioceptive afferents within the mesentery, on peritoneal surfaces, and within the mucosa and muscularis of hollow organs. These afferents respond to both mechanical and chemical stimuli producing dull, crampy, poorly localized pain sensations. The principal mechanical stimulus is stretch, while a variety of chemical stimuli including substance P, serotonin, prostaglandins, and hydrogen ions are perceived as noxious by visceral chemoreceptors [2]. Abdominal pain occurs in three broad patterns: visceral, parietal, and referred. Visceral nocioception typically involves stretch and distension of the abdominal organs, although torsion, and contraction also contribute. The pain stimulus is carried on slow-conducting C-fibers. Patients often describe pain of visceral origin as a dull ache. The location of visceral pain is often midline, because visceral innervation of abdominal organs is typically bilateral. Pain location corresponds to those dermatomes that match the innervation of the injured organ [2]. Generally, visceral pain from organs proximal to the Ligament of Treitz, including the hepatobiliary organs and spleen, is felt in the epigastrum. Visceral pain from abdominal organs between the Ligament of Treitz and the hepatic flexure of the colon is felt in the periumbilical region. Visceral pain generated from organs distal to the hepatic flexure is perceived in the midline lower abdomen. Parietal pain is typically sharp and well localized, resulting from the direct irritation of the peritoneal lining. Parietal peritoneal afferents are A-d fibers, with a rapid conduction velocity and result in sharp pain sensation similar to skin and muscle pain. Because parietal innervation is unilateral, lateralization of pain occurs [2]. Referred pain occurs when visceral afferents carrying stimuli from a diseased organ enter the spinal cord at the same level as somatic afferents from a remote anatomic location. It is typically well localized. A single diseased organ may produce all three types of pain. For example, when a patient develops cholecystitis, gallbladder inflammation is initially experienced as a visceral pain in the epigastric region. Eventually, the inflammation extends to the parietal peritoneum, and the patient will experience parietal pain that lateralizes to the right upper quadrant. Gallbladder pain may also refer to the right shoulder. Awareness of the anatomy and innervation of the abdominal viscera allows one to formulate a differential diagnosis of abdominal pain based on the location and distribution of the pain (Table 1). However, there is
ACUTE ABDOMINAL PAIN
661
significant overlap between abdominal pain presentations. Furthermore, disease processes from organs outside of the abdominal cavity can present with abdominal pain. To considerably narrow the differential diagnosis, it is crucial to approach each patient in a systematic, logical, and deliberate manner. Having already elicited a chief complaint of abdominal pain, this begins with the patient’s history and physical examination. The history should not only include a thorough assessment of the present condition, but also a detailed assessment of underlying medical problems, medications, family history, substance abuse history, recent travel history and occupational history. Important clues to the etiology of the pain should be determined in the patient’s history by inquiring about the nature of the pain, which includes its quality, location, rapidity of onset, chronicity, radiation, intensity, exacerbating factors, alleviating factors, and associated symptoms. Chronicity of symptoms is an important factor in evaluating abdominal pain. Generally, patients with chronic symptoms can be evaluated on an outpatient basis, as the process underlying the pain rarely requires acute intervention. On the other hand, patients with new-onset of symptoms are more likely to have a significant disease process that can bring them harm in the hours to days ahead. An exception to this is an acute worsening of chronic or intermittent abdominal pain symptoms. Examples include acute mesenteric ischemia superimposed on a history of chronic intestinal angina or development of acute cholecystitis in a patient with a history of biliary colic. Additionally, identifying high-risk patients such as the elderly, pregnant, and those with immunodeficiency syndromes proves invaluable in triaging patients. Following a thorough history, a focused physical examination should be performed. The generation of a differential diagnosis will help the practitioner tailor the examination, whose purpose is to provide confirmatory or contradictory data for each disease process on their differential. Although centered on the abdomen, the examination should also focus on extraabdominal organ systems when indicated. For example, a patient with suspected mesenteric ischemia should have a cardiovascular examination assessing for arrhythmias and evidence of atherosclerotic disease. Overall, the abdominal examination should begin with general observation of the patient, followed by abdominal inspection. A patient with peritonitis often lies completely still as movement further irritates the peritoneum. Their abdomen will be visually rigid. On the other hand, a patient with renal colic may writhe in pain, may not be able to be consoled to comfort, and have a nonrigid abdomen. Once initial observation is complete, a review of the vital signs is imperative. Any significant abnormality of vital signs should prompt the physician to consider a more life threatening process. Auscultation of the abdomen determines whether intestinal peristalsis is appropriate and whether any abdominal bruits are present. Next, palpation of the abdomen should be performed to distinguish pain, a subjective sensation, from tenderness, which is an objective finding.
662
FLASAR
et al
Table 1 Diagnosis of abdominal pain based on location and distribution Right upper quadrant
Middle upper abdomen
Left upper quadrant
Peptic ulcer disease Biliary disease Biliary colic Choledocholithiasis Cholecystitis Cholangitis Liver disease Hepatitis Neoplasm Abscess Congestive hepatopathy Lung disease Pneumonia Subphrenic abscess Pulmonary embolism Pneumothorax Abdominal wall Herpes zoster Muscular Stain Kidney disease Pyelonephritis Perinephric abscess Nephrolithiasis Colonic causes Colitis Right sided diverticulitis
Peptic ulcer disease Pancreatitic disease Pancreatitis Pancreatic neoplasm Biliary disease Biliary colic Choledocholithiasis Cholecystitis Cholangitis Esophageal disease Reflux esophagitis Infectious esophagitis Pill esophagitis Cardiac disease Myocardial ischemia or infarction Pericarditis AAA rupture/aortic dissection Mesenteric ischemia
Peptic ulcer disease Splenic disease Splenic rupture Splenic infarct Pancreatic disease Pancreatitis Pancreatic neoplasm Lung disease Pneumonia Subphrenic abscess Pulmonary embolism Pneumothorax Kidney disease Pyelonephritis Perinephric abscess Nephrolithiasis
Periumbilical Appendicitis (early) Small bowel obstruction Gastroenteritis Mesenteric ischemia AAA rupture Aortic dissection Right lower quadrant
Suprapubic
Left lower quadrant
Appendicitis Inflammatory bowel disease OB-GYN causes Ovarian tumor Ovarian torsion Ectopic pregnancy Pelvic inflammatory disease (PID) Kidney disease Pyelonephritis Perinephric abscess Nephrolithiasis Intestinal disease Right sided diverticulitis Ileocolitis Gastroenteritis Hernia
Inflammatory bowel disease OB-GYN causes Ovarian tumor Ovarian torsion Ectopic pregnancy PID Dysmenorrhea Colonic disease Proctocolitis Diverticulitis Urinary tract disease Cystitis Nephrolithiasis Prostatitis
Inflammatory bowel disease OB-GYN causes Ovarian tumor Ovarian torsion Ectopic pregnancy PID Kidney disease Pyelonephritis Perinephric abscess Nephrolithiasis Intestinal disease Sigmoid diverticulitis Ileocolitis Gastroenteritis Hernia
ACUTE ABDOMINAL PAIN
663
Table 1 (continued) Diffuse Gastroenteritis Bowel obstruction Peritonitis Mesenteric ischemia Inflammatory bowel disease Diabetic ketoacidosis Porphyria Uremia Hypercalcemia Sickle cell crisis Vasculitis Heavy metal intoxication Opiate wihdrawal Familial mediterranean fever Hereditary angioedema
When performing palpation, the location of tenderness should be used to narrow the differential diagnosis. Additionally, the presence of guarding or rebound tenderness should be noted, as these findings imply peritoneal irritation. Furthermore, palpation can determine the presence of visceral enlargement, masses, or fluid. It is often useful to begin the abdominal examination by palpating distant from, then working toward the site of pain, palpating with the fingertips. It is helpful to keep the patient’s hips and knees in a flexed position during a supine examination to help relax the abdominal musculature. Techniques such as patient distraction during examination or palpation during auscultation with the stethoscope head may help discriminate functional from organic pain. The importance of a properly executed history and physical examination cannot be underestimated. Although the sensitivity and specificity of a history and physical may not match that of an abdominal CT scan, there is no risk, minimal time required, and essentially no cost. In fact, one observational study revealed that, based on history and physical alone, physicians were able to correctly differentiate between organic and nonorganic causes of abdominal pain nearly 80% of the time [3]. Furthermore, historic features such as pain location have been shown in prospective investigation to be specific for certain disease states [4]. Once the history and physical is completed, the practitioner will usually be armed with sufficient information to sharply narrow the differential diagnosis in the majority of patients presenting with abdominal pain. The detection of the warning signs of a life-threatening process in a patient with abdominal pain is often up to the primary physician long before the emergency room physician, surgeon, gastroenterologist, or other specialist encounters the patient. Certain historic and examination findings should raise red flags that a life-threatening abdominal process is present, and
664
FLASAR
et al
prompt early triage to an emergency department or inpatient hospital bed (Table 2). Red flags from the history include fever, vomiting, inability to move the bowels (obstipation), syncope, concomitant chest or back pain, respiratory distress, excessive acute vaginal bleeding, and overt gastrointestinal bleeding. Red flags from the physical examination include any significant abnormality of the vital signs, mental status changes, involuntary guarding, rebound abdominal tenderness, complete absence of bowel sounds, and pain out of proportion to abdominal tenderness on physical examination. In patients without signs or symptoms of an acute abdominal catastrophe, an acute expedited outpatient evaluation should be performed. Selection of initial laboratory and imaging studies to evaluate abdominal pain should be guided by the differential diagnoses generated from the primary evaluation. Historically, plain abdominal radiographs have been the initial imaging test of choice. They can be obtained rapidly at a relatively low cost, and are often available to primary practitioners in the office. However, with the evolution of more sensitive and specific modalities such as CT and ultrasound, the utility of the plain abdominal series has been debated. Nonetheless, the authors feel that plain films should still be the initial imaging modality in patients with suspected visceral perforation or obstruction. The abdominal plain film series should include supine and upright abdominal films in conjunction with an upright chest (or lateral decubitus abdominal) film. Plain abdominal imaging has been estimated to be diagnostic in up to 60% of cases of suspected small bowel obstruction [5], although sensitivity is more limited in cases of low-grade obstruction [6]. The location, volume, and distribution of intraluminal air, presence and distribution of air–fluid levels, and luminal diameter can often be helpful in differentiating between an obstructive and nonobstructive process, such as partial or complete bowel obstruction, ileus, or intestinal pseudoobstruction. Unfortunately, overlap in the radiographic appearance of obstructive and nonobstructive processes limits the sensitivity and specificity of plain films in this setting. The ability of plain films to detect free air depends on the volume of free air within the peritoneal cavity. For the detection of large volumes, as would be expected with a perforated viscus, sensitivity of plain films is reported to be as high as 100%. Sensitivity is maximized if the patient is placed in the upright or decubitus position for 5 to 10 minutes before obtaining an upright chest or lateral decubitus film. This allows small volumes of air to redistribute to and collect within nondependent areas. Detection of volumes as small as 1 to 2 cc of air has been reported using this method [6,7]. CT is an imaging tool that is sensitive for the diagnosis of many etiologies of abdominal pain. Because of its widespread availability, CT is often accessible to primary care providers in the outpatient setting for same-day interpretation. With newer rapid helical scanning methods, advances in intravenous and oral contrast agents, three-dimensional reformatting, and
Table 2 Red flags in abdominal pain Physical exam
Labs
Radiography
* * * * * *
* * * * * * * * *
* * * * *
* * * * * *
Inability to maintain po intake Projectile vomiting Overt gastrointestinal blood loss Syncope Pregnancy Recent surgery or endoscopic procedure * Fever * Caustic or foreign body ingestion
Pathologic changes in vital signs Bloody, maroon, or melanotic stool Hernia (incarcerated and tender) Hypoxia Cyanosis Altered mentation Jaundice Peritoneal signs Abdominal pain out of proportion to examination
Renal failure Metabolic acidosis Leukocytosis Elevated transaminases Elevated alkaline phosphatase and bilirubin * Anemia or polycythemia * Hyperlipasemia/hyperamylasemia * Hyperglycemia/hypoglycemia
* * * *
Abdominal free air Gallbladder wall thickening Pericholecystic fluid Dilated biliary tree Bowel obstruction Dilated small bowel loops air fluid levels Intraabdominal abscess Bowel wall thickening Air in the portal venous system Pneumatosis intestinalis
ACUTE ABDOMINAL PAIN
History
665
666
FLASAR
et al
other advanced software capabilities, CT has become the initial imaging modality of choice for the evaluation of most presentations of acute abdominal pain. For example, CT diagnoses acute appendicitis with a reported sensitivity and specificity as high as 98% and 97%, respectively [8]. In fact, the superior diagnostic capability of CT is rendering plain films obsolete. Even in situations where plain films have proven diagnostic accuracy, such as perforated viscus or small bowel obstruction, many physicians now opt for CT as the initial imaging study. Computed tomography has proven to be more sensitive and specific for nearly all etiologies of acute abdominal pain [9–11]. In patients with abdominal pain suspected to be from a hepatobiliary source, abdominal ultrasonography should be considered as an initial imaging option. It is accurate for the detection of gallstones and dilation of the biliary tree. However, ultrasound is less sensitive for stones in the common bile duct. Although MRI can be highly accurate in the diagnosis of acute abdominal pain, high cost and lack of immediate availability limit its use in the primary care setting. Following clinical evaluation of patients with abdominal pain, the primary physician must appropriately triage the patient. In addition to red flags revealed by the history and physical examination, there are laboratory and radiographic ‘‘red flags’’ that should alert the physician of a potentially more serious cause of the abdominal pain (Table 2). Although cardiac, pulmonary, urologic, musculoskeletal, and gynecologic causes of abdominal pain will not be specifically addressed in this review, it is important to keep these extraabdominal disease processes in the differential diagnosis of abdominal pain. Red flags that a life-threatening extraabdominal cause of abdominal pain is present include chest pain, back pain, shortness of breath, vaginal bleeding, and hemodynamic instability. Finally, there is a spectrum of systemic medical disorders, such as adrenal insufficiency, diabetic ketoacidosis, porphyria, and sickle cell pain crisis, that can present with abdominal pain. Evidence of these disorders in the past medical history, medications, or physical examination should prompt their consideration as the cause of the patient’s pain. Although a detailed discussion of all of the potential etiologies of acute abdominal pain is beyond the scope of this review, there are some etiologies that merit a more detailed discussion. What follows is an overview of those gastrointestinal etiologies of abdominal pain that can often be seen in adults in the outpatient setting, with a focus on those etiologies prone to more serious or life-threatening complications.
Hepatopancreatobiliary Biliary colic When patients with cholelithiasis present with abdominal pain, they will most commonly have biliary colic as the cause [12]. Biliary colic occurs when
ACUTE ABDOMINAL PAIN
667
then one or more gallstones transiently occlude the cystic duct. Tonic cystic duct spasm ensues, causing pain. Patients typically experience visceral epigastric or right upper quadrant abdominal pain, often with radiation to the right shoulder or scapula. The pain is sometimes postprandial, but often there is no trigger; it is not uncommon for patients to have nocturnal pain. The term biliary colic is a misnomer, as the pain is not colicky in nature. The classic time course of pain from biliary colic is one that builds over 15 to 60 minutes, lasting up to several hours before slowly dissipating. Physical examination is typically benign, with tenderness sometimes elicited in the epigastrum or right upper quadrant of the abdomen. Patients suspected of having biliary colic should have further evaluation with liver function tests and abdominal ultrasonography. Transaminases, bilirubin, and alkaline phosphatase are usually normal, and an ultrasound can verify the presence of gallstones. Although biliary colic usually resolves without sequelae, it identifies those patients whom are at higher risk for complications of gallstone disease, such as pancreatitis or cholangitis. Therefore, referral to a general surgeon for elective cholecystectomy is recommended. Cholecystitis More than 90% of cases of acute cholecystitis are caused by gallstones. The remainder of cases are termed acalculous cholecystitis, typically occur in critically ill patients, and are rarely seen in the outpatient setting [13]. Acute cholecystitis is most commonly caused by the obstruction of the cystic duct by the offending gallstone. Prolonged obstruction of the cystic duct (O6 hours) impairs gallbladder emptying, leading to inflammation of the gallbladder mucosa. Secondary bacterial infection of the gallbladder may ensue, leading to possible empyema, gallbladder necrosis, and perforation. Acute cholecystitis results in gallbladder perforation in up to 12% of cases, with a subsequent mortality rate of 20% [7]. Emphysematous cholecystitis, characterized by air in the wall of the gallbladder, is most often seen in patients with diabetes mellitus. Approximately 75% of patients who develop acute cholecystitis have a prior history of biliary colic [14]. The pain is similar to that of biliary colic, but with a longer duration. Pain lasting longer than 6 hours signifies cholecystitis rather than biliary colic. As acute gallbladder inflammation irritates the parietal peritoneum, the pain may shift from the epigastrum to the right upper quadrant. The physical examination of patients with acute cholecystitis reveals right upper quadrant tenderness. An inspiratory arrest during deep right upper quadrant palpation is referred to as Murphy’s sign. Laboratory abnormalities include leukocytosis with a predominance of neutrophils, and elevation of alkaline phosphatase, and transaminases. Hyperbilirubinemia generally does not occur with acute cholecystitis owing to the unimpeded flow of bile through the common bile duct. An exception is Mirizzi’s syndrome, where a large stone in the cystic duct compresses or
668
FLASAR
et al
erodes into the common hepatic duct, resulting in variable degrees of biliary obstruction. Right upper quadrant ultrasonography is the test of choice to diagnose acute cholecystitis, with reported sensitivity, specificity, and accuracy approaching 95% [7]. Common findings include cholelithiasis, gallbladder wall thickening, pericholecystic fluid, and a sonographic Murphy’s sign. The latter finding occurs when ultrasound transducer pressure on the gallbladder results in tenderness. The finding of cholelithiasis and a positive sonographic Murphy’s sign has a positive predictive value of 92% for acute cholecystitis. Conversely, when both of these findings are absent, the negative predictive value is 95% [7]. Radionuclude cholescintigraphy scans, such as the HIDA scan, can be used to confirm the diagnosis of acute cholecystitis when ultrasound findings are equivocal. The sensitivity, specificity, and positive predictive value for acute cholecystitis are 95%, 99%, and 97%, respectively [15]. Patients with suspected acute cholecystitis should have an expedited evaluation with early surgical consultation. Cholecystectomy within 24 to 48 hours of presentation has been shown to reduce mortality and shorten length of hospital stay, compared with surgery performed after weeks of conservative management aimed at ‘‘cooling off’’ the gallbladder [16–18]. The benefits of early cholecystectomy have been prospectively validated for the laparoscopic approach as well [19–23]. Cholangitis Ascending cholangitis is a potentially lethal entity that occurs when the bile duct become obstructed. Once bile flow is impeded, superinfection of the stagnant bile occurs. As pus builds up under pressure, the infection can rapidly ascend into the liver and spread into the blood stream. Common pathogens include Escherichia coli, Klebsiella species, Bacteroides, Enterococcus, and other enteric pathogens [24]. The most common cause of obstruction in the United States is choledocholithiasis, accounting for approximately 85% of cases. Benign biliary strictures, choledochal cysts, biliary parasites, and neoplasms are less common causes of cholangitis [25]. The classic presentation of cholangitis is fever, jaundice, and right upper quadrant pain. These findings are collectively referred to as Charcot’s triad, which has a reported sensitivity for cholangitis as high as 75% [26]. If the obstruction is not relieved, mental obtundation and shock can occur. The combination of Charcot’s triad with these findings is known as Reynold’s pentad, which is associated with a higher morbidity and mortality rate [13]. Laboratory findings include leukocytosis with a predominance of neutrophils, elevated alkaline phosphatase, and elevation of the transaminases. An elevation of pancreatic enzymes can be seen in about one third of patients, especially with concomitant gallstone pancreatitis [27]. As the pathophysiology of this disorder involves common bile duct obstruction, conjugated hyperbilirubinemia is invariably present.
ACUTE ABDOMINAL PAIN
669
The diagnosis of cholangitis is often made clinically, and should be confirmed with cholangiography. Although ultrasonography may suggest the presence of biliary obstruction, its sensitivity for choledocholithiasis is poor [7]. Therefore, when the clinical diagnosis of ascending cholangitis is suspected, patients should undergo cholangiography even in the setting of an unremarkable right upper quadrant ultrasound. Patients suspected of having acute cholangitis should be referred quickly to an emergency department or hospitalized, as the clinical course can be rapidly progressive and fatal if left untreated. Patients with suspected cholangitis should have blood cultures drawn at presentation, so that therapy can be directed toward the offending pathogen. The definitive therapy for cholangitis is decompression of the obstructed biliary system. Endoscopic retrograde cholangiopancreatography (ERCP) is the diagnostic and therapeutic procedure of choice, and is successful in relieving the obstruction in more than 95% of cases [15]. This is typically accomplished by stone extraction or placement of a stent into the common bile duct. In cases where therapeutic ERCP is not available or is unsuccessful, options include percutaneous transhepatic cholangiography or surgical decompression. If choledocholithiasis is the cause of ascending cholangitis, patients should undergo elective cholecystectomy once the infection resolves. Acute pancreatitis Acute pancreatitis is an inflammatory disease of the pancreas that not only may cause significant morbidity but also carries a case fatality rate of 5% to 9% [28]. Gallstones and alcohol account for more than 80% of cases in the United States [15]. Less common causes of pancreatitis include medications, trauma, hypercalcemia, severe hypertriglyceridemia (O1000 mg/dL), malignancy, sphincter of Oddi dysfunction, infections, iatrogenic (ERCP), and congenital abnormalities of the pancreas such as pancreas divisum. The remainder are termed idiopathic, although as many as 75% of cases of idiopathic pancreatitis may actually be due to biliary sludge or microlithiasis [29]. Regardless of the etiology, diffuse pancreatic inflammation and edema develop. In severe cases, necrosis of pancreatic and peripancreatic tissue occurs, and multiorgan failure may ensue. Necrotizing pancreatitis occurs in up to 25% of patients with pancreatitis and has a mortality rate of 15% to 20% [28,30]. Patients typically present with the acute onset of abdominal pain, nausea, and vomiting. The pain is steady and usually in the epigastrium, although patients may also note discomfort in the right or left upper quadrants of the abdomen. Pain is classically described as a boring sensation that radiates into the back. Patients are often unable to get comfortable when lying supine, and they will lean forward in an attempt to find relief. Because of marked fluid shifts, patients may become severely volume depleted. Resultant tachycardia and hypotension with orthostatic changes may develop. Other vital sign abnormalities include low-grade fever and
670
FLASAR
et al
tachypnea. The latter is a poor prognostic sign, and may herald the development of sepsis or acute respiratory distress syndrome. The abdominal examination may reveal distension and diminished or absent bowel sounds secondary to the development of a paralytic ileus. With palpation, focal tenderness in the epigastrum is seen, although the abdomen may also be diffusely tender. With more severe cases, voluntary guarding and rebound tenderness may also be appreciated. Signs of hemorrhagic pancreatitis such as Gray Turner’s sign (flank ecchymosis), Cullen’s sign (periumbilical ecchymosis), or Fox’s sign (inguinal ecchymosis) are seen in less than 1% of cases. When acute pancreatitis is suspected clinically, levels of serum amylase or lipase should be determined. In the setting of suspected acute pancreatitis, levels greater than three times the normal values have a high specificity for acute pancreatitis. Serum lipase remains elevated for longer durations than serum amylase and is more specific for acute pancreatitis [31]. It is important to note that the magnitude of serum amylase and lipase elevation does not correlate well with disease severity. Because of the marked intravascular volume depletion secondary to third spacing, hemoconcentration often occurs in acute pancreatitis. Hematocrit levels higher than 44% are associated with a worse prognosis, indicating potentially dangerous fluid shifts [32]. Hyperbilirubinemia, elevations of the alkaline phosphatase, and alanine aminotransferase levels O150 mg/dL suggest gallstones as the etiologic cause of the pancreatitis [33]. Because of marked fluid shifts that occur with acute pancreatitis, blood urea nitrogren, creatinine, and serum electrolytes including calcium should be assessed. Imaging of the pancreas with CT can confirm the diagnosis of acute pancreatitis, but is not necessary in all cases. The authors feel that CT scanning should be reserved for patients in whom the diagnosis is in question, in cases of suspected pancreatic necrosis, or in cases of clinical deterioration despite adequate medical therapy. Use of intravenous contrast is highly recommended, and CT should therefore be deferred until the patient has received adequate volume resuscitation to prevent nephrotoxicity. Because the care of patients with acute pancreatitis is complicated by the difficulty in differentiating whether a patient’s course will be mild or severe, several scoring systems have been developed to assess the severity in acute pancreatitis and determine prognosis. The most well known of these criteria is Ranson’s criteria, which was originally developed for alcoholic pancreatitis and later modified and validated for gallstone pancreatitis. Ranson’s criteria has limited clinical value because it takes 48 hours to determine. The Imrie/Glasgow criteria and APACHE II score are two other prospective systems, but both are cumbersome to perform. A prognostic CT scoring system, known as the Balthazar criteria, has been validated for predicting severity in acute pancreatitis. The score is heavily weighted on the presence of pancreatic necrosis [34]. The cornerstone of therapy in acute pancreatitis is intravenous volume resuscitation coupled with the prevention of pancreatic stimulation. Patients should be kept strictly nothing by mouth, and
ACUTE ABDOMINAL PAIN
671
therefore require a hospital setting for treatment. Very aggressive intravenous fluid repletion is necessary to maintain intravascular volume and allow adequate perfusion of the pancreas and extrapancreatic organs such as the kidneys.
Luminal and vascular disorders Acute appendicitis Acute appendicitis is the most common abdominal surgical emergency in the United States, with over 250,000 appendectomies performed annually [35]. Most cases of appendicitis are believed to result from obstruction of the appendiceal lumen by fecaliths. Following obstruction, increased intraluminal pressure causes local ischemia, leading to transmural inflammation. Secondary bacterial infection occurs, and gangrene and perforation of the appendix may result. A thorough history and physical examination are all that is required to make a clinical diagnosis of appendicitis, thereby heightening the importance of the initial care provider’s assessment. As a result of appendiceal hypertension and distension, a crampy visceral type pain is initially felt in the peri-umbilical region. There is often associated nausea, vomiting, and fever. As the inflammatory process progresses, and directly irritates the parietal peritoneum, the quality of the pain becomes sharp and shifts to the right lower quadrant (RLQ). Almost all patients with appendicitis will lose their appetite, and if a patient exhibits hunger, the clinical diagnosis of appendicitis should be questioned. Auscultation of the abdomen reveals diminished or absent bowel sounds. Classically, the examination of patients with appendicitis reveals tenderness to palpation at McBurney’s point, anatomically located two thirds of the way from the umbilicus to the anterior superior iliac spine. Guarding, rigidity, and rebound tenderness may be present from peritoneal irritation. Rovsing’s sign may be present, which is RLQ pain upon left lower quadrant palpation. The obturator and iliopsoas signs can be elicited by internal rotation of the right hip and extension of the right hip, respectively. The finding of abdominal pain during either maneuver occurs as a result of the inflammatory process, irritating the respective muscles during passive movement. Patients who present with acute abdominal pain that migrates from the umbilicus to the right lower quadrant, whom also exhibits RLQ tenderness on palpation, should be referred for emergent appendectomy. The accuracy of the clinical diagnosis in this situation has been estimated to be nearly 95% [36]. However, this classic presentation of appendicitis occurs only in two thirds of patients [37]. Atypical presentations account for the remainder. They result from either anomalous appendiceal anatomy, or certain populations of patients that are more prone to atypical presentations of common diseases, such as the elderly, immunocompromised, or pregnant
672
FLASAR
et al
patients. For example, a retrocecal appendix that becomes inflamed may produce right flank rather than abdominal pain. Patients older than 55 years of age may present with vague symptoms and exhibit more subtle examination findings, thereby causing a delay in diagnosis. This delay can result in a higher rate of perforation compared with their younger counterparts [37]. Finally, misdiagnosis is more common in premenopausal women owing to a broadened gynecologic differential diagnosis and confusing presentations [38]. Therefore, the importance of considering this diagnosis at any age remains important. Patients with suspected appendicitis should be made nothing by mouth (NPO) and started on intravenous fluids. The prophylactic use of antibiotics is not supported by the literature, and should only be used in cases of suspected perforation. Because of the potential perforation risk, patients with a clinical diagnosis of appendicitis should undergo emergent surgical intervention. Historically, a 20% presurgical false positive rate has been considered acceptable. In patients where the clinical diagnosis is uncertain, imaging studies and observation admissions for serial abdominal examinations may decrease this false positive rate. In any woman of childbearing age, pregnancy should be ruled out with a serum or urinary b-human chorionic gonadotropin (b-HCG) test before imaging or appendectomy. Diverticulitis Nearly a third of patients over the age of 50, and two thirds by the age of 80 have diverticular disease [39]. Diverticulitis, a complication caused by the perforation of a diverticulum, affects up to 25% of patients with diverticular disease [40]. Inspissated food, stool, and increased intraluminal pressure are believed to be involved in the pathogenesis of diverticular perforation. The clinical presentation of patients with diverticulitis is dependent on the extent of the perforation. Small perforations are walled off by surrounding mesentery and pericolonic fat, while larger perforations can result in extensive intraperitoneal abscess formation and frank peritonitis. The location of abdominal pain in patients symptomatic with diverticulitis is dependent on the location of the perforated diverticulum. Diverticular disease most commonly affects the sigmoid colon, so patients most often present with crampy, left lower quadrant abdominal pain. However, patients with a redundant sigmoid colon or diverticular disease involving the right colon may complain of RLQ abdominal pain [41]. Patients may additionally complain of nausea, vomiting, fever, and anorexia. Physical examination often reveals tenderness over the inflamed area, and an inflammatory mass may be palpable in some patients. In patients with free perforation, peritoneal signs such as rebound, guarding, and rigidity may be present. Although the diagnosis can often be made on clinical grounds alone, an imaging study should be performed during a patient’s initial presentation to confirm the presence of diverticula. This can be done as an outpatient,
ACUTE ABDOMINAL PAIN
673
provided none of the aforementioned red flags are present. CT of the abdomen and pelvis with intravenous and oral contrast is the diagnostic modality of choice, with a reported sensitivity as high as 98% [42]. Colonoscopy should not be performed in patients with suspected diverticulitis, as perforation is a contraindication for endoscopy. Management of mild, uncomplicated diverticulitis can occur on an outpatient basis, and consists of a clear liquid diet and the administration of oral antibiotics that cover typical gastrointestinal pathogens. Complicated diverticulitis occurs when patients develop intraabdominal abscesses, fistula, free perforation, or intestinal obstruction. Patients with complicated diverticulitis or those with mild disease who fail to respond to above therapies require hospitalization. Patients should be started on intravenous antibiotics, made NPO, and be evaluated by a surgeon. Intraabdominal abscesses can often be managed with percutaneous drainage catheters, but surgery is sometimes required [43]. Free perforation or intestinal obstruction usually mandates emergent surgery. Obstruction Bowel obstruction occurs when the normal flow of intestinal contents is interrupted by a mechanical blockage. In patients with a history of abdominal surgery, nearly 75% of cases of small bowel obstruction (SBO) are the result of adhesive peritoneal bands [44–46]. In fact, up to 15% of patients who undergo laparotomy will be readmitted within 2 years with SBO from adhesions, while up to 3% will require operative intervention as a result [46]. Furthermore, it is estimated that the 10-year risk of developing recurrent SBO from adhesions is around 40% [47]. Hernias are the second most common cause of SBO, and account for up to 25% of cases [48]. The remainder of cases of SBO result from a number of etiologies, including Crohn’s disease, volvulus, neoplasm, intussusception, gallstones, and ischemia. The clinical presentation of large bowel obstruction (LBO) is very similar to that of SBO. Nearly 60% of cases of LBO are the result of malignancy, with colon cancer being the most common. Other causes include diverticular strictures and colonic volvulus [49]. The cecum and the sigmoid colon are the most common locations of colonic volvulus [50]. Once the bowel is obstructed, the segment of bowel proximal to the obstruction becomes increasingly distended by swallowed air, gas from bacterial fermentation, and luminal secretions. Bacterial overgrowth, bowel edema, and loss of absorptive function follow. If the obstruction is not promptly treated, then ischemia, necrosis and perforation may occur. The pain caused by SBO is a colicky, diffuse pain that waxes and wanes over 5- minute intervals. Nausea, vomiting, distention, and obstipation are often associated with the abdominal pain. Emesis is often feculent due to bacterial overgrowth. The passage of stool and flatus do not eliminate SBO from the differential diagnosis, as luminal contents distal to the
674
FLASAR
et al
blockage can still pass. Patients will often exhibit physical signs of volume depletion. Abdominal examination reveals a distended abdomen with either hyperactive high-pitched or hypoactive bowel sounds. Rushes of luminal fluid can often be heard. The abdomen is usually diffusely tender, with findings of rigidity, rebound tenderness, or guarding suggesting peritonitis. A ventral, inguinal, or periumbilical hernia should be sought as a potential etiology for the obstruction. Laboratory analysis is usually nonspecific, but common abnormalities include hemoconcentration, leukocytosis, and electrolyte imbalances. An abdominal plain film series should be the initial diagnostic imaging test in patients with suspected obstruction. Typical findings include air–fluid levels, small bowel distention, and a paucity of air in the rectal vault. In addition, evidence of complications such as intraperitoneal free air can be seen. Although most diagnoses can be made clinically with the confirmatory assistance of plain films, there are instances where plain films are not sufficient. In these instances, CT may be helpful for both diagnosing SBO and determining the etiology, with a reported sensitivity of 100% and accuracy of 90% [51,52]. Patients with evidence of bowel obstruction should be admitted to the hospital, both for decompression and observation. Patients are initially managed with strict restriction of oral intake, nasogastric tube decompression, intravenous fluids, and electrolyte repletion. Early surgical evaluation is mandatory given the perforation risk if left unattended. The philosophy that ‘‘the sun should neither rise nor set on a bowel obstruction,’’ still remains true today. Peptic ulcer disease Peptic ulcer disease (PUD) is a common affliction that significantly impacts quality of life. In 1989, more than 5 billion dollars were spent on the care of patients with PUD [53]. Helicobacter pylori infection, the most common cause of PUD, has been associated with 75% to 95% of duodenal ulcers (DU) and 65% to 95% of gastric ulcers (GU) [54–56]. Nonsteroidal anti-inflammatory medications (NSAIDs) are the second most common cause of PUD, with an estimated yearly incidence of clinically significant gastric or duodenal ulceration of approximately 1.5% [57]. Use of NSAIDs presents a particular challenge, as up to 40% of patients will not report the use of NSAIDs [58]. Acid hypersecretory syndromes such as Zollinger Ellison syndrome accounts for the majority of the remaining cases. The clinical presentation of PUD depends on the location of the ulcer, and whether complications from the ulcer develop. Complications of PUD include bleeding, obstruction, perforation, and penetration into adjacent structures. Patients with uncomplicated peptic ulcers may be asymptomatic, or they may present with pain in the upper abdomen [59,60]. The pain is typically described as a burning or gnawing pain, but patients may occasionally describe it as
ACUTE ABDOMINAL PAIN
675
crampy. Nausea and vomiting may also be seen. Relation of pain to meals is unreliable to differentiate DU from GU. Bleeding from PUD may present with melena, hematochezia, or hematemesis, with or without hemodynamic compromise. Bleeding can generally be managed medically with intravenous fluid, blood transfusions, antisecretory therapy, and endoscopic therapy. Endoscopy is also useful to determine the risk for recurrent bleeding [61]. Surgical or angiographic intervention is reserved for bleeding refractory to endoscopic therapies. Pyloric channel and duodenal bulb ulcers may cause gastric outlet obstruction. In addition to epigastric pain, patients with outlet obstruction may present with nausea, projectile vomiting, early satiety, anorexia, and weight loss. Conservative measures are often successful, although many patients will require surgery or endoscopic dilatation therapy [62,63]. Most ulcers that perforate are located in the duodenal bulb, and are often associated with NSAID use [64,65]. Patients present with the sudden onset of epigastric pain that quickly becomes diffuse as generalized peritonitis ensues. Patients can sometimes develop paradoxic improvement in their pain following perforation despite a markedly rigid and diffusely tender abdomen. Plain films are usually adequate to confirm the diagnosis of ulcer perforation. Perforations require immediate surgical evaluation. Ulcer penetration into adjacent structures occurs in up to 20% of cases of PUD, but only a small proportion become clinically apparent [66]. The most common sites of ulcer penetration include the pancreas, omentum, hepatobiliary system, colon, and adjacent vasculature. Patient presentation reflects the structure that is involved, and the therapy is site-specific. Ischemic bowel disease Depending on the location, degree, and acuity of the vascular compromise, ischemic bowel disease is classified into three distinct syndromes: acute mesenteric ischemia, chronic mesenteric ischemia, and colonic ischemia. Acute mesenteric ischemia results from the rapid loss of blood supply to the portion of the intestines supplied by the celiac, superior mesenteric, or inferior mesenteric arteries. The cause is most commonly thromboembolic disease. The consequences of acute mesenteric ischemia are severe, and include bowel necrosis, infarction, and death. Chronic mesenteric ischemia results from the gradual loss of blood supply to the portion of the intestines supplied by the celiac, superior mesenteric, or inferior mesenteric arteries. The cause is usually atherosclerosis. Patients with chronic mesenteric ischemia present with chronic postprandial abdominal pain, which is termed intestinal angina. Because eating worsens the pain, patients develop a fear of eating (sitophobia), and significant weight loss may occur. Colonic ischemia, also known as ischemic colitis, is the most commonly encountered intestinal vascular disorder [67]. Colonic ischemia occurs when there is a decrease in colonic mucosal oxygenation. In the vast majority of patients, colonic
676
FLASAR
et al
ischemia does not result from an occlusive vascular process, but rather occurs when the oxygen requirements to a specific portion of the colon are not met by the vascular supply. Colonic ischemia occurs in the portions of the colon where blood flow is least redundant, such as the splenic flexure, and rectosigmoid junction. Lower gastrointestinal bleeding, rather than abdominal pain, is the most common presenting symptom. The disorder is self-limited in the majority of cases, and the prognosis is good. Of the three ischemic bowel syndromes, acute mesenteric ischemia is the disease that presents with acute abdominal pain, and will be further discussed below. Acute interruption of blood supply in the mesenteric vasculature results from either thromboembolic disease or vasospasm. The major risk factors include advanced age, hypercoaguability, vascular disease, and cardiac disorders such as atrial fibrillation or valvular disease. Once the blood supply to the mesenteric vascular is interrupted, acute ischemia ensues. If the vascular compromise persists, bowel infarction, necrosis, and perforation may occur. Patients with acute mesenteric ischemia present with acute onset, severe periumbilical abdominal pain. Early in the disease course, the pain is often out of proportion to tenderness produced during the physical examination. If the patient presents after bowel infarction has already occurred, peritoneal signs may develop. The stool may be positive for occult blood, but hematochezia is uncommon with acute mesenteric ischemia. Common laboratory abnormalities seen with acute mesenteric ischemia include leukocytosis and an elevated hematocrit from hemoconcentration. A low serum bicarbonate, metabolic acidosis, and elevated lactate level are seen once bowel infarction has occurred. Retrospective studies evaluating the role of elevated plasma D-dimer levels in the diagnosis of early mesenteric ischemia showed initial promise, although subsequent prospective evaluations have shown D-dimer to be less helpful [68,69]. Several imaging modalities, including plain films, Doppler ultrasound, conventional CT, and MRI have been studied for the diagnosis of acute mesenteric ischemia. Unfortunately, these imaging techniques lack sensitivity and specificity to accurately make the diagnosis [70]. Mesenteric angiography is the ‘‘gold standard’’ test for diagnosing occlusive arterial mesenteric ischemia. Its sensitivity and specificity are 75% to 100% and 100%, respectively [67]. In addition to its diagnostic capabilities, angiography offers the potential for treatment. Several studies demonstrate a decreased mortality in patients who undergo angiography for suspected occlusive mesenteric arterial ischemia [71,72]. The mortality rate for patients with acute mesenteric ischemia in whom the diagnosis is not made before the onset of bowel infarction is reported to be as high as 90% [67]. Therefore, early diagnosis is crucial. Because laboratory findings may be nonspecific early in the disease course, a high index of suspicion, based upon predisposing risk factors, and clinical presentation is required. Patients with suspected acute mesenteric ischemia should have prompt angiography and surgical evaluation [73].
ACUTE ABDOMINAL PAIN
677
Abdominal aortic aneurysm It is estimated that as many as 10% of patients over the age of 65 have an abdominal aortic aneurysm (AAA). Rupture of the AAA carries an overall mortality rate of nearly 90%, which only falls to 70% if patients survive to reach the operating room [74,75]. Because of such high mortality and rapid course, a ruptured AAA is unlikely to present in an outpatient clinic. However, any severe abdominal pain complaint in a patient with a known AAA mandates immediate referral to an emergency department for evaluation. Inflammatory bowel disease Inflammatory bowel disease (IBD) encompasses ulcerative colitis (UC), Crohn’s disease (CD), and indeterminate colitis. All three disorders are chronic, characterized by disease-free intervals, followed by flares of disease. These disorders are generally managed in the outpatient setting, and abdominal pain is often a component of active disease. Nonetheless, there are several acute, potentially life-threatening complications from IBD that may present as abdominal pain in the outpatient setting. These include fulminate colitis, toxic megacolon, bowel obstruction, bowel perforation, and abscess formation. Fulminate colitis is typically associated with UC, and is the initial presenting scenario in up to 10% of patients with UC [76–78]. It is defined as abdominal pain, O10 bloody bowel movements per day, volume depletion, anemia, and any two of the following: white blood count O10,500 cells/mL, fever O38.6 C, tachycardia, and hypoalbuminemia [79]. Patients with known UC will complain of increasingly severe, crampy, generalized abdominal pain in addition to the typical complaints of bloody diarrhea, urgency, and tenesmus. Toxic megacolon occurs in colitis patients when there is pathologically dilated large bowel in conjunction with evidence of systemic toxicity. Early series reported mortality rates of 19%, although more recent series estimate overall mortality approaching 0%, owing to earlier recognition and improved management strategies [80–82]. Toxic megacolon was originally described, and is most commonly seen in the setting of UC, but can also occur with Crohn’s colitis, infectious colitis (especially Clostridium difficile), ischemic colitis, diverticulitis, and colon cancer. Toxic megacolon generally occurs early in the course of UC, with 30% of cases occurring within the first 3 months of diagnosis and 60% of cases within 3 years of diagnosis [83]. Physical examination classically reveals abdominal distension with tympany to percussion, as well as tenderness above the underlying colon. However, examination findings are less reliable in the setting of active corticosteroid therapy. In patients with peritoneal signs, perforation should be strongly suspected. The diagnosis is made based on the presence of colonic distension (O6 cm) on imaging plus any three of the following: fever
678
FLASAR
et al
(O38.6 C), leukocytosis (O10,500 cells/mL), anemia, or tachycardia (O120 beats/min), plus any one of the following: altered mental status, volume depletion, hypotension, or electrolyte abnormalities [84]. The patient with either fulminate colitis or toxic megacolon should be hospitalized immediately for aggressive medical care and surgical evaluation. They should be made NPO, started on intravenous hydration with appropriate electrolyte repletion, and in cases of toxic megacolon, should have a nasogastric tube placed to facilitate decompression. Bowel obstruction is common in CD, and is most frequently seen in the terminal ileum [85]. Obstruction most commonly results from active inflammatory intestinal strictures, postinflammatory fibrotic intestinal strictures, or peritoneal adhesive disease resulting from previous abdominal surgeries. Although far more common in CD, strictures occur in about 5% of patients with UC, with up to 30% representing malignant disease [86]. Perforation can occur both with CD and UC. In patients with UC, it is most commonly the result of toxic megacolon, and carries a mortality rate as high as 50% [87]. In patients with CD, perforation results from unrelieved small bowel obstructions. Because of the powerful immunosuppressive medications that IBD patients are frequently taking, the clinical severity of a perforation may be muted. Therefore, a high index of suspicion is needed. Intraabdominal abscess formation is common in CD, occurring in approximately 25% of patients [88–90]. They result from microperforations in patients with penetrating or stricturing disease. Patients typically present with fever, leukocytosis, and abdominal pain. Additionally, they may experience back or groin pain if the abscess involves the ileopsoas or pelvic structures, respectively. A special subset of abscess patients includes those with perianal disease, where the abscess occurs in the perirectal fascia, musculature, or adipose tissue. Approximately 30% of patients with perianal fistulizing CD will develop a perirectal abscess [91]. These patients may note low pelvic/perineal pain, defecatory urgency, or tenesmus in addition to constitutional symptoms. Severity may be blunted secondary to concomitant immunosuppressive medications. Regardless, patients with suspected perirectal or intraabdominal abscess with a history of CD warrant hospital admission for antibiotics, surgical evaluation, and advancement of medical therapy when appropriate. Irritable bowel syndrome Irritable bowel syndrome (IBS) has an estimated prevalence in the United States as high as 22%, though only one third will ever present for medical evaluation [92]. It is estimated to account for 12% of primary care visits [93]. The hallmark of the disorder is chronic abdominal pain or discomfort that is associated with altered bowel movements. The pathogenesis is not fully understood, but altered motility, visceral hypersensitivity, luminal factors, and psychologic etiologies are felt to all play a role. Affected patients
ACUTE ABDOMINAL PAIN
679
exhibit variation in the description of the abdominal pain, including cramping, bloating, aching, and even sharp localized pain. Although IBS is troubling to the patient and detrimental to everyday functioning and quality of life, IBS is not a life-threatening process. Therefore, it should be managed on an outpatient basis. Associated abnormalities such as anemia, poor nutritional status, weight loss, evidence of infection, fever, and electrolyte or metabolic abnormalities suggest an alternative diagnosis. Special populations There are several populations of patients who warrant special consideration when presenting with abdominal pain because of either underlying processes that are unique to these groups, or because presentations tend to be atypical in these groups. These include the elderly, immunosuppressed (including patients with AIDS), patients on analgesics, women of childbearing age, and pregnant women. Additionally, patients presenting under the influence of alcohol or illicit substances often exhibit atypical presentations of common disorders. Patients O65 years of age represent the fastest growing population demographic in the United States [94–97]. The elderly often delay seeking medical care, causing them to present at a potentially more dangerous point in their disease course. Compounding this, the history and physical examination have less reliability in the elderly. Many factors contribute to this, including underlying central nervous system disorders, fear of losing independence, hearing loss, depression, complex medical histories, vague description of the discomfort, polypharmacy, and change in normal physiology (eg, inability to mount leukocytosis or pyrexic response to infection). As a result, diagnostic accuracy has been reported as low as 40% in elderly patients with acute abdominal pain [98]. An important point to consider when evaluating elderly patients is that common disorders may manifest with uncommon presentations. For example, both coronary ischemia and urinary tract infections have been well-described causes of abdominal pain in the elderly. Also, it is not uncommon for an elderly patient to present with altered mental status as the lone sign of an acute abdominal process. Immunosuppressed patients and patients with immunodeficiency syndromes who present with abdominal pain generate a more expansive differential diagnosis (especially infectious causes) for their pain owing to their inability to mount a normal immune response. As in the elderly, the physical examination may be less accurate owing to an abnormal inflammatory response to pathologic processes. The differential diagnosis and evaluation of gynecologic and obstetric processes manifesting as acute abdominal pain are beyond the scope of this review. However, a few points need to be stressed. All women of childbearing age require a pelvic examination and evaluation for elevated b-HCG in the workup of acute abdominal pain. Additionally, as the gravid uterus
680
FLASAR
et al
enlarges in a pregnant patient, the normal topography of the small and large intestines may be altered due to mass effect, thereby making the location of abdominal tenderness atypical. This is classically true with appendicitis [99]. Abdominal pain is a common complaint in the outpatient setting. Although many etiologies have a benign course, some have potentially morbid and lethal complications. Care providers must understand the basis of the perception of abdominal pain, and develop a focused approach to the initial evaluation of these patients. Performing a thorough history and physical evaluation will allow the practitioner to generate a differential diagnosis that will guide further laboratory, imaging, and management decisions. References [1] Woodwell DA, Cherry DK. National Ambulatory Medical Care Survey: 2002 summary. Adv Data 2004;(346):1–44. [2] Feldman M, Friedman LS, Sleisenger MH. Sleisenger and Fordtran’s gastrointestinal and liver disease: pathophysiology/diagnosis/management. 7th edition. Philadelphia: Saunders; 2002. [3] Benedict M, Bucheli B, Battegay E, et al. First clinical judgment by primary care physicians distinguishes well between organic and nonorganic causes of abdominal or chest pain. J Gen Intern Med 1997;12(8):459–65. [4] Yamamoto W, Kono H, Maekawa M, et al. The relationship between abdominal pain regions and specific diseases: an epidemiologic approach to clinical practice. J Epidemiol 1997;7(1):27–32. [5] Maglinte DDT, Balthazar EJ, Kelvin FM, et al. The role of radiography in the diagnosis of small bowel obstruction. AJR Am J Roentgenol 1997;168(5):1171–80. [6] Miller RE, Nelson SW. The roentographic demonstration of tiny amounts of free intraperitoneal gas: experimental and clinical studies. AJR Am J Roentgenol 1971;112(3):574–85. [7] Billittier AJ, Abrams BJ, Brunetto A. Radiographic imaging modalities for the patient in the emergency department with abdominal complaints. Emerg Med Clin North Am 1996;14(4): 789–850. [8] Gupta H, Dupuy D. Advances in imaging of the acute abdomen. Surg Clin North Am 1997; 77(6):1245–63. [9] Ahn SH, Mayo-Smith WW, Murphy BL, et al. Acute nontraumatic abdominal pain in adult patients: abdominal radiography compared with CT evaluation. Radiology 2002;225(1): 159–64. [10] Tsushima Y, Yamada S, Aoki J, et al. Effect of contrast-enhanced computed tomography on diagnosis and management of acute abdomen in adults. Clin Radiol 2002;57(6):507–13. [11] Stapakis JP, Thickman D. Diagnosis of pneumoperitoneum: abdominal CT vs. upright chest film. J Comput Assist Tomogr 1992;16(5):713–6. [12] Berger MY, van der Velden JJ, Lijmer JG, et al. Abdominal symptoms: do they predict gallstones? A systematic review. Scand J Gastroenterol 2000;35(1):70–6. [13] Yusoff IF, Barkun JS, Barkun AN. Diagnosis and management of cholecystitis and cholangitis. Gastroenterol Clin North Am 2003;32(4):1145–68. [14] Raine PA, Gunn AA. Acute cholecystitis. Br J Surg 1975;62(9):697–700. [15] Kadakia SC. Biliary tract emergencies. Med Clin North Am 1993;77(5):1015–36. [16] van der Linden W, Sunzel H. Early versus delayed operation for acute cholecystitis: a controlled trial. Am J Surg 1970;120(1):7–13. [17] McArthur P, Cuschieri A, Sells R, et al. Controlled clinical trial comparing early with interval cholecystectomy for acute cholecystitis. Br J Surg 1975;62(10):850–2.
ACUTE ABDOMINAL PAIN
681
[18] Jarvinen H, Hastbacka J. Early cholecystectomy for acute cholecystitis: a prospective randomized trial. Ann Surg 1980;191(4):501–5. [19] Lai PB, Kwong KH, Leung KL, et al. Randomized trial of early versus late laparoscopic cholecystectomy for acute cholecystitis. Br J Surg 1998;85(6):764–7. [20] Lo CM, Liu CL, Fan ST, et al. Prospective randomized study of early versus late laparoscopic cholecystectomy for acute cholecystitis. Ann Surg 1998;227(4):461–7. [21] Chandler CF, Lane JS, Ferguson P, et al. Prospective evaluation of early versus delayed laparoscopic cholecystectomy for treatment of acute cholecystitis. Am Surg 2000;66(9): 896–900. [22] Brodsky A, Matter I, Sabo E, et al. Laparoscopic cholecystectomy for acute cholecystitis: can the need for conversion and the probability for complications be predicted? Surg Endosc 2000;14(8):755–60. [23] Pessaux P, Teuch JJ, Rouge C, et al. Laparoscopic cholecystectomy in acute cholecystitis: a prospective comparative study in patients with acute vs. chronic cholecystitis. Surg Endosc 2000;14(4):358–61. [24] Maluenda F, Csendes A, Burdiles P, et al. Bacteriologic study of choledocal bile in patients with common bile duct stones, with or without acute suppurative cholangitis. Hepatogastroenterology 1989;36(3):132–5. [25] Saharia PC, Zuidema GD, Cameron JL. Primary common duct stones. Ann Surg 1977; 185(5):598–604. [26] Saik RP, Greenburg AG, Farris JM, et al. Spectrum of cholangitis. Am J Surg 1975;130(2): 143–50. [27] Lipsett PA, Pitt HA. Acute cholangitis. Surg Clin North Am 1990;70(6):1297–312. [28] Banks PA. Practice guidelines in acute pancreatitis. Am J Gastroenterol 1997;92(3):377–86. [29] Ros E, Navarro S, Bru C, et al. Occult microlithiasis in ‘‘idiopathic’’ acute pancreatitis: prevention of relapses by cholecystectomy or ursodeoxycholic acid therapy. Gastroenterology 1991;101(6):1701–9. [30] Gullo L, Migliori M, Olah A, et al. Acute pancreatitis in five European countries: etiology and mortality. Pancreas 2002;24(3):223–7. [31] Steinberg WM, Goldstein SS, Davis ND, et al. Diagnostic assays in acute pancreatitis. A study of sensitivity and specificity. Ann Intern Med 1985;102(5):576–80. [32] Brown A, Baillargeon J-D, Hughes MD, et al. Can fluid resuscitation prevent pancreatic necrosis in severe acute pancreatitis? Pancreatol 2002;2(2):104–7. [33] Tenner S, Dubner H, Steinberg W. Predicting gallstone pancreatitis with laboratory parameters: a meta-analysis. Am J Gastroenterol 1994;89(10):1863–6. [34] Simchuk EJ, Traverso LW, Nukui Y, et al. Computed tomography severity index is a predictor of outcomes for severe pancreatitis. Am J Surg 2000;179(5):352–5. [35] Owings MF, Kozak LJ. Ambulatory and inpatient procedures in the United States, 1996. Vital and health statistics. Series 13. No. 139. Hyattsville (MD): National Center for Health Statistics; 1998. p. 26. [36] Paulson EK, Kalady MF, Pappas TN. Suspected appendicitis. N Engl J Med 2003;348(3): 236–42. [37] Graffeo CS, Counselman FL. Appendicitis. Emerg Med Clin North Am 1996;14(4):653–71. [38] Rothrock SG, Green SM, Dobson M, et al. Misdiagnosis of appendicitis in nonpregnant women of childbearing age. J Emerg Med 1995;13(1):1–8. [39] Freeman SR, McNally PR. Diverticulitis. Med Clin North Am 1993;77(5):1149–67. [40] Parks TG. Natural history of diverticular disease of the colon. Clin Gastroenterol 1975;4: 53–69. [41] Markham NI, Li AK. Diverticulitis of the right colon: experience from Hong Kong. Gut 1992;33(4):547–9. [42] Ambrosetti P, Jenny A, Becker C, et al. Acute left colonic diverticulitis: compared performance of computed tomography and water-soluble contrast enema: prospective evaluation of 420 patients. Dis Colon Rectum 2000;43(10):1363–7.
682
FLASAR
et al
[43] Schecter S, Eisenstat TE, Oliver GC, et al. Computerized tomographic scan-guided drainage of intra-abdominal abscesses: preoperative and postoperative modalities in colon and rectal surgery. Dis Colon Rectum 1994;37(10):984–8. [44] Bizer LS, Liebling RW, Delany HM, et al. Small bowel obstruction: the role of nonoperative treatment in simple intestinal obstruction and predictive criteria for strangulation obstruction. Surgery 1981;89(4):407–13. [45] Greene WW. Bowel obstruction in the aged patient: a review of 300 cases. Am J Surg 1969; 118(4):541–5. [46] Beck DE, Opelka FG, Bailey HR, et al. Incidence of small-bowel obstruction and adhesiolysis after open colorectal and general surgery. Dis Colon Rectum 1999;42(2): 241–8. [47] Landercasper J, Cogbill TH, Merry WH, et al. Long-term outcome after hospitalization for small-bowel obstruction. Arch Surg 1993;128(7):765–70. [48] Mucha P Jr. Small intestinal obstruction. Surg Clin North Am 1987;67(3):597–620. [49] Kahi CJ, Rex DR. Bowel obstruction and pseudo-obstruction. Gastroenterol Clin North Am 2003;32(4):1229–47. [50] Ballantyne GH, Brandner MD, Beart RW Jr, et al. Volvulus of the colon: incidence and mortality. Ann Surg 1985;202(1):83–92. [51] Frager D. Intestinal obstruction: role of CT. Gastroenterol Clin North Am 2002;31(3): 777–99. [52] Frager D, Baer JW, Medwid SW, et al. Detection of intestinal ischemia in patients with acute small-bowel obstruction due to adhesions or hernia: efficacy of CT. AJR Am J Roentgenol 1996;166(1):67–71. [53] Sonnenberg A, Everhart JE. Health impact of peptic ulcer in the US. Am J Gastroenterol 1997;92(4):614–20. [54] Borody TJ, George LL, Brandl S, et al. Helicobacter pylori-negative duodenal ulcer. Am J Gastroenterol 1991;86(9):1154–7. [55] Ciociola AA, McSorley DJ, Turner K, et al. Helicobacter pylori infection rates in duodenal ulcer patients in the United States may be lower than previously estimated. Am J Gastroenterol 1999;94(7):1834–40. [56] Tytgat G, Langenberg W, Rauws E, et al. Campylobacter-like organism (CLO) in the human stomach. Gastroenterology 1985;88(5):1620 [abstract]. [57] Silverstein FE, Graham DY, Senior JR, et al. Misoprostol reduces serious gastrointestinal complications in patients with rheumatoid arthritis receiving nonsteroidal anti-inflammatory drugs. A randomized, double-blind, placebo-controlled trial. Ann Intern Med 1995; 123(4):241–9. [58] Lanas AI, Remacha B, Esteva F, et al. Risk factors associated with refractory peptic ulcers. Gastroenterology 1995;109(4):1124–33. [59] Kuipers EJ, Thijs JC, Festen HP. The prevalence of Helicobacter pylori in peptic ulcer disease. Aliment Pharmacol Ther 1995;9(Suppl 2):59–69. [60] Hilton D, Iman N, Burke GJ, et al. Absence of abdominal pain in older persons with endoscopic ulcers: a prospective study. Am J Gastroenterol 2001;96:380. [61] Laine L, Cohen H, Brodhead J, et al. Prospective evaluation of immediate versus delayed refeeding and prognostic value of endoscopy in patients with upper gastrointestinal hemorrhage. Gastroenterology 1992;102(2):314–6. [62] Weiland D, Dunn DH, Humphrey EW, et al. Gastric outlet obstruction in peptic ulcer disease: an indication for surgery. Am J Surg 1982;143(1):90–3. [63] Boylan JJ, Gradzka MI. Long-term results of endoscopic balloon dilatation for gastric outlet obstruction. Dig Dis Sci 1999;44(9):1833–6. [64] Gunshefski L, Flancbaum L, Brolin RE, et al. Changing patterns in perforated peptic ulcer disease. Am Surg 1990;56(4):270–4. [65] Lanas A, Serrano P, Bajador E, et al. Evidence of aspirin use in both upper and lower gastrointestinal perforation. Gastroenterology 1997;112(3):683–9.
ACUTE ABDOMINAL PAIN
683
[66] Norris JR, Haubrich WS. The incidence and clinical features of penetration in peptic ulceration. JAMA 1961;178:386–9. [67] Walker JS, Dire DJ. Vascular abdominal emergencies. Emerg Med Clin North Am 1996; 14(3):571–92. [68] Acosta S, Nilsson TK, Bjorck M. Preliminary study of D-dimer as a possible marker of acute bowel ishaemia. Br J Surg 2001;88(3):385–8. [69] Acosta S, Nilsson TK, Bjorck M. D-dimer testing in patients with suspected acute thromboembolic occlusion of the superior mesenteric artery. Br J Surg 2004;91(8):991–4. [70] Lefkovitz Z, Cappell MS, Lookstein R, et al. Radiologic diagnosis and treatment of gastrointestinal hemorrhage and ischemia. Med Clin North Am 2002;86(6):1357–99. [71] Boos S. Angiography of the mesenteric artery 1976 to 1991: a change in the indications during mesenteric circulatory disorders. Radiologe 1992;32(4):154–7. [72] Clark RA, Gallant TE. Acute mesenteric ischemia: angiographic spectrum. Am J Radiol 1984;142(3):555–62. [73] Kaleya R, Sammartano R, Boley SJ. Aggressive approach to acute mesenteric ischemia. Surg Clin North Am 1992;35(6):613–23. [74] Multicentre Aneurysm Screening Study Group. The Multicentre Aneurysm Screening Study (MAS) into the effect of abdominal aneurysm screening on mortality in men: a randomised controlled trial. Lancet 2002;360:1531–9. [75] Ruttedge RA, Oller DW, Meyer AA, et al. A statewide, population-based, time-series analysis of the outcome of ruptured abdominal aortic aneurysm. Ann Surg 1996;223: 492–505. [76] Becker JM. Surgical therapy for ulcerative colitis and Crohn’s disease. Gastroenterol Clin N Am 1999;28:371–90. [77] Greenstein AJ, Kark AE, Dreiling DA. Crohn’s disease of the colon. III. Toxic dilatation of the colon in Crohn’s disease. Am J Gastroenterol 1975;63:117–28. [78] Present D. Toxic megacolon. Med Clin North Am 1993;77(5):1129–48. [79] Truelove SC, Witts LF. Cortisone in ulcerative colitis: final report on a therapeutic trial. BMJ 1955;2:1041–8. [80] Katzka I, Katz S, Morris E. Management of toxic megacolon: the significance of early recognition and medical management. J Clin Gastroenterol 1983;78:557–9. [81] Mungas JE, Mooja AR, Block GE. Treatment of toxic megacolon. Surg Clin N Amer 1976; 56:95–102. [82] Strauss RJ, Flint GW, Platt N, et al. The surgical management of toxic dilatation of the colon: a report of 28 cases and review of the literature. Ann Surg 1976;184(6):682–8. [83] Kleer CG, Appleman HD. Ulcerative colitis: patterns of involvement in colorectal biopsies and changes with time. Am J Surg Pathol 1998;22:983–9. [84] Jalan KN, Sircus W, Card WI, et al. An experience with ulcerative colitis: toxic dilation in 55 cases. Gastroenterology 1969;57:68–82. [85] Cheung O, Regueiro MD. Inflammatory bowel disease emergencies. Gastroenterol Clin North Am 2003;32:1269–88. [86] Chutkan R. Inflammatory bowel disease. Prim Care 2001;28(3):539–56. [87] Greenstein AJ, Aufses AH Jr. Differences in pathogenesis, incidence, and outcome of perforation in inflammatory bowel disease. Surg Gynecol Obstet 1985;160:63–9. [88] Cybulsky IJ, Tam P. Intra-abdominal abscesses in Crohn’s disease. Am Surg 1990;56: 672–82. [89] Lambiase RE, Cronan JJ, Dorfman GS, et al. Percutaneous drainage of abscesses in patients with Crohn’s disease. AJR. AJR Am J Roentgenol 1988;150:1043–5. [90] Ribeiro MB, Greenstein AJ, Yamazaki Y, et al. Intra-abdominal abscesses in regional enteritis. Ann Surg 1991;179:495–8. [91] Makowiec F, Jehle EC, Becker HD, et al. Perianal abscess in Crohn’s disease. Dis Colon Rectum 1997;40:443–50. [92] Borum ML. Irritable bowel syndrome. Prim Care 2001;28(3):523–38.
684
FLASAR
et al
[93] Mitchell CM, Drossman DA. Survey of the AGA membership relating to patients with functional gastrointestinal disorders. Gastroenterol 1987;92:1282. [94] Ciccone A, Allegra JR, Cochrane DG, et al. Age-related differences in diagnoses within the elderly population. Am J Emerg Med 1998;16:43–9. [95] Eliastam M. Elderly patients in the emergency department. Ann Emerg Med 1989;18: 1222–9. [96] Kamin RA, Nowicki TA, Courtney DS, et al. Pearls and pitfalls in the emergency department evaluation of abdominal pain. Emerg Med Clin North Am 2003;21:61–72. [97] deDombal FT. Acute abdominal pain in the elderly. J Clin Gastroenterol 1994;19:331–5. [98] Marco CA, Schoenfeld CN, Keyl PM, et al. Abdominal pain in geriatric emergency patients: variables associated with adverse outcomes. Acad Emerg Med 1998;5:1163–8. [99] Weingold AB. Appendicitis in pregnancy. Clin Obstet Gynecol 1983;26:801.
Prim Care Clin Office Pract 33 (2006) 685–695
Dermatologic Emergencies Brian J. Browne, MD, FACEP, Brian Edwards, MD*, Robert L. Rogers, MD, FAAEM, FACEP, FACP Department of Emergency Medicine, The University of Maryland School of Medicine, 110 South Paca Street, Sixth Floor, Suite 200, Baltimore, MD 21201, USA
Primary care physicians are the gatekeepers of the medical community. They are the physicians to whom patients first present, and they are often the physicians with whom patients have the longest lasting relationships. Primary care physicians, as a result of these long-term relationships, have been endowed with a unique responsibility to the health of their patients. By the very nature of their practice, primary care physicians do not have the resources to treat emergent life-threatening conditions. They must, however, be able to diagnose these potentially life-threatening conditions and be able to stabilize and appropriately refer a patient for urgent evaluation by specialists or emergency physicians. There are many types of emergencies encountered in the outpatient setting, ranging from cardiac to toxicologic. As important as recognizing signs and symptoms of cardiac ischemia is the ability to recognize potentially life-threatening dermatologic disorders or dermatologic manifestations of life-threatening systemic diseases. There are innumerable dermatologic conditions that can be potentially fatal. The primary care physician must be able to recognize a select few that demand immediate evaluation by emergency physicians, dermatologists, intensivists, and surgeons, among others. Included in this brief review will be the following conditions: Stevens-Johnson Syndrome/Toxic Epidermal Necrolysis, pemphigus vulgaris, scalded skin syndrome, Rocky Mountain Spotted fever (RMSF), and Lyme disease. Many of these conditions require aggressive inpatient treatment, and must be immediately recognizable to the primary care physician. Although RMSF, Lyme disease and pemphigus vulgaris may not require rapid transfer to a higher level of care, they do demand that the clinician remain vigilant and be able to rapidly diagnose these conditions to avoid potentially serious sequelae. * Corresponding author. E-mail address:
[email protected] (B. Edwards). 0095-4543/06/$ - see front matter Ó 2006 Elsevier Inc. All rights reserved. doi:10.1016/j.pop.2006.06.002 primarycare.theclinics.com
686
BROWNE
et al
Stevens-Johnson syndrome/toxic epidermal necrolysis Stevens-Johnson syndrome (SJS) and toxic epidermal necrolysis (TEN) are potentially serious conditions that need to be recognized so that patients may receive appropriate referral and care. SJS was first described in 1922, in two children, aged 7 and 8 years, who presented with ‘‘an extraordinary, generalized eruption with continued fever, inflamed buccal mucosa and severe purulent conjunctivitis’’ [1]. TEN was first described by Ruskin, in 1948, and again by Lyell, in 1956, although two of Lyell’s four patients actually had scalded skin syndrome [1]. Clinically, SJS and TEN often present with prodromal symptoms such as malaise, rhinitis, odynophagia, myalgias, and arthralgias, which may often be mistaken for a viral syndrome [2]. This prodromal state may last for up to 2 weeks, and is followed by the abrupt development of a macular rash that may or may not be targetoid as classically described. This macular exanthem will mostly start centrally, followed by a spread to the extremities. The exanthem becomes confluent and significant dermal–epidermal dissociation ensues, resulting in a positive Nikolsky’s signddenudation with shear stress. Along with the cutaneous manifestations of SJS/TEN, there is almost invariable involvement of the oral mucosa in terms of blistering and progression to severe erosion. Oropharyngeal SJS/TEN can likewise progress to involvement of the esophagus and trachea leading to increased risk of respiratory failure. It is thought that TEN is merely a severe form of SJS with severity measured in terms of body surface area (BSA) affected. Stevens-Johnson syndrome has been diagnosed when the extent of involvement is !10% and TEN diagnosed when BSA involved is O30% [2]. There remains debate concerning how to classify clinically significant disease affecting between 10% and 30% of BSA. The progression phase of SJS/TEN is variable in terms of length and in terms of ultimately what percentage of BSA is involved. Mucous membrane involvement becomes more apparent during the progression phase, and is found in between 92 and 100% of those with SJS/TEN [1]. As mentioned above, involvement of the respiratory tract can occur. It is characterized by sloughing of the tracheal and bronchial epithelium, resulting in obstructive symptoms. There may be expectoration of bronchial casts, and mechanical ventilation may be required. Involvement of the conjunctivae may also occur. After the progression phase, where there is continued sloughing of the epidermis, there is the ‘‘acme phase’’ where risk of infection and sepsis is at its greatest due to the extent of denudation [2]. Subsequent to this acme phase, the patients begin a healing process of reepithelialization that can last approximately 1 week to 6 months, depending on the extent of BSA involved. The skin generally heals without significant scarring but the new epidermis may be hypo- or hyperpigmented in relation to what native epidermis remains [1]. Study of SJS/TEN has been limited by both the rarity of the ailment and the lack of agreement concerning diagnostic criteria. There have been many
DERMATOLOGIC EMERGENCIES
687
classification systems proposed, but none have been broadly accepted [1]. The incidence of SJS and TEN is therefore quite variable, depending on the classification scheme used. The incidence of SJS runs anywhere between 1.1 and 7.1 cases per million person-years, with a mean age of patients between 25 and 47 years. Older age is associated with an increased degree of skin loss. Mortality has been reported at less than 5% in SJS and 30% to 50% for TEN [3]. The etiology and pathogenesis of SJS and TEN is incompletely worked out. It is thought that the majority of cases are drug induced, with the main culprits being sulfonamide antibiotics, anticonvulsants such as phenytoin or carbamazepine, allopurinol, and nonsteroidal anti-inflammatory medications from the oxicam family [4]. Histologic examination reveals full-thickness necrosis of the epidermis with subepidermal blister formation. From a primary care perspective, the initiation of treatment involves transfer to an appropriate referral center, generally a burn center for those with TEN. As the most common etiology is drug induced, the foundation of treatment is withdrawal of the offending agent. Primary care providers must intimately know which medications their patients take, including those prescribed by consultants. Prompt withdrawal of the offending medications (those recently prescribed or those with a known risk of SJS/TEN) must be instituted as soon as clinical suspicion is aroused [3]. The second tier of treatment involves supportive care such as the multidisciplinary care found at burn centers or other dermatologic referral centers. Although these patients are at risk for fluid and electrolyte abnormalities, there is not the same degree of fluid loss as is found in burn patients. Early institution of enteral feedings, mostly by nasogastric tube, has been advocated due to the fact that SJS/TEN is a hypercatabolic state with high nutritional demands. Appropriate care of the exposed dermis is not standardized. Aggressive debridement is recommended by some, but a more expectant strategy is adopted by others. Exposed dermis should be covered with hydrocolloid dressings or gauze or skin allo-transplantation. Patients with significant involvement of body surface area are at much higher risk for sepsis due to bacterial infection. Given this high risk, the patients should be protected from iatrogenic infections. Some recommend screening bacterial and fungal cultures and advocate prophylactic antibiotics [1–3]. A third tier of treatment involves active interventions to halt the continued epidermolysis. Because SJS/TEN is thought to be an immune-mediated process, many mechanisms of immune suppression, including corticosteroids, intravenous immunoglobulin (IVIG), cyclophosphamide, cyclosporine, and plasmapheresis have been attempted [3]. Studies of the effects of these interventions have been marred by both study size and presence or absence of control groups. Specific discussion of risks and benefits of these highly specialized treatments is beyond the scope of this article. SJS and TEN are rare but clinically important conditions that need to be recognized by the primary care provider. These conditions are mostly
688
BROWNE
et al
iatrogenic in nature, caused by the medications that we prescribe. Early recognition allows time for appropriate referral to a dermatologic hospital or burn center for aggressive multidisciplinary management.
Staphylococcal scalded skin syndrome Staphylococcal scalded skin syndrome (SSSS) is a rare entity in adult and adolescent populations, with the total number of reported cases just above 50. The first case was reported in 1972 [5]. Although rare, if this condition presents in adults, it must be immediately recognized by the primary care provider or emergency physician due to a reported mortality rate of approximately 60% [6]. Similar to TEN and SJS, SSSS encompasses a spectrum from localized bullous impetigo to full-blown SSSS, depending on the distribution of exfoliative toxins. In patients that will ultimately develop widespread SSSS, they often present with fever and other nonspecific symptoms such as malaise [6]. In children, this process can often be associated with a sore throat or a severe conjunctivitis [5]. The bullous eruption begins with areas of erythema that begin centrally. These areas of erythema evolve into flaccid bullae that rupture easily, leaving large ‘‘scalded’’ areas. Given the intraepidermal nature of the cleavage plane, SSSS is associated with less morbidity than TEN even when similar surface areas are involved. These superficial skin lesions often heal over 7 to 10 days, without significant scarring. The SSSS is mediated through toxins produced by Staphylococcus aureus. Approximately 5% of all S aureus produce these toxins [7]. Two exfoliative toxins have been identified, ETA and ETB. ETA is the most common, and therefore clinically significant of the exfoliative toxins, produced by 89% of the isolates. ETB has been found to be produced by 4% of isolates. The correct diagnosis of SSSS is essential, as it must be differentiated from TEN. The latter will generally demand a higher level of care as well as fundamentally different treatment modalities. A major difference between SJS/TEN and SSSS is that mucous membrane involvement is nearly universal in SJS/TEN while absent in SSSS. Beyond that, biopsy of the lesions is the ‘‘gold standard’’ in the differentiation between TEN and SSSS. In TEN, there is epidermal necrosis with dermal sparing, while in SSSS, there is an intraepidermal, subcorneal division. SSSS mostly presents in children younger than 5 years of age, but there are reports of adult involvement [8–10]. The higher prevalence in the very young is likely due to a relative lack of antibodies to ETA and ETB. Antibodies to the exfoliative toxins are found in only 30% of children aged 3 months to 2 years, increasing to 91%t in those over 40 years of age. The low prevalence in adults is secondary to this apparent immunocompetence. Those adults at risk have compromised immune function leading to increased proliferation of S aureus. Adult patients at risk include those with advanced HIV, other deficits in cell-
DERMATOLOGIC EMERGENCIES
689
mediated immunity, malignancy, malnutrition, and others. Exfoliative toxins are renally excreted. Renal failure has been shown to be the most important risk factor for SSSS in adults [5]. Once SSSS is diagnosed, the treatment consists of two parts: treatment of the underlying infection, and supportive care. Because SSSS is, indirectly, an infectious process, the mainstay of treatment is administration of antibiotics. S aureus isolates are mostly sensitive to semisynthetic penicillins such as dicloxacillin, although there have been few reports of methicillinresistant S aureus resulting in SSSS [5,6]. Some physicians may initiate oral antibiotics in limited disease, but if the disease progresses or is already widespread, intravenous antibiotics should be used until progression ceases. Just as in SJS/TEN, supportive care is imperative. Given the intraepidermal nature of the lesions, the hemodynamic and electrolyte imbalances are not as severe as in SJS/TEN. That being said, patients with widespread denudation should be monitored closely for hypotension, electrolyte imbalances, thermal dysregulation, and bacterial superinfection of exposed surfaces.
Pemphigus vulgaris Although more of a historic dermatologic emergency, pemphigus vulgaris (PV) remains a condition associated with significant morbidity and some mortality. The term pemphigus is derived from pemphix, a Greek term meaning bubble or blister. The term pemphigus denotes a family of blistering diseases including pemphigus foliaceous, bullous pemphigoid, paraneoplastic pemphigus and, most importantly to this review, PV. The common thread linking these disorders is acantholysis, the separation of keratinocytes, mediated by host immune response. This brief review will focus mostly on PV, as it is the most potentially life-threatening PV is a bullous condition which, in 50% to 70% of patients, begins with the development of oral lesions [11]. These oral lesions begin as transient flaccid bullae that rupture easily leaving painful, superficial, and irregular nonhealing ulcerations arising from apparently healthy epithelium mainly in the buccal mucosa, lips, and palate [12,13]. Cutaneous bullous disease generally occurs subsequent to the onset of the oral lesions [14]. Flaccid bullae appear on otherwise healthy appearing skin. These bullae exhibit a positive Nikolsky’s sign and rapidly develop into painful erosions that a predilection for the scalp, chest, back and, often, the face and neck [13]. In addition to the oral and cutaneous involvement, involvement of esophageal, laryngeal, nasopharyngeal, vaginal, and rectal epithelium may also occur [15]. The natural history of this disease is one of chronicity with bullae progressing to painful erosions that heal poorly without intervention. Before the use of systemic corticosteroids, the fatality of PV reached 100%, generally due to dehydration or superinfection. Now mortality is now less than 10% [12,16].
690
BROWNE
et al
PV has has an annual incidence of 0.42 per 100,000, peaking between the fourth and sixth decades of life [16]. The disease is mostly prevalent in people of Jewish, Mediterranean, or south Asian ancestry [12]. Diagnosis of the condition is based on history and physical, combined with diagnostic studies such as biopsy and immunologic studies. The mainstay of treatment is topical corticosteroids for isolated cutaneous lesions and systemic corticosteroids for more widespread disease. Treatment has been subdivided into three basic stages: the control phase, the consolidation phase, and the maintenance phase [13]. The control phase involves aggressive titration of corticosteroids until a regression of symptoms is noted. In patients with recalcitrant disease, plasmapheresis and IVIG has been tried with success. Other immunosuppressants such azathioprine, methotrexate, mycophenolate mofetil, cyclophosphamide, and rituximab have been used with varying success [16]. The consolidation phase of treatment entails continuation of the effective dose of immunomodulators until most lesions are resolved. The maintenance phase involves tapering these medications until they can be safely stopped. While PV can be a chronic condition with a mortality rate of !10%, it is important that primary care physicians recognize and appropriately refer these patients to specialists for confirmatory testing and highly monitored treatment. Failure to do so in a timely fashion may negatively affect the health of the patient. Meningococcemia Meningococcemia in its most basic definition is the presence of Neisseria meningitidis in the blood. Invasive meningococcal disease is associated with a significant risk of mortality, and must be immediately recognized so that treatment may be initiated. Populations affected include small children and young adults [17]. Risk factors will be delineated below. The composition of the primary care provider’s practice will determine the likelihood that they will see primary presentations of meningococcemia. The clinical presentation of invasive meningococcal infection is varied, and is not necessarily associated with meningitis. Meningococcemia is often preceded by an upper respiratory tract infection. Some consider such infection, in addition to mycoplasma infection, a risk factor for meningococcemia [18]. The patient may present with nonspecific complaints including, but not limited to, flu-like symptoms such as abrupt onset of fever, lethargy, mental status changes, and a rash [17]. This rash is the pathognomonic feature of meningococcemia, the one sign that a primary care provider cannot miss. Although the flu-like prodrome occurs early in the disease process, the development of the rash is somewhat later. It begins with cutaneous pain followed by erythema, difficult to distinguish from a viral exanthem [19]. Petechiae occur in approximately 50% to 60% of patients, and often begin on the wrists, ankles, and axillae. From there, they spread to the rest of the body, generally sparing the head, palms, and soles [18]. From petechiae, the
DERMATOLOGIC EMERGENCIES
691
rash often progresses to the classic purpura, which results from microvascular thrombosis with subsequent perivascular hemorrhage [20]. Meningococcemia can rapidly progress to septic shock, characterized by hypotension, congestive heart failure, disseminated intravascular coagulation, and acute renal failure. This is associated with significant mortality. The mechanism behind invasive meningococcal disease is complex. There are three main serogroups of N meningitidis that account for the majority of infection: A, B, and C. Serogroups B and C account for approximately 90% of all cases in the United States [17]. N meningitidis is a Gram-negative diplococcus whose only known natural reservoir is the upper respiratory tract of humans. Transmission of the bacterium is mediated through transfer of pharyngeal secretions, and the majority of those colonized do not become clinically infected. The progression from colonization to infection is thought to be mediated by a multitude of factors that enable the bacterium to access the bloodstream. As mentioned above, some consider viral upper respiratory tract infection or infection with mycoplasma to be predisposing factors to invasive meningococcal infection. One of the best known risk factors is crowded living conditions such as those found in colleges, prisons, or the military. Host defense against invasive meningococcal disease is mediated by complement in the presence of a strain specific antibody. This mechanism explains the preponderance of disease burden in children less than 2 years of age, given their relative lack of immunocompetence. Young adults are likewise susceptible to meningococcal infection, but this may simply be due to the predominance of crowded living conditions in this age group. Once a meningococcal bacteremia is established in the absence of a functional bactericidal response, endotoxin derived from the lipopolysaccharide cell well initiates a cascade that ultimately leads to septic shock and the cutaneous manifestations of meningococcemia. Meningococcemia is a rare condition, occurring in approximately 1.3 of every 100,000 individuals in the United States [20]. Although rare, these patients will often present to outpatient offices. The cornerstone of treatment of meningococcemia is early recognition. Patients will present with nonspecific symptoms, and the primary care physician must be aware of any cutaneous manifestations such as petechiae or purpura. In patients that arouse the clinical suspicion of meningococcemia, the primary care physician must arrange transport to a referral center equipped with an intensive care unit for invasive monitoring and ventilatory support. If unable to affect immediate transfer, the primary care physician should draw blood cultures and, if possible, deliver antibiotics while awaiting transfer. Appropriate choices include either penicillin G or ceftriaxone, depending on local susceptibilities. Rocky Mountain spotted fever RMSF is a tick-borne illness caused by inoculation of the host with Rickettsia rickettsii by either the American dog tick, Dermacentor variabilis, or
692
BROWNE
et al
the wood tick, Dermacentor andersonii, in the East and West, respectively [21]. The clinical presentation is variable, but there are pathognomonic symptoms and signs that must be recognized. Classically, the patient will present with headache, fevers, and a characteristic rash. This triad of symptoms occurs in only 44% of patients [22]. Thirty percent of patients will have no history of a known tick bite [21]. The characteristic rash begins approximately 3 to 4 days after disease onset, and initially presents as blanchable erythematous macules. This classically begins at the wrists and ankles, spreading distally to the palms and soles, and finally migrating centrally creating the characteristic diffuse rash [23]. The macular exanthem evolves into a petechial rash and, in some instances, ecchymoses. Beyond the rash, there are a multitude of other nonspecific symptoms such as headache, myalgias, malaise, and abdominal pain (sometimes mimicking an acute abdomen) [21]. If untreated, RMSF will progress to significant end organ involvement, including skin necrosis, myocarditis, meningo-encephalitis, acute renal failure, and pulmonary manifestations such as hemorrhage, edema, and acute respiratory distress syndrome. The case fatality rate still approaches 25% [22]. The clinical manifestations of Rickettsia rickettsii are a direct result of this Gram- negative rods tropism for endothelial cells. Upon introduction of the bacteria via the salivary secretions of the tick, it invades the endothelial cytoplasm and spreads cell to cell, resulting ultimately in a lymphocytoplasmic vasculitis [21]. This vasculitis can affect multiple organ systems, and contributes directly to the variable pathology found in this illness. RMSF was first described in Montana, but it has since been found to be endemic to Maryland, North Carolina, Tennessee, Virginia, South Carolina, and Oklahoma. Infection mostly occurs in children between the ages of 5 and 9 years, and mostly from April to September [24]. Those at higher risk include those who live in or near wooded areas, those who pursue recreation in such areas, and those who have close contact with dogs. It is not necessary to have a history of a known tick bite to pursue a diagnosis of RMSF, especially if the patient resides in an endemic area and has classic symptoms such as those mentioned above. For a clinician to accurately diagnose RMSF a strong degree of clinical suspicion must be maintained, especially depending on the area of practice. If highly suspicious, primary care physicians must initiate treatment based solely on their history and physical examination. Before initiating treatment the physician should send samples to their laboratory for indirect fluorescent antibody testing. The mainstay of treatment is administration of tetracycline antibiotics, notably doxycycline, for 7 to 10 days. For those allergic or intolerant to doxycycline, chloramphenicol may be used [25]. Lyme disease Lyme disease, like RMSF, is a tick-borne illness. It results from the transmission of Borrelia burgdorferi to human hosts from mice and deer
DERMATOLOGIC EMERGENCIES
693
via the Ixodes scapularis and Ixodes pacificus species of tick [23]. Lyme disease progresses through definable stages. Stage I (early localized stage) occurs within 7 to 10 days after the tick bite. It is characterized by the development of erythema migrans at the site of the tick bite. This rash will occur in approximately 75% of those affected, and is described as a macular rash with a central area of clearing [24]. Alternately, erythema migrans has been described as a central red patch surrounded by seemingly normal skin that is, in turn, surrounded by migrating erythematous band [23]. Along with the above-mentioned rash, Lyme disease is associated with a number of nonspecific complaints similar to those found in RMSF, including flulike symptoms, headaches, and arthralgias. Stage II (early disseminated stage) occurs weeks after the initial exposure, and it is characterized by central nervous system symptoms such as meningitis or Bell’s palsy. Other complications noted at this stage can include myocarditis, atrioventricular nodal block, migratory arthralgias, and myalgias, motor or sensory radiculopathy, as well as disseminated annular lesions [23,24,26]. Stage III (late chronic disease) is characterized by profound fatigue, prolonged arthralgias, encephalopathy, polyneuropathy, and possibly leukoencephalitis [26]. The natural reservoir of B burgdorferi can be found in the white footed mouse and the expanding deer populations in the northeastern region of the United States, ranging primarily from Maine to Maryland. There are also foci of endemicity in the Midwest and Western states. According to Centers for Disease Control data, there are approximately 15,000 cases diagnosed yearly, mostly in the spring and summer months [27]. The age groups statistically at most risk are those between the ages of 2 to 15 and 30 to 59 years. The accurate and timely diagnosis of Lyme disease is imperative in the prevention of disseminated diseasedstages II and III. The ability to diagnose this disease rests on the primary care clinician’s knowledge of the prevalence of the Lyme disease in their area and their ability to establish a pretest probability based on that prevalence and the clinical presentation of the patient. Patients residing in endemic areas who present with symptoms concerning for Lyme disease and objective signs such as erythema migrans or atrioventricular block should be treated without serologic testing. Those from areas of high endemicity with symptoms of Lyme disease, but without erythema migrans should be referred for serologic testing with confirmatory Western blot testing before initiation of treatment. The value of testing individuals from areas with moderate to low endemicity with nonspecific symptoms without erythema migrans is controversial. Testing should be considered if symptoms persist longer than 2 weeks. In patients who experience a known tick bite, but who are asymptomatic, there is no indication for prophylactic treatment. These patients, however, do need to be monitored for the development of symptoms for approximately 30 days [26]. If the tick has been attached for less than 24 to 48 hours, the chances of disease transmission are extremely low.
694
BROWNE
et al
The treatment of stage I Lyme disease involves antibiotic therapy for 2 to 3 weeks. The antibiotic of choice is doxycycline 100 mg twice daily. For those allergic to or intolerant of doxycycline, amoxicillin 500 mg three times daily may be administered. Intravenous antibiotic therapy may be required for advanced disease with complications such as neurologic involvement, atrioventricular block, and meningitis [24,26,28,29]. Although not a true emergency, Lyme disease is a condition that must be recognized early in its presentation to avoid serious complications. The ability to diagnose this condition is founded upon a firm understanding of the prevalence of the disease in that area and maintenance of a high clinical suspicion. This brief review encompasses a variety of dermatologic diseases, some frankly emergent such as TEN, SSSS, and meningococcemia, and some merely urgent such as RMSF, Lyme disease, and pemphigus vulgaris. The common thread among these processes, beyond their dermatologic nature, is the importance of prompt recognition. Although all of these conditions do not require emergent transport to a higher level of care, they all require the primary care clinician to have a heightened level of awareness and suspicion concerning potentially serious dermatologic disease.
References [1] Letko E, Papaliodis DN, Papaliodis GN, et al. Stevens-Johnson syndrome and toxic epidermal necrolysis: a review of the literature. Ann Allergy Asthma Immunol 2005;94(4):419–36 [quiz 36–8, 56]. [2] Fritsch PO, Sidoroff A. Drug-induced Stevens-Johnson syndrome/toxic epidermal necrolysis. Am J Clin Dermatol 2000;1(6):349–60. [3] Chave TA, Mortimer NJ, Sladden MJ, et al. Toxic epidermal necrolysis: current evidence, practical management and future directions. Br J Dermatol 2005;153(2):241–53. [4] Bachot N, Roujeau JC. Differential diagnosis of severe cutaneous drug eruptions. Am J Clin Dermatol 2003;4(8):561–72. [5] Patel GK, Finlay AY. Staphylococcal scalded skin syndrome: diagnosis and management. Am J Clin Dermatol 2003;4(3):165–75. [6] Ladhani S. Recent developments in staphylococcal scalded skin syndrome. Clin Microbiol Infect 2001;7(6):301–7. [7] Ladhani S, Evans RW. Staphylococcal scalded skin syndrome. Arch Dis Child 1998;78(1): 85–8. [8] Levine G, Norden CW. Staphylococcal scalded-skin syndrome in an adult. N Engl J Med 1972;287(26):1339–40. [9] Patel GK, Varma S, Finlay AY. Staphylococcal scalded skin syndrome in healthy adults. Br J Dermatol 2000;142:1253–5. [10] Richard M, Mathieu-Serra A. Staphylococcal scalded skin syndrome in a homosexual adult. J Am Acad Dermatol 1986;15(2 Pt 2):385–9. [11] Keehn CA, Morgan MB. Clinicopathologic attributes of common geriatric dermatologic entities. Dermatol Clin 2004;22(1):115–23 [vii.]. [12] Black M, Mignogna MD, Scully C. Number II. Pemphigus vulgaris. Oral Dis 2005;11(3): 119–30. [13] Bystryn JC, Rudolph JL. Pemphigus. Lancet 2005;366(9479):61–73. [14] Bickle K, Roark TR, Hsu S. Autoimmune bullous dermatoses: a review. Am Fam Physician 2002;65(9):1861–70.
DERMATOLOGIC EMERGENCIES
695
[15] Blauvelt A, Hwang ST, Udey MC. Allergic and immunologic diseases of the skin. J Allergy Clin Immunol 2003;111(2 Suppl):S560–70. [16] Yeh SW, Sami N, Ahmed RA. Treatment of pemphigus vulgaris: current and emerging options. Am J Clin Dermatol 2005;6(5):327–42. [17] Ferguson LE, Hormann MD, Parks DK, et al. Neisseria meningitidis: presentation, treatment, and prevention. J Pediatr Health Care 2002;16(3):119–24. [18] Salzman MB, Rubin LG. Meningococcemia. Infect Dis Clin North Am 1996;10(4):709–25. [19] Darmstadt GL. Acute infectious purpura fulminans: pathogenesis and medical management. Pediatr Dermatol 1998;15(3):169–83. [20] Hazelzet JA. Diagnosing meningococcemia as a cause of sepsis. Pediatr Crit Care Med 2005; 6(3 Suppl):S50–4. [21] Sexton DJ, Kaye KS. Rocky mountain spotted fever. Med Clin North Am 2002;86(2): 351–60 [vii–viii.]. [22] Masters EJ, Olson GS, Weiner SJ, et al. Rocky Mountain spotted fever: a clinician’s dilemma. Arch Intern Med 2003;163(7):769–74. [23] McGinley-Smith DE, Tsao SS. Dermatoses from ticks. J Am Acad Dermatol 2003;49(3): 363–92 [quiz 93–6]. [24] Bratton RL, Corey R. Tick-borne disease. Am Fam Physician 2005;71(12):2323–30. [25] Drage LA. Life-threatening rashes: dermatologic signs of four infectious diseases. Mayo Clin Proc 1999;74(1):68–72. [26] DePietropaolo DL, Powers JH, Gill JM, et al. Diagnosis of Lyme disease. Am Fam Physician 2005;72(2):297–304. [27] Steere AC. Lyme disease. N Engl J Med 2001;345(2):115–25. [28] Dumler JS, Walker DH. Rocky Mountain spotted feverdchanging ecology and persisting virulence. N Engl J Med 2005;353(6):551–3. [29] Leaute-Labreze C, Lamireau T, Chawki D, et al. Diagnosis, classification, and management of erythema multiforme and Stevens-Johnson syndrome. Arch Dis Child 2000;83(4):347–52.
Prim Care Clin Office Pract 33 (2006) 697–710
Skin and Soft Tissue Infections Robert L. Rogers, MD, FAAEM, FACEP, FACP*, Jack Perkins, MD Department of Emergency Medicine, The University of Maryland School of Medicine, 110 South Paca Street, Suite 200, Emergency Medicine, Sixth Floor, Baltimore, MD 21201, USA
Skin and soft tissue infections represent a continuum of symptoms that range from uncomplicated cellulitis to the potentially lethal entity necrotizing fasciitis (NF). The primary care physician will see a myriad of infectious skin disorders in the outpatient setting, and must be capable of discerning which presentations warrant emergency department evaluation and inpatient admission. This article will highlight three entities under the broad umbrella of skin and soft tissue infections. Cellulitis, cutaneous abscess, and NF present not only to the emergency department but also to the outpatient setting. This article aims to help primary physicians recognize patterns of disease that herald significant illness and the need for more extensive evaluation than an outpatient setting can offer. This article discusses the etiology, presentation, evaluation, and management of cellulitis, cutaneous abscess, and NF. Particular attention will be spent addressing the emerging problem with community-acquired methicillin-resistant Staphylococcus aureus (ca-MRSA). Community-acquired MRSA has significantly changed the emergency department approach to cellulitis, and has altered the spectrum of antimicrobials that are used for outpatient treatment of cellulitis. Familiarity with ca-MRSA is essential for the primary care physician, because misdiagnosis and improper antibiotic selection can lead to significant morbidity and mortality [1].
Cellulitis In 1996, cellulitis was estimated as the 28th most common discharge diagnosis in the United States. The actual incidence and prevalence are * Corresponding author. E-mail address:
[email protected] (R.L. Rogers). 0095-4543/06/$ - see front matter Ó 2006 Elsevier Inc. All rights reserved. doi:10.1016/j.pop.2006.06.005 primarycare.theclinics.com
698
ROGERS & PERKINS
difficult to estimate due to the number of outpatient presentations; however, one study reported that cellulitis was responsible for 2.2% of outpatient office visits [2]. Cellulitis is defined as an infection of the dermis with variable extension into the subcutaneous tissues [2]. It usually develops on extremities, but has been documented on every area of the body. Cellulitis generally occurs after the protective barrier of the skin, the epidermis, has been compromised (eg, trauma, ulcers, eczema) allowing bacterial access to the subepidermal tissues. Often, the portal of bacterial entry is unknown. The predilection of cellulitis to occur on areas exposed to the environment (eg, hands, feet) supports the theory that disruption of the epidermis is essential in the development of cellulitis. Factors that predispose to cellulitis include tinea pedis, diabetes mellitus, peripheral vascular disease, peripheral edema, and prior history of cellulitis [3]. Bacterial etiology The majority of cases of cellulitis are caused by beta-hemolytic streptococci, mostly from subtypes A and B. S aureus, and more recently caMRSA, have become increasingly prominent pathogens throughout the United States. Other causes of cellulitis are associated with specific clinical scenarios and are summarized in Table 1 [5–7]. Table 1 Cellulitis key clinical features organism Periorbital cellulitis
Crepitant cellulitis
Must be distinguished from Orbital cellulitis Medical emergency hallmarked by limitation in extraocular movement Must evaluate for underlying Ischial fossa abscess May represent necrotizing fasciitis
Puncture wound Salt water exposure Fresh water exposure Dog bite/Cat bite Cat Scratch Human bite
Often in plantar aspect of foot History of exposure is key element History of exposure is key element Look for puncture wound Look for proximal lymphadenopathy Do NOT suture closed
Orbital cellulitis Perianal cellulitis
Calf cellulitis Foot cellulitis Cellulitis in infants
Suspect underlying DVT Suspect origin as Tinea Pedis Often hematogenous spread; consider osteomyelitis, meningitis, septic joint Immunocompromised HIV, Cancer patients Cell mediated immunity HIV, Leukemia patients, dysfunction may be disseminated Hot tub exposure Cellulitis in bathing suit distribution
Strep and staph in adults, H. influenza in children Strep, staph Group A strep (GAS) GAS, Anaerobes, Clostridia Pseudomonas Vibrio vulnificus Aeromonas hydrophilia Pasteurella multocida Bartonella henselae Anaerobes, Strep pyogenes, Eikenella corrodens Multiple Organisms Multiple organisms Predominantly Group B Strep Consider gram negatives Cryptococcus neoformans Pseudomonas aeruginosa
SKIN AND SOFT TISSUE INFECTIONS
699
Presentation Cellulitis can develop on any area of the body, but is most likely to occur on extremities that are susceptible to microtrauma. The hallmarks of cellulitis include erythema, warmth, swelling, and tenderness of the affected area. The absence of all of these signs would make cellulitis a highly improbable diagnosis [2]. The erythema in cellulites is generally confluent, and leading edges tend to be poorly demarcated. Sharply demarcated borders are more indicative of erysipelas, which is a distinct form of cellulitis due almost exclusively to beta-hemolytic streptococci [4]. Other clinical features of cellulitis include tender lymphadenopathy and occasional abscess formation, especially if S aureus is the causative organism. Although systemic features such as fever, chills, or rigors may be associated with cellulitis, evidence of systemic toxicity should raise suspicion for a more serious pathology such as NF or hematogenous dissemination of the cellulitic organism. Laboratory and radiographic evaluation Laboratory investigation of cellulitis is of limited value because it is unusual to definitively establish a causative organism [2]. Blood culture results are positive in less than 5% of patients [5]. Blood cultures are frequently obtained when patients are admitted to the hospital for cellulitis; however, they just as often represent contamination as a true pathogen. Sending a complete blood count or any other laboratory studies are not recommended in the outpatient setting, and have limited value in the inpatient setting. Occasionally, consideration is given to a punch biopsy of the cellulitic area; however, the results can be quite variable, with culture identifying a likely organism 5% to 40% of the time, depending on the patient population [5]. Radiographic evaluation is recommended in the diagnostic workup of cellulitis only when there is a question of whether another process may be involved. Occasionally, the clinical examination may suggest an occult abscess or perhaps NF. In such cases, it is reasonable to consider plain films to evaluate for gas in the soft tissue or even a CT scan of the area in question with intravenous contrast to evaluate for abscess or muscle involvement. Plain films are also indicated if a retained foreign body is considered as the source of the overlying cellulitis (see Fig. 1). Community acquired MRSA MRSA infections began in the early 1960s, a few years after methicillin was introduced to combat the problem of resistance to penicillin [8]. By the 1980s, MRSA had become a common problem throughout US hospitals. In the 1990s, patients who had no risk factors for hospital-acquired MRSA infection (eg, recent hospitalization, intravenous drug use, hemodialysis, or resident in a chronic care facility) were developing cellulitis from MRSA [9]. Interestingly, this strain of MRSA was susceptible to a broader range of antibiotics
700
ROGERS & PERKINS
Fig. 1. Left lower extremity cellulitis. (Copyright Ó Challenger Corporation (Memphis, TN). 2005. All rights reserved.)
than hospital-associated or -acquired MRSA. This new strain of MRSA was given the name ca-MRSA or community-associated MRSA. The incidence and prevalence of ca-MRSA has risen dramatically over the past decade, and its emergence has changed clinical practice in outpatient settings, urgent care facilities, emergency departments, and inpatient hospital settings. Although anyone may become colonized or present with an illness due to ca-MRSA, there appears to be certain populations that are more susceptible to outbreaks. Prisoners, children (especially in daycare centers), intravenous drug users, sports participants, soldiers, homosexual men, and homeless individuals are at particular risk of ca-MRSA colonization and infection [1]. All of these at-risk populations share the similarities of large numbers of people who convene or habitate in close proximity to others where hygiene may be difficult to maintain. There also seems to be a preponderance of ca-MRSA among children. A few studies have reported ca-MRSA isolates in up to 50% of children in whom S aureus is found on routine colonization cultures [10]. Clinical manifestations of ca-MRSA infection are varied, and include skin and soft tissue infections, osteomyelitis, pneumonia, joint infection, and sepsis. The spectrum of disease presentations can be diverse. Cellulitis is a common presentation of ca-MRSA although it appears less prevalent than abscess formation with or without associated cellulitis. Any patient who presents with cellulitis and an abscess should be considered to have ca-MRSA until proven otherwise. Patients who have cutaneous abscesses due to ca-MRSA tend to have frequent recurrences, and family members are prone to developing similar symptoms. ca-MRSA has recently been reported in association with NF [11]. These case reports are interesting in that ca-MRSA was the only organism isolated in what is generally a polymicrobial infection. Osteomyelitis, septic joints, and severe community-acquired pneumonias have all been documented [1,9]. Mortality in ca-MRSA pneumonia was close to 25% in one study, although most of the patients were previously healthy hosts [11]. Morbidity and mortality are dependent upon proper antibiotic selection, which makes
SKIN AND SOFT TISSUE INFECTIONS
701
it essential for the primary care provider to consider ca-MRSA as a possible pathogen in all skin or soft tissue infections presenting to the outpatient setting. Treatment The treatment of cellulitis has changed significantly in the past several years. Cephalexin (Keflex), amoxicillin-clavulanate (Augmentin), and dicloxacillin (Dynapen) have been staples of treatment for outpatient cellulitis in emergency departments, urgent care, and primary care clinics for many years. The emergence of ca-MRSA has led to a significant increase in the number of treatment failures for skin and soft tissue infections seen in the emergency department. The most important aspect of selecting an appropriate antibiotic involves being familiar with the patient population (eg, intravenous drug users, prisoners) and the susceptibility patterns in the area. Current infectious disease guidelines still recommend beta-lactam (eg, cephalexin, dicloxacillin) antibiotics as first-line treatment for cellulitis that will not require admission [1,12]. Those patients who are more likely to have ca-MRSA as a cause of their cellulitis (eg, high risk populations or cellulitis associated with abscess) should receive trimethoprim–sulfamethoxazole (TMP-SMZ), clindamycin, doxycycline, minocycline, or third- or fourthgeneration fluoroquinolones. Patients who have significant comorbid illnesses such as diabetes mellitus, congestive heart failure, or who are on dialysis should be given agents that will cover ca-MRSA as they may not tolerate 24 to 48 hours of improper antibiotic selection. Recently, clindamycin-inducible resistance has become a problem in some areas, prompting some clinicians to select TMP-SMZ as their initial choice when they suspect ca-MRSA is the pathogen. Clindamycin resistance should be suspected when a ca-MRSA susceptibility pattern reveals sensitivity to clindamycin but resistance to erythromycin. This indicates a high likelihood of ‘‘inducible clindamycin resistance’’ [8]. TMP-SMZ failures are rare when treating cellulitis, usually only occurring with isolates of methicillin-sensitive S aureus (MSSA) [1,5]. No definitive data supports the concept that administration of a single dose of an intravenous antibiotics before discharge from the emergency department or from outpatient clinic has any benefit over oral therapy alone [13]. If patients are well enough to be sent home for treatment, they are likely well enough for oral antibiotic therapy. Admission to the hospital and administration of parenteral antibiotics are recommended for patients with systemic symptoms (eg, fever, rigors, emesis), unstable vital signs (eg, hypotension, tachycardia), or those who have failed an appropriate course of oral antibiotic therapy that should have covered the suspected pathogen [1,5,12]. Consideration should also be given to patients who have serious comorbid conditions (diabetes, congestive heart failure, end-stage renal disease) or those who are immunocompromised from cancer, organ transplantation, or HIV infection.
702
ROGERS & PERKINS
Patients admitted to the hospital should receive parenteral antibiotics. Many authorities conclude that vancomycin is the medication of choice for skin and soft tissue infections when resistant S aureus is considered as a possible pathogen. Cultures from the blood and any wounds should be obtained in these circumstances to try to isolate a pathogen. Other appropriate inpatient medications include quinupristin–dalfopristin (Synercid), linezolid (Zyvox), or daptomycin (Cubicin). All of these antibiotics have strong activity against both hospital-acquired MRSA and ca-MRSA. These medications are prohibitively expensive for routine use, but are appropriate when vancomycin-resistant enterococcus is involved or when the patient has a vancomycin allergy.
Cutaneous abscess The incidence and prevalence of skin abscesses has risen in parallel with the emergence of ca-MRSA [5,14]. The diagnosis of a cutaneous abscess is usually not difficult, and often the patient may complain of spontaneous drainage. The treatment of these abscesses frequently involves incision and drainage. Because this is considered a minor surgical procedure, some primary care physicians will choose to refer for definitive management. In addition, procedures such as these may not be possible in the outpatient setting due to time constraints. The primary care office setting is, however, an ideal location for subsequent follow-up evaluation after a patient has undergone incision and drainage. Patients with these disorders need close followup, and primary care physicians are in a perfect position to play a pivotal role in the patient’s recovery. Microbiology and clinical features S aureus is by far the most common pathogen involved in cutaneous abscess formation [15]. Roughly 50% of cutaneous abscesses are due to S aureus; however, most abscesses are not cultured. Treatment is empirically directed against this organism [14]. Mot skin abscesses result from minor trauma, although in innercity populations, injection drug use is an extremely common portal of entry, especially for ca-MRSA. Occasionally, skin abscesses will be the result of bacteremia. Risk factors for recurrent abscess formation include obesity, diabetes mellitus, corticosteroid use, neutrophil dysfunction, and injection drug use [16]. Presentation of a skin abscess almost always involves pain, swelling, warmth, and erythema [14]. Local lymphadenopathy and signs of systemic involvement (eg, fever, chills) are unusual unless there is an associated cellulitis. Furuncles are a specific type of cutaneous abscess that occur around hair follicles, and are more common in African American populations. Carbuncles are defined as a series of connected furuncles, and are most commonly found in the posterior scalp area. Hidradenitis suppurativa is an
SKIN AND SOFT TISSUE INFECTIONS
703
inflammatory condition where recurrent abscess formation occurs in the buttocks or axillary area largely in connection with excessive perspiration and obesity. Abscesses often drain spontaneously before presentation to the clinic or emergency department, but occasionally may involve deeper tissues requiring CT imaging or ultrasound for diagnosis. Treatment The most important aspect of abscess treatment is drainage of the fluid collection [17,18]. Spontaneous drainage makes the decision for further incision rather straightforward; however, many abscesses will not spontaneously drain, nor will they be pointing and fluctuant. A recommended tactical maneuver to evaluate for a fluid collection is the insertion of a small bore needle (eg, 22 or 25 gauge) into the area of maximum induration and apparent fluctuance. Return of pus indicates that incision and drainage is indicated [14]. The initial incision is best made by use of a #11 blade scalpel. The tip of the scalpel is aimed at the center of fluctuance and directed perpendicular to the skin surface. The incision should be deep enough to fully evacuate the fluid collection. In addition, loculations should be probed and broken up with the use of a hemostat if possible [17,18]. If there is any question about whether the patient is to be admitted, or if they have failed previous antibiotic therapy, a culture of the fluid should be obtained to help identify a causative organism. The cavity should be irrigated after drainage and loosely packed with iodoform gauze. Often, pain is a rate-limiting factor of debridement, especially in intravenous drug users who may have increased tolerance to narcotic analgesics. As a rule of thumb, lidocaine does not work efficiently in acidic environments such as in areas of pus; thus, lidocaine (with epinephrine if in areas other than ears, fingers, toes, nose, or genitalia) should be injected generously throughout the surrounding tissue and in the subcutaneous area overlying the abscess [19]. Patients will sometimes present with an abscess that is not ready for incision and drainage. Induration may be present, but no discrete are of fluctuance may be appreciated. In these cases, antibiotics and warm compresses to the area are indicated to try to bring the abscess to a head for drainage. Although antibiotic therapy has never been shown to be helpful for an uncomplicated abscess, the induration that often precedes discreet abscess formation can be confused for uncomplicated cellulitis. Follow-up should be arranged in 24 to 48 hours, regardless of whether the abscess is drained, so that healing can be evaluated, or to see if drainage is now indicated. Risks of incision and drainage include bleeding, pain at the incision site, scar formation, and bacteremia [18]. The incidence of bacteremia following surgical incision of an abscess is unclear, as data is conflicting. The American Heart Association recommends one intravenous dose of antibiotics to those who are at moderate to high risk for bacterial endocarditis [20,21]. These patients would include
704
ROGERS & PERKINS
those with any type of implanted cardiovascular device, such as stents, grafts, porcine, or mechanical valves. A dose of an antistaphylococcal antibiotic such as cefazolin or nafcillin would be appropriate [20,21]. There has always been some controversy regarding whether antibiotics need to be administered following debridement of an abscess. In those patients who have no evidence of systemic infection and no surrounding cellulitis, antibiotics are not indicated. Available evidence supports withholding antibiotics in these clinical circumstances. Patients who have systemic symptoms, have surrounding cellulitis, are immunocompromised (e.g., diabetes mellitus, HIV), or whose abscess is not ready for drainage should receive an antistaphylococcal antibiotic such as cephalexin (Keflex) or dicloxacillin (Dynapen). However, strong consideration should be given to the presence of ca-MRSA in at risk populations. If ca-MRSA is a possible cause of the abscess and surrounding soft tissue infection, TMP-SMZ, doxycycline, clindamycin, or a third- or fourth-generation fluoroquinolone should be considered [22]. In general, a 7-day course of antibiotics is considered sufficient [17,18]. Abscesses in the oral, perirectal, or genital areas should be considered multibacterial in etiology, and consideration should be given to using agents with Gram-positive, Gram-negative, and anaerobic activity [23]. Suitable medications would include amoxicillin–clavulanate (Augmentin) or third- or fourth-generation fluoroquinolones. The importance of follow-up for any patient who has been treated for skin and soft tissue infections cannot be overstated. The presence of ca-MRSA has significantly increased the number of treatment failures. As many as 61% of ca-MRSA infections are initially treated inappropriately with beta-lactam antibiotics. This may lead to an increased morbidity and mortality rate [14,24].
Necrotizing fasciitis NF is a rare clinical entity accounting for only 500 to 1500 cases of severe soft tissue infections annually in the United States [25]. Although it is a relatively rare clinical entity, its enigmatic presentation and overlap with other more benign skin and soft tissue infections make this a frequently missed diagnosis. The clinical significance of recognizing this disease process early in its course is crucial because mortality is estimated at 24% to 34%. In addition, the associated morbidity (eg, amputations, need for dialysis, multiple surgical procedures, prolonged hospital courses) in those who survive is quite high [25,26]. Primary care physicians may never encounter NF in their clinical careers; however, it is a diagnosis that cannot be missed, as prompt surgical intervention can significantly affect patient outcome. Any case of suspected NF should be referred immediately to the nearest appropriate center that has the capability of caring for the patient. In many cases, the patient will be transferred to the nearest emergency department for surgical evaluation.
SKIN AND SOFT TISSUE INFECTIONS
705
Microbiology and clinical features NF is defined as a rapidly progressing soft tissue infection with fulminant tissue destruction, rapid bacterial spread along tissue planes, thrombosis of blood vessels, systemic signs of toxicity, and high rates of morbidity and mortality [27]. NF is typically divided into two separate categories based on involved organisms. Type I NF is the less common variety, and is due to infection with aerobic and anaerobic bacteria. Frequently, Type I NF is a mixed bacterial infection. Nearly 66% of cases in one study were attributed to a combination of aerobic and anaerobic bacteria [25]. Common isolates from Type I NF include streptococcus (other than Group A, which causes Type II NF), Staphylococcus aureus, enterococcus, peptostreptococcus, Escherichia coli, Bacteroides fragilis, and Clostridium species. Type I NF includes Fournier’s gangrene, which involves the perineal area and can expand rapidly to involve the genitalia and anterior abdominal wall. Patients at risk for Type I NF include diabetics, those with surgical wounds, peripheral vascular disease, alcohol abuse, obesity, hypoalbuminemia (used as a clinical marker of malnutrition) and illicit drug users who ‘‘skin pop’’ or inject into the muscle [25,26]. Type II NF is amonomicrobial infection caused almost exclusively by group A streptococcus (GAS or streptococcus pyogenes) [27]. In the past few years, cases of NF caused by MRSA (mostly ca-MRSA) have been reported [8,28]. The incidence of Type II NF infections increased in the 1990s for unknown reasons. It differs from Type I NF in that any age group is a susceptible host for infection. Some predisposing risk factors include injection drug use, surgical procedures, burns, surgical wounds, blunt trauma, varicella, and even perhaps nonsteroidal anti-inflammatory medications [26,27]. It is hypothesized that hematogenous spread of GAS from a pharyngeal source to the site of muscle or soft tissue injury is one possible explanation for the development of NF. Type II NF is often associated with streptococcal toxic shock syndrome, which is heralded by multiorgan failure and severe shock. The most important clinical symptom of NF is that patients often report pain that is out of proportion to findings during the clinical examination [29]. Patients may complain of extreme pain in an area with few or no cutaneous manifestations. The presence of warmth, erythema, tenderness, or skin discoloration is nonspecific, and their presence or absence does not help significantly in securing a diagnosis of NF. However, the development of bullae, especially hemorrhagic bullae, should be taken as an ominous clinical finding, and should prompt consideration of an aggressive soft tissue infection. Crepitus has often been heralded as a strong clinical sign of NF, but this is only found in 10% of all cases of NF [27]. Fever, malaise, myalgias, rigors, tachycardia, and hypotension may all be seen at presentation or develop in the first 24 hours of the clinical course. Because NF spreads rapidly along tissue planes, one may see rapid progression of erythema or significant
706
ROGERS & PERKINS
change in other clinical symptoms or signs (eg, tachycardia, pain) only hours after presentation (see Fig. 2, 3, and 4 for examples of NF). The key takehome point is to remember to be hypervigilant about skin and soft tissue infections, and to always consider NF in the differential diagnosis. Failure to do so may prove to be disastrous for the patient. Laboratory and radiographic evaluation NF, as opposed to uncomplicated cellulitis, necessitates casting a wider diagnostic net. Laboratory analysis as well as prompt radiographic evaluation, is indicated if NF is a possibility. A peripheral white blood count is frequently elevated, and is typically accompanied by a marked left shift. Hyponatremia, elevated BUN and creatinine, hypocalcemia, hypoalbuminemia, and hyperglycemia are found quite often in cases of NF. One study reported that the combination of a white blood cell count O14,000 cells/mL and serum sodium !135 mEq/L resulted in a 90% sensitivity and specificity of 76% for distinguishing between necrotizing and nonnecrotizing soft tissue infections [25]. Blood cultures are positive in 60% of cases of Type II NF due to the rapid ability of group A strep (GAS) to grow in available culture medium [27]. Type I NF has a lower return of positive cultures due to more fastidious organisms being involved. An elevated lactic acid level reflects the presence of anaerobic metabolism in the involved tissue and muscle. Elevated creatine phosphokinase levels are also a common finding that indicates muscle tissue death. An arterial blood gas will likely reveal a metabolic acidosis (elevated lactic acid level) and perhaps a compensatory respiratory alkalosis. Disseminated intravascular coagulation (elevated prothrombin time [PT], partial thromboplastin time [PTT], international normalized ratio [INR], D-dimer, fibrin split products) can accompany NF, especially if caused by GAS or streptococcal toxic shock syndrome. Radiographic imaging is indicated as soon as NF is suspected to look for evidence of subcutaneous gas.This finding, however, only has a sensitivity of
Fig. 2. Lower extremity necrotizing fasciitis. (Copyright Ó Challenger Corporation (Memphis, TN). 2005. All rights reserved.)
SKIN AND SOFT TISSUE INFECTIONS
707
Fig. 3. Necrotizing fasciitis with hemorrhagic bullae. (Copyright Ó Challenger Corporation (Memphis, TN). 2005. All rights reserved.)
39%, specificity of 95%, and negative predictive value of 88% [25]. CT is superior to plain film for detection of gas. Additionally, CT can reveal fluid collections, muscle necrosis, fat stranding, and fascial thickening suggestive of NF. The potential problem with CT is that it requires the patient to leave the resuscitation area. Also, evaluating for an abscess or fluid collection requires the administration of intravenous dye. This could lead to the development of contrast nephropathy, especially if the patient is already hypotensive. MRI has even higher sensitivity and specificity than CT; however, it is time consuming and will delay definitive treatment. Treatment The treatment of NF can be considered similar to the philosophy that underlies the treatment of an acute ST elevation myocardial infarction, !time equals tissue.’’ No antibiotic or supportive measure will be an adequate substitute for definitive operative debridement. The role of the primary care physician is to identify that NF is a possible diagnosis and refer for rapid,
Fig. 4. Extensive upper extremity necrotizing fasciitis. (Copyright Ó Challenger Corporation (Memphis, TN). 2005. All rights reserved.)
708
ROGERS & PERKINS
definitive treatment. The key mistake made in outpatient clinical practice is assigning a benign diagnosis, such as cellulitis. Again, evidence of systemic toxicity, pain out of proportion, or presence of bullae or crepitus, should be considered NF until proven otherwise. Antibiotics play a key supportive role in treatment but do not provide the same definitive therapy as they do in cellulitis. Intravenous broad-spectrum antibiotics are indicated for either type of NF, and early administration of antibiotics is recommended. An extended spectrum beta-lactam penicillin (eg, piperacillin/tazobactam, ampicillin/sulbactam), combined with clindamycin, is noted as the regimen of choice from most trials [27,30]. Clindamycin is important, as it has been established as effective in suppression of bacterial toxin production (especially GAS toxin). It also enhances synthesis of penicillin-binding protein [29]. Other supportive measures such as aggressive fluid resuscitation and invasive monitoring are important adjunctive measures [30,31]. Summary The spectrum of skin and soft tissue infections seen by the primary care physician can be as benign as folliculitis to as life-threatening as NF. Cellulitis remains the most common skin and soft tissue infection seen in primary care. The ever present danger of ca-MRSA, however, has changed the way primary care physicians approach the common problem of cellulitis. The presence of risk factors for colonization with ca-MRSA and a history or examination finding of skin abscess should raise the suspicion of ca-MRSA, and antibiotic therapy should include TMP-SMZ, clindamycin, doxycycline, or minocycline. Skin abscess may occur independently of cellulitis, and often may safely be incised and drained in the primary care setting as long as timely follow-up is assured to assess for wound healing. Available evidence suggests that abscess formation without accompanying cellulitis does not require oral antibiotic therapy. Finally, although NF is rare as an outpatient clinical presentation, it is a diagnosis that the primary care physician should be familiar with. Failure to consider the diagnosis and refer may lead to significant morbidity and even mortality. References [1] Kowalski TJ, Berbari EF, Osmon DR. Epidemiology, treatment, and prevention of community-acquired methicillin-resistant Staphylococcus aureus infections. Mayo Clinic Proc 2005; 80(9):1201–7. [2] Baddour LM. Epidemiology, clinical features and diagnosis of cellulitis. In: Rose BD, editor. Wellesley (MA): Up-to-Date; 2005. [3] Lutomski DM, Trott AT, Runyon JM, et al. Microbiology of adult cellulitis. J Fam Prac 1988;26:45–8. [4] Bisno AL, Stevens DL. Streptococcal infections of the skin and soft tissues. N Engl J Med 1996;334:240–6.
SKIN AND SOFT TISSUE INFECTIONS
709
[5] Stevens DL, Bisno AL, Chambers HF, et al. Practice guidelines for the diagnosis and management of skin and soft-tissue infections. Clin Infect Dis 2005;41:1373–406. [6] Swartz MN. Cellulitis. N Engl J Med 2004;350:904–12. [7] Ginsberg MB. Cellulitis: analysis of 101 cases and review of the literature. South Med J 1981; 74:530–3. [8] Shrock S. Community-associated methicillin-resistant Staphylococcus aureus. Senior resident grand rounds. Baltimore: University of Maryland Medical Systems; 2005. [9] Fridkin SK, Hageman JC, Morrison ME, et al. Methicillin-resistant Staphylococcus aureus disease in three communities. N Engl J Med 2005;352(22):1436–44. [10] Boyce JM. Epidemiology; prevention; and control of methicillin-resistant Staphylococcus aureus in adults. In: Rose BD, editor. Wellesley (MA): Up-to-Date; 2005. [11] Chanbers HF. Community-associated MRSA – resistance and virulence converge. N Engl J Med 2005;352(14):1485–7. [12] Braun LO, Craft DA, Williams RO, et al. Increasing clindamycin resistance among methicillin-resistant Staphylococcus aureus in 57 northeast United States military treatment facilities. Pediatr Infect Dis J 2005;24(7):622–6. [13] Choo EK, Mitchell P, Mehta SD, et al. The utility of initial dose intravenous antibiotics for cellulitis prior to discharge on oral antibiotics from the emergency department. Ann Emerg Med 2005;46(3 Suppl):84. [14] Baddour LM. Skin abscess. In: Rose BD, editor. Wellesley (MA): Up-to-Date. 2005. [15] Stevens DL. Cellulitis, pyoderma, abscesses and other skin and subcutaneous infections. In: Cohen JO, Powderly WI, editors. Infectious diseases. 2nd edition. Philadelphia (PA): Mosby; 2004. [16] Bryan CS, Hawes SJ. Infections of the skin and its appendages, muscle, bones, and joints. In: Bryan CS, editor. Infectious diseases in primary care. Philadelpia (PA): W.B. Saunders; 2002. [17] Halvorson GD, Halvorson JE, Iserson KV. Abscess incision and drainage in the emergency departmentdPart I. J Emerg Med 1985;3:227–32. [18] Halvorson GD, Halvorson JE, Iserson KV. Abscess incision and drainage in the emergency departmentdPart (II). J Emerg Med 1985;3:299–305. [19] Fine BC, Sheckman PR, Bartlett JC. Incision and drainage of soft-tissue abscesses and bacteremia [letter]. Ann Intern Med 1985;103:645. [20] Dajani AS, Taubert KA, Wilson W, et al. Prevention of bacterial endocarditis. Recommendations by the American Heart Association. JAMA 1997;277:1448–58. [21] Bobrow BJ, Pollack CV Jr, Gamble S, et al. Incision and drainage of cutaneous abscesses is not associated with bacteremia in afebrile adults. Ann Emerg Med 1997;29: 404–8. [22] Llera JL, Levy RC. Treatment of cutaneous abscess: a double-blind clinical study. Ann Emerg Med 1985;14(1):15–9. [23] Meislin HW, Lerner SA, Graves MH, et al. Cutaneous abscesses. Anaerobic and aerobic bacteriology and outpatient management. Ann Intern Med 1977;87:145–9. [24] Veser F III, McGary WB, Bush J. Incidence, susceptibility, and antibiotic use in emergency department community-acquired methicillin-resistant Staphylococcus aureus (CA-MRSA) abscesses. Ann Emerg Med 2005;46(3)(Suppl):1–122. [25] Kuncir EJ, Tillou A, St. Hill CR, et al. Necrotizing soft-tissue infections. Emerg Med Clin North Am 2003;21:1075–87. [26] Anaya DA, McMahon KE, Nathans AB. Predictors of mortality and limb loss in necrotizing soft tissue infections. Arch Surg 2005;140(2):151–7. [27] Stevens DL. Necrotizing infections of the skin and fascia. In: Rose BD, editor. Wellesley (MA): Up-to-Date; 2005. [28] Miller LG, Perdreau-Remington FR, et al. Necrotizing fasciitis caused by community-associated methicillin-resistant Staphylococcus aureus in Los Angeles. N Engl J Med 2005; 352(14):1445–53.
710
ROGERS & PERKINS
[29] Kessenich CR, Bahl AS. Necrotizing fasciitis: understanding the deadly results of the uncommon ‘‘flesh-eating bacteria.’’ Am J Nurs 2004;104(9):51–5. [30] Wang TL, Hung CR. Role of tissue oxygen saturation monitoring in diagnosing necrotizing fasciitis of the lower limbs. Ann Emerg Med 2004;44(3):222–8. [31] Wang CH. Tissue oxygen saturation monitoring in diagnosing necrotizing fasciitis of the lower limbs: a valuable tool but only for a select few. Ann Emerg Med 2005;45(4):461–2.
Prim Care Clin Office Pract 33 (2006) 711–725
Emergencies in Diabetic Patients in the Primary Care Setting Susan D. Wolfsthal, MDa,*, Rebecca Manno, MDa, Evonne Fontanilla, MDa a
Department of Medicine, University of Maryland School of Medicine, Education Office, Room N3E09, 22 South Greene Street, Baltimore, MD 21201, USA
The diabetic patient poses a unique set of challenges for the primary care physician. Because these patients often have underlying hyperlipidemia, hypertension, diminished peripheral circulation, and a decreased capacity to combat infections, they have a propensity to develop cardiovascular complications and infections. The primary care physician frequently deals with problems related to glycemic control, heart disease, autonomic dysfunction, and a variety of infections. These patients may present in subtle ways that may belie the severity of the impending condition. Glycemic control Hypoglycemia Current practice guidelines exist to clearly delineate the goals for glycemic control in diabetic patients. Despite best efforts of patients and caregivers, either episodic or chronic problems with glycemic control occur. In response to evidence from the Diabetes Control and Complications Trial Group [1,2] and other studies, caregivers strive to tightly control their patients’ blood glucose levels. Tight control, along with other factors such as decreased oral intake, intercurrent illness, or noncompliance with the timing and amount of medications, may result in hypoglycemia. This problem is more common in Type I rather than Type II diabetes, given the propensity of insulin-dependent diabetics to have deficits in counter-regulatory responses of glucagon and epinephrine [3].
* Corresponding author. E-mail address:
[email protected] (S.D. Wolfsthal). 0095-4543/06/$ - see front matter Ó 2006 Elsevier Inc. All rights reserved. doi:10.1016/j.pop.2006.06.008 primarycare.theclinics.com
712
WOLFSTHAL
et al
Diabetics who have recurrent hypoglycemia may present with a variety of symptoms, from mild cognitive dysfunction to syncope, tachycardia, diaphoresis, or more subtly with nocturnal symptoms of night sweats, nightmares, and early morning hyperglycemia. This latter findingdcalled the Smogyi phenomenondresults from the surge of counter-regulatory hormones after an episode of hypoglycemia occurring during sleep. The hypoglycemia causes release of glucagon, cortisol, and epinephrine that ultimately results in hyperglycemia and elevated early morning blood glucose readings. Because the episode of nocturnal hypoglycemia may also be associated with accumulation of acetone from relative starvation, this condition may be differentiated from chronic hyperglycemia by checking for acetone in an early morning urine sample. The primary care physician should consider the possibility of intermittent hypoglycemia in any patient presenting with episodic nondescript symptoms of tachycardia, confusion, or other vague symptoms accompanied by a relief of symptoms with administration of glucose. Patients at risk for hypoglycemia include those who have had a previous episode of hypoglycemia, adolescents, males, those consuming alcohol, those using higher doses of insulin and oral hypoglycemic agents, and those using beta-blockers [4]. Baseline glycohemoglobin levels do not predict the incidence of hypoglycemic episodes [5]. Although obtaining a blood glucose level at the time of symptoms helps to solidify the diagnosis, this is often not available. For patients presenting with classic acute hypoglycemia in the office setting, the management is straightforward, and includes oral or intravenous administration of high concentrations of glucose (25–50 mg of 50% dextrose) or 0.5 to 1 mg of subcutaneous or intramuscular glucagon. Most patients will respond rapidly, and can be observed in the office with hourly glucose levels for 2 or 3 hours before allowing them to go home. Hypoglycemia can recur if it is caused by overuse of sulfonylureas and excess longacting insulin. For patients who have recurring hypoglycemia observed in the office setting, management in an emergency department or inpatient setting is prudent. For patients in whom hypoglycemia seems a likely cause of their intermittent symptoms, closer scrutiny of the glucose pattern is indicated. Patients should monitor their glucose levels before every meal and at bedtime, being certain to obtain levels during a symptomatic phase. Obtaining an early morning urine specimen to identify ketonuria helps to evaluate for nocturnal hypoglycemia. Those diabetics who have very tight control and recurrent hypoglycemia may have dulled their counter-regulatory response, and hence have more difficulty raising their blood sugar during a crisis. These patients may benefit from allowing more loose control of their blood sugars, thus allowing a return of critical counter-regulatory hormone responses. Patients should be counseled to always carry glucose tablets to stave off symptoms at first onset, and to have glucagons kits for more severe episodes. In patients who have syncope or dizziness, further evaluation for arrhythmias and silent ischemia is indicated. Family and
EMERGENCIES IN DIABETIC PATIENTS
713
friends should be trained in the management of hypoglycemia, and patients should be counseled against driving or using heavy machinery until their glycemic control is stabilized. Hyperglycemia Patients who have Type II diabetes are particularly susceptible to hyperglycemia. Although some patients have chronically elevated blood glucose levels, others develop hyperglycemia in response to infections or other severe stress. The primary care physician must distinguish early between these two scenarios. As discussed in the coming sections on cardiovascular disease and infectious problems in diabetic patients, hyperglycemia may accompany these complications. The actual level of hyperglycemia is less important than the etiology or associated abnormalities of the elevated blood sugars. In addition, early diagnosis of diabetic ketoacidosis (DKA) or hyperosmolar hyperglycemia early is critical to successful patient management [6]. DKA is a serious complication that occurs most commonly in Type I diabetics, but can also be seen in Type II diabetics who have severe hyperglycemia, infections, use of drugs such as steroids or thiazide diuretics, alcohol or substance abuse, or underlying pancreatic pathology, such as a pancreatic cancer [7]. These patients frequently present with shortness of breath caused by the metabolic acidosis, dizziness due to volume depletion and severe polyuria and polydipsia, nausea with vomiting, abdominal pain, and weight loss [6]. Rapid diagnosis involves obtaining serum glucose levels, serum acetone and determination of the anion gap. Once diagnosed, these patients must be admitted to an acute care facility for further management. A search for precipitating factors such as myocardial infarction (MI) or an infectious process should be undertaken. Hyperosmolar nonketotic hyperglycemia is more common in Type II diabetics and may also present with polyuria and polydipsia. Precipitating causes are similar to those noted for DKA. During the early stages, the patient may have few symptoms other than those associate with volume depletion; however, during more advanced stages, altered mental status and confusion may ensue. Type II diabetics who have severe hyperglycemia usually respond promptly to volume expansion with normal saline and small doses of short acting insulin. Although not clearly suffering from DKA, these patients may have a small amount of acetone in the serum, with a minimally elevated anion gap. In the presence of any cognitive dysfunction or orthostatic hypotension, these patients should be managed in the inpatient setting. A less serious degree of hyperglycemia is frequently encountered in the office setting. Diabetic patients who have historically poor glycemic control often present with minimal systemic symptoms, but with moderately elevated glucose levels. These can range from the high 100s mg/dl to 400 to 500 mg/dl. Patients should be queried as to the presence of symptoms of infectious disease or cardiovascular symptoms, including anginal
714
WOLFSTHAL
et al
equivalents such as shortness of breath or fatigue. Orthostatic vital signs, an electrocardiogram, and a urinalysis should be obtained. If the patient is reliable and has no evidence of significant hypovolemia, episodes of modest hyperglycemia can be treated on an outpatient basis. The patient should monitor blood glucose levels before meals and at bedtime, administering short-acting insulin (eg, regular humulin or lyspro) on a sliding scale. Modifying diet by reducing excess sugars and carbohydrates in addition to increasing physical activity will assist in returning blood glucose levels to normal. Those patients who have less severely elevated glucose levels (eg, 200–300 mg/dl) may be treated with an increase dose of their usual hypoglycemic medications. In all instances, close home and office monitoring is essential. Consideration should also be given to having the patient see a dietician or nutritionist, and engage in a graduated exercise program. Cardiovascular and peripheral vascular disease The relationship between diabetes and atherosclerotic disease is well-established. Complications from vascular disease account for a high burden of morbidity and mortality among diabetic patients. Vascular pathology in diabetics involves multiple organ systems. Some of the most devastating effects can be seen as the result of coronary heart disease and peripheral vascular disease. In the primary care setting, practitioners frequently focus on prevention. The goal is to halt the progression of atherosclerotic disease, while managing chronic end organ damage caused by established vascular disease. A review of diabetic office based emergencies would not be complete without a discussion of the number one cause of death in diabetic patientsd cardiovascular disease [8]. Clearly acute myocardial ischemia or infarction must be promptly referred to the emergency department; however, diabetic patients will frequently present to their primary care physician first with signs and symptoms of impending cardiac catastrophe. It is our job as primary care clinicians not only to have tried our best to prevent these events by counseling on the dangers of smoking and aggressively decreasing cholesterol levels, but also to maintain a high index of suspicion for cardiovascular complications in our diabetic patients. Although cardiovascular disease is responsible for many deaths, peripheral arterial disease (PAD) is credited with much morbidity [9]. PAD can lead to devastating limb ischemia and amputation, dramatically decreasing patients’ mobility and independence. Given the extent and severity of atherosclerotic complications in diabetic patients, in a primary care office setting practitioners should be prepared for vascular emergencies in this special population of patients. Coronary artery disease and silent ischemia The incidence of coronary artery disease among diabetics is staggering. It is estimated that 65% of diabetics will succumb to a cardiovascular event
EMERGENCIES IN DIABETIC PATIENTS
715
leading to death [10]. Furthermore, patients who have diabetes tend to have poorer outcomes from cardiovascular events compared with their nondiabetic counterparts [10]. MI is one of the most dreaded complications of atherosclerotic coronary artery disease. Mortality after acute MI is 1.2 to 2 times higher in diabetic patients compared with nondiabetics [11]. The detection of acute MI, particularly in the ambulatory setting, is a challenging and serious problem. Patients who have acute MI, especially those who have no history of coronary artery disease, can be easily misdiagnosed in the office setting with a gastrointestinal-related disorder, upper respiratory infection, or musculoskeletal pain [12]. This becomes even more difficult when managing the diabetic patient. Traditionally, angina pectoris is the classic symptom of myocardial ischemia that leads clinicians to suspect underlying coronary artery disease; however, this is not always the case with diabetic patients. Diabetics are among the population of patients who present with acute MI without classic substernal chest pain [13]. This can lead to delay in diagnosis and treatment. In the primary care setting, the authors try to predict a patient’s risk of MI by evaluating the risk factors and by careful history-taking to assess for not-so-obvious symptoms of coronary artery disease. This becomes paramount with diabetic patients, because they are part of a group of patients who experience silent ischemia. Silent ischemia is the presence of objective findings suggestive of myocardial ischemia that is not associated with angina or anginal equivalent symptoms [8]. True silent ischemia or silent MI is evidence of reversible ischemia on stress testing, or documentation of myocardial scar, in the absence of known MI. The Framingham Study originally predicted that almost 25% of heart attacks were silent or had gone unrecognized [8]; however, if atypical symptoms, such as those that are frequently experienced by diabetics (eg, nausea, vomiting, and shortness of breath) are accounted for, it is estimated that only approximately 12% of heart attacks were truly silent. Diabetic patients tend to account for a disproportionately high number of these silent MIs [14]. Cardiac autonomic neuropathy The underlying pathophysiology of silent ischemia is unknown. Impaired symptom perception, increased pain threshold, and endorphin excess are all proposed mechanisms [8]; however, cardiac autonomic dysfunction is thought to play a key role in silent ischemia of the diabetic patient, as well as contributing to several other cardiovascular complications [8]. Diabetic autonomic neuropathy encompasses many disorders and organ systems, including the gastrointestinal tract, genitourinary system, and sweat glands. Dysfunction of the autonomic nervous system can impact profoundly on quality of life issues for diabetic patients; however, cardiac autonomic dysfunction can have life-threatening consequences. Cardiac autonomic dysfunction is thought to occur in 17% of Type I and 22% of
716
WOLFSTHAL
et al
Type II diabetics. In addition to silent ischemia, cardiac autonomic neuropathy has been associated with orthostatic hypotension, exercise intolerance, intraoperative cardiovascular lability, and fatal arrhythmias leading to sudden cardiac death [15]. Many of the signs and symptoms of cardiac autonomic dysfunction are caused by inappropriate autonomic regulation of heart rate, cardiac output, and peripheral vascular resistance. It has been proposed that loss of cardiac parasympathetic tone occurs initially. This can be seen as a resting sinus tachycardia caused by unopposed, increased sympathetic activity [14]. Loss of appropriate heart rate variability is one of the earliest signs of cardiac autonomic dysfunction. This can be seen when sinus tachycardia persists despite activities that would normally increase parasympathetic vagal tone, such as a Valsalva maneuver. As sympathetic dysfunction progresses, the tachycardia may gradually decrease, and a fixed heart rate is the result [16]. Decreased exercise tolerance is multifactorial. There is impaired sympathetic response, and cardiac output does not increase appropriately to meet systemic demands with exercise. This effect is exacerbated by reduced ejection fraction and impaired diastolic filling, which are not unusual findings in the diabetic patient. For these reasons, cardiac stress testing by primary care providers may be considered before the initiation of exercise programs for diabetics [17]. Sudden cardiac death is another dreaded cardiac complication that is associated with cardiac autonomic dysfunction [8]. A prolonged QT interval may indicate an imbalance between right and left cardiac sympathetic innervation. Patients who have this condition are at the greatest risk for fatal ventricular tachyarrhythmias, particularly in the setting of impaired left ventricular function [15]. Postural hypotension caused by autonomic dysfunction, a fall in systolic blood pressure greater than 20 mmHg upon standing, is usually caused by damage of efferent sympathetic vasomotor fibers, particularly in the splanchnic vasculature. Patients may present with dizziness, blurry vision, and presyncopal symptoms [15]. When patients who have autonomic dysfunction rise from a seated position, they may feel dizzy and lightheaded. Impaired autonomic signaling results in an inadequate increase in cardiac output and peripheral vascular resistance leading to a drop in blood pressure. This is a late complication of autonomic neuropathy, and is indicative of somatic as well as cardiac autonomic dysfunction [17]. Tests for cardiac autonomic dysfunction may be able to identify those diabetics at highest risk for the above mentioned complications before they present to the office or emergency department with an event. It is important to identify these patients, because we know that diabetic patients who have cardiac autonomic neuropathy have increased fatal cardiac complications [17]. There are three key tests for office evaluation of diabetic autonomic neuropathy. These include R-R variation, heart rate response to Valsalva
EMERGENCIES IN DIABETIC PATIENTS
717
maneuver, and postural blood pressure response [14,17]; however, further investigation still needs to be pursued regarding the appropriate patient population to use these tests in and the best use of the results. In addition to providing comprehensive primary cardiovascular care for our diabetic patients, in an office-based setting practitioners must be cognizant of these patients’ high risk of morbidity and mortality from this disease. Cardiac autonomic neuropathy and coronary artery disease can have dangerous and fatal consequences. The primary care physician must be extremely observant and have a high index of suspicion for clues to these complications of diabetes before silent ischemia and fatal arrhythmias occur. Peripheral arterial disease and limb-threatening ischemia Twenty percent of diabetic patients have symptomatic PAD. An even higher percentage of patients have asymptomatic PAD or occlusive atherosclerotic disease of the lower extremities [9]. It is a major risk factor for lower extremity amputation, particularly when combined with peripheral neuropathy and infection [18]. Diabetes and smoking are the two strongest risk factors for PAD. Diabetes is more strongly associated with femoral-popliteal and tibial PAD compared with other risk factors such as smoking and hypertension, which tend to have more proximal disease. The most common symptom of PAD is claudication with pain, generally in calves, thighs, or buttocks, with ambulation, and it is relieved by rest. In the office setting, primary care physicians should be on the lookout for the limb-threatening manifestations of PAD, which require urgent intervention. The consensus statement from the American Diabetes Association has grouped the signs and symptoms of rest pain, tissue loss, gangrene, and acute ischemia into a category of critical limb ischemia (CLI) [9]. CLI is particularly concerning because it predicts poor outcomes: 30% of patients who have CLI will have amputations, and 20% will die within 6 months [9]. Diabetic patients presenting to the office with known PAD and new, severe, extremity pain should raise great concern. Physical examination findings such as absent or markedly diminished pulses and color changes are even more worrisome indications that a vascular complication has occurred. These patients are more susceptible to arterial thrombosis and acute limb ischemia [9]. These patients need immediate vascular surgery evaluation for possible angiography and thrombectomy. Toe gangrene can have a similar and equally emergent presentation. Patients may have sudden onset of toe pain with new discrete discoloration of the underperfused ischemic area. This may occur in the setting of palpable dorsalis pedis and posterior tibialis pulses. Gangrene of the toes can occur as the result of ruptured atherosclerotic plaque with thrombus formation, microthrombi from infection, or cholesterol emboli that has dislodged from ulcerated plaques [18]. If gangrene is bilateral, a more proximal source
718
WOLFSTHAL
et al
of cholesterol emboli (ie, aorta) should be considered [18]. Treatment should focus on evaluation of the gangrenous toe itself, and treating any complications such as infection and debridement of surrounding tissue, but also on immediate vascular evaluation of the embolic source and possible removal of ulcerated plaques [18]. A thorough examination of the extremities should be a routine part of all office visits for diabetic patients. This includes a vascular examination to evaluate pulses and skin examination to assess for any developing ulcers. Infection should be treated early and aggressively with antibiotics and surgical debridement if necessary. Diabetic patients must be educated not to walk barefoot and to wear appropriate, nonconstricting footwear if ulcers develop [9]. Medical therapy with antiplatelet agents such as aspirin or clopidogrel should be initiated in all diabetic patients. Further symptomatic treatment of claudication can be managed with pentoxifylline, which reduces blood viscosity, or cilostazol, an oral phosphodiesterase Type III inhibitor. Finally, these patients must be recognized early and referred for revascularization before limb-threatening ischemia occurs. Indications for revascularization include claudication unresponsive to the above medications and CLI. Aggressive and early intervention can prevent limb amputation, although amputation may be unavoidable if life-threatening infection or ischemia occurs [9].
Infectious diseases in the diabetic patient Risk of diabetes and infection Although it is generally accepted that strict blood glucose control with insulin therapy among critically ill patients in the surgical intensive care unit reduces morbidity and mortality [19], the question arises as to whether there is a correlation between diabetes and the risk of infection in the ambulatory care setting. Studies have depicted the inhibitory effects of elevated glucose concentrations on leukocyte phagocytosis. In addition, protein glycosylation can impair polymorphonuclear function. The sum of these factors has been shown to contribute to altered host defense [20]. Although diabetics are susceptible to infections that affect the general population, they are at increased risk for serious complications. Likewise, there are certain infections that occur more frequently and almost exclusively in the diabetic population. The following discussion focuses on a system based approach to early recognition of infectious emergencies in the ambulatory care setting and treatment options. Head and neck infections Invasive ‘‘malignant’’ otitis externa, most often caused by Pseudomonas aeruginosa, is an uncommon but potentially life-threatening infection that
EMERGENCIES IN DIABETIC PATIENTS
719
affects the external auditory canal and adjacent soft tissues, mastoid, temporal bone, and eventually the base of the skull. Patients often present with severe ear pain, especially at night, foul-smelling or purulent otorrhea, hearing loss, and fever. Concomitant cellulitis with edema or granulation tissue in the external auditory canal especially at the bone-cartilage junction may also be noted [21]. Unfortunately, these symptoms may be confused with a simple noninvasive otitis externa. One should suspect necrotizing external otitis if symptoms do not improve after 6 to 8 weeks of treatment. Work-up includes prompt referral to an otolaryngologist, debridement of necrotic tissue, and obtaining tissue samples for Gram stain and culture. Further imaging using MRI with gadolinium is helpful in defining the extent of bony and soft-tissue involvement. In addition, CT has been shown to aid in early diagnosis and to follow-up resolution. Intravenous antipseudomonal antibiotics should be administered for 4 to 6 weeks [22]. Mucormycosis is an acute and often fatal fungal infection, most often caused by Rhizopus, Absidia, and Mucor spp. The majority of cases are associated with an underlying disorder such as diabetes complicated by ketoacidosis and other immunocompromised states. At the onset, patients may complain of facial or ocular pain, and nasal discharge with or without rhinorrhea. Proptosis, chemosis, generalized headache, fever, and lethargy may then occur [22]. In one Canadian study [23], patients initially presented with complaints of a sinus infection. Progression of the disease resulted in the development of palatal necrotic ulcers, black eschars, or unilateral blindness from cranial nerve involvement. Early diagnosis is essential in the prevention of intracranial extension of the infection, which can lead to death in 80% of cases. It is considered to be the most fatal fungal infection known to man because it is rapidly disseminated by the blood vessels [24]. Workup includes biopsy and culture of tissue from the nasal passages or palate or direct sampling of sinus tissue and administering appropriate systemic antifungal agents, such as amphotericin B. MRI can delineate the extent of involvement of the sinuses, orbit, and central nervous system. Some cases have been documented to be cured by systemic antifungal therapy with minimal or no surgery; however, exenteration until the surgical margins are free of infection is also critical for a successful outcome. Pulmonary infections The incidence of pulmonary infections is not increased in the diabetic population. One study found that diabetes was not a significant independent risk factor for death in elderly patients who had pneumonia [22]. Respiratory infections in diabetics have been associated with significant morbidity and mortality, however. According to one Italian study [25], diabetes affects the pulmonary microcirculation by increasing vessel wall thickness and impairing gas exchange, which can lead to both loss of respiratory function and efficiency. In addition, they are more susceptible to lower respiratory
720
WOLFSTHAL
et al
tract infections by atypical microorganisms [25]. When compared with respiratory infections affecting the nondiabetic patient, infections caused by Streptococcus pneumoniae, Haemophilus influenzae, Legionella spp, and influenza are often more severe, and may require diabetic patients to be hospitalized. They are also more likely to have pulmonary infections caused by Staphylococcus aureus, gram-negative bacilli, Mycobacterium tuberculosis, and fungi [26]. It is important to vaccinate all diabetic patients for influenza and pneumococcal disease. Genitourinary infections The prevalence of bacteriuria in diabetic women is approximately 7% to 13%, three times higher when compared with nondiabetic women. This has been attributed to autonomic neuropathy leading to impaired bladder voiding. Symptomatic urinary tract infections may be more severe in diabetic than nondiabetic women; however, long-term prospective studies of asymptomatic bacteriuria in diabetic women suggest that asymptomatic bacteriuria is not harmful [27]. Treatment with antibiotics has also been shown not to decrease the frequency or severity of symptomatic urinary infections [27]. On the other hand, emphysematous cystitis is a necrotizing infection resulting in gas in the urinary bladder wall. Emphysematous pyelonephritis (EPN) is an infection that involves the renal parenchyma, perinephric tissues, and collecting system. Four factors involved in the pathogenesis of EPN are the presence of gas-forming bacteria, high tissue glucose, impaired tissue perfusion, and a defective immune response. Diabetics compose 70% to 90% of all cases, and 100% of bilateral EPN episodes. Escherichia coli has been found to be the causative agent, but Klebsiella pneumoniae, Proteus mirabilis, Pseudomonas aeruginosa, Aerobacter aerogenes, Citrobacter spp, and rarely, yeast such as Candida albicans or Cryptococcus neoformans, have been implicated as well. Patients typically present with symptoms similar to those of pyelonephritis: fever, chills, nausea, vomiting, and flank pain. Occasionally pneumaturia may be present with emphysematous cystitis, not EPN. Persistence of symptoms despite a 3 to 4 day course of antibiotic therapy should lead to further work-up [28]. Every diabetic who has a urinary tract infection and who appears severely ill should have at a minimum an abdominal radiograph to screen for emphysematous complications. CT scans are the most sensitive imaging test of the kidney, and can be obtained to confirm the presence of gas and diffuse thickening of the urinary bladder wall [29]. Left untreated, EPN is fatal. Patients who are treated with medical therapy alone (ie, hyperglycemic control and intravenous antibiotics) have a higher incidence of mortality when compared with those undergoing surgical therapies such as nephrectomy. By itself, antibiotic therapy is usually ineffective, and prompt nephrectomy in the setting of spreading gas is necessary.
EMERGENCIES IN DIABETIC PATIENTS
721
Skin and soft-tissue infections Foot infections are the most commonly encountered infection in diabetics in the office setting. Approximately one fifth of diabetes-related hospital admissions are secondary to foot infections. Complications associated with foot infections range from local involvement to cellulitis, osteomyelitis, amputation, and possibly death. Fifteen percent of all diabetic patients will develop foot ulcers, and of those, 16% will proceed to lower extremity amputation [30]. The presence of sensory, motor, and autonomic neuropathy also places the diabetic patient at higher risk for development of foot infections. There are many ways to categorize diabetic infections: the Wagner wound classification system, or non-limb threatening, limb-threatening, or life-threatening infections. A much simpler and widely accepted approach involves identifying infections as mild, moderate, or severe (limb threatening). Mild infections involve superficial ulcers without deep tissue or bone involvement. A majority of these infections can be treated as outpatients. Unlike mild infections, moderate infections do involve deep tissue, but there is no evidence of systemic toxicity. Severe infections, as the name implies, are likely to require hospitalization, because deep tissue necrosis, bone infections, gangrene, ischemia, and systemic toxicity can occur [30]. Most infections begin as a result of trauma, with the formation of a foot ulcer. Concern for infection should be raised when a patient presents to the office with a foot that has at least two of the following characteristics: redness, warmth, swelling, induration, pain, tenderness, or exuding pus. In patients who have significant neuropathy, pain may not be the presenting symptom. Like most infections in diabetics, systemic signs of toxicity such as leukocytosis, fever, or chills can be absent or appear late in the course of an infection. Worsening glycemic control is also a sign of significant infection. Infections can occur around the paronychia, interdigital areas of the foot, or neuropathic/ischemic ulcers. If there is concern for infection, one should first determine the extent of tissue involvement and the presence of underlying sinus tracts or abscesses. A prospective study comparing culture results [31] from various sources in 32 diabetics scheduled for amputation found that cultures obtained by swabs of superficial exudates or sinus tracts were less likely to agree with deep tissue cultures. In the office, wound cultures should be obtained either by biopsy, ulcer curettage, or aspiration [31]. Proper wound debridement and care, in addition, to antibiotics and glycemic control, are also important to the in-office treatment of foot infections. According to the Infectious Diseases Society of America (IDSA) guidelines, antibiotics should be chosen according to the severity of the infection and likely causes. The most common organisms isolated from diabetic foot wound cultures are aerobic gram-positive cocci. Approximately 25% of cases are caused by aerobic gram-negative bacilli in the setting of a polymicrobial infection. Anaerobic organisms are
722
WOLFSTHAL
et al
a causative factor in fewer than 20% of all cases, and always in conjunction with another organism. For mild to moderate cases, antibiotics such as ofloxacin, pipercillin-tazobactam, levofloxacin, clindamycin, and linezolid can be used. Recommended length of treatment includes 1 to 2 weeks for mild infection, 2 to 4 weeks for moderate to severe infections, and at least 4 to 6 weeks for patients who have osteomyelitis [31]. Treatment guidelines published by the IDSA recommend hospitalization if any of the following occur: systemic toxicity; metabolic instability; rapidly progressive or deep tissue infection; considerable necrosis, gangrene, or progression to critical limb ischemia; the necessity for urgent diagnostic or therapeutic interventions; and the inability to care for one’s self or lack of home support. Like most diabetic infections, care of foot infections requires a multidisciplinary approach. Surgical evaluation is necessary if there is extensive bone or joint involvement, or significant necrosis or gangrene [32]. As mentioned earlier, one feared complication of diabetic foot infections is the development of osteomyelitis. Osteomyelitis should be considered in the diabetic patient who, despite 2 weeks of antibiotics, presents with prolonged soft tissue infections, especially over bony prominences, ulcers deeper than 3 mm, and the ability to pass a probe directly through an ulcer to bone. If suspected, radiographs of the involved extremity may be taken initially. Radiographs by themselves are not very sensitive at detecting osteomyelitis; however, they can be very specific if erosive or lytic changes are present in conjunction with a positive probe test. MRI, however, is both sensitive (90%–100%) and specific (80%–100%) for osteomyelitis, and can be helpful in determining the extent of tissue penetration. Radionuclide bone scans, another imaging modality, are 90% sensitive, but only 50% specific [30]. Crepitant anaerobic cellulitis, nonclostridial anaerobic myonecrosis, clostridial myonecrosis (gas gangrene), and necrotizing fasciitis are all distinct types of infections that affect the diabetic foot; however, their clinical characteristics are difficult to differentiate. It is important to recognize that each is a medical emergency. Perhaps one of the most concerning skin infections that diabetics may rapidly develop is necrotizing fasciitis, an infection that starts in the subcutaneous space and spreads along fascial planes. Fournier’s gangrene refers to necrotizing fasciitis that involves the perineal area. The most common causes are from genitourinary infections and trauma. A known complication is scrotal gangrene. Initially, necrotizing fasciitis begins as an infection caused by either streptococci, specifically group A b-hemolytic (monomicrobial); or by facultative gram-negative bacilli such as E coli, Bacteroides fragilis, or clostridia (polymicrobial). Arms, legs, and the abdominal wall are areas often involved. Typically the patient presents with pain out of proportion to the physical examination, erythema, swelling, and tenderness of the involved body part. The skin will then appear shiny, smooth, and tense as the swelling continues. As the disease progresses, development of bullous lesions, eschar, or crepitus may occur. The bullae
EMERGENCIES IN DIABETIC PATIENTS
723
are initially serous, and then become hemorrhagic. Gangrene and sepsis are potentially fatal complications. If crepitus is suspected, this can be confirmed with a plain radiograph. CT scanning, on the other hand, is even more effective at noting soft-tissue gas, and can be helpful in demarcating the extent of infections. Ultrasound can be used similarly in Fournier’s gangrene. Because the associated mortality of necrotizing fasciitis is approximately 40% to 75%, prompt surgical evaluation and initiation of antibiotics is important. One should initially start broad-spectrum antibiotics targeting aerobic gram-positive and gram-negative organisms and anaerobes. A combination of penicillin or cephalosporin, clindamycin or metronidazole, and gentamicin should be started pending cultures. Surgical debridement is necessary to offset the high mortality [33]. According to the American Diabetes Association, approximately 18 million people (about 6.3% of the US population) have diabetes. The overall risk of death in diabetics is two times greater when compared with nondiabetics. Although it can be debated if diabetics are more susceptible to infections than nondiabetics, there is no doubt that they have a higher risk of complications and a worse prognosis if prompt recognition and treatment do not occur. It is often the task of the primary care physician to be the first responder and to recognize these medical urgencies in the acute ambulatory setting. Summary The diabetic patient poses special problems in the primary care setting. Symptoms that are relatively unimpressive on initial presentation, such as polyuria or dizziness, may actually be the beginning of serious medical complications. With careful evaluation and follow-up, some patients, such as those who have mild hypo- and hyperglycemia and certain infections, can be managed as an outpatients; however, many cardiovascular conditions, such as cardiac ischemia or limb-threatening peripheral vascular disease, require immediate transfer to an acute care facility. In all situations, close monitoring of glucose levels during all phases of caredin the office, in the hospital and at homedis essential to achieving target glycemic control and rapid detection of clinical conditions that often first manifest as alterations in glycemic control. References [1] The Diabetes Control and Complications Trial Research Group. The effect of intensive treatment of diabetes on the development and progression of long-term complications in insulin-dependent diabetes mellitus. N Engl J Med 1993;329:977–86. [2] The Diabetes Control and Complications Trial Research Group. Epidemiology of severe hypoglycemia in the diabetes control and complications trial. Am J Med 1991;90:450–9. [3] Meneilly GS, Cheung E, Tuokko H. Counterregulatory hormone responses in hypoglycemia in the elderly patient diabetes. Diabetes 1994;43:403–10.
724
WOLFSTHAL
et al
[4] Pedersen-Bjergaard U, Reubsaet JL, Nielsen SL, et al. Psychoactive drugs, alcohol and severe hypoglycemia in insulin-treated diabetes: analysis of 141 cases. Am J Med 2005;118: 307–10. [5] Cox DJ, Kovatchev BP, Julin DM, et al. Frequency of severe hypoglycemia in insulin-dependent diabetes mellitus can be predicted from self-monitoring blood glucose data. J Clin Endocrinol Metab 1994;79:1659–62. [6] American Diabetes Association. Position statement. Hyperglycmic crises in diabetes. Diabetes Care;27(Suppl 1):S94–102. [7] Kitabchi AE, Umpierrez GE, Murphy MB, et al. Hyperglycemic crises in patients with diabetes mellitus. Diabetes Care 2003;26(Suppl 1):S109–17. [8] Tabibiazar R, Edelman SV. Silent ischemia in people with diabetes: a condition that must be heard. Clinical diabetes 2003;21:5–9. [9] American Diabetes Association. Peripheral arterial disease in diabetes. Diabetes Care 2003; 26:3333–41. [10] Gavin JR, Peterson K, Warren-Boulton E. Reducing cardiovascular disease risk in patients with type 2 diabetes: a message from the national diabetes education program. Am Fam Physician 2003;68(8): 1569–74, 1577–8. [11] Reddy MS, Gupta SC. Diabetes and cardiovascular disease. Emerg Med 2002;34(4):28–35. [12] Sequist TD, Bates DW, Cook EF, et al. Prediction of missed myocardial infarction among symptomatic outpatients without coronary heart disease. Am Heart J 2005;149(1):74–81. [13] Canto JG, Shlipak MG, Rogers WJ, et al. Prevalence, clinical characteristics, and mortality among patients with myocardial infarction presenting without chest pain. JAMA 2000; 283(24):3223–9. [14] Chaudhuri D, Hopkins WE. Cardiac disease in diabetes mellitus. In: Leahy JL, Clark NG, Cefalu WT, editors. Medical management of diabetes mellitus. New York: Marcel Dekker; 2000. p. 367–84. [15] Vinik AI, Freeman R, Erbas T. Diabetic autonomic neuropathy. Semin Neurol 2003;23(4): 365–72. [16] Stevens MJ. Diabetic autonomic neuropathy. Up To Date. Available at: http://www. utodol.com/utd/content/topic.do?topicKey¼neuropat/7978&type¼A&selectedTitle¼1w12. Accessed September 26, 2005. [17] Maser RE, Lenhard MJ. Cardiovascular autonomic neuropathy due to diabetes mellitus: clinical manifestations, consequences, and treatment. J Clin Endocrinol Metab 2005; 90(10):5896–903. [18] Levin ME. Preventing amputation in the patient with diabetes. Diabetes Care 1995;18(10): 1383–94. [19] Van den Berghe G, Wouters P, Weekers F, et al. Intensive insulin therapy in critically ill patients. N Engl J Med 2001;345:1359–67. [20] Marhoffer W, Stein M, Maeser E, et al. Impairment of polymorphonuclear leukocyte function and metabolic control of diabetes. Diabetes Care 1992;15:256–60. [21] Handzel O, Halperin D. Necrotizing (malignant) external otitis. Am Fam Physician 2003;68: 309–12. [22] Nirmal J, Caputo G, Weitekamp M, et al. Infections in patients with diabetes mellitus. N Engl J Med 1999;341:1906–12. [23] Safar A, Marsan J, Marglani O, et al. Early identification of rhinocerebral mucormycosis. J Otolaryngol 2005;34(3):166–71. [24] Hilal A, Taj-Aldeen S, Mirghani A. Rhinoorbital mucormycosis secondary to Rhizopus oryzae: a case report and literature review. Ear Nose Throat J 2004;83(8): 556: 558–60, 562. [25] Ardigo D, Valtuena S, Zavaroni I, et al. Pulmonary complications in diabetes mellitus: the role of glycemic control. Curr Drug Targets Inflamm Allergy 2004;3(4):455–8. [26] D’Agata E, Eliopoulos G. Clinical updates in infection disease: infections in the diabetic patient. Available at: http://www.nfid.org/%5Fold/publications/clinicalupdates/id/ diabetic.html. Accessed October 17, 2005.
EMERGENCIES IN DIABETIC PATIENTS
725
[27] Nicolle LE. Asymptomatic bacteriuria: when to screen and when to treat. Infect Dis Clin North Am 2003;17(2):367–94. [28] McHugh T, Albanna S, Stewart N. Bilateral emphysematous pyelonephritis. Am J Emerg Med 1998;16(2):166–9. [29] Perlemoine C, Neau D, Ragnaud JM, et al. Emphysematous cystitis. Diabetes Metab 2004; 30(4):377–9. [30] Frykberg R. An evidence-based approach to diabetic foot infections. Am J Surg 2003;186: 44S–54S. [31] Weintrob A, Sexton D. Management of diabetic foot infections. Up To Date. Available at: http://www.utdol.com/utd/content/topic.do?topicKey¼skin_inf/8559&type¼A&selected Title¼1w10. Accessed September 27, 2005. [32] Hellekson K. IDSA releases guidelines on the diagnosis and treatment of diabetic foot infections. Am Fam Physician 2005;71:1429. Available at: http://www.aafp.org/afp/ 20050401/practice.html#p1. Accessed August 30, 2006. [33] Green R, Dafoe D, Raffin T. Necrotizing fasciitis. Chest 1996;110:219–29.
Prim Care Clin Office Pract 33 (2006) 727–750
Common Problems and Emergencies in the Obstetric Patient Kevin S. Ferentz, MD*, LaQuandra S. Nesbitt, MD Department of Family Medicine, University of Maryland School of Medicine, 29 South Paca Street, Baltimore, MD 21201, USA
More than 6 million pregnancies occur in the United States annually. Although most physicians in America do not provide obstetric care, most physicians deal with pregnant patients. True obstetric emergencies occur infrequently; but many normal changes and urgent conditions in pregnancy may be perceived as an emergency to the expectant mother. This article is designed to provide primary care practitioners with a review of obstetric emergencies and common obstetric problems.
Physiology Many of the symptoms of pregnancy are related to normal physiologic changes. Essentially, every organ system is affected. Most of the physiologic changes in pregnancy described below are detailed in any standard textbook of obstetrics. Being aware of these physiologic processes makes physicians better prepared to provide nonobstetric care to pregnant patients. Cardiac–respiratory Maternal plasma volume, cardiac output, and stroke volume increase during pregnancy, accompanied by a decrease in systemic vascular resistance. These changes manifest as mild increases in heart rate and transient decrease in blood pressure during the first and second trimester. The expanding uterus may lead to compression of the inferior vena cava. The enlarged uterus causes a decreased total lung capacity and residual volume, resulting in an increase in tidal volume. Progesterone increases carbon
* Corresponding author. E-mail address:
[email protected] (K.S. Ferentz). 0095-4543/06/$ - see front matter Ó 2006 Elsevier Inc. All rights reserved. doi:10.1016/j.pop.2006.06.003 primarycare.theclinics.com
728
FERENTZ & NESBITT
dioxide sensitivity causing mild hypoventilation leading to a compensated respiratory alkalosis. Gastrointestinal The gastrointestinal tract is affected from the oral cavity to the rectum. Dietary cravings and pica may negatively impact nutritional status. Increased salivation (ptyalism) may occur. Hyperemia from increased plasma volume may cause discolored and bleeding gums, while engorged rectal blood vessels cause hemorrhoids. Morning sickness, while typically benign, can progress to hyperemesis gravidarum (see below). Gastric reflux may be precipitated or worsened by uterine compression of the stomach and a decrease in lower esophageal sphincter tone secondary to progesterone. Decreased gastric motility from progesterone may result in indigestion or constipation. Constipation is often due to oral iron administration. Genitourinary The enlarging uterus may compress the ureters, causing transient and subclinical hydronephrosis. Urinary tract infections are more likely to result in ascending infections. Mild glucosuria is often normal in pregnancy, but proteinuria and ketonuria usually require further investigation. Musculoskeletal A common complaint of pregnancy is low back pain. Lumbar curvature is altered by the change in the woman’s center of gravity, and causes pain that may be relieved by massage or heat. The enlarging uterus stretches its supporting structures, primarily the round ligaments, which causes lower quadrant and groin pain in the second trimester.
Diagnosing pregnancy Physicians should screen all women of childbearing age for pregnancy before treatment. In cases of serious trauma this is not always possible, but in nonlife-threatening situations quick screening may prevent exposure to teratogens. History and physical Few physical findings in early pregnancy yield a definitive diagnosis of pregnancy, making the history essential. A patient should be questioned about the date of her last normal menstrual period (LMP) and any intermenstrual bleeding. Signs and symptoms include fatigue, sleepiness, nausea, vomiting, breast tenderness, and urinary frequency. Physical findings may include a bluish discoloration of the cervix and vagina (Chadwick’s sign),
COMMON PROBLEMS AND EMERGENCIES IN THE OBSTETRIC PATIENT
729
softening of the cervix (Goodell’s sign), and softening of the lower uterine segment (Hegar’s sign). Laboratory studies Pregnancy is diagnosed by detecting the b-subunit of human chorionic gonadotropin (HCG) in the urine or serum. HCG is produced almost exclusively by the placenta. The urine HCG test is a qualitative measurement that is occasionally positive before the missed menstrual period, and can detect levels as low as 20 mIU/mL [1]. Serum HCG is a quantitative measurement, and may be elevated before the missed menstrual period. HCG levels double approximately every 48 hours in early pregnancy. Ultrasound Ultrasound is the study of choice to diagnose and date pregnancy. A gestational sac is usually visualized by transvaginal ultrasound at 5 weeks gestation (from date of LMP), which correlates with a b-HCG of R1000 mIU/mL [2,3]. The yolk sac is typically seen at a b-HCG level of 2500 mIU/mL, corresponding to a gestational age of 6 weeks. The fetal pole is first identified on transvaginal ultrasound at 7 weeks and a b-HCG of R5000 mIU/mL [2,4]. Fetal heart activity can be detected by ultrasound at approximately 5 weeks, and can be heard with a handheld Doppler by 10 weeks [5].
Antepartum emergencies Trauma Trauma is the most common cause of nonobstetric death among pregnant patients [6–8]. Significant trauma occurs in 6% to 7% of US pregnancies [9–11]. The incidence of trauma increases throughout pregnancy, with half occurring in the third trimester [12]. Most blunt trauma is caused by motor vehicle collisions, domestic violence, or physical abuse, and falls [6]. Fetal mortality after maternal blunt trauma ranges from 3.4% to 38%, and is usually secondary to placental abruption, maternal shock, or maternal death [13–18]. Other factors associated with increased fetal mortality include maternal pelvic fracture, ejection from a motor vehicle, history of alcohol use, young maternal age, smoking history, and uterine rupture [12,16,17,19–21]. Because there cannot be fetal survival without maternal survival, initial efforts should focus on stabilizing the expectant mother. Airway and breathing Initial stabilization includes attention to airway, breathing, and circulation. Pregnant trauma patients should be transported to facilities that provide both maternal and fetal monitoring. Oxygen and intravenous (IV)
730
FERENTZ & NESBITT
fluids should be given to decrease the risk of anoxia and circulatory collapse. Fetal anoxia may progress more rapidly in the pregnant patient due to the increase in oxygen consumption that occurs with pregnancy [9]. Forceful administration of oxygen with a ball-valve mask should be avoided, as pregnant women are at increased risk of aspiration due to delayed gastric emptying. For oral intubations, rapid-sequence induction with cricoid pressure and gastric decompression is suggested [22,23]. If tension pneumothorax is suspected, needle aspiration at the second intercostal space along the midclavicular line or closed-tube thoracostomy at the third or fourth intercostal space should be performed [23,24]. Circulation The mother should be kept in the left lateral decubitus position (to prevent compression of the inferior vena cava) while stabilizing the cervical spine. Folded sheets or pillows can be used as a wedge under the right side if the patient is on a backboard. Uterine compression of the inferior vena cava may cause a decrease of 30 mmHg in maternal systolic pressure and a 30% decrease in stroke volume, resulting in decreased uterine blood flow [9]. Pregnant women have a 30% to 50% increase in blood volume, predisposing to severe hemorrhage. As uterine blood flow reaches almost 600 mL/min [23,25], blunt trauma or penetrating injury may result in exsanguination. Up to 30% of blood volume, or 2 L, can be lost before the patient develops hypotension or tachycardia; therefore, maternal hypotension can be an ominous sign. Ringer’s lactate should be given at a ratio of 3:1 of estimated blood loss. Intravenous lines should not be placed in the groin or lower extremity to avoid pooling in injured pelvic veins or inferior vena cava syndrome. When blood transfusion is necessary, typed and crossmatched Rh-compatible blood or autotransfusion should be used. When time does not allow for proper crossmatching, type-specific or type O-negative blood should be used. Vasopressors do not substitute for volume resuscitation. Norepinephrine and epinephrine may increase maternal blood pressure, but they cause severe decreases in uterine blood flow. Low-dose dopamine (less than 5 mg/kg) may increase maternal blood pressure with little effect on uterine blood flow [23]. Secondary assessment After addressing the ABCs, fetal status should be assessed. If fetal heart tones are absent, fetal resuscitation should not be attempted due to the low probability of survival. If heart tones are present, fetal viability should be determined based on gestational age and fetal weight. A fetus with a gestational age of 24 to 26 weeks and estimated fetal weight of 500 g is deemed potentially viable [9]. Leopold maneuvers, fundal height, history, and ultrasound help determine gestational age and fetal weight. If the fetus is viable, fetal heart monitoring should be continued while completing the maternal
COMMON PROBLEMS AND EMERGENCIES IN THE OBSTETRIC PATIENT
731
assessment. Contractions are commonly seen and resolve within a few hours in 90% of cases. Eight or more contractions per hour for more than 4 hours is often associated with placental abruption [9,26]. Severe fetal bradycardia (!100 beats/min) or repetitive late decelerations unresponsive to intrauterine resuscitation require immediate delivery. Although most abnormal fetal heart patterns are not predictive of fetal outcome, a normal tracing combined with normal physical examination has a negative predictive value of 100% [9,11]. The duration of continuous monitoring is controversial. A prospective trial to determine outcomes after obstetric trauma shows that patients discharged after 4 or 24 hours had similar outcomes [9]. Patients with persistent contractions, vaginal bleeding, or a nonreactive stress test, or who were involved in high-risk trauma (vehicle versus pedestrian or high-speed crashes) should be monitored continuously for 24 hours [14]. Diagnostic imaging The risk of fetal death secondary to radiation exposure is less than 1% within the first 2 weeks after conception [27]. A risk of malformation exists for 2 to 15 weeks after conception, but the risk is small for exposures limited to 100 mGy. Radiologic studies should not be withheld in the pregnant patient if they may aid in a definitive diagnosis. Low-dose radiology is defined as conventional X-rays of the extremity, head, or thorax, CT of the head or chest (if direct exposure to the fetus is avoided), MRI, and ultrasound. High-dose studies include abdominal or pelvic X-rays, CT of the abdomen, and nuclear studies (depending on the contrast media). Typical abdominal and pelvic studies involve a dose of 1 to 50 mGy [6,28]. Initial radiographs may include chest, abdominal, and cervical spine X-rays, fetal ultrasound, and maternal abdominal ultrasound. Abdominopelvic ultrasound provides quick fetal cardiac assessment, placenta location, gestational age, and amniotic fluid index, but may miss up to 80% of abruptions [21,26,29]. If initial diagnostic studies are equivocal, additional studies may include peritoneal lavage, CT, MRI, and angiography to identify bleeding sources (and embolization if required), and cystourethrography to evaluate urethral bleeding. If the patient is unstable, exploratory laparotomy should be preformed in lieu of radiologic studies. Surgery for trauma patients has not been associated with increased fetal loss [23]. Laboratory studies Labs should include a complete blood count (CBC), blood type, and crossmatch, basic metabolic panel, coagulation studies, and toxicology. The Kleihauer-Betke test to determine fetal versus maternal bleeding has become obsolete, and is used only for determining the quantity of fetal–maternal hemorrhage for accurate dosing of immune globulin [11,30]. The Kleihauer-Betke test is not predictive of fetal outcome, may be positive in low-risk nontrauma patients, and has no impact on clinical management.
732
FERENTZ & NESBITT
Disseminated intravascular coagulation (DIC) may occur with placental abruption, and requires immediate intervention due to its association with poor fetal outcomes. One retrospective study found that maternal acidosis is associated with poor fetal outcomes. Administering bicarbonate with IV fluids may improve tissue perfusion and fetal oxygenation [19]. Thirteen percent of pregnant trauma patients have positive serum alcohol levels, and 12% to 16% have positive toxicology screens [9,15,23]. Prevention Although trauma is unpredictable, there is a role of primary prevention when providing prenatal care. Proper seat belt use is associated with decreased maternal and fetal injury [31,32]. Unrestrained women are nearly three times more likely to abort [9]. Common reasons for lack of restraint use are lack of comfort and forgetfulness [33]. Proper seatbelt use should be reviewed at the initial prenatal visit. The lap belt should be placed under the gravid abdomen, with the shoulder harness off to the side of the uterus, between the breasts, and over the midline of the clavicle. A domestic violence assessment should also be completed. Domestic violence affects up to 25% of pregnant women but physicians identify only 4% to 10% of victims [13,34]. In 1999, homicide was the third leading cause of injury-related death for all women of reproductive age, and the pregnancy-associated homicide ratio was 1.7 per 100,000 live births. Risk factors for death by homicide are age younger than 20 years, Black race, and late or no prenatal care [35].
Ectopic pregnancy Ectopic pregnancy is the implantation of a fertilized ovum in a site other than the intrauterine cavity. Fallopian tube pregnancies account for 95% of ectopics, followed by the cornua of the uterus, ovary, cervix, and abdominal cavity [36,37]. Ectopic pregnancy is the most common cause of maternal mortality in the first trimester, and causes 10% to 15% of all maternal deaths. Ectopic pregnancies occur in the United States in approximately 20 per 1000 pregnancies [38]. Risk factors include history of pelvic inflammatory disease (PID), previous ectopic, tubal or uterine surgery or instrumentation, uterine or tubal anomalies, and cigarette smoking. Intrauterine devices have not been shown to be a risk factor for ectopic pregnancy or pelvic inflammatory disease [36]. A woman with a history of one ectopic has a recurrence rate of 15% to 20% that increases to 32% with two previous ectopics [36,39]. Ectopic pregnancy is more frequently diagnosed in women over 35 and non-White women [37]. Any woman of childbearing age who presents to a physician’s office or emergency department. could have an ectopic pregnancy, and physicians
COMMON PROBLEMS AND EMERGENCIES IN THE OBSTETRIC PATIENT
733
should have a high index of suspicion. Approximately 50% are missed at the time of initial presentation [36,40,41]. Presenting complaints include signs and symptoms of pregnancy, abdominal pain with spotting, pain radiating to the shoulder, or syncope. Physical findings include cervical motion tenderness, slightly enlarged uterus, adnexal mass, and abdominal tenderness. In cases of leaking or ruptured ectopic pregnancies, hypotension and shock may be present. Initial evaluation should include a quantitative b-HCG along with an ultrasound. Ectopic pregnancy is suggested when the b-HCG is O1500 mIU/mL in the absence of a gestational sac on transvaginal ultrasound. If abdominal ultrasound is used, absence of gestational sac with b-HCG O 6500 mIU/mL is suspicious for ectopic [36]. If ultrasound is not available, a culdocentesis may be performed. The presence of nonclotting bloody fluid is suggestive of ectopic pregnancy. Culdocentesis may help differentiate between ectopic pregnancy and ruptured ovarian cyst, as the latter will yield clear straw colored fluid. In some instances, a definitive diagnosis cannot be made on the initial evaluation. If the patient is hemodynamically stable and compliant with medical management, serial b-HCG levels can be monitored as an outpatient. A b-HCG should be done in 48 hours, at which time the level should be doubled. Inadequate rise or falling b-HCG levels at 48 hours indicates a nonviable pregnancy [42]. If a patient opts for serial b-HCG monitoring, the signs and symptoms of ruptured ectopic pregnancy should be reviewed with the patient and another responsible adult. Treatment has evolved over time, and depends on the stability of the patient. Sixty-eight to 77% of ectopic pregnancies resolve spontaneously [36]. Expectant or medical management may be used in a hemodynamically stable and compliant patient. Noninvasive therapies are preferred over surgery to prevent scarring of fallopian tubes, which increases the risk of subsequent ectopic pregnancies [43]. Expectant management is most effective if the ectopic pregnancy is !3.5 cm in diameter with declining b-HCG levels [36]. For medical management, methotrexate in a single IV or intramuscular dose of 50 mg/m2 is used when b-HCG levels are between 6000 and 15,000 mIU/mL [44]. Success rates range from 64% to 94% [45–47]. Side effects include nausea, vomiting, urinary frequency, and diarrhea [48]. An injection of 1 mg/kg into the ectopic sac may be as effective with fewer side effects [49]. A 5-day course of oral methotrexate is no more effective than placebo [50]. Failure of methotrexate therapy is associated with a b-HCG O10,000 mIU/mL [51,52]. If the patient is hemodynamically unstable, has a b-HCG O10,000 mIU/mL, or coexisting intrauterine pregnancy, surgical exploration is required. Salpingectomy and salpingostomy have similar fertility and recurrence rates. Salpingectomy is preferred with uncontrolled bleeding, severe tubal damage, or previous ectopic in the same tube. All patients should have serial b-HCG levels until it is undetectable or less than 5 mIU/mL [36].
734
FERENTZ & NESBITT
Gastrointestinal surgical conditions Appendicitis Appendicitis occurs in 1 in 1500 pregnancies, and is the most common nontraumatic and nonobstetric cause for surgery during pregnancy [53]. In pregnancy, the appendix migrates upward, closer to the gallbladder than McBurney’s point, reaching the level of the iliac crest by 24 weeks gestation [54]. Patients typically present with colicky abdominal pain. Pain may be epigastric, periumbilical, or localized to the right side. Anorexia, vomiting, rebound tenderness, and guarding may be present. Fever and leukocytosis may aid in diagnosis, and a calcified appendolith on ultrasound is diagnostic. Serial examinations should be performed and patients should be taken to surgery promptly when appendicitis is suspected. Fetal loss is 3% to 5% without perforation, and increases to 36% with perforation [53]. Antibiotic coverage for Gram-negative and anaerobic organisms should be given if perforation is suspected. Open or closed appendectomy may be performed, with laparoscopic removal increasingly popular in the first trimester [55]. Small bowel obstruction Intestinal obstruction complicates 1 in 1500 to 1 in 3000 pregnancies, with the incidence being higher in the third trimester as the uterus expands into the abdomen [56]. Maternal mortality may be as high as 20%, and fetal mortality can reach 40% if perforation occurs [23]. Adhesions from previous surgery are the most likely cause. Other causes include PID, volvulus, and intussusception. Patients typically present with abdominal pain, vomiting (possibly feculent), and obstipation. Lab studies should include a metabolic panel, as hypochloremic metabolic alkalosis may occur due to emesis. Serum lactate increases with ischemic bowel. Serial upright and flat-plate abdominal radiographs are 82% sensitive [56]. Dilated bowel loops and air fluid levels suggest obstruction. Initial conservative management includes bowel decompression via nasogastric suction. IV solutions with dextrose should be given as well as electrolytes. If conservative management fails, exploratory laparotomy is indicated.
Vaginal bleeding Vaginal bleeding is a common reason for concern in pregnancy. Causes include implantation, trauma from coitus, cervicitis, spontaneous abortion, or placental abnormalities. Hemodynamic status should be assessed. With significant bleeding, a type and cross should be obtained in anticipation of transfusion and to determine Rh status. Causes of life-threatening bleeding include spontaneous abortion, placenta previa, and placental abruption.
COMMON PROBLEMS AND EMERGENCIES IN THE OBSTETRIC PATIENT
735
Abortion Abortion is defined as the termination of pregnancy by any means before the fetus is sufficiently developed to survive (usually 20 weeks), corresponding to a fetal weight of less than 500 g [57]. Spontaneous abortion, or miscarriage, is the emptying of products of conception without medical intervention. Spontaneous abortion occurs in 15% to 20% of all diagnosed pregnancies [58]. Spontaneous abortion can be further divided into subgroups: threatened, inevitable, incomplete, missed, and recurrent. A speculum examination and ultrasound aid in differentiating types of abortion. Vaginal bleeding before 20 weeks gestation is termed threatened abortion. Vaginal bleeding occurs in 20% to 25% of women during the first half of pregnancy [59]. Bleeding can manifest as spotting or menstrual-like bleeding, and may persist for days to weeks. In the case of spontaneous abortion, bleeding typically preceeds the development of cramping abdominal pain, low back pain, pelvic pressure, and suprapubic pain. The combination of bleeding and pain yields a poorer prognosis. Speculum exam may reveal dark red blood or active bleeding from a closed cervical os. On ultrasound, a mean sac diameter of at least 17 mm without an embryo or 13 mm without a yolk sac has a specificity and positive predictive value of 100% in predicting nonviability [60]. Additional criteria are an empty gestational sac with a diameter of 15 mm at 7 weeks and 21 mm at 8 weeks (specificity of 90.8% in predicting miscarriage) [61]. The detection of fetal cardiac activity improves prognosis and has been shown to be associated with a loss rate of only 3.4% to 5.5% [62]. Maternal serum markers aid in prognosis. A free serum b-HCG level of 20 ng/mL has an 88.3% sensitivity and 82.6% positive predictive value for differentiating a viable and nonviable pregnancy [63]. A serum progesterone level of at least 25 ng/mL is associated with a 97% likelihood of viability, whereas a level of less than 14 ng/mL is associated with an adverse prognosis (sensitivity, 87.6%; specificity, 87.5%) [64]. Inevitable abortion occurs when vaginal bleeding or rupture of membranes occur before 20 weeks gestation in the presence of cervical dilatation. Incomplete abortion typically occurs after 10 weeks gestation when the fetus is expelled and the placenta is retained in the uterus. Incomplete abortion may cause severe hemorrhage but has a low associated mortality. Missed abortion is defined as retention of nonviable products of conception for several weeks. Missed abortions often occur without bleeding or pain. Uterine growth does not continue after a missed abortion and breast tenderness may resolve. Prolonged retention of products may be associated with DIC [65]. The management of spontaneous abortion has evolved. The standard of care had been surgical uterine evacuation. This remains the treatment of choice in patients with hemodynamic instability. Recent studies compared conservative, medical, and surgical management in the stable patient. Conservative management consists of monitoring for evidence of hemodynamic instability, sepsis, or coagulopathy [66]. Medical management uses
736
FERENTZ & NESBITT
misoprostol 400, 800, or 1200 mg orally or intravaginally, or mifepristone 600 mg or 1200 mg orally or intravaginally, to cause expulsion of uterine contents [66–68]. Surgical methods include vacuum aspiration or dilatation and curettage. In the stable patient, clinical criteria should be used instead of ultrasonographic criteria to determine whether surgical evacuation is indicated. A randomized trial showed that in inevitable and incomplete abortions, spontaneous resolution occurred within 3 days in 80% of women treated conservatively, with bleeding lasting only 1 day longer than those treated surgically [69,70]. A randomized trial comparing expectant and medical management yielded results in favor of expectant management. The groups were similar in pain scores, vaginal bleeding, and complications; however, convalescence was 1.8 days longer after medical treatment, likely secondary to the side effects of misoprostol [68]. Four studies have compared the use of medical management versus surgical intervention. In medically managed patients, results showed failure rates and complications increased as uterine size increased [71–74]. Although expectant management is a reasonable alternative to surgery, medical management has not been proven to have any added benefit [65]. All unsensitized Rh-negative pregnant women should be given Rh immune globulin within 72 hours of an abortion. Fetal–maternal transmission has been shown as early as 5 to 6 weeks of gestation, and 0.25 mL of fetal blood can cause isoimmunization. Although the incidence of isoimmunization after a first trimester abortion ranges from 0% to 3% in clinical trials, the benefits of preventing hydrops fetalis in subsequent pregnancies is higher than the risks of giving Rh immune globulin. The standard dose of 50 mg of Rh immune globulin intramuscularly (IM) provides prophylaxis for a fetomaternal hemorrhage of 2.5 mL [75]. Placenta previa Placenta previa is the implantation of the placenta over the cervical os. When placenta previa is present, cervical dilation disrupts the placental attachment causing bleeding. Placenta previa occurs in 1 in 200 births, and should be suspected in any patient presenting with vaginal bleeding after 24 weeks gestation [76]. Risk factors include previous cesarean, multiparity, age over 35 years, tobacco, and cocaine abuse. The risk of placenta previa increases from 1% to 4% after one Cesarean to 10% after four or more [77–81]. Placenta previa commonly presents as bright red, painless vaginal bleeding. Patients may occasionally complain of uterine contractions. Evaluation should include maternal vital signs, assessment of fetal status (heart rate), and palpation of the uterus for tenderness and contractions. A digital vaginal examination should not be performed! A sterile speculum should be gently inserted into the vagina to evaluate bleeding from the cervix. Placenta previa can be diagnosed by transabdominal ultrasound with an accuracy of approximately 95% [82]. Transvaginal ultrasound may improve accuracy but may
COMMON PROBLEMS AND EMERGENCIES IN THE OBSTETRIC PATIENT
737
precipitate bleeding [83]. If the distance between the lower edge of the placenta and the internal os is greater than 5 cm, placenta previa is excluded [84]. The course of placenta previa is variable. Of cases diagnosed before 20 weeks gestation, 90% will resolve by term [85]. Total placenta previa diagnosed in the second trimester will resolve in 74% of cases. Partial and marginal previas will resolve in 97.5% of cases [86]. The treatment of choice is usually observation. Cesarean delivery is indicated for severe hemorrhage, regardless of gestational age. General anesthesia is preferred, as regional anesthesia induces a sympathetic block that may inhibit the maternal response to acute blood loss [84]. Magnesium sulfate should be used if tocolysis is needed as beta agonists may worsen maternal hypotension and tachycardia in the hypovolemic patient [87]. Corticosteroids (betamethasone 12 mg IM every 24 hours times two doses or dexamethasone 6 mg IM every 12 hours times four doses) should be given to aid fetal lung maturity if delivery is expected before 34 weeks gestation. In a stable patient, initial therapy is bed rest in the inpatient setting for a minimum of 48 hours after the bleeding has ceased [83]. The patient may then be discharged home with bed rest and pelvic rest. Several studies have shown no significant difference in maternal and neonatal morbidity between inpatient and outpatient bed rest, with a 50% reduction in hospital days and savings of $15, 000 per case with outpatient management [88,89]. If outpatient management is chosen, the patient should be counseled that recurrent bleeding occurs in 60% of patients between 24 and 36 weeks gestation. At 36 weeks gestation, amniocentesis should be performed to assess fetal lung maturity. Elective Cesarean can be performed if the lungs are mature. Vaginal delivery may be attempted if the placenta is anterior as the fetal head will compress the bleeding placental bed against the pubic symphysis during delivery [83]. Placental abruption Placental abruption is a life-threatening cause of third trimester bleeding, which occurs when the placenta separates from its implantation before fetus delivers. Abruption may be partial or complete. Once separation occurs, prostaglandins are released, causing contractions that further compromise placental perfusion. Placental abruption complicates 1% to 2% of all pregnancies, with an incidence of 5.9 to 6.5 per 1000 singleton births and 12.2 per 1000 twin births [90]. Abruption accounts for 20% of all perinatal mortality. Abruptions involving over 50% of the placenta significantly increase the likelihood of stillbirth. Risk factors include gestational hypertension, maternal age, increasing parity, multiple gestation, polyhydramnios, chorioamnionitis, prolonged rupture of membranes, trauma, thrombophilias, and drug use, primarily cocaine and tobacco [91]. Several studies have shown a dose response to tobacco, with a 20% increased risk of abruption for every 10 cigarettes smoked per day and a 40% increased risk per year of maternal smoking [92].
738
FERENTZ & NESBITT
Patients with abruption typically present with painful vaginal bleeding. Other signs and symptoms are uterine tenderness, contractions, shock, and nonpalpable fetal parts. Making the diagnosis can be difficult, as the presence of classic signs and symptoms is highly variable. Blood may be retained between the placenta and uterus (concealed hemorrhage) so vaginal bleeding is present in only 78% of cases [76]. Uterine tenderness or pain is present in 66% and contractions or uterine hypertonus in 34%. Although ultrasound is insensitive and unreliable (hemorrhage detected in only 50% of cases), it should be performed to exclude placenta previa [90]. Placental abruption is a clinical diagnosis based on symptoms and postpartum evaluation of the placenta. Certain laboratory studies may aid in determining prognosis and treatment plan such as a CBC with platelet count, prothrombin time, International Normalized Ratio (INR), and fibrin degradation products to evaluate the severity of hemorrhage and presence of DIC. Management is aimed at avoiding fetal–neonatal mortality, maternal death, massive hemorrhage, DIC, infection, and Couvelaire uterus (blood in the uterine muscle). Hemodynamic status should be assessed continuously, with transfusions given as needed. Fetal heart rate should be monitored continuously, as fetal status determines management. Immediate delivery is recommended with a mature fetus or if the maternal or fetal status is compromised. A trial of labor and vaginal delivery is recommended as tolerated. Vaginal delivery is often successful, as the abruption tends to cause forceful frequent contractions that aid in rapid delivery. Cesarean delivery is indicated for failure to progress in labor, an unstable mother, or fetal distress. Disseminated intravascular coagulation may occur within 1 to 2 hours of an abruption, and delivery is the only way to stop its progression. Expectant management is acceptable in clinically stable maternal–fetal pairs with preterm placental abruption. Magnesium sulfate is the tocolytic of choice due to its limited cardiovascular effects, and may delay delivery without increasing neonatal or maternal morbidity [76,90]. If the fetus is less than 34 weeks gestation, corticosteroids may aid in fetal lung maturity. Hypertension Hypertension complicates 12% to 22% of pregnancies and is the third leading cause of pregnancy-related death [93,94]. Hypertensive management is the reason for 25% of antenatal hospital admissions [95]. Complications from hypertension in pregnancy include myocardial infarction, stroke, and renal failure and must be treated aggressively. Gestational hypertension During the first trimester, vasodilation occurs via prostacyclin and nitric oxide, causing a fall in systemic blood pressure. Blood pressure falls until 22 to 24 weeks gestation then gradually increases to prepregnancy levels [96].
COMMON PROBLEMS AND EMERGENCIES IN THE OBSTETRIC PATIENT
739
Gestational hypertension is defined as systolic blood pressure of 140 mmHg or higher or diastolic blood pressure O90 mmHg after 20 weeks gestation in a woman with previously normal blood pressure. Blood pressure typically returns to prepregnancy levels within the first 12 weeks postpartum. Patients with gestational hypertension are typically asymptomatic, and should be monitored closely for the development of proteinuria with urine dipstick measurements during prenatal visits. The first line agent for management of gestational hypertension is methyldopa, partly due to its history of safety [95]. The starting dose is 250 mg by mouth twice a day, titrating to a maximum of 3 g/d. Second-line agents include nifedipine and hydralazine. Caution should be used when combining nifedipine with magnesium sulfate as hypotension may occur. Thiazide diuretics and a- and b-adrenergic blockers are considered third-line agents. Labetalol is the most commonly used b-blocker, with doses starting at 100 mg by mouth twice a day. There is concern for intrauterine growth restriction with b-blockers, and they should be avoided in the first trimester. Thiazide diuretics can be continued in pregnant patients with chronic hypertension; however, they cause decreased plasma volume expansion and may worsen preeclampsia [95,97]. Preeclampsia Preeclampsia is new-onset hypertension (blood pressure R140/90 mmHg) with proteinuria (O0.3 g/24 h) after 20 weeks gestation in a woman with previously normal blood pressure [93,98,99]. Severe preeclampsia occurs with blood pressure O160/90 mmHg and 5 g or more of proteinuria/24 h. Preeclampsia complicates 5% to 7% of pregnancies (23.6 cases/1000 deliveries) [100]. Risk factors for preeclampsia are age !20 or O35 years, Black race, nulliparity, personal or family history of preeclampsia, and preexisting chronic medical conditions [98]. Left untreated, preeclampsia may progress to eclampsia, which is associated with a higher morbidity and mortality. The preeclamptic patient may be asymptomatic or complain of headache, blurry vision, abdominal pain, and hand or facial edema. Physical findings are nonspecific. Blood pressure is measured on two separate occasions before a diagnosis of preeclampsia can be made. The correct sized cuff should be used with the patient upright or in the left lateral recumbent position after 10 minutes of rest [101]. A urine dipstick should be obtained to identify proteinuria. If 1 þ protein is present, a 24-hour urine collection should be obtained to quantify the proteinuria [99]. A spot urine protein:creatinine ratio of !0.2 excludes preeclampsia (sensitivity, 90%; specificity, 70%) [102]. A CBC and liver enzymes should be obtained to evaluate for HELLP syndrome (hemolysis, elevated liver enzymes, and low platelets), which occurs in 4% to 14% of preeclamptic women [103]. Disseminated intravascular coagulation may occur, and can be monitored by fibrin degradation products. Decreased renal urate excretion occurs in preeclampsia, and the use of uric
740
FERENTZ & NESBITT
acid levels has become increasingly popular. In a study of 135 patients, a uric acid level of 5.9 mg/dL or greater was predictive of preeclampsia; however, the positive predictive value was only 33% [104]. Delivery is the definitive treatment. In the preterm patient, the risk and benefits of delivery should be assessed. If the mother is stable and there is no evidence of fetal compromise, the hypertension can be treated as an outpatient, and close monitoring may continue until the fetus is near term. Close monitoring should include serial CBCs and liver enzymes as well as weekly or biweekly nonstress tests. Corticosteroids may be given to aid in lung maturity. Delivery should occur in patients at term. Vaginal delivery is preferred over Cesarean section if the fetus can tolerate labor. Induction with prostaglandins for cervical ripening or oxytocin for augmentation is appropriate. Magnesium sulfate should be given during labor to prevent seizures. Randomized controlled trails in severe preeclampsia showed magnesium sulfate was associated with a lower incidence of seizure than antihypertensives [105]. The loading dose is 6 g IV, followed by a continuous infusion of 2 g per hour [106]. Magnesium sulfate is continued for at least 24 hours postpartum. Patients should be monitored frequently for the development of hyporeflexia, pulmonary edema, and oliguria. Antihypertensive therapy should be continued during labor with a goal blood pressure of 160/90 mmHg. Blood pressure should be lowered gradually using systemic antihypertensives, primarily hydralazine and labetalol. Hydralazine can be given as 5 mg IV or 10 mg IM every 20 minutes until blood pressure is controlled or the maximum dose of 400 mg per day is reached [99]. If hydralazine fails, labetalol should be given as a 20 mg IV bolus. If the hypertension is refractory to the initial dose, 40 mg IV may be given 10 minutes after the initial dose, followed by 80 mg every 10 minutes, for a maximum of 220 mg. If hydralazine and labetalol fail, nifedipine 10 to 20 mg orally every 30 minutes can be used to a maximum dose of 50 mg, or sodium nitroprusside should be considered. To date, there is no consistent data to support the use of low-dose aspirin or calcium supplementation in the prevention of preeclampsia, and their use is not recommended [99]. Eclampsia Eclampsia is the onset of seizures in a preeclamptic patient without a history of seizure or other identifiable cause. Eclamptic seizures are rare, and occur in less than 1% of preeclamptic women, but are associated with a high mortality, most commonly due to intracranial hemorrhage [105]. Eclampsia is also associated with increased rates of placental abruption, DIC, acute renal failure, and cardiopulmonary arrest [107]. Eclampsia may occur at any time during pregnancy, with 40% of seizures occurring before delivery and 16% of eclamptic seizures occurring within 48 hours postpartum [106]. Eclampsia may occur as late as 4 weeks postpartum [108].
COMMON PROBLEMS AND EMERGENCIES IN THE OBSTETRIC PATIENT
741
During the seizure the patient should be placed in the left lateral decubitus position with maintenance of the airway and suctioning of oral secretions. Oxygen should be given. Magnesium sulfate is more effective in treating eclamptic seizures than phenytoin or diazepam [106]. The loading dose is 6 g IV followed by a continuous infusion of 2 g per hour to be continued until 24 hours postpartum or 24 hours after the last seizure. If convulsions recur while receiving the maintenance infusion, an additional 2 g IV bolus should be given over 3 to 5 minutes. If the convulsions are refractory to adequate doses of magnesium sulfate, consider giving sodium amobarbital 250 mg IV over 3 to 5 minutes [106]. Asthma exacerbations Asthma complicates 1% to 4% of pregnancies, with status asthmaticus seen in about 0.2% of pregnancies [109,110]. One third of asthmatic pregnant women will experience worsening of their asthma during pregnancy. Increased incidence of preeclampsia, preterm labor, low birth weight, and perinatal mortality has been associated with pregnancies complicated by asthma [111]. Fetal hypoxia may occur before maternal signs and symptoms develop. Fetal compromise is caused by decreased uterine blood flow and decreased maternal venous return. Improving maternal oxygenation reverses these effects. The treatment of asthma exacerbations should focus on maintaining maternal oxygenation. Pulse oximetry, and, potentially, arterial blood gases, should be obtained, and supplemental oxygen given with a goal PaO2 O 60 mmHg and oxygen saturation O95%. b-Agonists are the first line of therapy, typically albuterol nebulized treatments (2.5 mg/3 mL or 0.083%) every 20 minutes times three. If the patient does not respond, corticosteroids should be given as methylprednisolone 60 mg IV or prednisone 60 mg orally. As the onset of action of corticosteroids is several hours, b-agonist treatment should continue. If albuterol is ineffective, the patient can be given epinephrine 1:1000 solution 0.1 to 0.5 mg subcutaneously every 10 to 15 minutes. Preventing respiratory failure is the ultimate endpoint of therapy, and intubation should occur at the first signs of respiratory distress. Prevention of asthma exacerbations in pregnancy is critical. All patients should have an action plan based on their baseline peak flow value. If the patient has a peak flow of less than 80% of expected, therapy with a bagonist should be initiated as needed. If the patient uses their albuterol multi-dose inhaler (MDI) more than twice a week, a controller medication (typically an inhaled corticosteroid) should be added. Leukotriene modifiers are another form of controller therapy. Special considerations during labor and delivery in the pregnant asthmatic patient are recommended. If the patient has an acute exacerbation requiring oral corticosteroids within 4 weeks of delivery, consider giving IV corticosteroids (hydrocortisone 100 mg every 8 hours) during labor. The
742
FERENTZ & NESBITT
use of F series prostaglandins should be avoided in the asthmatic patient, as they may cause bronchoconstriction. Methylergonovine (Methergine) should be given instead of synthetic prostaglandin F2a (carboprost tromethamine, Hemabate) in patients with postpartum hemorrhage [112]. Cervicitis/vaginitis Vaginitis and cervicitis are frequently seen during pregnancy. Theoretically, lower genital tract infections can become ascending infections that result in endometritis or chorioamnionitis, which have been correlated to preterm labor. Approximately 40% of spontaneous preterm labor is believed to be caused by infection [113,114]. Bacterial vaginosis Bacterial vaginosis (BV) is a risk factor for premature labor and perinatal infection [115,116]. BV is caused by an increase in normal vaginal flora such as Gardnerella vaginalis, Mycoplasma hominis, and anaerobes [117]. It occurs in 10% to 30% of women of childbearing age and up to 40% of pregnant women [118]. The gold standard for diagnosis is the Amsel criteria: milky, homogenous discharge; vaginal pH O4.5; positive Whiff test (fishy smell when KOH added); and clues cells visualized on microscopy [119]. Three out of four of these findings correlate with a 90% likelihood of BV. Oral metronidazole is often avoided during the first trimester secondary to theoretic concerns for teratogenicity [120]. Topical treatments are typically less effective [117]. Metronidazole vaginal gel can be used twice a day for 5 days, and has no effect on preterm labor [121]. Topical clindamycin cream may increase the risk of prematurity and neonatal infections [122]. It may be prudent to withhold treatment of the asymptomatic patient until the second trimester when metronidazole 250 mg orally three times daily for 7 days may be used. Vulvovaginal candidiasis Vulvovaginal candidiasis often manifests as a thick white discharge accompanied by pruritis, vulvovaginal swelling, and dysuria. Diagnosis should be based on microscopic evaluation of a vaginal sample with KOH, which reveals pseudohyphae in most cases. In rare cases, vaginal culture may be necessary for definitive diagnosis. Topical azoles are the treatment of choice during pregnancy, with a clinical cure rate of approximately 80% [123]. Fluconazole is categorized as a pregnancy class C drug, and can be used when azole therapy fails. Trichomonas vaginalis Trichomonas vaginalis is associated with preterm labor and low birth weight [124]. Women with trichomonas often complain of vaginal
COMMON PROBLEMS AND EMERGENCIES IN THE OBSTETRIC PATIENT
743
irritation and a copious, frothy, foul smelling discharge. Cervical examination may yield a friable, erythematous cervix (‘‘strawberry cervix’’). Microscopic examination of a saline wet mount usually shows motile trichomonads, but may be negative in up to 50% of culture positive infections [125]. As Trichomonas vaginalis is considered a sexually transmitted infection, it should be treated in both symptomatic and asymptomatic women to prevent further transmission. Treatment in the asymptomatic gravid woman, however, has not been shown to alter pregnancy outcome [124]. The Centers for Disease Control recommends treating symptomatic women with metronidazole 2 g orally [122]. Treatment of asymptomatic women should be withheld until the second or third trimester. Intravaginal clotrimazole has a cure rate of 48%, and can be used as a first line therapy during the first trimester [126]. If treatment is deferred, patients should be counseled to use condoms until curative treatment is provided. Neisseria gonorrhoeae Neisseria gonorrhoeae cervicitis has been linked to premature delivery [113]. Although often asymptomatic, women may present with vaginal discharge, vaginal bleeding, or dysuria. Cervical examination often reveals a mucopurulent discharge. Cervical motion tenderness may be present. Although the ‘‘gold standard’’ for diagnosis is culture, DNA tests are commonly used because of the ability to concurrently test for chlamydial infections. Treatment of gonococcal cervicitis is associated with decreased risk for preterm delivery [124]. Treatment is a single dose of ceftriaxone 125 mg IM [120]. Treatment for coexisting chlamydial infection is recommended. High-dose azithromycin 2 g can also be used to treat both gonococcal and nongonococcal cervicitis, but is less tolerated secondary to gastrointestinal side effects [127]. Doxycycline is a pregnancy category D drug, and should be avoided during pregnancy. Chlamydia trachomatis Chlamydia trachomatis remains the most common sexually transmitted infection, and occurs in 5% to 26% of pregnant women. Conflicting evidence exists regarding the impact of chlamydial cervicitis on pregnancy. Patients may present with a mucopurulent discharge, cervical motion tenderness, or be asymptomatic. DNA probe testing for Chlamydia trachomatis is 90% sensitive and 97% specific. Concurrent treatment for N gonorrhoeae is not recommended unless testing reveals the presence of N gonorrhoeae. Treatment regimens in pregnancy include: single-dose azithromycin 1 g orally, amoxicillin 500 mg three times a day for 7 days, or erythromycin base 500 mg orally four times daily for 7 days [127].
744
FERENTZ & NESBITT
Urinary tract infections Urinary tract infections have been shown to contribute to preterm births. Asymptomatic bacteriuria, defined as greater than 100,000 colonies of a single species isolated from a clean catch specimen, is present in 2% to 7% of pregnant women [128]. If untreated, pyelonephritis may develop in 20% to 40% of pregnant patients [129]. Approximately 40% to 80% of pregnancies complicated by acute pyelonephritis could be prevented by screening for and treating asymptomatic bacteriuria [128]. Routine screening should occur at the first prenatal visit. Subsequent urinalysis should be performed for symptomatic patients only. The most common pathogen is Escherichia coli, but the presence of group B streptococcus (GBS) may prove to be more concerning. If GBS is present, the patient should be treated at the time of diagnosis, and during the intrapartum period to prevent neonatal GBS sepsis. Therapeutic options include nitrofurantoin 100 mg orally twice daily, amoxicillin 250 mg orally three times daily, or cephalexin 250 mg orally four times daily [130]. Length of treatment ranges from 3 to 7 days for acute uncomplicated urinary tract infections. Suppressive therapy can be given with nitrofurantoin 100 mg orally at bedtime or cephalexin 250 mg orally at bedtime for patients who have recurrent of persistent bacteriuria [131]. Nitrofurantoin is contraindicated from 38 to 42 weeks of gestation. Hyperemesis gravidarum Nausea and vomiting of pregnancy, or morning sickness, typically begins between 4 and 7 weeks after the last menstrual period. It occurs in 80% of pregnant women, and will resolve by 20 weeks gestation in 90% of affected women. Hyperemesis gravidarum is defined as persistent vomiting causing dehydration, ketosis, electrolyte disturbances, and weight loss of more than 5% of body weight. This form of severe nausea and vomiting occurs in 1 in 200 pregnancies [132]. If nausea is mild, the patient should be encouraged to avoid noxious stimuli and replace large meals with frequent small meals consisting of simple carbohydrates. Antiemetics such as promethazine 25 mg every 4 hours can be given orally or rectally for symptomatic treatment. Refractory nausea and vomiting should be treated with other phenothiazines (prochlorperazine, chlorpromazine) or ondansetron. Dehydration, ketonuria, and ketosis should be treated with aggressive intravenous hydration using dextrose with normal saline or lactated Ringers solution. In severe cases of hyperemesis gravidarum, parenteral nutrition may be necessary. Summary Physicians not used to caring for pregnant patients may feel uncomfortable dealing with the many routine problems that can occur during
COMMON PROBLEMS AND EMERGENCIES IN THE OBSTETRIC PATIENT
745
a pregnancy. Other than true obstetric emergencies, which are usually cared for by obstetricians and family physicians, and the common problems of pregnancy can often be cared for by any primary care physician. Given the litigious nature of our society, especially in the realm of obstetrics, it does behoove the physician caring for pregnant women to be aware of the standards of care. When in doubt, it would be prudent to consult with a physician that routinely provides care to pregnant women.
References [1] QuickView þ product instructions. San Diego: Pacific Biotech, Inc. (a subsidiary of Quidel Corporation); 2005. [2] Dart RG. Role of pelvic ultrasonography in evaluation of symptomatic first-trimester pregnancy. Ann Emer Med 1999;33:310–20. [3] Bree R, Edwards M, Bohm-Velez M, et al. Transvaginal sonography in the evaluation of normal early pregnancy: correlation with HCG level. ARJ Am J Roentgenol 1989;153: 75–9. [4] Fossum G, Davajan V, Kletzky O. Early detection of pregnancy with transvaginal ultrasound. Fertil Steril 1988;49:788–91. [5] Bastian LA, Piscitelli JT. Is this patient pregnant?: can you reliably rule in or rule out early pregnancy by clinical examination? JAMA 1997;278:586–91. [6] Goldman SM, Wagner LK. Radiologic ABCs of maternal and fetal survival after trauma: when minutes may count. Radiographics 1999;19:1349–57. [7] Doan-Wiggins L. Trauma in pregnancy. In: Benrubi GI, editor. Obstetric and gynecologic emergencies. Philadelphia (PA): Lippincott; 1994. p. 57–76. [8] Doan-Wiggins L. Trauma in pregnancy. In: Harwood-Nuss A, editor. Obstetric emergencies. New York: Churchill Livingstone; 1990. p. 45–64. [9] Grossman NB. Blunt trauma in pregnancy. Am Fam Physician 2004;70:1303–10, 1313. [10] Rogers FB, Rozycki GS, Osler TM, et al. A multi-institutional study of factors associated with fetal dealth in injured patients. Arch Surg 1999;134:1274–7. [11] Connolly AM, Katz VL, Bash KL, et al. Trauma and pregnancy. Am J Perinatol 1997;14: 331–6. [12] Baerga-Varela Y, Zietlow SP, Bannon MP, et al. Trauma in pregnancy. Mayo Clin Proc 2000;75:1243–8. [13] Poole GV, Martin JN Jr, Perry KG, et al. Trauma in pregnancy: the role of interpersonal violence. Am J Obstet Gynecol 1996;174:1873–8. [14] Curet MJ, Schermer CR, Demarest GB, et al. Predictors of outcome in trauma during pregnancy: identification of patients who can be monitored for less than 6 hours. J Trauma 2000; 49:18–25. [15] Kissinger DP, Rozycki GS, Morris JA Jr, et al. Trauma in pregnancy. Predicting pregnancy outcome [published correction appears in Arch Surg 1991;126:1524]. Arch Surg 1991;126: 1079–86. [16] Theodoru DA, Velmahos GC, Souter I, et al. Fetal death after trauma in pregnancy. Am Surg 2000;66:809–12. [17] Ali J, Yeo A, Gana TJ, et al. Predictors of fetal mortality in pregnant trauma patients. J Trauma 1997;42:782–5. [18] Bochicchio GV, Napolitano LM, Haan J, et al. Incidental pregnancy in trauma patients. J Am Coll Surg 2001;192:566–9. [19] Scorpio RJ, Esposito TJ, Smith LG, et al. Blunt trauma during pregnancy: factors affecting fetal outcome. J Trauma 1992;32:213–6.
746
FERENTZ & NESBITT
[20] Weiss HB, Songer TJ, Fabio A. Fetal deaths related to maternal injury. JAMA 2001;286: 1863–8. [21] Reis PM, Sander CM, Pearlman MD. Abruptio placentae after auto accidents. A case– control study. J Reprod Med 2000;45:6–10. [22] El Kady D, Gilbert WM, Anderson J, et al. Trauma during pregnancy: an analysis of maternal and fetal outcomes in a large population. Am J Obstet Gynecol 2004;190:1661–8. [23] Coleman MT, Trianfo VA, Rund DA. Nonobstetric emergencies in pregnancy: trauma and surgical conditions. Am J Obstet Gynecol 1997;177:497–502. [24] Esposito TJ. Trauma during pregnancy. Emerg Med Clin North Am 1994;12:167–99. [25] Knuppel RA, Hatangad SB. Acute hypotension related to hemorrhage in the obstetric patient. Obstet Gynecol Clin North Am 1995;22:111–29. [26] Pearlman MD, Tintinalli JE, Lorenz RP. A prospective controlled study of outcome after trauma during pregnancy. Am J Obstet Gynecol 1990;162:1502–10. [27] Stovall M, Blackwell CR, Novada DH, et al. Fetal dose from radiotherapy with photon beams: report of AAPM Radiation Therapy Committee Task Group no. 36. Med Phys 1995;22:63–82. [28] Wagner LK, Lester RG, Saldana LR. Exposure of the pregnant patient to diagnostic radiation: a guide to medical management. 2nd edition. Madison (WI): Medical Physics; 1997. [29] Aitokallio-Tallberg A, Halmermaki E. Motor vehicle accident during the second or third trimester of pregnancy. Acta Obstet Gynecol Scand 1997;76:313–7. [30] Dhanraj D, Lambers D. The incidences of positive kleihauer-betke test in low-risk pregnancies and maternal trauma patients. Am J Obstet Gynecol 2004;190:1461–3. [31] Pearlman MD, Klinich KD, Schneider LW, et al. A comprehensive program to improve safety for pregnant women and fetuses in motor vehicle crashes: a preliminary report. Am J Obstet Gynecol 2000;182:1554–64. [32] Hyde LK, Cook LJ, Olson LM. Effect of motor vehicle crashes on adverse fetal outcomes. Obstet Gynecol 2003;102:279–86. [33] McGwin G Jr, Russell SR, Rux RL, et al. Knowledge, beliefs, and practices concerning seat belt use during pregnancy. J Trauma 2004;56:670–5. [34] Hedin LW, Janson PO. Domestic violence during pregnancy. The prevelance of physical injuries, substance use: aborttions and Miscarriages. Acta Obstet Gynecol Scand 2000; 79:625–30. [35] Chang J, Berg CJ, Saltzman LE, et al. Homicide: a leading cause of injury deaths among pregnant and postpartum women in the United States, 1991–1999. Am J Public Health 2005;95:471–7. [36] Tenore JL. Ectopic pregnancy. Am Fam Physician 2000;61:1080–8. [37] Hankins GD, Clark SL, Cunningham FG, et al. Ectopic pregnancy. In: Operative obstetrics. Norwalk (CT): Appleton & Lange; 1995. p. 437–56. [38] Centers for Disease Control and Prevention. Current trends in ectopic pregnancydUnited States, 1990–1992. MMWR Morb Mortal Wkly Rep 1995;44:46–8. [39] Pisarska MD, Carson SA, Buster JE. Ectopic pregnancy. Lancet 1998;351:1115–20. [40] Abbott J, Emmans LS, Lowenstein SR. Ectopic pregnancy: ten common pitfalls in diagnosis. Am J Emerg Med 1990;8:515–22. [41] Kaplan BC, Dart RG, Moskos M, et al. Ectopic pregnancy: prospective study with improved diagnostic accuracy. Ann Emerg Med 1996;28:10–7. [42] Ankum WM, Van der Veen F, Hamerlynck JV, et al. Suspected ectopic pregnancy. what to do when human chorionic gonadotropin levels are below the discriminatory zone. J Reprod Med 1995;40:525–8. [43] Clinical Policy. Critical issues in the initial evaluation and management of patients presenting to the emergency department in early pregnancy. Ann Emerg Med 2003;41:123–33. [44] American College of Obstetrics and Gynecologists. Medical management of tubal pregnancy. ACOG Practice Bulletin 3. Washington (DC): ACOG; 1998.
COMMON PROBLEMS AND EMERGENCIES IN THE OBSTETRIC PATIENT
747
[45] Hajenius PJ, Engelsbel S, Mol BWJ, et al. Randomized trial of systemic methotrexate versus laparoscopic salpingostomy in tubal pregnancy. Lancet 1997;350:771–9. [46] Lipscomb GH, Bran D, McCord ML, et al. Analysis of three hundred of fifteen ectopic pregnancies treated with single-dose methotrexate. Am J Obstet Gynecol 1998;178:1354–8. [47] Pietrzak Z, Augustyniak T, Szram K, et al. Efficacy of methotrexate in the treatment of nonruptured ectopic pregnancy. Acta Obstet Gynecol Scand 1999;78:258–9. [48] Stitka CS, Anderson L, Fredericksen MC. Single dose methotrexate for the treatment of ectopic pregnancy: northwestern memorial hospital three-year experience. Am J Obstet Gynecol 1996;174:1846–8. [49] Fernandez H, Bourget P, Ville Y, et al. Treatment of unruptured tubal pregnancy with methotrexate: pharmokinetic analysis of local versus intramuscular administration. Fertil Steril 1994;62:943–7. [50] Korhonen J, Stenman UH, Ylostalo O. Low-dose oral methotrexate with expectant management of ectopic pregnancy. Obstet Gynecol 1996;88:775–8. [51] Lipscomb GH, McCord ML, Stovall TG, et al. Predictors of success of methotrexate treatment in women with tubal ectopic pregnancies. N Engl J Med 1999;341:1974–8. [52] Hung TH, Shau WY, Hsieh TT, et al. Prognostic factors for an unsatisfactory primary methotrexate treatment of cervical pregnancy: a quantitative review. Hum Reprod 1998; 13:2636–42. [53] Babaknia A, Parsa H, Woodruff JD. Appendicitis during pregnancy. Obstet Gynecol 1977; 50:40–4. [54] Nathan L, Huddleston JF. Acute abdominal pain in pregnancy. Obstet Gynecol Clin North Am 1995;22:55–67. [55] Fallon WF, Newan JS, Fallon GL, et al. The surgical management of interabdominal inflammatory conditions during pregnancy. Surg Clin North Am 1995;75:15–31. [56] Connolly MM, Unti JA, Nova P. Bowel obstruction in pregnancy. Surg Clin North Am 1995;75:101–13. [57] Cunningham FG, Grant NF, Leveno KF, et al. Abortion. Williams obstetrics, 21st edition. New York: McGraw-Hill; 2001. [58] Makrydimas G, Sebire NJ, Lolis D, et al. Fetal loss following ultrasound diagnosis of a live fetus at 6–10 weeks of gestation. Ultrasound Obstet Gynecol 2003;22:368–72. [59] Sotiriadis A, Papatheodorou S, Makrydimas G. Threatened miscarriage: evaluation and management. BMJ 2004;329:152–5. [60] Tongsong T, Wanapirak C, Srisomboon J, et al. Transvaginal ultrasound in threatened abortions with empty gestational sacs. Int J Gynaecol Obstet 1994;46:297–301. [61] Falco P, Zagonari S, Gabrielli S, et al. Sonography of pregnancies with first-trimester bleeding and a small intrauterine gestational sac without a demonstrable embryo. Ultrasound Obstet Gynecol 2003;21:62–5. [62] Tannirandorn Y, Sangsawang S, Manotaya S, et al. Fetal loss in threatened abortion after embryonic/fetal heart activity. Int J Gynaecol Obstet 2003;81:263–6. [63] al-Sebai MA, Diver M, Hipkin LJ. The role of a single free beta-human chorionic gonadotropin measurement in the diagnosis of early pregnancy failure and the prognosis of fetal viability. Hum Reprod 1996;11:881–8. [64] al-Sebai MA, Kingsland CR, Diver M. The role of a single progesterone measurement in the diagnosis of early pregnancy failure and the prognosis of fetal viability. Br J Obstet Gynaecol 1995;102:364–9. [65] Ankum WM, Wieringa-de Waard M, Bindels PJE. Management of spontaneous miscarriage in the first trimester: an example of putting informed shared decision making into practice. BMJ 2001;322:1343–6. [66] Sotiriadis A, Makrydimas G, Papatheodorou S, et al. Expectant, medical, or surgical management of first-trimester miscarriage: a meta-analysis. Am J Obstet Gynecol 2005;105: 1104–13.
748
FERENTZ & NESBITT
[67] Demetroulis C, Saridogan E, Kunde D, et al. A prospective randomized control trial comparing medical and surgical treatment for early pregnancy failure. Hum Reprod 2001;16: 365–9. [68] Nielsen S, Hahlin M, Platz-Christensen J. Randomised trial comparing expectant with medical management for first trimester miscarriages. Br J Obstet Gynaecol 1999;106:804–7. [69] Nielsen JH, Hahlin M. Expectant management of first-trimester spontaneous abortion. Lancet 1995;345:84–6. [70] Chipchase J, James D. Randomised trial of expectant versus surgical management of spontaneous miscarriage. Br J Obstet Gynaecol 1997;104:840–1. [71] de Jonge ET, Makin JD, Madefeldt E, et al. Randomised clinical trial of medical evacuation and surgical curettage for incomplete miscarriage. BMJ 1995;311:662. [72] Chung TK, Lee DT, Cheung LP, et al. Spontaneous abortion: a randomized, controlled trial comparing surgical evacuation with conservative management using misoprostol. Fertil Steril 1997;71:1054–9. [73] Johnson N, Priestnall M, Marsay T, et al. A randomised trial evaluating pain and bleeding after first-trimester miscarriage treated surgically or medically. Eur J Obstet Gynecol Reprod Biol 1997;72:213–5. [74] Hinshaw HKS. Medical management of miscarriage. In: Grudzinskas JG, O’Brien PMS, editors. Problems in early pregnancy. London: RCOG Press; 1997. p. 284–95. [75] Jabara S, Barnhart K. Is Rh immune globulin needed in early first-trimester abortion? A review. Am J Obstet Gynecol 2003;188:623–7. [76] Baron F, Hill WC. Placenta previa: placenta abruption. Clin Obstet Gynecol 1998;41: 527–32. [77] Clark S, Koonings P, Phelan J. Placenta previa/accreta and prior cesarean section. Obstet Gynecol 1985;66:89–92. [78] Cunningham FG, Leveno KJ. Pregnancy after 35. Williams obstetrics, 18th editor. Norwalk (CT): Appleton & Lange; 1989. [79] Iyasu S, Saftlas A, Rowley D, et al. The epidemiology of placenta previa in the United States. Am J Obstet Gynecol 1993;168:1424–9. [80] Handler AS, Mason ED, Rosenberg DL, et al. The relationship between exposure during pregnancy to cigarette smoking and cocaine use and placenta previa. Am J Obstet Gynecol 1994;170:884–9. [81] Williams MA, Mittendorf R, Lieberman E, et al. Cigarette smoking in pregnancy in relation to placenta previa. Am J Obstet Gynecol 1991;165:28–32. [82] Bowie JD, Rochester D, Cadkin AV, et al. Accuracy of placenta localization by ultrasound. Radiology 1978;128:177–80. [83] Chamberlain G, Steer P. ABC of labour care: obstetric emergencies. BMJ 1999;318:1342–5. [84] Neilson JP. Interventions for suspected placenta previa. Cochrane Database Syst Rev 2005;3. [85] Rizos N, Doran T, Miskin M, et al. Natural history of placenta previa ascertained by diagnostic ultrasound. Am J Obstet Gynecol 1979;133:287–91. [86] Zelop C, Bromley B, Frigeletto FJ, et al. Second trimester sonographically diagnosed placenta previa: prediction of persistent previa at birth. Int J Gynecol Obstet 1994;44:207–10. [87] Watson P, Cefalo R. Magnesium sulfate tocolysis in selected patients wih symptomatic placenta previa. Am J Perinatol 1990;7:251–3. [88] Wing DA, Paul RH, Millar LK. Management of the symptomatic placenta previa: a randomized, controlled trial of inpatient versus outpatient expectant management. Am J Obstet Gynecol 1996;175:806–11. [89] Rosen DM, Peek MJ. Do women with placenta previa without antepartum haemorrhage require hospitalization? Aust N Z J Obstet Gynaecol 1994;34:130–4. [90] Hladky K, Yankowitz J, Hansen W. Placental abruption. Obstet Gynecol Surv 2002;57: 299–305.
COMMON PROBLEMS AND EMERGENCIES IN THE OBSTETRIC PATIENT
749
[91] Ananth CV, Smulian JC, Dimissie K, et al. Placental abruption among singleton and twin births in the US: risk factors. Am J Epidemiol 2001;153:771–8. [92] Raymond E, Mills J. Placental abruption: maternal risk factors and associated fetal conditions. Acta Obstet Gynecol Scand 1993;72:633–9. [93] ACOG Committee on Obstetric Practice. ACOG practice bulletin. diagnosis and management of preeclampsia and eclampsia. No 33, January 2002. Am College of Obstetricians and Gynecologists. Obstet Gynecol 2002;99:159–67. [94] Walker JJ. Pre-eclampsia. Lancet 2000;356:1260–5. [95] James PR, Nelson-Piercy C. Management of hypertension before, during, and after pregnancy. Heart 2004;90:1499–504. [96] Sibai BM. Treatment of hypertension in pregnancy. N Engl J Med 1996;335:257–65. [97] Hannaford P, Ferry S, Hirsh S. Cardiovascular sequelae of toxemia of pregnancy. Heart 1997;77:154–8. [98] Wagner LK. Diagnosis and management of preeclampsia. Am Fam Physician 2004;70: 2317–24. [99] Report of the National High Blood Pressure Education Program Working Group on High Blood Pressure Pregnancy. Am J Obstet Gynecol 2000;183:S1–22. [100] Samadi AR, Mayberry RM, Zaidi AA, et al. Maternal hypertension and associated pregnancy complications among African-American and other women in the United States. Obstet Gynecol 1996;87:557–63. [101] Garovic VD. Hypertension in pregnancy: diagnosis and treatment. Mayo Clin Proc 2000; 75:1071–6. [102] Neithardt A, Dooley SL, Borensztajn J. Prediction of 24-hour protein excretion in pregnancy with a single voided urine protein-to-creatinine ratio. Am J Obstet Gynecol 2002; 186:883–6. [103] Sibai BM, Ramadan MK, Chari RS, et al. Pregnancies complicated by HELLP syndrome: subsequent pregnancy outcome and long-term prognosis. Am J Obstet Gynecol 1995;172: 125–9. [104] Jacobson SL, Imhof R, Manning N, et al. The value of doppler assessment of the uteroplacental circulation in predicting preeclampsia or intrauterine growth retardation. Am J Obstet Gynecol 1990;162:110–4. [105] Sibai BM. Diagnosis, prevention, and management of eclampsia. Obstet Gynecol 2005;105: 402–10. [106] Witlin AG, Sibai BM. Magnesium sulfate therapy in preeclampsia and eclampsia. Obstet Gynecol 1998;92:883–9. [107] Lopez-Llera M. Main clinical types and subtypes of eclampsia. Am J Obstet Gynecol 1992; 166:4–9. [108] Chanes MC, Livingston JC, Invester TS, et al. Late postpartum eclampsia: a preventable disease? Am J Obstet Gynecol 2002;186:1174–7. [109] Cunningham FG, Grant NF, Leveno KF, et al. Pulmonary disorders. In: Williams obstetrics. 21st edition. New York: McGraw-Hill; 2001. [110] Mabie WC, Barton JR, Wasserstrum N, et al. Clinical observations in asthma in pregnancy. J Matern Fetal Med 1999;1:45. [111] Schatz M. Asthma and pregnancy. Lancet 1999;353:1202. [112] Hankins GDV, Berryman GK, Scott RT, et al. Maternal arterial desaturation with 15methyl prostaglandin f2 alpha for uterine atony. Obstet Gynecol 1998;72:367. [113] Cram LF, Zapata MI, Toy EC, et al. Genitourinary infections and their association with preterm labor. Am Fam Physician 2002;65:241–8. [114] Lettieri L, Vintzileos AM, Rodis JF, et al. Does ‘‘idiopathic’’ preterm labor resulting in preterm birth exist? Am J Obstet Gynecol 1993;168:1480–5. [115] Hillier SL, Nugent RP, Eschenbach DA, et al. Association between bacterial vaginosis and preterm delivery of a low-birth weight infant. The Vaginal Infections and Prematurity Study Group. N Engl J Med 1995;333:1737–42.
750
FERENTZ & NESBITT
[116] ACOG Committee Opinion. Bacterial vaginosis screening for prevention of preterm delivery. Number 198dFebruary 1998. Committee on Obstetric Practice. American College of Obstetricians and Gynecologists. Int J Gynaecol Obstet 1998;61:311–2. [117] Owen MK, Clenney TL. Management of vaginitis. Am Fam Physician 2004;70:2125–32, 2139–40. [118] Glantz JC. Screening and treatment of bacterial vaginosis during pregnancy: a model for determining benefit. Am J Perinatol 1997;14:487–90. [119] Amsel R, Totten PA, Spiegel CA, et al. Nonspecific vaginitis. Diagnostic criteria and microbial and epidemiologic associations. Am J Med 1983;74:14–22. [120] Centers for Disease Control and Prevention. 1998 Guidelines for treatment of sexually transmitted diseases. MMWR Morb Mortal Wkly Rep 1998;47:1–115. [121] Joesoef MR, Schmid GP, Hiller SL. Bacterial vaginosis: review of treatment options and potential clinical indications for therapy. Clin Infect Dis 1999;28:S57–65. [122] Sexually Transmitted Diseases Treatment Guidelines 2002. Centers for Disease Control and Preventive MMWR Recomm Rep 2000;51:1–80. [123] Sobel JD, Brooker D, Stein GE, et al. Single oral dose fluconazole compared with conventional clotrimazole topical therpay of candida vaginitis. Fluconazole Vaginitis Study Group. Am J Obstet Gynecol 1995;172:1263–8. [124] Gibbs RS, Romero R, Hillier SL, et al. A review of premature birth and subclinical infection. Am J Obstet Gynecol 1992;166:1515–28. [125] Krieger JN, Tam MR, Stevens CE, et al. Diagnosis of trichomoniasis. comparison of conventional wet-mount examination with cytologic studies, cultures, and monoclonal antibody staining of direct specimens. JAMA 1988;259:1223–7. [126] ACOG Technical Bulletin. Vaginitis. Number 226dJuly 1996. Committee on Technical Bulletins of the American College of Obstetricians and Gynecologists. Int Gynaecol Obstet 1996;54:293–302. [127] ACOG Technical Bulletin. Gonorrhea and chlamydial infections. Number 190dMarch 1994. Int J Gynaecol Obstet 1994;45:169–74. [128] Kinningham RB. Asymptomatic bacteriuria in pregnancy. Am Fam Physician 1993;47: 1232–8. [129] Beckmann CR, Ling FW, Barzansky BM, et al. Medical and surgical conditions of pregnancy. In: Obstetrics and Ggnecology. 2nd edition. Baltimore (MD): Williams & Wilkins, 1995. p. 82–3. [130] ACOG Educational Bulletin. Antimicrobial therapy for obstetric patients. Number 245d March 1998. American College of Obstetricians and Gynecologists. Int J Gynaecol Obstet 1998;61:299–308. [131] Patterson TF, Andriole VT. Detection, significance, and therapy of bacteriuria in pregnancy. Update in the managed health care era. Infect Dis Clin North Am 1997;11:593–608. [132] Quinlan JD, Hill DA. Nausea and vomiting of pregnancy. Am Fam Physician 2003;68: 121–8.
Prim Care Clin Office Pract 33 (2006) 751–777
Common Urgent Musculoskeletal Injuries in Primary Care Yvette L. Rooks, MD*, Brian Corwell, MD Department of Family Medicine, University of Maryland Medical Center, 29 South Paca Street, lower level, Baltimore, MD 21201, USA
Approximately 25% of patients seen in the primary care office complain of problems related to the musculoskeletal system. General knowledge of the musculoskeletal system is important for timely diagnosis and comprehensive treatment of patients, either by a primary care provider or in conjunction with an orthopedist. Accurate diagnosis, not definitive treatment, is the primary role of the primary care physician in the care of musculoskeletal injuries. Although musculoskeletal injuries are rarely life-threatening, they may be associated with other injuries that are. Injuries to muscle, bones, and joints may cause permanent disability that can, however, be minimized with timely appropriate care. This article is designed to provide primary care providers with a review of urgent treatment of specific common musculoskeletal injuries.
Evaluation of the patient who has acute shoulder pain Shoulder pain is a common complaint in both primary and emergency care. The pain may be traumatic or insidious, chronic or disabling. Patients who have true acute or traumatic pain are more likely to come to the emergency department than to the primary care office, whereas those whose pain occurs after minor falls and athletics mishaps, including fractures, dislocations, and soft-tissue injuries, may come to any medical facility. Acute pain originating within the shoulder usually appears suddenly after some traumatic event. Determining when and how the injury occurred is very important, as is determining the precise location, intensity, and pattern of radiation. The initial step in evaluating is to obtain a good history. Questions
* Corresponding author. E-mail address:
[email protected] (Y.L. Rooks). 0095-4543/06/$ - see front matter Ó 2006 Elsevier Inc. All rights reserved. doi:10.1016/j.pop.2006.06.009 primarycare.theclinics.com
752
ROOKS & CORWELL
should first be general. The second group of questions should be directed toward specific clinical entities suggested by the patient’s responses to the general questions. The clinical examination should include visual inspection for deformity, ecchymosis, atrophy, and asymmetry. Active and passive range of motion of the shoulder girdle should be assessed, and provocative tests that will provide a more focused evaluation should be used as well. A neurovascular evaluation of the upper extremity is essential to a thorough shoulder examination. To properly diagnose and treat shoulder pain, and to know when to refer for specialty consultation, the clinician must understand and be familiar with the functional anatomy of the shoulder, the common mechanisms of injury, the appropriate radiological studies, and available treatments. The shoulder is composed of the humerus, glenoid, scapula, acromion, clavicle, and surrounding soft-tissue structures. The shoulder region includes the glenohumeral joint, the acromioclavicular joint, the glenoclavicular joint, and the scapulothorasic articulation. Glenohumeral stability is the result of a combination of ligamentous and capsular constraints, surrounding musculature, and the glenoid labrum. Static joint stability is provided by the joint surfaces and the capsulolabral complex, and dymnamic stability by the rotator cuff muscles and scapular rotators [1]. Acromioclavicular joint injury A common injury among athletes and active patients is acromioclavicular (AC) sprain, also referred to as ‘‘shoulder separation.’’ AC joint injuries account for 12% of all dislocations of the shoulder girdle, and are more common in men than in women [2]. They are usually associated with contact sports, motor vehicle accidents, and falls. Acromioclavicular joint injury is associated with subsequent local pain. Tenderness, swelling, and often a deformity with prominence of the distal clavicle are seen. AC joint injuries are graded in severity according to the extent of ligamentous injury. Grade I represents an AC ligament sprain; Grade II, a complete ligament rupture, with a widened joint space. In a Grade III ligamentous injury, the AC and coracoid (CC) ligaments and their muscle attachments are totally disrupted, and the joint space is significantly widened. Grade IV injuries include the previous injuries, and, in addition, the clavicle is displaced posteriorily into or through the trapezius. In a Grade V injury, the distal clavicle is displaced superiorly, and in a Grade VI injury, the distal clavicle is displaced inferiorly. On examination, the AC joint is tender to palpation and painful on attempted shoulder motion. AC radiographs help determine the severity of the injury. Radiographs are likely to be normal with a Grade I injury. For appropriate management, one should evaluate for associated injuries, including sternoclavicular (SC) dislocation, and clavicle or coracoid fractures. All patients should be placed in a sling, both for comfort and to avoid further disruption of the injured joint.
COMMON URGENT MUSCULOSKELETAL INJURIES
753
Patients who have Grade I and II injuries can be managed with rest, ice, and analgesics for 1 to 3 weeks. The patient may begin active range of motion of the arm when pain-free. Patients who have a Grade III sprain or higher should be referred to an orthopedist for consultation and possible surgical repair. Clavicle fracture Clavicle fractures account for 5% of all types of fractures, and are the most common type among children. Most clavicle fractures occur through the middle third of the clavicle. An indirect or direct blow to the shoulder causes fracture of the clavicle. Because the clavicle is directly beneath the skin, these fractures are relatively easy to diagnose. A radiograph should be taken to determine whether the patient has a displaced fracture of the proximal or distal clavicle, which may require a swifter referral to orthopedic care. A complete neurovascular examination of the upper extremities needs to be performed to rule out associated brachial plexus or vascular injury. The figure-of-eight bandage is the traditional method of treatment for fractures that occur through the middle third of the clavicle, but a simple sling is sufficient and can be an effective alternative. The primary intent is to reduce motion at the fracture site, in order to reduce the patient’s discomfort. Patients should wear the sling until a series of films indicate evidence of callus formation. Acute shoulder dislocations Glenohumeral dislocation is the most common shoulder dislocation, representing 95% of all shoulder injuries [2]. Acute disorders of the glenohumeral joint are usually the result of indirect forces applied to the joint. The shoulder can dislocate either anteriorly, posteriorly, or inferiorly. Anterior dislocations compose approximately 95% of such injuries, posterior and inferior dislocations making up the remaining 5%. In acute anterior dislocations the humeral head is forced out of the glenoid anteriorly and comes to rest beneath the coracoid process, clavicle, or glenoid. The humeral head usually is palpable anteriorly, and the diagnosis is often confirmed by locating a dimple in the skin beneath the acromion. Posterior dislocation of the shoulder joint is rare, and is frequently missed during initial evaluation. Typically, the patient holds the arm close to the body in abduction and internal rotation. Forward elevation is extremely limited. The physician should always look for possible related injuries, including proximal humeral fractures, avulsion of the rotator cuff, and injuries to the adjacent neurovascular structures. The axillary nerve is the most likely to be involved. The diagnosis of glenohumeral dislocation is confirmed with at least two radiographic views of the affected shoulder. A true anteroposterior view usually reveals an anterior dislocation. An axillary lateral view
754
ROOKS & CORWELL
provides a more accurate picture of anteroposterior position of the humeral head, and can exclude a posterior dislocation. The shoulder dislocation should be reduced as soon as possible. Many maneuvers have been described, and the clinician should use the method most familiar. Postreduction radiographs can be obtained to confirm realignment. After any reduction procedure, the practitioner should perform a second, well-documented neurovascular examination. Immediate orthopedic consultation is required if reduction is not achieved and for any dislocation associated with fracture. In all cases, the shoulder should be immobilized and appropriate analgesia administered. Orthopedic referral should be obtained even for uncomplicated dislocations. Associated injuries, such as a ligamentous injury, may not be immediately obvious and are difficult to recognize in the acute evaluation. These potential injuries should be re-evaluated at time of follow-up.
Elbow and forearm injuries Elbow dislocation Swelling, deformity, and limited elbow motion suggest significant elbow injury. After the shoulder, the elbow is the second most frequently dislocated major joint [3]. Elbow dislocations are usually the result of a fall on an outstretched hand with the elbow extended. Most are simple dislocations with no concomitant fractures. Complex dislocations are less common, and involve major fractures that require emergency orthopedic consultation. Initial assessment of the dislocated elbow is critical, because it can affect the final outcome. Possible clinical findings include swelling, pain, loss of range of motion, and obvious deformity. Evaluation of other injuries to the ipsilateral upper extremities is necessary. Neurologic and vascular damage with displaced elbow injuries may occur, after which compartment syndrome of the forearm may develop; therefore it is mandatory to assess motor and sensory function of the neurovascular structures of the elbow and forearm. After the assessment, the practitioner must decide whether to reduce the elbow or to splint the joint and send the patient to the emergency department or a consulting orthopedist. This decision should be based on the physician’s experience and comfort with reduction. Radiographs should be examined for occult fractures. Closed reduction of the noncomplicated elbow dislocation is usually possible, but this procedure becomes more difficult if delayed. If the primary care provider attempts reduction, confirmation that the elbow joint has regained full range of motion and stability and a repeat neurovascular examination are essential. Post-reduction radiographs are necessary to confirm reduction. A long arm posterior splint is used to immobilize the elbow at 90 of flexion. Close orthopedic followup is necessary. Any complicated dislocation involving fracture, neurovascular compromise, or open dislocation should be referred immediately.
COMMON URGENT MUSCULOSKELETAL INJURIES
755
Forearm fractures Fractures and dislocations occurring near the elbow and forearm are common in adults and children. These are severe and complex injuries, and frequently require orthopedic referral. These injuries have the potential for deforming complications. In the primary care setting, the initial diagnosis and treatment of these injuries is key to a gratifying recovery and avoiding permanent disability. Most forearm fractures are associated with a history of a fall on an outstretched arm or a direct blow to the forearm. Radial and ulnar fractures cause swelling, tenderness, and deformity of the forearm. Open fractures of this type are common. Isolated ulna or radius fractures cause localized swelling or tenderness over the fracture site. Less deformity is noticed with an isolated radial or ulnar fracture. In evaluating for a fracture, the physician should assess the function of the radial, median, and ulnar nerves, and the extensor and flexor tendons of the forearm. Radiographs of the forearm are needed to confirm the diagnosis. Additional radiographs of the elbow and wrist joints are warranted to rule out any additional injury. Subluxation of the head of the radius Radial head subluxation, or nursemaid’s elbow, is a very common injury in young children. It generally results from a sudden pull on the upper limb, such as that exerted by an adult to prevent a child from falling. The radial head is traumatically subluxed with forceful traction on the hand, with the elbow extended and the forearm pronated. The annular ligament either tears or slips over the radial head, allowing for radial head subluxation. With the release of traction on the arm, the ligament remains interposed between the radial head and the capitellum. The child’s injured elbow will usually be partially flexed and the forearm pronated and supported close to the trunk of the body. There will be anterolateral tenderness over the radial head. Radiographs are typically normal. The physician treats this condition by holding the elbow in one hand with the thumb overlying the head of the radius and slowly flexes the elbowdwhile the forearm is rotated into full supination. The physician may hear a click, signifying reduction, and the child will be quickly comfortable. The child should not be immobilized or restricted in any way. Distal radial fracture A Colles’ fracture, the most common distal radius fracture, is a closed fracture of the distal radial metaphysis. Colles’ fractures are common in adults and rare in children, who tend to fracture through the distal radial physis. The cause of injury is usually a fall on the outstretched hand. Examination of the forearm reveals the classic ‘‘dinner fork’’ deformity of the wrist, which is produced by the dorsal displacement of the fracture. The
756
ROOKS & CORWELL
examiner should look for associated injuries, including fracture of the ulnar styloid and median nerve. Nondisplaced fractures can be immobilized in a long arm cast or sugar tong splint and referred to an orthopedist for follow-up. Displaced fractures require prompt referral for reduction, manipulation, and immobilization. Wrist injuries Injuries to the wrist are often misdiagnosed as a sprain. Although wrist sprains are common, that diagnosis should be considered only after careful physical and radiographic examinations have ruled out fracture and dislocation in this anatomic region. An accurate diagnosis of wrist injuries presupposes thorough knowledge of the topographical anatomy of the wrist and a careful, systematic evaluation of the extremity and appropriate radiographs. Ascertaining the specific point of tenderness within the carpus is the most important diagnostic test in assessing injuries to the wrist. Understanding the surface anatomy of the hand and wrist allows the physician to evaluate common injuries and appreciate less common injuries that might otherwise get overlooked. Fractures of the scaphoid result in point tenderness in the anatomic snuffbox and over the scaphoid tuberosity. Scapholunate and lunate injuries are maximally tender just distal to Lister’s tubercle on the dorsum of the wrist. Hamate hook fractures are diagnosed clinically by eliciting tenderness over the area just distal and radial to the pisiform. Scaphoid fracture Fractures of the scaphoid are the most common fracture of the carpal bones [4]. They are also the most commonly missed of all wrist injuries. The fracture is caused by a fall on the outstretched, dorsiflexed hand and wrist. Patients who have pain over the anatomical snuffbox should be treated for a possible scaphoid fracture, and a scaphoid view radiograph should be obtained to assist in the diagnosis. If the patient has scaphoid tenderness without radiographic evidence of a fracture, the wrist should be immobilized in a thumb spica splint/cast and have follow-up radiographs in 10 to 14 days. Orthopedic follow-up is prudent in all cases because: (1) vascular supply is through the bone’s distal portion, and missed diagnosis of a midscaphoid fracture can result in avascular necrosis of the proximal fragment; (2) nonunion is frequent following scaphoid fracture; and (3) there may be related injuries to the forearm, wrist, or hand. Fractures of the hook of the hamate A fractured hook of the hamate is a less common wrist injury that often is not diagnosed, because it is not apparent on standard radiographic views of the wrist. The injury occurs when a patient falls while holding an object, and the object lands between the ground and the ulnar side of the palm, or when
COMMON URGENT MUSCULOSKELETAL INJURIES
757
there is a direct blow to the hypothenar eminence. A delay in diagnosis usually leads to nonunion. Patients will generally have full range of motion and minimal swelling, but will have a decreased grip as compared with the opposite side. The definitive physical diagnostic method is deep palpation to identify the point of maximal tenderness. In addition to standard radiographs of the wrist, a carpel tunnel view should be obtained. The ulnar nerve should also be evaluated during the examination. Patients should be splinted in an ulnar gutter or volar splint and referred to an orthopedist for definitive treatment. Carpal dislocations Scapholunate dislocation is commonly missed in the initial examination. It is the most common form of carpal instability. The mechanism of injury is hyperextension. Patients will complain of pain localized to the scapholunate joint and weakness, especially while gripping. Radiographs demonstrate a space between the scaphoid and lunate that is wider than 3 mm. In the physician’s office, a thumb spica splint is applied, and the patient is referred to an orthopedist for surgical repair. Lunate and perilunate dislocations are the most common carpal bone dislocations. Patients who fall on the heel of the hand are particularly susceptible to these injuries. Common symptoms include wrist deformity and limitation of motion. The deformity may be subtle on an anterior-posterior radiograph. The key to making the diagnosis is a true lateral radiaograph, in which close attention must be paid to the relationship between the radius, lunate, and capitate. In a lunate dislocation, the capitate is aligned with the radius and the lunate dislocates volar to the radius. In a perilunate dislocation, the lunate remains aligned with the radius and the capitate dislocates dorsally relative to the lunate. Perilunate dislocations may occur with or without an associated scaphoid fracture. Patients who have both types of dislocations need to be promptly referred to an orthopedist. Hand injuries Proper evaluation of hand injuries requires simultaneous consideration of the structure and function of all components of the hand. The close structural proximity of nerves, arteries, and bones increases the chances of associated lesions and skeletal injuries of the hand. Treatment cannot be initiated without regard for maintenance of sensibility and mobility of the hand, and the integrity of the skin overlying the injury [5]. Tendon injuries Tendons can be broadly divided into two categoriesdflexor and extensor. Flexor tendon injuries cause less impairment of hand function than extensor tendon injuries. This is mainly because of the redundancy of the hand
758
ROOKS & CORWELL
flexor tendons, whereas only one extensor tendon exists for the second through fifth fingers [6]. The challenge for the primary care provider is to recognize the injury and make the proper diagnosis. Initially, the examiner needs to observe the resting posture of the hand. Any change in the normal flexion cascade may be a sign of a tendon injury. Secondly, if a laceration is present, the physician needs to examine the wound with the hand/finger in the position it was in at the time of the injury. Finally, the wound/tendon needs to be examined as the hand/finger is taken through its full range of motion. Flexor tendon injuries Although closed traumatic disruption can occur, these types of tendon injuries most often occur with lacerations. It is imperative to test the function of these tendons throughout their range of motion by evaluating the strength of these tendons against resistance; partial tears will cause weakness against resistance. Many flexor tendon injuries are treated with protective splinting alone. Only an orthopedist or hand specialist should perform surgical repair of flexor tendon injuries. Extensor tendon injuries The mallet finger is a common avulsion injury of the terminal extensor tendon of the distal interphalangeal joint (DIP). This injury occurs when there is an acute, forceful passive flexion of the DIP joint while the joint is in active extension. On examination there is no active extension of the joint. Radiographic examination is important to rule out bony avulsion. Treatments range from simple splinting of the joint in slight hyperextension for 6 to 8 weeks to open reduction and internal fixation. Boutonnie´re deformity results from an injury at the dorsal surface of the proximal interphalangeal (PIP) joint. A sharp force against the tip of a partially extended finger will result in hyperflexion of the middle joint. On examination the clinician should evaluate for point tenderness about the base of the middle phalanx, and for diminished extensor tendon strength with increased pain when the middle joint is extended against resistance. Unfortunately, this injury is often missed and diagnosed as a ‘‘jammed’’ or ‘‘sprained’’ finger. Radiographs should be obtained to evaluate for avulsion fracture. Treatment involves splinting the joint in complete extension. The patient should then be referred to an orthopedic/hand surgeon for possible operative repair. PIP joint dislocations are the most common ligamentous injury of the hand [7]. Fracture-dislocations of the PIP joint occur frequently and are potentially disabling. Hyperextension is the most common cause. The usual deformity is dorsal displacement of the middle phalanx on the proximal phalanx. After a careful history and examination for swelling, neurovascular
COMMON URGENT MUSCULOSKELETAL INJURIES
759
status, and range of motion, a true lateral and posteroanterior radiograph should be obtained. The radiograph will demonstrate any joint displacement, angulation, and incongruity. Dorsal and lateral reduced injuries should be splinted at 30 of flexion following reduction and referred. The definitive treatment of PIP joint injuries depends on the type of injury. Hand fractures General principles of orthopedic management apply to fractures of the hand. Most closed phalanx fractures can be stabilized with splinting and managed on an outpatient basis. Open, intra-articular, or unstable phalanx fractures require more immediate consultation with a specialist. Metacarpal fractures of the head, neck, shaft, or base should be referred promptly for emergency room or orthopedic evaluation because of the difficulty with reduction, rotational deformity, and angulation that is common in these types of fractures. Phalangeal fractures Proximal phalangeal injuries can result in significant functional disability if not recognized and treated appropriately. Injuries to the fingers are usually a result of direct trauma to the digit. After a detailed history and physical examination, the possible presence of a rotational deformity must be carefully considered. Often the oblique radiograph does not show the true nature of the fracture, and a true lateral radiograph is needed to show angulation of the fracture. Stable, nondisplaced middle and proximal phalangeal fractures without angulation or rotational deformity can be managed with dynamic splinting to the adjacent finger. Follow-up radiographs should be obtained in 7 to 14 days to assess angulation and healing. Unstable fractures of the proximal and middle phalanges and displaced intra-articular fractures generally require operative reduction and fixation. Metacarpal fractures Metacarpal neck fractures are very common, particularly neck fractures of the fourth and fifth metacarpals (boxer’s fracture). They are usually the result of a direct blow or crush injury to the metacarpal. Radiographs demonstrate the proximal fragment angulating in the dorsal direction, whereas the distal fragment angulates in the volar direction. The amount of angulation that is acceptable varies directly with the normal physiologic mobility of the involved metacarpal. Rotational deformities, if present, must be completely corrected. Having the patient slowly make a fist with the palm facing upward can help assess for rotational deformities. Each digit should point toward the scaphoid tuberosity and should not overlap. Reduction of these types of fractures is usually beyond the scope of most primary care physicians. The patient should be splinted in a volar or gutter splint, depending
760
ROOKS & CORWELL
on the injured metacarpal, and referred to an orthopedic surgeon. Most of these metacarpal neck fractures can be treated with closed reduction and casting.
Acute lower back pain Acute low back pain is one of the most common symptom-related complaints for visits to primary care physicians. A majority of these patients first present to a primary care physician, not a subspecialist. Though back pain has a benign course in over 90% of patients, the primary care physician must be vigilant and comfortable identifying the few patients with warning signs of a neurologically impairing or life-threatening etiology. The goal of the office assessment of patients with acute lower back pain is to evaluate for potentially dangerous etiologies first, which, if not promptly recognized, could result in significant morbidity and mortality. The approach recommended by these authors and others is to obtain from all patients who have lower back pain a systematic history, and to perform a careful physical examination and rely on the presence of so-called ‘‘red flags’’ or ‘‘alarm symptoms’’ to guide further diagnostic tests, specialty evaluation, and treatment. A summary of the red flag signs and symptoms can be found in Box 1. A careful and comprehensive history and physical examination are used to
Box 1. History and physical examination red flags Signs and symptoms concerning for infection or malignancy Age under 18 or over 50 Pain lasting for more than 6 weeks History of cancer Fever, chills, night sweats, weight loss Unremitting pain, night pain (may awaken patient from sleep) Intravenous drug users, immunocompromised Fever Signs and symptoms concerning for epidural compression Bowel or bladder incontinence Saddle anesthesia Decreased or loss of anal sphincter tone Severe or progressive neurologic defect Motor weakness Signs and symptoms concerning for fracture Major trauma Minor trauma in the elderly
COMMON URGENT MUSCULOSKELETAL INJURIES
761
identify the small percentage of patients who have serious pathology as the underlying cause of their back pain. This approach can also be used to differentiate the most common entity, benign nonspecific back pain, from syndromes that require more immediate identification and evaluation: (1) nerve root compression, which refers to pain with accompanying sensory and motor deficits in the distribution of a lumbosacral nerve root; and (2) epidural compression syndrome, which consists of cauda equina syndrome, the closely related conus medullaris syndrome, and spinal cord compression. These disorders are grouped because they share a similar presentation and initial evaluation. Different causes of acute low back pain tend to have different distinguishing characteristics. For example, peripheral nerve pain may be described as ‘‘pins and needles’’ or ‘‘burning,’’ as opposed to nerve root pain, which is transient and very sharp, relieved with recumbent positioning and exacerbated by Valsalva-type maneuvers. Discogenic pain is typically worse with flexion, whereas pain from spondylolisthesis is aggravated by facet loading, which occurs in extension. Spinal stenosis is characterized by lower extremity radicular pain, which also becomes worse with extension (standing) and improves with flexion (sitting). Typical nonspecific back pain is unilateral. It may radiate to the buttocks or posterior thigh but not past the knee. Pain is increased with movement, improved with rest, and there are no complaints of numbness, weakness, or bowel or bladder dysfunction. Sciatica is sharp and burning, and radiates posteriorly down the leg past the knee. There may also be associated numbness or weakness. Epidural compression syndrome is associated with numbness, weakness, bilateral leg pain, incontinence, and saddle anesthesia. The importance of the clinical history cannot be overemphasized. As with any patient who complains of pain, symptoms should be characterized by the basic historical elements of the episode, such as the onset, character, severity, location, exacerbating, and alleviating factors, and the presence of radiation. Further questions, related to the aforementioned red flags, must also be asked to identify high-risk patients. Physicians must be especially cautious if the patient is under age 18 or over age 50, because tumors and infection appear with higher frequency in these age groups. In the older patient, one must always consider an abdominal aortic aneurysm (AAA) as a potential etiology of back pain. The presence of hematuria can make this entity appear as the classic kidney stone. The elderly may sustain fractures, including pathologic fractures, with relatively minor trauma. In other patients, any history of moderate to major trauma should increase suspicion of fracture. The immunocompromised and intravenous drug users are at increased risk of spinal bacterial infections. In fact, any patient who has intravenous drug use and back pain should be assumed to have an abscess or osteomyelitis until proven otherwise. Those who have a past medical history of cancer, especially cancers that are known to metastasize to the spine, are also at high risk. Most episodes of lower back pain will resolve within 4 to
762
ROOKS & CORWELL
6 weeks. Lack of significant improvement in 6 to 8 weeks, or pain of greater duration are other red flags [8,9]. It should then be determined whether the patient is primarily describing back or leg symptoms. Radicular pain in a dermatomal distribution, often into the lower hamstring, knee, and foot, indicates nerve root compression or irritation (sciatica), whereas the more common nonspecific back pain without dermatomal radiculopathy implies muscle or ligamentous strain. The clinician should attempt to elicit historical features that are suggestive of infection or malignancy, such as unintentional weight loss, fevers, chills, dysuria, and night sweats. These screening questions are very important, because a majority of patients whose back pain is found to be caused by spinal malignancy have no known history of cancer. Because most benign back pain is tolerable, worsened with activity, and improved with rest, symptoms such as night pain and severe unrelenting pain that is not relieved by rest and recumbency should also raise a red flag. Symptoms that are suggestive of epidural compression syndrome are mainly neurologic in origin, and include any loss of bowel or bladder function, urinary retention, saddle anesthesia, or distal leg numbness or weakness. Pain that is worsened by coughing, sneezing, prolonged sitting and standing, or Valsalva maneuvers raises suspicion of disc herniation. The purpose of the physical examination is to evaluate neurologic complaints discovered in the history, to identify potential additional neurologic defects, and to continue to uncover any red flags. The neurologic examination systematically tests the reflexes and the motor and sensory components of the most commonly affected nerve roots. The clinician must account for abnormal or unstable vital signs, which may indicate an extraspinal cause of the back pain such as an AAA. The presence of fever (infection) and focal vertebral tenderness to percussion (fracture, infection) are important findings. Digital rectal examination should be performed in anyone in whom epidural compression syndrome is being considered (severe pain, hard neurologic findings) to assess sphincter tone and perianal sensation, to check for masses, and for a possible perirectal abscess. The straight leg raise (SLR), which tests for the presence of a herniated disc causing nerve root compression, is one of the most important tests for evaluating back pain (Appendix). Pain referred to the affected leg when the opposite leg is tested, called a positive crossed straight leg raise, is highly indicative of nerve root irritation and has a very high specificity. Reproduction of pain in the back, hamstring, or buttock region does not constitute a positive test. The clinician must also be able to distinguish between nonorganic back pain and true back pathology. Waddell’s signs are physical examination findings that can aid in making this important distinction [10]. Superficial, nonanatomic, or variable tenderness and gross overreaction during the physical examination suggest a nonorganic cause. The clinician may also simulate back pain through provocative maneuvers such as axial loading of the head or passive rotation of the shoulders and
COMMON URGENT MUSCULOSKELETAL INJURIES
763
pelvis. Neither maneuver should elicit low back pain. There may be a discrepancy between the supine and sitting SLR. The seated version of the test, sometimes termed the distracted SLR, can be performed while distracting the patient or appearing to focus on the knee. Further, radicular pain elicited at a leg elevation of less than 30 is suspicious because the nerve root and surrounding dura do not move in the neural foramen until an elevation of more than 30 is reached. Sensory and motor findings suggestive of a nonorganic cause include nondermatomal sensory loss or cogwheel or give-way weakness. Waddell’s signs, especially if three or more are present, correlate with malingering and functional complaints. These signs can be used to evaluate select patients, and are merely a component of a comprehensive physical examination. They should never be used independently because they lack the sensitivity and specificity to rule out true organic pathology. A detailed distal neurologic examination must be performed that targets the three most common locations for disc herniation: L4, L5, and S1. Over 95% of herniated discs affect the L4-L5 or L5-S1 interspaces [11,12]. Those who have pathology at the higher lumbar spine will have hip flexion weakness and anterior thigh sensory changes in the corresponding dermatome. Those with pathology at the lower sacral levels (S2-S5) can have abnormal perianal sensation, anal wink (S2-S4), rectal tone (S2-S5), and bladder function. An understanding of the key physical examination components of the targeted L4-S1 neurologic examination is essential (Table 1). Though unnecessary for office practice, urinary catherization can be helpful in evaluating select patients. Measurement of a postvoid residual volume tests for the presence of urinary retention with overflow incontinence, suggesting compromised neurologic function. This is a very sensitive and specific finding for cauda equina syndrome. Laboratory testing, consisting of a complete blood count (CBC), erythrocyte sedimentation rate (ESR), and urinalysis (UA), is indicated in cases in which infection or tumor is suspected. Blood cultures, prostate-specific antigen (PSA), C-reactive protein (CRP), calcium and alkaline phosphatase may also be considered in appropriate cases. The emergency or specialty physician who will be assuming care of the patient should order the appropriate diagnostic imaging. Plain films (anteroposterior [AP] and lateral) are often the first step in cases of suspected infection, fracture, malignancy, or neurologic compromise. Additional views are only indicated if spondylolysis Table 1 L4-S1 neurologic examination (wf) Neurologic level
Motor
Sensory
Reflex
L4 L5
Knee extension Heel walking Great toe dorsiflexion Toe walking Foot eversion
Anteromedial thigh/knee Great toe web space
Patellar None
Lateral foot
Achilles
S1
764
ROOKS & CORWELL
or spodylolisthesis, problems common in children, are suspected,. A bone scan is useful when looking for spinal stress fractures and spinal metastatic disease. Computed tomography scan is the study of choice for vertebral fractures and other bony disease. CT with myelography may be used in those patients who are unable to have magnetic resonance imaging. With the exception of the acute traumatic evaluation, MRI is the most informative investigative modality. An MRI is the modality of choice for evaluation of spinal infectious lesions, malignancy, herniated discs, and epidural compression syndrome. Disc disease is a very nonspecific finding, and also a component of normal aging. In fact, one in four of all asymptomatic persons younger than 60 will have positive MRI findings of a herniated disc. That number jumps to one in three in people over the age of 60. In one study, over 50% of patients studied were identified as having a disc bulge [13]. Most patients with back pain have the classic nonspecific lumbosacral back pain that may radiate into the buttocks or thigh, and that is worsened with activity and relieved with rest. Examination does not reveal any red flags or neurologic abnormalities. These patients are managed conservatively, which may include ice, heat, prescribed analgesia, or muscle relaxants, as well as close follow-up to ensure that the episode resolves. These types of patients should be told to continue daily activities as tolerated and should not be prescribed bed rest. Sciatica is defined as radicular pain into the leg in the distribution of a lumbar or sacral nerve root, which is often accompanied by a motor or sensory deficit. This is the classic clinical scenario of a herniated disc, which occurs when the tough outer disc layerdthe annulus fibrosisdtears, and the inner gelatinous materialdthe nucleus pulposusdprolapses, inflames, and presses on a nerve root. Patients complain less of back pain and more of lower extremity pain and radicular symptoms, because of the anatomic distribution of the nerve roots involved. Diagnosis consists of localizing the pain and neurologic dysfunction to an isolated nerve root. Multi-nerve– root pathology is a potential indicator of a spinal mass lesion or central disc herniation. A positive straight leg test further supports the diagnosis. Patients who have an otherwise normal examination and no hard neurologic findings do not require an urgent MRI or subspecialty referral. In fact, many patients who have a herniated disc can be managed by their primary care physician without specialty referral. Surgery is unlikely ever to be considered until a patient has failed a period of conservative therapy. Only epidural compression syndromes require emergent spinal decompression surgery. Epidural compression syndrome is characterized by lower back pain, unilateral or bilateral sciatica, lower extremity sensory or motor deficits, decreased or asymmetric deep tendon reflexes, and bowel or bladder dysfunction. Causes are multiple and include midline disc herniation, trauma, or mass effect from an abscess, hematoma, or spine metastases. These lesions are a surgical emergency and require immediate diagnosis and treatment. In
COMMON URGENT MUSCULOSKELETAL INJURIES
765
cauda equina syndrome, urinary retention with overflow incontinence (greater than 50–100 mL) is the most common finding. Sensory loss, commonly in the buttocks or perineal region (saddle anesthesia), is the most common sensory deficit. Anal sphincter tone is often decreased. These patients require immediate specialty evaluation, MRI, and steroid administration to minimize the ongoing neurologic damage from compression and edema. Studies have shown that the ultimate functional outcome is largely dependent on the duration of symptoms and the condition at the time of presentation. Patients who receive rapid surgical decompression have a good prognosis. Spinal infections are an uncommon though emergent cause of back pain. Infection can affect the disc space or the vertebral body, or form an epidural abscess. Infection occurs via hematogenous spread, skin and soft-tissue infection, and following spinal surgery or epidural anesthesia. Spinal abscess of unknown etiology is not uncommon. Predisposing factors for epidural abscess are immunocompromised states such as intravenous drug abuse, diabetes mellitus, chronic renal failure, and alcoholism. Patients who have spinal infections often do not present acutely, because prolonged symptoms are common. Multiple visits to physicians often precede the correct diagnosis. Patients are often misdiagnosed as having a musculoskeletal strain. The key to diagnosis is not overlooking the disease as a possible cause. Outcome is related to the speed of diagnosis, and that diagnosis is made before the development of myelopathic signs. Missed diagnosis carries serious potential morbidity and mortality. Patients present with back pain, fever, chills, weight loss, unrelenting pain, and eventual neurologic deficits. Spinal cord damage occurs because of direct compression, local venous thrombosis, and interruption of the arterial supply. Focal neurologic deficits are late findings. Patients require MRI with contrast and immediate evaluation by a spine surgeon. Another infrequent though emergent etiology of back pain is spinal malignancy. Pain is the initial symptom in a vast majority of these cases. Those who have new or progressive neurologic signs or symptoms (weakness, sensory changes, decreased or absent reflexes, and so forth) should have emergent specialty evaluation and imaging with MRI. Unlike cauda equina syndrome, which requires only a focal MRI of the lumbosacral spine, suspicion of malignancy requires a screening MRI of the entire spine to evaluate for falsely localizing lesions, because there is a 10% risk of distant asymptomatic metastases, which may affect treatment. A metastatic workup, including CT of the chest, abdomen, and pelvis, should be obtained to identify the primary malignancy. Spinal cord compression secondary to neoplasm will also require consultation for possible emergent radiation therapy. Cancers known to cause spinal malignancies include primary cancers, myeloma, and metastatic cancers (breast, lung, thyroid, kidney, prostate). Metastatic disease is far more frequent than primary bone tumors. Unlike adults, children are more likely to have an established diagnosis of their back pain. Etiologies such as infection, including discitis and
766
ROOKS & CORWELL
osteomyelitis, and tumor must be considered. Also common in preadolescent and adolescent children is spondylolysis, a defect in the pars interarticularis often secondary to repeated lumbar stress. It affects young athletes such as football players and gymnasts. Unlike nonspecific muscle pain, pain is exacerbated by spine extension (facet loading) and improved with flexion. The defect of the pars may be seen on oblique radiographs. Spondylolisthesis is the slippage of one vertebra anteriorly in relation to the vertebral body below it, and is best visualized on the lateral spine radiograph. The proper clinic disposition of patients with lower back pain is very important. Most patients will leave the office with a diagnosis of nonspecific back pain, and the problem will resolve spontaneously. Those with sciatica and no acute neurologic deterioration can be managed conservatively. Though the acute pain is often debilitating, pain often abates with nonsteroidal anti-inflammatory drugs (NSAIDs), analgesics, and physical therapy. MRI can be delayed for approximately 4 to 6 weeks. Those who have a herniated disc with symptoms in a single nerve root distribution can be managed conservatively as outpatients and re-evaluated in 1 week. Patients with multiroot or bilateral neurologic findings, or signs and symptoms of malignancy or infection should be sent to an emergency department that is equipped to perform the appropriate imaging study, most likely an MRI, and which has access to specialty consultation. Unless there is concern for acute decompensation en route to the emergency department, patients may be transported from the office via ambulance with basic life support (BLS) trained personnel. Patients who have back pain and a past medical history of cancer are a unique group. Those with isolated back pain without neurologic findings suggestive of cord compression can be closely followed for improvement and lack of progression, and should be re-examined within 5 to 7 days. All back pain patients evaluated in the primary care doctor’s office should be given clear ‘‘discharge’’ instructions, with unambiguous indications to return or go to the nearest emergency department with symptoms such as new or progressive leg weakness, bowel or bladder dysfunction, or saddle anesthesia. When sufficient concern exists for one of the red flag diagnoses, the patient’s workup should be transitioned to the local emergency department. The clinical pitfall to avoid is diagnosing an emergent back pain episode as ‘‘just a back strain.’’ The clinician should check for the presence of red flags in all patients who have back pain. To summarize, the patients who have low back pain emergencies are: (1) those who have a past medical history of malignancy and new back pain with neurologic findings, (2) those who have back pain and symptoms of epidural compression syndrome, (3) those who have back pain with symptoms suggesting an infectious etiology, (4) those who have back pain with gross muscle weakness or paralysis, and (5) those who have back pain and multiple nerve root involvement correlating with the clinical examination.
COMMON URGENT MUSCULOSKELETAL INJURIES
767
Foot and ankle injury Patients with injuries to the foot and ankle are common in both physician’s offices and outpatient clinics. The common clinical question arises as to which patients need plain films of their foot or ankle. Clinical decision guidelines called the Ottawa Ankle Rules were developed to help the physician determine whether to obtain radiographs of the midfoot or ankle (Box 2). The rules were derived from an initial series of studies that were later prospectively validated. These rules have been reported to have a sensitivity of 97% to 100% [14]. All office patients presenting with a recent ankle injury should be sent for imaging if they meet the Ottawa ankle criteria. Ankle sprains are some of the most frequent sports-related orthopedic injuries. Common orthopedic injuries may mask themselves as simple ankle sprains, delaying both accurate diagnosis and appropriate treatment, and at times causing a worse long-term prognosis. An Achilles tendon injury can be easily overlooked unless it is specifically considered. This condition has been misdiagnosed in as many as 25% of cases [15,16]. The Achilles tendon represents the distal confluence of the gastrocnemius and soleus muscles, which insert on the calcaneus. The typical patient is a middle-aged male, poorly conditioned or sedentary, or a ‘‘weekend’’ athlete. The usual mechanism of injury is sudden or unexpected dorsiflexion of the foot or ankle. Disruption usually occurs 2 to 6 cm proximal to the tendon insertion, corresponding to an area of poor vascular supply. A common false assumption is that plantar flexion at the ankle is controlled solely by the Achilles tendon. In fact, many muscles can cause plantar flexion, including the toe
Box 2. Ottawa ankle and foot rules An ankle radiograph series is only required if there is any pain in the malleolar zone and any of these findings: Bone tenderness at the posterior edge of the distal 6 cm or tip of the fibula Bone tenderness at the posterior edge of the distal 6 cm or tip of the tibia Total inability to bear weight, both immediately and in the office (weight-bearing is defined as the ability to take four steps regardless of limping) A foot radiograph series is only required if there is any pain in the midfoot zone and any of these findings: Bone tenderness at the base of the fifth metatarsal Bone tenderness at the navicular Total inability to bear weight, both immediately and in the office
768
ROOKS & CORWELL
flexors (flexor digitorum and flexor hallucis longus), tibialis posterior, and the peroneal muscles. Factors predisposing a person to rupture of the Achilles tendon include certain systemic inflammatory diseases, fluoroquinolone antibiotics, prior tendon steroid injections, and Achilles tendonitis. The diagnosis of Achilles tendon rupture is primarily clinical. Patients may report a ‘‘pop’’ or ‘‘snap’’ at the time of injury associated with sudden severe pain, similar to a ‘‘gunshot wound.’’ The acute pain may resolve quickly, and the episode may be misdiagnosed as an ankle sprain. The clinician may note swelling of the lower calf. There may be a palpable defect in the tendon 2 to 6 cm proximal to its insertion, called a positive ‘‘gap sign,’’ though this may be missed if swelling is significant. Patients may still be able to plantar flex the foot, albeit weakly, as described above. The patient often has difficulty bearing weight. The test to diagnose a complete rupture of the Achilles tendon is called the Thompson test (see Appendix). Imaging is unnecessary in cases of obvious tendon rupture. MRI or ultrasound are the diagnostic modalities of choice if the clinical diagnosis is in doubt. Office treatment following diagnosis involves rest, ice, compression and elevation (RICE). In addition, the clinician should provide crutches and pain-free immobilization in a posterior splint in passive equines position (plantar flexion). Patients may be sent home and should be referred for orthopedic follow-up within 72 hours. Prolonged delay may result in retraction of the tendon. Both operative and nonoperative treatment options are available, and depend on individual patient characteristics. The patient’s age, level of activity, degree of tendon retraction, and general medical health guide definitive care. Conservative treatment involves cast immobilization in plantar flexion, so as to approximate the two ends of the torn tendon. Surgical repair decreases the incidence of rerupture and allows an earlier return to activities [17]. The Maisonneuve fracture complex is defined as a fracture of the proximal third of the fibula, with associated syndesmosis disruption and deltoid ligament tear or medial malleolar fracture. Although less common than other types of ankle fractures, the Maisonneuve fracture is often initially misdiagnosed, which may result in long-term disability. The patient may present with pain in the region of the ankle without pain in the proximal fibula. This fracture will be missed if the clinician concentrates solely on the injured ankle and does not examine the proximal fibula. The latter is very important. Unless the diagnosis is considered at the time of the examination, only ankle films will be ordered, and the fracture will be missed. The mechanism is a pivot stress to the ankle, with external foot rotation in which the force vector travels upward, damaging the syndesmosis complex and fracturing the proximal fibula. This diagnosis should be suspected in patients who have a history of ankle eversion, together with medial malleolar and proximal fibular tenderness on examination. These patients require ankle and lower extremity plain films. The proper diagnosis will not be missed if the proximal fibula is palpated in all patients who have an ankle sprain.
COMMON URGENT MUSCULOSKELETAL INJURIES
769
Patients should be referred to an orthopedist for open reduction and internal fixation. The base of the fifth metatarsal is a common site of fracture. Both the Jones’ fracture and the pseudo-Jones’ fracture involve pain over the fifth metatarsal base. These two entities should be managed very differently. They have different prognoses that must be distinguished by the primary care physician. A Jones’ fracture is a transverse fracture of the proximal diaphysis of the fifth metatarsal. It may be incorrectly diagnosed, unless the base of the fifth metatarsal is palpated for tenderness in all patients with ‘‘ankle sprains.’’ A true Jones’ fracture occurs approximately 1.5 cm distal to the base of the fifth metatarsal (the junction of the diaphysis and metaphysis). This must be differentiated from the more common pseudo-Jones’ fracture, in which the peroneal brevis tendon avulses its osseous attachment at the base of the fifth metatarsal. The avulsion fracture is associated with an inversion injury mechanism, as opposed to the Jones’ fracture, in which a strong adduction force is applied to the forefoot with the ankle plantarflexed. The difference is easily visible on plain films. A common clinical error is that only ankle films are ordered, which may not include or provide clear visualization of the fifth metatarsal. Proper clinical examination of the patient should always include palpation of the fifth metatarsal and, if it is tender, a foot imaging series should be ordered. Patients who have a Jones’ fracture should be made non-weight–bearing with crutches, and immobilized in a posterior splint. They should be referred to an orthopedist within 24 to 48 hours, who will place them in a non-weight–bearing, short leg cast for approximately 6 weeks and follow them closely. Competitive athletes or other highly active individuals may be offered surgery. Because this area is the weight-bearing lateral margin of the foot and since it corresponds to a watershed area of the blood supply, fracture is frequently complicated by nonunion, malunion, or recurrence. These complications occur in almost one fourth of patients treated conservatively [18]. Avulsion fractures can be treated with a hard-sole shoe, walking cast, or compression wrap. Patients should be made weight bearing as tolerated, and will usually heal well without sequelae. The clinical pitfall is misdiagnosing a potentially serious lower extremity injury as ‘‘just an ankle sprain.’’ The clinician should check all patients for other injuries in a systematic manner. Specific injuries not to be missed include Achilles tendon rupture, proximal fibular fracture, and fifth metatarsal fractures. Knee injuries Knee injuries are among the most common musculoskeletal disorders. The primary care physician is frequently called on to evaluate patients with acute bony and soft-tissue knee injuries. Being knowledgeable about and comfortable with evaluating knee injuries is an invaluable part of
770
ROOKS & CORWELL
a physician’s clinical armamentarium. Clinical decision guidelines similar to those used for the ankle exist to aid in the determination of whether knee radiographs are necessary. Although not as widely used as the Ottawa ankle rules, two popular decision rules exist: the Ottawa and Pittsburgh knee rules (Box 3). Both rules have a high sensitivity (97%–99%), but the Pittsburgh knee rules have greater specificity [19]. The standard imaging series consists of three views: AP, lateral, and oblique radiographs. A sunrise view should be included to detect nondisplaced vertical patella fractures that may be missed in the conventional three-view series. Rupture of the quadriceps tendon results from forceful contraction of the quadriceps muscle, or can be secondary to a fall on a flexed knee. This injury is most often seen in older patients and younger individuals involved in jumping activities such as high jump or basketball. Rupture occurs just proximal to the patella. Just as in rupture of the Achilles tendon, patients will report severe pain, a loud ‘‘pop,’’ and an immediate inability to bear weight or extend the knee. Examination often reveals a palpable soft-tissue defect proximal to the superior pole of the patella. This may be obscured if sufficient edema has already developed by the time of examination. Patients will be unable to perform a seated straight leg raise, or to extend the knee from a fully flexed position. Knee films should be obtained to rule out associated fracture. Patients should be placed in a knee immobilizer and provided crutches. Early diagnosis and surgical repair within 48 to 72 hours are necessary to preserve the extensor mechanism of the knee. Two commonly confused musculoskeletal disorders warrant discussion because of their varied management and treatment: patellar dislocation and knee dislocation. Both disorders may have spontaneously reduced before evaluation, thereby increasing the potential for missed diagnosis. A
Box 3. Ottawa and Pittsburgh knee rules Ottawa knee rules. A knee radiograph series is required for acute knee injuries and any of these following conditions: Age over 55 Fibular head tenderness Inability to flex to 90 Inability to bear weight for four steps both immediately and in the office Pittsburgh knee rules. A knee radiograph series is only required if there is blunt trauma or a fall as mechanism, plus either of the following conditions: Age under 12 or over 50 Inability to bear weight for four steps both immediately and in the office
COMMON URGENT MUSCULOSKELETAL INJURIES
771
true knee dislocation is an orthopedic emergency. It is a rare injury because of the strength of the supporting ligaments. Knee dislocations usually require a great amount of force, such as would be generated by motor vehicle accidents, falls, or sports injuries. Dislocations with relatively minor trauma have been reported, often in obese patients. Anterior and posterior dislocations, named by the direction of the displacement of the tibia relative to the femur, are most common. Dislocations involve disruption of at least two of the major knee ligaments with the anterior and posterior cruciate ligaments (ACL, PCL) being most frequently involved. These injuries are typically associated with a large hemarthrosis. Because of the potentially devastating and limb-threatening vascular consequences, all office physicians need to be aware of the presentation, diagnosis, and management of knee dislocations. The nature of the injury is not always clinically evident because of spontaneous reduction at the time of evaluation. Further, swelling may be negligible because of tearing of the capsular structures and resulting dissipation of the acute hemarthrosis into the adjacent soft tissue. Therefore, a significant mechanism of injury in the setting of a multidirectional, severely unstable knee may indicate a spontaneously reduced dislocation. In any suspected case, a thorough neurovascular examination must be performed before and after reduction. Up to one third of injuries involve the tightly tethered, popliteal artery. Though the absence of distal pulses suggests vascular injury, the presence of pulses cannot be used as evidence of the lack of a vascular injury. Also, as many as 20% to 30% of cases involve injury to the common peroneal nerve, which controls ankle dorsiflexion and sensation over the first dorsal web space [20]. In a small study of knee dislocations [21], two thirds were reduced on presentation. The distribution of ligamentous injuries favored ACL (84%), PCL (87%), and combination (71%) injuries over injuries to the medial collateral (44%) and lateral collateral (62%) ligaments. Especially in cases of vascular compromise, knee dislocations should be reduced using longitudinal traction as soon as possible. The patient should then be placed in a long posterior splint in 15 of flexion and immediately sent to the nearest hospital for further evaluation. Whereas in the past all patients received arteriography, today a selective approach is employed. All patients showing hard signs of a vascular injury, such as an absence of pulses or presence of a bruit, require arteriography. Note that such hard signs may not manifest until after 24 to 48 hours. Patients with a normal vascular examination and ankle-brachial index may be admitted overnight for serial neurovascular examinations. Patella dislocations are relatively benign compared to knee dislocations. The usual mechanism of injury is a twisting movement about the knee. Direct trauma to the knee can also cause dislocation. Patellar dislocations are more common in women, because of their greater physiologic laxity, and in those who have connective tissue disorders. Dislocations predominately dislocate laterally and have a high rate of recurrence. The trauma results in a tearing of the medial patellofemoral ligament and the patella being
772
ROOKS & CORWELL
displaced over the lateral condyle. The resulting pain, hemarthrosis, inability to extend the knee, and obvious deformity are very troubling to the patient. The patella may spontaneously reduce in the field with simple leg straightening. Otherwise, acute dislocation of the patella is usually obvious on clinical evaluation, because the dislocated patella causes a large bulge in the lateral leg with a prominent medial femoral condyle. If the dislocation is recent and did not spontaneously reduce, immediate reduction should be attempted. Reduction is moderately uncomfortable and may occasionally require procedural sedation. The patient should be placed in the supine position with the hip flexed (to relax the quadriceps muscles). The knee is then slowly hyperextended while gentle medial pressure is applied to the lateral aspect of the dislocated patella. Successful reduction is indicated by the return of the patella to the trochlear groove, absence of deformity, and normal knee function. Knee radiographs should be obtained after the reduction to assess for avulsion or fracture. Following reduction, the patient should receive crutches, be placed in a knee immobilizer in locked extension, and should follow the RICE protocol. The patient may be made partially weight-bearing as tolerated, and later be advanced to full weightbearing. Patients should return for follow-up in 1 week with either their primary care physician or an orthopedic surgeon. They will benefit from improving their range of motion and strengthening the quadriceps, and may be given a referral for physical therapy. Severe dislocations with associated ligamentous injury in competitive athletes and anyone who has a recurrent dislocation may benefit from orthopedic referral. Most athletes should wear a supportive knee brace and can expect to return to play within 4 to 6 weeks. Although the majority of knee injuries that present to the primary care physician are overuse injuries, acute traumatic soft tissue injuries of the knee are very common. Soft-tissue injuries of the knee are better classified as urgencies than emergencies. It is more important for the primary care physician to identify the presence of an internal knee derangement than to identify the specific injured structure. Another important category of patient presentation is the patient with acute knee trauma, hemarthrosis and a negative radiograph series. These patients are likely to have one of the following three entities: (1) an acute ligamentous tear, (2) a meniscal tear, or (3) an osteochondral fracture. Obtaining a detailed history and a careful physical examination can help distinguish among these conditions. The specific mechanism of injury (eg, a football player having been tackled from the side) is one of the most important aspects of the history. The physician should inquire about the direction and degree of the traumatic stress and the position of the knee at the time of injury. Also relevant is whether there was an audible ‘‘pop’’ at the time of injury, if there was immediate or delayed swelling, if the patient can continue to ambulate, and whether the knee locks or ‘‘gives way.’’ Always consider that pain referred from the hip or back may present as knee pain. Because of the ACL’s rich blood supply, sudden onset of a large effusion suggests an ACL injury.
COMMON URGENT MUSCULOSKELETAL INJURIES
773
The physical examination is relatively straightforward, and includes inspection for deformity, edema, ecchymosis, and erythema. The examination can be limited in patients with severe pain or those with a large, tense effusion. The physician should always palpate for bony tenderness and order plain films if appropriate. The knee should not be significantly stressed if a fracture is suspected. The physician must assess neurovascular stability and rule out associated injuries. If radiographs are negative or not indicated, assessment of the anterioposterior and varus-valgus stability of the knee is the next step. A review of the physical examination of the knee can be found in the Appendix. Ligamentous injuries are classified according to a threelevel grading system. A Grade I sprain involves tenderness, mild, if any swelling, and no joint laxity. Grade II injuries involve a partial ligament tear with some laxity of the joint. Grade III injuries are complete ligamentous tears without a firm end point during stress testing. Important points to consider include always examining the uninjured knee first to establish a baseline clinical stability and dispel the patient’s fears, thereby improving cooperation. One must compare any irregularities to the uninjured knee, because abnormalities may be subtle. Side-to-side differences are more important than absolute laxity. The physician should consider aspirating large, tense joint effusions to increase the patient’s comfort, aid in diagnosis, and to improve the accuracy of the physical examination. There is a significant risk of a false-negative Lachman test in the presence of a large effusion. If the aspiration of a hemarthrosis reveals fat globules (lipohemarthrosis), a fracture is likely. The cruciate ligaments, the ACL and PCL, are primary stabilizers of the knee, and resist anteriorly and posteriorly directed stress, respectively. The ACL is a very frequently injured major knee ligament. It is often injured in contact sports, skiing, basketball, and soccer. The mechanisms of injury include acute extreme deceleration, noncontact- and contact-related hyperextension or valgus force, and external rotation. Patients typically report the knee ‘‘giving way,’’ with associated sudden pain, an audible ‘‘pop,’’ immediate swelling (hemarthrosis usually develops within 2 hours of injury), and an inability to continue activities. Diagnostic tests include the anterior drawer, Lachman, and pivot shift tests. The Lachman test is the single best component of the physical examination for testing the integrity of the ACL, and is more sensitive and accurate than the anterior drawer test. Meniscal tears frequently accompany, ACL tears. Definitive treatment of an ACL injury depends on the patient’s age, desired activity level, and whether there are other associated injuries. The ACL requires elective surgical repair for anyone planning return to sports activities. The PCL, by comparison, is broader and stronger than the ACL, and is less frequently injured. Injury is most frequently caused by a direct blow to the proximal tibia (classically, a knee-against-dashboard injury in a motor vehicle accident) or a fall on a flexed knee with a plantar-flexed foot. Hyperextension, a varus/valgus directed force while in full extension, and
774
ROOKS & CORWELL
hyperflexion are less common mechanisms of injury. PCL injuries require an increased amount of force, and though isolated injuries do occur, PCL injuries are often part of a combined ligamentous injury. Unlike the ACL injury, PCL injury does not usually involve a distinct pop or tear. Patients often report only vague symptoms, such as gait unsteadiness, and they usually do not have severe pain. Swelling, if present, is often mild. Isolated injuries are often missed on initial evaluation. The posterior drawer test is diagnostic. Treatment of isolated Grade I and II PCL injuries is conservative. Grade III injuries, combined injuries, avulsion fractures, and failure of nonsurgical treatment (ie, recurrent symptomatic instability) often require surgical reconstruction. The medial collateral ligament (MCL) and the lateral collateral ligament (LCL) are the medial and lateral stabilizers of the knee and resist valgus and varus stress, respectively. The MCL is more frequently injured than the LCL. The cause of injury to the MCL is usually a lateral blow to the knee with the foot in a fixed position, or a valgus-directed force to the knee with external tibial rotation. LCL injuries are caused by varus directed stress with internal tibial rotation. Isolated LCL injuries are relatively rare. Varus and valgus stress tests examine the integrity of the collateral ligaments. The degree of injury can be estimated by the amount of joint line opening. The ligaments may also be tender to palpation along their entire course. With the possible exception of complete LCL tears, isolated tears usually heal without surgical intervention, with the possible exception of complete LCL tears. If the physician discovers a high-grade injury to the posterior or lateral side of the knee, referral for orthopedics is always prudent because of the need to rule out a commonly missed entity called a posterolateral corner injury. The posterolateral corner is a complex anatomical region that contributes to the static and dynamic stability of the knee. It is comprised of the LCL, popliteus muscle and tendon, the popliteofibular ligament, the arcuate ligament, and the posterolateral capsule. Posterolateral corner injuries are associated with posterolateral knee pain, tenderness, and swelling. As the acute swelling subsides, the patient may notice instability of the extended knee. Accurate diagnosis is important because this injury may require early surgical intervention for a satisfactory result. The menisci are fibrocartilaginous pads positioned on the articulating surface of the tibia. The blood supply enters from the periphery and does not fully penetrate the entire meniscus, which explains the notoriously poor healing following injury. The medial meniscus is injured more frequently than the lateral meniscus. Tears may occur in isolation, or in association with a major ligament injury such as an ACL tear. Injury results from twisting maneuvers or a rapid change of direction on a weight-bearing knee. Unlike other high-grade ligament injuries, meniscal injuries are usually characterized by only mild swelling, significantly less disability, and a continued ability to participate in sports activity. Patients may complain of joint line pain, delayed mild swelling, and locking (indicating the presence of
COMMON URGENT MUSCULOSKELETAL INJURIES
775
a displaced meniscal fragment). Knee effusion develops more gradually than in ACL injuries, usually over 12 to 24 hours following injury. Joint line tenderness and McMurray’s and Apley’s tests aid in diagnosis. Treatment is primarily symptom-based. In the absence of mechanical symptoms, conservative therapy should be initiated. In cases involving younger patients, athletes, and those with large tears, mechanical symptoms, or a locked knee, surgical consultation is warranted. After any significant knee injury, the knee should be protected from further injury by placing the patient in a knee immobilizer and beginning the RICE protocol. The patient should be instructed to remove the knee immobilizer in order to perform daily range-of motion-exercises to avoid contracture and to maintain mobility. Orthopedic referral should occur within 1 week. MRI is the study of choice for noninvasive evaluation of ligament or meniscal injury and osteochondral fractures. The clinical pitfall to avoid is misdiagnosing a potentially serious knee injury as ‘‘just a sprain.’’ The clinician should check patients for other injuries in a systematic manner. Specific injuries not to be missed include quadriceps tendon rupture, patellar dislocation, knee dislocation, and meniscal or ligamentous injuries.
Appendix Physical examination tests Abduction (valgus) stress testdThis test assesses the integrity of the MCL. It is performed in extension and 30 of flexion. Place the patient in a supine position. One hand is placed on the lateral knee and the other hand grasps the forefoot and pulls it away from the midline. Instability indicates rupture of the MCL. Adduction (varus) stress testdThis test assesses the integrity of the LCL. It is performed in extension and 30 of flexion. Place the patient in a supine position. One hand is placed on the medial knee and the other hand grasps the forefoot and pulls it toward the midline. Instability indicates rupture of the MCL. Anterior drawer testdThis test assesses the integrity of the ACL. It is performed in 45 of hip flexion and 90 of knee flexion. Place the patient in a supine position. Sit on patient’s foot to anchor the lower extremity. Place hands around the knee, with the thumbs at the tibial tubercle, palpating the hamstring tendons with the fingers to ensure that they are relaxed. Grasp the proximal tibia with both hands and attempt anterior translation of the tibia. Assess for laxity and the presence of a discrete endpoint. Instability indicates rupture of the ACL. Apley’s grind testdThis test assesses the integrity of the menisci. Place the patient in a prone position with the knee flexed to 90 . Place one
776
ROOKS & CORWELL
hand on the foot and the fingers of the other hand on the medial and lateral joint lines in turn. While partially flexing and extending the knee, rotate the tibia and apply downward pressure on the foot. Painful clicking during internal (lateral meniscus) or external rotation (medial meniscus) constitutes a positive test. Lachman testdThis test assesses the integrity of the ACL. It is performed in 20 to 30 of flexion. Place the patient in a supine position. Place one hand on the distal femur and grasp the proximal tibia with the other hand and attempt anterior translation. Assess for laxity and the presence of a discrete endpoint. Instability indicates rupture of the ACL. This is the single best test for the ACL, and it is more sensitive and accurate than the anterior drawer test. McMurray’s testdThis test assesses the integrity of the menisci. Place the patient in a supine position with the knee maximally flexed. Place one hand on the foot and the fingers of the other hand on the medial and lateral joint lines, in turn. With valgus stress, slowly extend the knee with the tibia either internally (lateral meniscus) or externally rotated (medial meniscus). Painful clicking constitutes a positive test. Pivot shift testdThis test assesses the integrity of the ACL. Place the patient in a supine position with the knee in full extension. Place one hand on the foot and with the other hand grasp the knee with the thumb behind the fibular head. Rotate the foot and tibia internally. Apply valgus stress to the knee as the knee is slowly flexed to 40 . The anterior tibia is subluxed on the femur with extension and initial flexion. It suddenly reduces (shifts) with further flexion, noticed by the ‘‘clunk’’ of reduction at 20 to 30 of knee flexion. Posterior drawer testdThis test assesses the integrity of the PCL. It is performed in 45 of hip flexion and 90 of knee flexion. Place the patient in a supine position. Sit on patient’s foot to anchor the lower extremity. Place hands around the knee, with both thumbs on top of the medial and lateral tibial plateaus. Attempt posterior translation of the tibia. Assess for laxity and the presence of a discrete endpoint. Instability indicates rupture of the PCL. Straight leg raise testdThis test assesses for the presence of a herniated disc. The SLR can be performed with the patient in the supine or seated position. Place one hand under the heel and grasp the knee with the other hand, keeping it fully extended. Monitor for the presence of radicular pain below the knee of the affected leg when the leg is elevated between 30 and 60 , especially if that pain is worsened by ankle dorsiflexion. Thompson’s testdThis test assesses the integrity of the Achilles tendon. For accuracy, the patient must be kneeling in a chair or in the prone position with the feet hanging over the edge. Squeeze the calf at its midportion and note the presence or absence of plantar flexion. Under normal circumstances, squeezing the calf causes plantar flexion. Absence
COMMON URGENT MUSCULOSKELETAL INJURIES
777
of plantar flexion is a positive test and confirms a complete Achilles tendon rupture.
References [1] Woodward TW, Best TM. The painful shoulder: Part I. Clinical evaluation. Am Fam Physician 2000;61(10):3079–88. [2] Johns KL, Counselman FL. Evaluation and treatment of shoulder injuries. Emerg Med 2001;33(8):20–40. [3] Ross G. Acute elbow dislocation. Phys Sportsmed 1999;27(2):140–3. [4] Ballas MT, Tyko J, Mannarino F. Commonly missed orthopedic problems. Am Fam Physician 1998;57(2):267–74. [5] Trafton PG. Orthopedic emergencies. In: Saunders CE, Ho MT, editors. Current emergency diagnosis & treatment. 4th edition. East Norwalk (CT): Appleton and Lange; 1992. p. 329–42. [6] Lang J, Counselman FL. Common orthopedic hand and wrist injuries. Emerg Med 2003; 35(9):20–38. [7] Abrahamsson SO, Sollerman C, Lundborg G, et al. Diagnosis of displaced ulnar collateral ligament of the metacarpophalangeal joint of the thumb. J Hand Surg [Am] 1990;15(3): 457–60. [8] Bigos S, Bowyer O, Braen G, et al. Acute low back problems in adults. Clinical practice guideline. AHCPR Pub. No. 95–0643. Rockville (MD): US Department of Health and Human Services, Public Health Service, Agency for Health Care Policy and Research; 1994. [9] Deyo R, Weinstein J. Primary care: low back pain. N Engl J Med 2001;344(5):363–70. [10] Waddell G, McCulloch JA, Kummel E, et al. Nonorganic physical signs in low-back pain. Spine 1980;5(2):117–25. [11] Spangfort EV. The lumbar disc herniation. A computer-aided analysis of 2504 operations. Acta Orthop Scand Suppl 1972;142:1–95. [12] Jonsson B, Stromqvist B. Symptoms and signs in degeneration of the lumbar spine. A prospective, consecutive study of 300 operated patients. J Bone Joint Surg Br 1993;75(3): 381–5. [13] Jensen MC, Brant-Zawadzki MN, Obuchowski N, et al. Magnetic resonance imaging of the lumbar spine in people without back pain. N Engl J Med 1994;331(2):69–73. [14] Markert RJ, Walley ME, Guttman TG, et al. A pooled analysis of the Ottawa ankle rules used on adults in the ED. Am J Emerg Med 1998;16(6):564–7. [15] Inglis AE, Scott WN, Sculco TP, et al. Ruptures of the tendo achillis. An objective assessment of surgical and non-surgical treatment. J Bone Joint Surg Am 1976;58(7):990–3. [16] Ho K, Abu-Laban RB. Ankle and foot. In: Marx JA, Hockberger RS, Walls RM, editors. Rosen’s emergency medicine. 5th edition. St. Louis (MO): Mosby, Inc.; 2002. p. 714–8. [17] Cetti R, Christensen SE, Ejsted R, et al. Operative versus nonoperative treatment of Achilles tendon rupture. A prospective randomized study and review of the literature. Am J Sports Med 1993;21(6):791–9. [18] Josefsson PO, Karlsson M, Redlund-Johnell I, et al. Jones fracture. Surgical versus nonsurgical treatment. Clin Orthop Relat Res 1994;(299):252–5. [19] Seaberg DC, Yealy DM, Lukens T, et al. Multicenter comparison of two clinical decision rules for the use of radiography in acute, high-risk knee injuries. Ann Emerg Med 1998; 32(1):8–13. [20] Wood MB. Peroneal nerve repair. Surgical results. Clin Orthop Relat Res 1991;(267):206–10. [21] Twaddle BC, Bidwell TA, Chapman JR. Knee dislocations: where are the lesions? A prospective evaluation of surgical findings in 63 cases. J Orthop Trauma 2003;17(3):198–202.
Prim Care Clin Office Pract 33 (2006) 779–793
Acute Monarthritis: Diagnosis and Management Erinn E. Maury, MD*, Raymond H. Flores, MD Division of Rheumatology and Clinical Immunology, University of Maryland School of Medicine, 10 South Pine Street, MSTF 834, Baltimore, MD 21201, USA
The patient who presents with an acute painful synovitis of a single join provides a significant diagnostic and therapeutic challenge to the primary care physician. An aggressive approach is required to differentiate a potential infectious arthritis, with its attendant morbidity and potential mortality, from other causes of monarthritis that are not immediately life-threatening. The article reviews the common causes of acute monarthritis in the adult, including their presentation; as well as guidelines for rapid and efficient diagnosis and management. Common causes include infections (bacterial/ Lyme/mycobacterial/viral), microcrystalline disease (gout/pseudogout), and traumatic and reactive arthropathy. In addition, guidelines are suggested for the management approach to acute monarthritis when initial diagnostic testing is unrevealing of a specific diagnosis. The Appendix has a list of diseases that may present with acute monarthritis. Bacterial arthritis Background Non-gonococcal bacterial arthritis has a 10% to 20% mortality rate in adults [1]. Significant risk factors for developing bacterial arthritis include age 80 or above, diabetes, malignancy, immunosuppressive drugs, rheumatoid arthritis (RA) [2,3], joint replacement [3] and high-risk sexual behavior. Significant risk factors for poor outcome include older age, pre-existing joint disease (especially RA), and joint replacement or other hardware [1]. Bacterial arthritis should also be a consideration in the patient who has an inflammatory arthritis such as RA and who is presenting with an acute monarthritis flare [4].
* Corresponding author. E-mail address:
[email protected] (E.E. Maury). 0095-4543/06/$ - see front matter Ó 2006 Elsevier Inc. All rights reserved. doi:10.1016/j.pop.2006.06.010 primarycare.theclinics.com
780
MAURY & FLORES
Presentation Bacterial arthritis commonly presents with acute onset, and medical care is generally sought within a mean of 10 days after symptoms begin [2]. Patients generally present with severe pain and inability to bear weight on the affected joint. Constitutional symptoms such as fever, chills, and malaise may be present. There may or may not be a history of trauma; however, there is frequently a prior history of arthritis in the involved joint [1]. Physical examination The involved joint is usually erythematous, swollen, warm, and tender to touch. There is usually exquisite pain with both active and passive movement, indicating pathology within the joint capsule. A full physical examination is warranted to find the source of the bacterial infection. Splinter hemorrhages, Janeway’s lesions, or Osler’s nodes may be present on the hands or feet, indicating possible endocarditis. Other potential sources for infection include the urinary tract, respiratory tract, and skin infections [4]. Imaging Plain radiographs of the affected joint may show the presence of an effusion or soft-tissue swelling, or evidence of underlying arthritis [5]. Imaging such as MRI, CT, or bone scan should be reserved for suspicion of osteomyelitis, occult fracture, or other causes of acute monarthritis. Laboratory data The most diagnostically useful information comes from joint fluid analysis. Whenever accessible, joint fluid should be obtained for culture, Gram’s stain, cell count, and crystal evaluation. When only a small amount of fluid is obtainable, priority should be given to evaluation of Gram’s stain and culture. Additionally, only a drop of fluid is necessary for crystal analysis by polarizing light microscopy. In bacterial arthritis, cell counts generally exceed 50,000 cells/mm3, with a preponderance (O90%) of polymorphonuclear cells (PMNs) [5]. Cell counts of fewer than 50,000 cells/mm3 can be seen in patients who are immune compromised, or who have been partially treated with antibiotics before aspiration of the joint. The most common bacterial pathogens are Staphylococci (aureus and epidermitis) and Streptococci. Gram-negative bacilli, mycobacteria, and Neisseria gonorrhoeae are less common pathogens [6]. The absence of organisms on Gram’s stain examination does not rule out the presence of bacterial arthritis. Culture-negative bacterial arthritis may be caused by gonococcal infection. Additionally, cultures of blood and urine or other potential sites of infection may be helpful for diagnosis and management. Serum white blood cell count (WBC) should be elevated, with a preponderance of PMNs [5].
ACUTE MONARTHRITIS: DIAGNOSIS AND MANAGEMENT
781
Management The diagnosis is most effectively made by joint aspiration, preferably before starting antibiotics; however, antibiotic therapy should never be delayed if fluid is difficult to obtain and the index of suspicion is high for a bacterial arthritis. For example, it may be difficult to obtain fluid from the hip or sacroiliac joints. These joints typically require aspiration under fluoroscopy to obtain fluid [5]. Care should be taken not to aspirate a joint through an overlying cellulitis. Orthopedic consultation is mandatory when suspected bacterial arthritis involves a prosthetic joint. The patient should be admitted to the hospital for empiric broad-coverage intravenous antibiotic therapy for gram-positive organisms. If there is suspicion for a gram-negative or other bacterial infection by Gram’s stain or clinical history, antibiotic coverage should be inclusive. Coverage for methicillin-resistant S aureus should be considered in high-risk patients or if it is known to be present in the community. Once culture and sensitivity information are available, antibiotics should be adjusted accordingly. Orthopedics should be consulted immediately for evaluation for open drainage of the joint. Generally, bacterial infections of the hip and shoulder joints require surgical intervention, as do prosthetic joint infections. If open surgical drainage is not indicated, closed drainage or daily needle aspiration of the joint should be done until there is no longer any fluid to drain. Daily aspiration of joint fluid can be evaluated for cell count and culture to monitor for improvement [4,5]. Parenteral antibiotics should continue for a minimum of 2 weeks, and may need to be continued for up to a month, depending on the patient comorbidities and complications such as underlying osteomyelitis or endocarditis [4]. Viral arthritis Viral arthritis usually presents as an inflammatory polyarthritis and is rarely limited to one joint. Rarely coxsackie-B can present as a self-limited monarthritis. Fungal arthritis Fungal arthritis presents much like bacterial arthritis; however, it is rare, and underlying comorbidities such as malignancy and immune suppression from chemotherapy are usually factors. Spirochetes: Lyme disease Lyme disease is caused by Borrelia burgdorferi carried by the deer tick Ixodes scapularis, and should be suspected in endemic areas of the Northeast and in the central northern United States. Lyme disease most commonly presents with the target lesion of erythema chonicum migrans (ECM), although it can on occasion present with an acute monarthritis, usually
782
MAURY & FLORES
involving the knee. Generally, the infection with Lyme has been present for several months before joint manifestations appear [7]. Physical examination reveals an erythematous swollen joint with pain on passive and active motion. Attacks of this arthritis can be recurrent [7]. The most common finding on plain radiography is a knee effusion. Imaging of affected joints in chronic Lyme may reveal effusion, osteoporosis, loss of cartilage, and possibly periarticular erosions [8,9]. Aspiration of the joint may reveal inflammatory fluid with a cell count range of 500 to greater than 50,000 cells/mm3. Synovial fluid can be sent for Lyme polymerase chain reaction (PCR) for diagnostic purposes [9]. B burgdorferi is rarely cultured from synovial or other body fluids. Antibody testing by enzyme-linked immunosorbent assay (ELISA) may be positive, but there is a high rate of false-positives. Thus a positive test should be confirmed with a Western blot analysis [10,11]. If the pretest probability of Lyme is high, a positive test is more helpful diagnostically [12]. According to the Infectious Disease Society of America (IDSA), Lyme arthritis should be treated with doxycycline 100 mg twice a day, or if tetracyclines are contraindicated, amoxicillin 500 mg three times a day for a minimum of 28 days; doses should be adjusted for children [13]. Lyme arthritis may persist for months to years in approximately 10% of patients, despite eradication of the spirochete [7]. Gonococcal arthritis Background The prevalence of N gonorrhoeae has been in decline since the 1980s. Still, those who engage in high-risk sexual behaviors are at risk for contracting the infection. Although staphylococcus and streptococcus are overall the most common pathogens found in bacterial arthritis, N gonorrhoeae is still the most common bacterial arthritis found in sexually active young people aged 18 to 24 [9,14]. Presentation/physical Patients who have disseminated gonococcal infection can present anywhere from 1 day to 3 months after initial infection [9]. Gonoccal arthritis typically presents as an asymmetric arthritis of the joints of the upper extremities, particularly the wrist and extensor tendons; however, any joint may be involved [15,16]. Patients may also complain of myalgias, arthralgias, fever, malaise, and dermatitis. Nonpruritic, nonpainful papules, vesicles, or pustules on the trunk and extremities characterize the dermatitis. A full physical examination including a genitourinary examination with cultures is warranted [16]. Imaging Plain radiography will likely show soft-tissue swelling and possibly an effusion. If evidence of cartilage destruction or osteomyelitis exists, consider the possibility of longstanding infection or resistant bacteria if previously treated [9].
ACUTE MONARTHRITIS: DIAGNOSIS AND MANAGEMENT
783
Laboratory Diagnosis is made primarily by history and physical examination, because N gonorrhoeae is difficult to culture and is seen on Gram’s stain only 25% to 50% of the time. To increase the yield of a positive result, urethral (men), cervical (women), rectal, pharyngeal, and pustule swabs as well as joint fluid aspirate and blood should be obtained for Gram’s stain and culture. All cultures should be inoculated at bedside using techniques specific to the culturing of N gonorrhoeae. The medium used for the swabs is a prewarmed to 37 C, modified Thayer-Martin chocolate agar for synovial fluid and blood. Within 15 minutes of obtaining samples, the cultured medium should be incubated at 37 C in a 5% CO2 environment [16]. Screening for chlamydia and other sexually transmitted diseases should be done, because patients are commonly coinfected. Additionally, the serum WBC count will often be elevated, with a preponderance of PMNs. Management According to the Centers for Disease Control guidelines, patients should be admitted to the hospital and started on intravenous ceftriaxone 50 mg/ kg/day (if allergic, a fluoroquinolone) to be continued until 24 to 48 hours after improvement of symptoms and then switched to cefixime 400 mg orally twice daily (if allergic, a fluoroquinolone) to complete a total of 7 days of antibiotic treatment. Empiric treatment for chlamydia is included in the regimen if the patient switches to the fluoroquinolone. Otherwise the patient should also be treated for chlamydia with a single oral dose of 1g of azithrymycin [17]. Generally, daily joint aspiration or orthopedic consult is not necessary unless the patient is not responding to therapy. Crystalline arthritis The most common forms of crystalline arthritis are gout and calcium pyrophosphate dihydrate (CPPD or pseudogout) and calcific periarthritis or tendonitis with basic calcium phosphate (BCP). Keep in mind that idiopathic, crystalline arthritis may be the initial presentation of an underlying metabolic disease. For instance, CPPD can occur in the setting of hyperparathyroidism, hemochromatosis, sarcoidosis, or acromegaly. BCP crystals usually occur in patients on long-term hemodialysis. Although the acute arthritis associated with crystalline diseases are not life-threatening, long-term undertreatment can lead to joint destruction and increased morbidity [9,18]. Gout Background Gout has been around for thousands of years. It is one of the most painful forms of arthritis, but it is also one of the most treatable forms. Gout affects approximately 3.8% of men and 1.6% of women in the United States [19], and has a male to female ratio of around 7:1 [9,19]. Gout is generally
784
MAURY & FLORES
not seen in premenopausal women. The prevalence of gout increases with increasing age [9]. Patients who have gout generally have an elevated uric acid level (O7 mg/dl in men or O6 mg/dl in women); however, an elevated serum uric acid level does not always translate into developing gout attacks. There are many individuals who have asymptomatic hyperuricemia who will never develop gout and who do not need treatment. There are a number of conditions directly linked to hyperuricemia and gout, including lead intoxication, hematopoietic malignancy, and renal impairment among others. There are also comorbid conditions associated but not directly linked to gout and hyperuricemia, including diabetes, hypertension, obesity, hypothyroidism, and hyperlipidemia. Gout attacks can be precipitated by dietary indiscretions (ingestion of alcohol or shellfish), use of diuretics or dehydration, alteration of the uric acid level through introduction or discontinuation of hypouricemic therapy, and trauma or severe illness [9]. Presentation Gout typically presents with acute onset of severe pain of a single joint, most commonly the first metatarsal phalangeal (MTP) joint. After the first MTP, the most common joints involved, in decreasing order of frequency, are the foot and ankle, the knee, fingers, elbows, and wrists; however, any diarthrodial joint can be involved [20]. The involved joint may be so tender that it is even painful to have the bed sheets touch the skin. Physical examination On physical examination, the joint will appear erythematous and swollen, and will be warm to the touch. It is usually exquisitely painful. Evidence of tophaceous deposits indicates that the patient has had longstanding hyperuricemia and chronic gout. Uric acid typically deposits on the helix of the ear, the elbow, and Achilles tendon; however, these deposits can occur elsewhere, such as the fingers, prepatellar bursa, and other joints that may not be involved in this particular attack. The hands of a patient who has chronic tophaceous gout may resemble those of a patient who has rheumatoid arthritis [9]. Imaging In an acute gout attack, plain films may be useful to evaluate for the presence of an effusion for aspiration. With recurrent attacks, periarticular erosions may be seen. Further imaging modalities such as MRI, CT, or bone scan are not warranted for the evaluation of gout; however, they may be useful if the diagnosis is unclear. Laboratory The most definitive test to diagnose gout is evaluation of synovial fluid for the presence of uric acid crystals and the absence of infection [21].
ACUTE MONARTHRITIS: DIAGNOSIS AND MANAGEMENT
785
Uric acid crystals may also be obtained from tophaceous material [9]. Under light microscopy, monosodium urate crystals are needle-shaped and are strongly negatively birefringent, appearing as bright yellow needle-shaped crystals when parallel to the axis of slow vibration on the first-order compensator. The crystals may be either intra- or extracellular, although they are usually intracellular during acute attacks [18]. Synovial fluid WBC counts range from 200 to 50,000, but may be higher, and may appear purulent [22]. Blood serum urate levels may be normal during an acute attack [9,18,21]; however, it is helpful to check the serum urate level along with a chemistry panel to evaluate renal function, because some gout treatments are contraindicated or require renal dosing. Follow-up uric acid levels are indicated with uric-acid–lowering agents. Synovial fluid culture and Gram’s stain must always be done to rule out a coexistent infection. Management There are two strategies for management of gout: the first is acute treatment, the second is long-term chronic lowering of serum uric acid levels. Acute attacks are best managed by starting treatment at the very first sign of an attack. Once the acute gout flare is well established, it is more difficult to abort the attack. If a patient is already on a urate lowering agent, the agent should not be stopped in the event of an acute attack. In uncomplicated patients, who have no contraindications, nonsteroidal anti-inflammatory drugs (NSAIDs) are the first-line treatment. The maximum dose of the particular NSAID chosen should be used for 2 to 3 days, then taper from days 3 to 5. Prednisone is an option for those who have contraindications to NSAIDS or colchicine. Contraindications include but are not limited to renal disease, heart failure, and gastrointestinal bleeding. Prednisone may be used as a second-line agent and should be given at a daily oral dose of 30 to 50 mg tapered over 7 to 10 days. Intra-articular steroids may also be of benefit if the diagnosis of gout and absence of infection is certain [9,23]. Colchicine may be used as a third-line agent, and if used, should be used at the first sign of an attack. It is less likely to be effective when give after the first 24 to 48 hours of the attack. Colchicine is given at a dose of 0.6 mg orally every 2 hours until the pain subsides, until diarrhea or nausea develop, or until the maximum dose (6 mg) is reached. Colchicine is seldom used because of its gastrointestinal side effects and potential for toxicity. It is also necessary to dose this medication according to renal function, and it is not used in acute attacks in patients who have renal insufficiency for failure. Long-term prophylaxis with urate lowering agents is reserved for patients who have recurrent gout attacks, uric acid kidney stones, or tophaceous gout. The goal is to lower the urate level to less than 6 mg/dl, which should reduce the number of attacks and prevent and resolve tophi. Treatment may be lifelong, especially in the instance of chronic tophaceous gout, because it
786
MAURY & FLORES
may take years for tophi to resolve. Before starting long-term urate-lowering therapy, the patient should be completely symptom-free from gout. Alteration of the serum urate level can precipitate an acute attack of gout; therefore, an NSAID therapy or occasionally low dose prednisone or daily oral colchicine (renally dosed) should be given as prophylaxis against an acute attack. A drug-induced myopathy can be seen in individuals who are not renally dosed for prophylactic therapy with colchicine. Prophylaxis should be continued until a therapeutic dose of the urate-lowering drug has been achieved. There are two types of drugs currently available for reduction of serum urate: uricosurics and xanthine oxidase inhibitor [9,21,23]. Probenecid and sulfinpyrazone are uricosurics, and are best used in younger patients who have no history of kidney stones or renal insufficiency and who underexcrete uric acid, as measured with a 24-hour urine [23]. Probenecid should start at a twice-daily oral dose of 250 mg and gradually increase every 3 to 4 weeks to a total daily dose not to exceed 2g/day. The slow increase in dosing may reduce attacks caused by establishing the new medication. Side effects include urolithiasis, acute gout flare upon institution of therapy, and possible inhibition of platelet function [9,21]. Allopurinol is a xanthine oxidase inhibitor, and is frequently used in favor of probenecid because of daily dosing. This medication can be used in patients who have renal insufficiency if dosed properly. Dosing begins at a daily oral dose of 100 mg and is slowly increased to achieve a maximum dose of 300 mg daily. The most concerning side effect of allopurinol is a hypersensitivity reaction, in which case the drug should be stopped. Severe cases require supportive care in the hospital. Other preventive measures include patient avoidance of alcohol and shellfish. Weight loss may also be helpful [21]. Calcium pyrophosphate dihydrate disease Calcium pyrophosphate dihydrate disease (CPPD) can present similarly to gout or infectious arthritis, with acute onset of severe pain and disability in a single joint. Again, the joint appears erythematous and swollen, and is tender to touch with pain on active and passive motion. The most common joints involved are the ankles, knees, wrists, shoulders, and metacarpalphalangeal (MCP) joints, other joints include the hips and elbows [9,18]. Plain radiography is helpful in diagnosing CPPD. Characteristic findings on plain films include chondrocalcinosis that appears as linear calcification of the cartilage, especially seen in fibrocartilage in the menisci of the knee, symphysis pubis, hips, annulus fibrosus of the intervertebral discs, and the triangular cartilage of the wrist, but sometimes seen elsewhere in hyaline cartilage [9,24]. Calcification of the tendons and ligaments, subchondral cysts, and osteophytes are also seen on plain films [25]. Synovial fluid analysis may be helpful diagnostically. Rhomboid-shaped calcium pyrophosphate dihydrate crystals may be seen in the synovial fluid,
ACUTE MONARTHRITIS: DIAGNOSIS AND MANAGEMENT
787
and are weakly positively birefringent under polarized light microscopy [9,18]. Infection can coexist with CPPD and as such, culture and Gram’s stain should be done to guide treatment. Treatment of acute attacks is similar to treatment of gout. High dose NSAIDs for 2 to 3 days, then tapered over 3 to 5 days are beneficial. Colchicine may be used but is not as effective. Prednisone at a daily oral dose of 30 to 50 mg tapered over 7 to 10 days will relieve the attack, as will intraarticular steroids [9,18]. Basic calcium phosphate BCP crystalline disease, also called calcific periarthritis, tendonitis, or bursitis, presents very similarly to gout and infectious arthritis. The thought is that deposits of BCP in the periarticular soft tissues rupture either into the surrounding soft tissue or into the joint, causing acute onset of pain, swelling, erythema, and warmth of the joint. The joint most commonly involved, and in which it was first described, is the shoulder. Over time severe glenohumoral joint disease and large effusions can accumulate, resulting in the ‘‘Milwaukee’’ shoulder syndrome [26]. Plain films of the shoulder will reveal calcific deposits in the supraspinatus tendon or the subdeltoid bursa, and degenerative changes of the glenohumoral and acromioclavicular joints. Joint fluid aspiration will likely reveal serosanguinous or milky white fluid with mononuclear cell predominance [9]. The cell count tends to be less than 1000/cm2. The BCP crystals are too small to be appreciated with light microscopy, but can be seen with scanning electron microscopy. Treatment of basic calcium phosphate includes NSAIDs or colchicine, joint aspiration, or intra-articular steroid injection. Surgical intervention is also an option if pain is unremitting after conservative medical treatment [26]. Trauma Meniscal and ligament tears in the knee A meniscal tear may or may not present with a history of trauma; however, the onset of joint pain is still acute. Meniscal tears are caused by sudden deceleration, change in direction, or landing from jumping. The inflammatory response (warmth, erythema, swelling, and effusion) occurs over several hours. A large effusion that develops quickly may indicate a hemarthrosis and likely an anterior cruciate ligament (ACL) tear [27]. The patient who has a meniscal tear will complain of the knee ‘‘giving out’’ during normal walking, locking, or catching, and may be unable to fully extend the leg without significant pain. The examiner may feel a clunk as the torn meniscus is released from the joint space on full extension of the knee. McMurray’s test may be positive: with the patient supine, the practitioner extends the patient’s knee from
788
MAURY & FLORES
a flexed position while externally rotating the foot and palpating the knee medially, at which point the practitioner should feel a clunk. To evaluate the lateral meniscus, rotate the foot internally during extension and feel for a lateral clunk at the knee [9,27]. The anterior drawer sign will likely be positive if there is damage to the ACL. Here the practitioner has the patient in a supine position with the knee flexed. The lower leg is pulled forward and should stop if the ACL is intact. If it is not intact, the lower leg at the tibial plateau will have excessive forward movement. Conversely, the posterior cruciate ligament (PCL) is tested with the posterior drawer sign, in which the practitioner pushes the lower leg posterior when the knee is flexed [28]. Imaging should begin with plain radiography of the knee to evaluate for arthritis and loose bodies in the joint. MRI or arthroscopy is adequate for imaging a meniscal tear. The patient should be referred to an orthopedic surgeon for evaluation. The location of the tear in relation to the blood supply is important for prognosis and potential for healing. The peripheral area (outer one third) of the meniscus has a good blood supply, and therefore is more likely to heal without intervention. Tears in the inner avascular area (inner 2/3) heal poorly, and require arthroscopy for debridement of the torn cartilage. As much of the normal cartilage as possible is left intact. Removal of parts or the entire meniscus predisposes for early osteoarthritis of the knee [9,27]. Occult fracture Fatigue fractures can occur in young adult athletes and patients who have osteoporosis, osteomalacia, and fibrous dysplasia. These patients typically have no history of trauma because the fracture is caused by either repetitive muscular forces, as seen with athletes, or from normal physiologic forces, which can cause fractures in those who have bone mineralization abnormalities [29]. Those at risk include track and field athletes, dancers, military recruits, and those who have menstrual disturbances, eating disorders, and other metabolic conditions leading to bone mineralization abnormalities [30]. Patients who have fatigue fractures within the joint present acutely with a hot, swollen, painful joint. A full physical examination should be performed to evaluate for an underlying cause. Plain radiographs will likely be negative [29,31]. Joint fluid will be negative for crystals and organisms; however there may be evidence of bone marrow elements [32]. MRI and bone scans have comparable sensitivity for diagnosing stress fractures; however, MRI is used more frequently for the additional soft-tissue information it can provide [29,33]. The underlying cause of a fatigue fracture should be pursued. Consider checking a dual energy radiograph absorptiometry (DXA) scan, a 25-hydroxy vitamin D level, complete blood count (CBC), basic chemistry, and liver function tests. Treatment consists of treating the underlying disease, if present, and
ACUTE MONARTHRITIS: DIAGNOSIS AND MANAGEMENT
789
avoidance of weight-bearing in the affected joint until the fracture is healed. Referral to an orthopedic surgeon is warranted, because the patient may need casting [34]. Systemic inflammatory arthritis Systemic inflammatory arthropathies are typically considered to present with multiple inflamed joints with constitutional symptoms and morning stiffness; however, they can present with an acute painful monarthritis, and will have important extra-articular features. For example, juvenile idiopathic arthritis (JIA) frequently presents with a swollen painful joint with an effusion. Children who have JIA with a positive antinuclear antibody (ANA) should follow-up with an ophthalmologist every 3 months to monitor for asymptomatic uveitis that can cause blindness [35]. Similarly, patients who have seronegative spondyloarthropathies such as reactive arthritis (ReA), psoriatic arthritis (PSA) and ankylosing spondylitis (AS) are at risk for ocular involvement. Reactive arthritis in particular can present with an acute monarthritis and have asymptomatic uveitis. This section covers ReA, because it is the most common of the inflammatory arthropathies to present with an acute monarthritis in adults. Background ReA is considered to be in the category of seronegative spoldyloarthropathies, which also includes psoriatic arthritis and ankylosing spondylitis. It is an arthritic reaction to an infectious process, usually occurring after a diarrheal illness or genitourinary infection has run its course. The arthritis is an aseptic arthritis. Throughout history, outbreaks of reactive arthritis have been noted to occur after outbreaks of infectious diarrhea, as has non-gonococcal arthritis. The enteric pathogens associated with ReA are a number of Salmonella serotypes, Shigella flexneri, Clostridium difficile, Vibrio parahaemolyticus, and Yersinia enterocolitica. Chlamydia trachomatis is the only sexually transmitted disease commonly associated with ReA. ReA typically occurs in young adults aged 20 to 40, male to female ratio is 1:1, and there may be a family history [9]. Patients who have the human leukocyte antigen (HLA)-B27 are much higher risk of developing ReA; however, this genotype is not a requirement for the disease [9,36]. Presentation Typically, a young adult will present with a lower extremity asymmetric mono- or oligo-articular arthritis, most commonly in the knee; however, joints of the upper extremities, sacroiliacs, and spine may be involved. Constitutional symptoms such as fatigue, morning stiffness, fevers, and malaise may be present. Upon questioning, the history should reveal that the arthritis was preceded by an enteric or urogenital infection within the previous
790
MAURY & FLORES
month. The enteric infection can be very mild to severe. Prostatitis or urethritis and conjunctivitis may be a complaint [9,36]. Physical examination On musculoskeletal examination, any one joint may be involved, most commonly the knee or other large weight-bearing joint. Frequently the joint is warm, tender, and swollen with an effusion. Look for dactylitis, or fusiform swelling of a finger or toe, also called a ‘‘sausage digit.’’ The patient may also have enthesitis, inflammation of the insertion point of the tendon into the bone, especially the Achilles tendon and plantar fascia. The palms and soles of the feet may reveal pustular psoriatic type lesions called keratoderma blenorrhagicum. A painless, well-circumscribed erythematous lesion c called circinate balanitis an be found on the glans penis. Conjunctivitis occurs in up to 30% of patients [36]. Imaging Plain radiography early on typically shows a normal joint, soft-tissue swelling, or effusion. Asymmetric sacroiliitis is present in 4% [37] of patients early on, and up to one third of patients who have chronic disease [36]. Later on, abnormalities encountered may include reactive bone proliferation, fluffy periosteal reaction, and erosions, especially at tendon insertion sites [9,37]. Laboratory data Erythrocyte sedimentation rate (ESR) and C-reactive protein (CRP) may be elevated, but are nonspecific markers of inflammation. Mild anemia and mild leukocytosis may also be present [9,37]. Synovial fluid is typically inflammatory and WBC count ranges from 2000 to 50,000, with a variable differential, but cultures and crystals are negative [22]. Rheumatoid factor and antinuclear antibodies are negative. Testing for chlamydia with a urethral or cervical smear should be done. Management The majority of patients have resolution of symptoms within 3 to 6 months, although symptoms may recur even years later [36]. If there is a positive test or strong suspicion for chlamydia infection, a single oral dose of 1g of azithromycin should be given to the patient and the patient’s sexual partners [17]. The patient may benefit from therapeutic drainage of the joint (if indicated). NSAIDs are the mainstay of treatment because of their anti-inflammatory effects, and should be taken on a daily basis (dosed depending on the chosen drug). Recovery may be delayed if the NSAID is taken only as needed for pain. Treatment with NSAIDs may last for several months. Prednisone may be added for more debilitating cases of ReA,
ACUTE MONARTHRITIS: DIAGNOSIS AND MANAGEMENT
791
starting at a daily oral dose of 30 to 50 mg and tapering slowly over time as symptoms improve [9]. Patients who have HLA-B27 or keratoderma blenorrhagicum may have a prolonged or severe course with more extra-articular features; such patients should be referred to a rheumatologist. Referral to an ophthalmologist for slit lamp examination is prudent to evaluate for uveitis that can lead to irreversible vision loss [36]. Summary The diagnosis for an acute monarthritis may still be elusive, even after an extensive initial evaluation. For example, what should be done for a patient who has a paucity of extra-articular findings on physical examination and an inflammatory synovial fluid with negative Gram’s stain, cultures, and crystals? Conservative management is always prudent. Assume the joint is infected and treat as such until proven otherwise, because infection carries the highest morbidity and mortality of all the common acute monarthopathies. Appendix List of diseases that may present with acute monarthritis Infectious Bacterial (non-gonococcal & gonococcal)a Viral Spirochete (Lyme, syphillis) Fungal Microcrystalline Gouta Calcium pyrophosphate dihydrate (CPPD) or ‘‘pseudogout’’a Basic calcium phosphate (BCP, hydroxyapatite) Calcium oxalate Traumatic Occult fracture Meniscal or ligamentous injurya Foreign body Blunt trauma Inflammatory Juvenile idiopathic arthritis (JIA)a Reactive arthritis Psoriatic arthritis a
Indicates the most common causes.
792
MAURY & FLORES
Rheumatoid arthritis Lupus Metabolic Hyperparathyroidism Hemochromatosis Acromegaly Hypothyroidism Amyloidosis Benign tumors Osteochondroma Osteoid osteoma Pigmented villonodular synovitis Other Osteoarthritis Malignancy (primary or metastatic) Coagulopathy Osteonecrosis Sarcoidosis Behcet’s
References [1] Kaandorp CJ, Krijnen P, Moens HJ, et al. The outcome of bacterial arthritis: a prospective community-based study. Arthritis Rheum 1997;40(5):884–92. [2] Yu LP, Bradley JD, Hugenberg ST, et al. Predictors of mortality in non-post-operative patients with septic arthritis. Scand J Rheumatol 1992;21(3):142–4. [3] Kaandorp CJ, Van Schaardenburg D, Krijnen P, et al. Risk factors for septic arthritis in patients with joint disease. A prospective study. Arthritis Rheum 1995;38(12):1819–25. [4] Goldenberg DL. Infectious arthritis complicating rheumatoid arthritis and other chronic rheumatic disorders. Arthritis Rheum 1989;32(4):496–502. [5] Goldenberg DL, Reed JI. Bacterial arthritis. N Engl J Med 1985;312(12):764–71. [6] Dubost JJ, Soubrier M, De Champs C, et al. No changes in the distribution of organisms responsible for septic arthritis over a 20 year period. Ann Rheum Dis 2002;61(3):267–9. [7] Steere AC. Lyme disease. N Engl J Med 2001;345(2):115–25. [8] Buchmann RF, Jaramillo D. Imaging of articular disorders in children. Radiol Clin North Am 2004;42(1):151–68 [vii]. [9] Hochberg M, Silman AJ, Smolen JS, et al. Rheumatology. 3rd edition. St. Louis: Elsevier; 2003. [10] Kalish RA, McHugh G, Granquist J, et al. Persistence of immunoglobulin M or immunoglobulin G antibody responses to Borrelia burgdorferi 10–20 years after active Lyme disease. Clin Infect Dis 2001;33(6):780–5. [11] Blaauw AA, van Loon AM, Schellekens JF, et al. Clinical evaluation of guidelines and twotest approach for Lyme disease. Rheumatology (Oxford) 1999;38(11):1121–6. [12] Tugwell P, Dennis DT, Weinstein A, et al. Laboratory evaluation in the diagnosis of Lyme disease. Ann Intern Med 1997;127(12):1109–23.
ACUTE MONARTHRITIS: DIAGNOSIS AND MANAGEMENT
793
[13] Wormser GP, Nadelman RB, Dattwyler RJ, et al. Practice guidelines for the treatment of Lyme disease. The Infectious Diseases Society of America. Clin Infect Dis 2000;31(Suppl 1): 1–14. [14] Garcia-De La Torre I. Advances in the management of septic arthritis. Rheum Dis Clin North Am 2003;29(1):61–75. [15] Manshady BM, Thompson GR, Weiss JJ. Septic arthritis in a general hospital 1966–1977. J Rheumatol 1980;7(4):523–30. [16] Shirtliff ME, Mader JT. Acute septic arthritis. Clin Microbiol Rev 2002;15(4):527–44. [17] Sexually transmitted diseases treatment guidelines 2002. Centers for Disease Control and Prevention. MMWR Recomm Rep 2002;51(RR-6):1–78. [18] Schumacher HR. Crystal-induced arthritis: an overview. Am J Med 1996;100(2A):46S–52S. [19] Kramer HM, Curhan G. The association between gout and nephrolithiasis: the National Health and Nutrition Examination Survey III, 1988–1994. Am J Kidney Dis 2002;40(1): 37–42. [20] Grahame R, Scott JT. Clinical survey of 354 patients with gout. Ann Rheum Dis 1970;29(5): 461–8. [21] Terkeltaub RA. Clinical practice. Gout. N Engl J Med 2003;349(17):1647–55. [22] Harris ED, Ruddy S, Kelley WN. Kelley’s textbook of rheumatology. 7th edition. Philadelphia: Elsevier/Saunders Elsevier; 2005. [23] Emmerson BT. The management of gout. N Engl J Med 1996;334(7):445–51. [24] Resnick D, Niwayama G, Goergen TG, et al. Clinical, radiographic and pathologic abnormalities in calcium pyrophosphate dihydrate deposition disease (CPPD): pseudogout. Radiology 1977;122(1):1–15. [25] Abreu M, Johnson K, Chung CB, et al. Calcification in calcium pyrophosphate dihydrate (CPPD) crystalline deposits in the knee: anatomic, radiographic, MR imaging, and histologic study in cadavers. Skeletal Radiol 2004;33(7):392–8. [26] Halverson PB, McCarty DJ. Clinical aspects of basic calcium phosphate crystal deposition. Rheum Dis Clin North Am 1988;14(2):427–39. [27] Rath E, Richmond JC. The menisci: basic science and advances in treatment. Br J Sports Med 2000;34(4):252–7. [28] Solomon DH, Simel DL, Bates DW, et al. The rational clinical examination. Does this patient have a torn meniscus or ligament of the knee? Value of the physical examination. JAMA 2001;286(13):1610–20. [29] Garcia-Porrua C, Armesto V, Manana P, et al. Fatigue fractures suggesting clinical monarthritis. Arthritis Rheum 2003;49(6):741–4. [30] Bennell K, Matheson G, Meeuwisse W, et al. Risk factors for stress fractures. Sports Med 1999;28(2):91–122. [31] Reeder MT, Dick BH, Atkins JK, et al. Stress fractures. Current concepts of diagnosis and treatment. Sports Med 1996;22(3):198–212. [32] Guidelines for the initial evaluation of the adult patient with acute musculoskeletal symptoms. American College of Rheumatology Ad Hoc Committee on Clinical Guidelines. Arthritis Rheum 1996;39(1):1–8. [33] Deutsch AL, Coel MN, Mink JH. Imaging of stress injuries to bone. Radiography, scintigraphy, and MR imaging. Clin Sports Med 1997;16(2):275–90. [34] Sanderlin BW, Raspa RF. Common stress fractures. Am Fam Physician 2003;68(8): 1527–32. [35] Weiss JE, Ilowite NT. Juvenile idiopathic arthritis. Pediatr Clin North Am 2005;52(2): 413–42 [vi]. [36] Colmegna I, Cuchacovich R, Espinoza LR. HLA-B27-associated reactive arthritis: pathogenetic and clinical considerations. Clin Microbiol Rev 2004;17(2):348–69. [37] Amor B. Reiter’s syndrome. Diagnosis and clinical features. Rheum Dis Clin North Am 1998;24(4):677–95 [vii].