sports medicine Volume 41 Issue 3 March 2011

Sports Med 2011; 41 (3): 177-183 0112-1642/11/0003-0177/$49.95/0

LEADING ARTICLE

ª 2011 Adis Data Information BV. All rights reserved.

Transdermal Patch Drug Delivery Interactions with Exercise Thomas L. Lenz and Nicole Gillespie Department of Pharmacy Practice, Creighton University, Omaha, Nebraska, USA

Abstract

Transdermal drug delivery systems, such as the transdermal patch, continue to be a popular and convenient way to administer medications. There are currently several medications that use a transdermal patch drug delivery system. This article describes the potential untoward side effects of increased drug absorption through the use of a transdermal patch in individuals who exercise or participate in sporting events. Four studies have been reported that demonstrate a significant increase in the plasma concentration of nitroglycerin when individuals exercise compared with rest. Likewise, several case reports and two studies have been conducted that demonstrate nicotine toxicity and increased plasma nicotine while wearing a nicotine patch in individuals who exercise or participate in sporting events compared with rest. Healthcare providers, trainers and coaches should be aware of proper transdermal patch use, especially while exercising, in order to provide needed information to their respective patients and athletes to avoid potential untoward side effects. Particular caution should be given to individuals who participate in an extreme sporting event of long duration. Further research that includes more medications is needed in this area.

1. Introduction Previous reports have demonstrated that physiological changes due to exercise can alter the pharmacokinetics of certain medications.[1] Pharmacokinetics is a discipline of pharmacology that studies the drug parameters of absorption, distribution, metabolism and elimination. When researchers design a drug, these parameters are most often studied under controlled resting conditions that do not expose the drug or the subjects to non-resting physiological situations. As a result, the absorption, distribution, metabolism and elimination data for many medications on the market are only known when patients take the medications under ‘normal’, non-stressful

conditions.[1] Exercise can have a significant effect on one or more of these pharmacokinetic parameters. The specific type and degree of effect is dependent upon the individual characteristics of each drug and the specific type and duration of exercise being performed by the patient.[1] A previously published systematic review of the pharmacokinetic changes to medications resulting from exercise summarized several commonly used medications for their drug/exercise interactions.[1] The summary showed that the serum concentrations of two b-blocking agents (atenolol and propranolol) and one antibiotic (doxycycline), increased as a result of exercise.[2-6] Also, patients who exercise after taking digoxin experience a decreased serum digoxin concentration with an

Lenz & Gillespie

178

increased skeletal muscle concentration.[7] Theophylline clearance has also been shown to decrease resulting from an increase in plasma half-life during exercise.[8] The risk of hypoglycaemia may increase when patients with diabetes mellitus inject insulin into a muscle just prior to exercising that muscle.[9] Additionally, increasing physical activity in a patient taking warfarin has been shown to decrease the international normalized ratio.[10] The pharmacokinetic parameter drug absorption can occur at a number of different sites. These sites include the gastrointestinal tract, subcutaneous, intramuscular and transdermal tissues, and in the lungs through inhalation. This article addresses the evidence published to date regarding the altered drug absorption of transdermal drug delivery patches resulting from exercise and pertinent safety information that should be provided to athletes and patients who use a transdermal patch and exercise. 2. Transdermal Drug Delivery Transdermal drug delivery represents a convenient alternative to oral drug delivery and will most likely provide an alternative to hypodermic injections in the near future.[11] Drug manu-

facturers are currently producing the third generation of transdermal drug delivery systems. Second- and third-generation systems include ultrasound and iontophoresis as well as microneedles, thermal ablation, microdermabrasion, electroporation and cavitational ultrasound, respectively.[11] The first-generation systems, however, are responsible for most of the transdermal systems in clinical use today through the use of a patch and passive absorption into the skin.[11] The first transdermal system was approved for use in the US in 1979 and delivered scopolamine to treat motion sickness via a 3-day patch.[11] Nearly 10 years later, a nicotine transdermal drug delivery patch was introduced to the market and became a highly profitable method of drug delivery that was widely accepted by the public. Medications delivered via this system are included in table I. There are two basic designs to a transdermal patch drug delivery system, a membrane-controlled reservoir system and a monolithic matrix system.[12] In the reservoir system, the four-layer patch is designed to store the medication in a liquid or gel-based reservoir that is enclosed on one side with an impermeable backing and has an adhesive that contacts the skin on the other

Table I. Transdermal patch drug delivery systems approved for use in the US[11,12] Drug

Use

Clonidine

Hypertension

Transdermal patch delivery system Reservoir

Estradiol

Hormone replacement/treatment

Reservoir and matrix

Estradiol/norethindrone

Hormone replacement/treatment

Matrix

Estradiol/levonorgestrel

Contraceptive

Matrix

Fentanyl

Pain

Reservoir and matrix

Granisetron

Nausea/vomiting associated with cancer treatment

Matrix

Lidocaine

Pain

Matrix

Methylphenidate

ADHD

Matrix

Nicotine

Tobacco cessation

Reservoir and matrix

Nitroglycerin

Angina pectoris

Reservoir and matrix

Norelgestromin/ethinyl estradiol

Contraception

Matrix

Oxybutynin chloride

Overactive bladder

Matrix

Rivastigmine

Alzheimer’s disease, Parkinson’s disease

Matrix

Scopolamine

Motion sickness

Reservoir

Selegiline

Depression

Matrix

Testosterone

Hormone replacement/therapy

Reservoir

ADHD = attention-deficit hyperactivity disorder.

ª 2011 Adis Data Information BV. All rights reserved.

Sports Med 2011; 41 (3)

Exercise and Transdermal Patch Drug Delivery

side.[11,13] This patch design employs a semipermeable membrane to control the drug absorption rate. Other transdermal patch designs incorporate the drug into a solid polymer matrix within the adhesive layer of the patch. This patch design has three layers, which eliminate the semipermeable membrane. In either case, the medication contained within the patch is passively absorbed through the skin. The absorption of the medication is designed to occur consistent with the patch design when the skin is at a normal temperature and hydration.[11,12] 3. Transdermal Drug Delivery Interactions with Exercise A limited number of studies have been performed that show the effect of exercise on drug absorption. Two of the most common medications that are formulated to be delivered via a transdermal patch are nitroglycerin and nicotine. These drugs were also among two of the first medications to use a transdermal drug delivery system. As a result, the information known to date on transdermal patch interactions with exercise focus on the research and case studies of these two medications. The patch design of these two medications, however, is consistent with other medications listed in table I. 3.1 Nitroglycerin Transdermal Patch and Exercise

In as early as 1986, a study of 12 healthy volunteers applied a 10 mg nitroglycerin transdermal patch for 6 hours on each of 3 days.[14] The 3 days consisted of a control day, an exercise day of riding a bicycle ergometer and a day where the subjects sat in a sauna for 20 minutes. The results showed that the plasma concentration of nitroglycerin increased from 1.0 to 1.5 nmol/L at rest, to 3.1 nmol/L during exercise (p < 0.001) and to 7.3 nmol/L while in the sauna (p < 0.001). The authors suggest that the increased transdermal absorption observed during exercise was a result of increased subcutaneous circulation, which could increase nitroglycerin transport from a subcutaneous reservoir.[14] ª 2011 Adis Data Information BV. All rights reserved.

179

In 1987, researchers designed a different study to evaluate the influence of moderate, sustained exercise on nitroglycerin pharmacokinetics administered via transdermal patch versus intravenous administration.[15] Six healthy male volunteers received a 10 mg nitroglycerin patch 3 hours prior to exercise or a 7 mg/min infusion of nitroglycerin 30 minutes prior to exercise, 1 week apart and in random order. The study protocol involved three consecutive 1-hour periods of rest, exercise and recovery. The exercise period of the study consisted of riding a bicycle ergometer for 1 hour where the workload progressively increased every 10 minutes. Nitroglycerin plasma levels were measured every 5 minutes during each of the last 20 minutes of the resting, exercise and recovery periods. The results showed that plasma nitroglycerin levels increased significantly during exercise with both the transdermal patch (p < 0.05) and intravenous administration (p < 0.05). Although not statistically significant, the transdermal patch increased the plasma nitroglycerin level by 93% (mean – SD 0.15 – 0.12 to 0.29 – 0.19 ng/mL) compared with a 61% (mean – SD 0.31 – 0.22 to 0.50 – 0.27 ng/mL) increase from the intravenous infusion. The researchers concluded that the increased subcutaneous blood flow resulting from the increased workload during exercise, alters the pharmacokinetic absorption of nitroglycerin.[15] In a third study, researchers again studied the effects of exercise on the absorption of nitroglycerin by comparing plasma concentrations during a resting supine position versus a resting supine position that was interrupted by a 20-minute exercise period, in nine healthy subjects during three randomized sessions separated by 1-week intervals.[16] The room temperature was kept constant (mean – SD 23 – 1C) throughout the study. The protocol consisted of sitting from 0 to 2 hours supine, from 2 to 3 hours supine, sitting or exercising from 3 hours to 3 hours 20 minutes and then supine from 3 hours 20 minutes to 4 hours. The exercise period consisted of riding a bicycle ergometer for 20 minutes at 50% maximum workload. The maximum workload was pre-determined within 2 weeks before the study. The method to determine the Sports Med 2011; 41 (3)

Lenz & Gillespie

180

workload was not stated, nor was a comparison between the participants for maximum workload. The results showed that after 15 and 20 minutes of sitting, nitroglycerin concentrations increased (p < 0.05, within treatment) but were not significantly different from that observed in the supine session. During the exercise session, plasma concentrations increased significantly (p < 0.01, within treatment) peaking 5 minutes after stopping the exercise. One hour after stopping exercise, concentration levels were still significantly higher compared with levels before exercise (p < 0.01, within treatment). The authors concluded that exercise alters the pharmacokinetics of the nitroglycerin transdermal patch delivery system independent of the subject’s postural changes.[16] Finally, a study published in 1991 looked at the effects of nitroglycerin concentrations during exercise comparing a 10 mg nitroglycerin patch, which had been worn for 24 hours versus a patch worn for just 2 hours.[17] Ten healthy subjects participated in the study, where the exercise session consisted of riding a cycle ergometer for 20 minutes with a load adjusted to give a heart rate of 110–120 beats/min. The results showed that plasma nitroglycerin levels increased by 19% (p < 0.05) after the patch was worn for 24 hours and increased by 56% (p < 0.001) after the patch was worn for 2 hours. In addition, the study showed that nitroglycerin concentration was significantly elevated (30%) when exercise was performed for a 20-minute period immediately after patch removal. The authors’ concluded that the timing of the patch placement may be an important factor when considering the pharmacokinetic absorption alterations resulting from exercise when wearing a nitroglycerin patch. It is also important to note the clinical importance of exercise-induced increases in nitroglycerin concentration even after patch removal.[17] 3.2 Nicotine Transdermal Patch and Exercise

Nicotine transdermal drug delivery systems offer tobacco-cessation patients an alternative to nicotine-containing chewing gum. In 1996, a case report was published describing three patients who experienced nicotine toxicity while wearing a ª 2011 Adis Data Information BV. All rights reserved.

nicotine patch and participating in various types of strenuous physical activity.[18] Case one described a 29-year-old female who experienced symptoms of nausea, vomiting and disorientation while playing squash just after applying her second 21 mg nicotine patch. The nicotine patch strength was appropriately dosed based on her 20–25 cigarette per day habit. The women then went for a run in the hope of eliminating the nicotine from her body. As a result, she was seen in the emergency room a short time later with tachycardia and hives on her face, arms and chest. It is unclear if these symptoms were due to nicotine drug toxicity or from an allergic reaction to the nicotine patch. The second case report describes a 28-year-old male, using a 15 mg nicotine patch. This individual experienced palpitations, nausea, chest heaviness, severe fatigue and tremor while participating in his usual karate class on the second day after starting the patch. This individual had been a smoker for 11 years and had been participating in karate for 2 years prior to this event. The third case described a 33-year-old man wearing his first 14 mg nicotine patch. He experienced symptoms of chest pain, nausea, vomiting and insomnia following an active hockey game and a hot shower.[18] In an effort to study the effects of exercise on the plasma concentrations of nicotine during the application of a nicotine patch, Klemsdal et al.,[19] enrolled eight healthy subjects with an average age of 38 years. The subjects were treated with a 14 mg nicotine patch on a control and an exercise day. After 11 hours of patch application, plasma nicotine concentrations were measured before and after exercise and after 20 minutes of rest. The exercise routine consisted of riding a cycle ergometer for 20 minutes at a heart rate of approximately 130 beats/min. The results showed that mean plasma nicotine concentrations increased from 9.8 to 11.0 ng/mL during exercise (p = 0.015) and decreased from 10.5 to 10.2 ng/mL while at rest (non-significant change). The authors concluded that the increase in blood flow to the skin while exercising caused the increase in nicotine drug absorption.[19] It should be noted that although plasma nicotine concentrations Sports Med 2011; 41 (3)

Exercise and Transdermal Patch Drug Delivery

increased at a statistically significant level, they are likely not clinically significant. Smoking just one cigarette can increase plasma nicotine concentrations above those observed in this study as a result of exercise. As previously described (in section 2), there are two different transdermal patch designs. The reservoir system, with semi-permeable layer design, is designed to control the release of the medication and is thought to be less prone to variations of skin temperature. The matrix system, however, without the semi-permeable layer design, is thought to be influenced by skin conditions such as temperature, humidity and blood flow.[20] A more recent study, published in 2005, was conducted to directly compare the two transdermal nicotine patch delivery systems on nicotine release at rest and during exercise.[20] Ten male smokers who were otherwise healthy were enrolled, and randomly received a 21 mg/day dose patch of either the reservoir or matrix systems at rest and at exercise. The exercise session consisted of a cycle ergometer test at a workload building from 50 to 150 W in a 30-minute time period that began 8 hours after attaching the nicotine patch. The results showed that both systems increased nicotine release during exercise compared with rest; however, there was not a significant difference between the two formulations with respect to the change in nicotine serum concentration.[20] These results indicate that exercise increases nicotine concentrations and that these changes occur regardless of transdermal patch design. 4. Transdermal Patch Patient Information The information presented above (in section 3) reports data collected from just two medications that are formulated to deliver drugs via a transdermal patch. To date, there are 16 different medications in 32 different brand names and generic products available in a transdermal patch for use in the US.[11,12] Published reports measuring the effects of exercise on the drug absorption of each of the medications is not available. However, the transdermal patch design of each of these medications is the same (reservoir or maª 2011 Adis Data Information BV. All rights reserved.

181

trix). Therefore, it is possible that the pharmacokinetic drug absorption of each medication delivered via a transdermal patch may be affected by exercise. From data published to date, the pharmacokinetic changes related to transdermal patch drug delivery systems with exercise are largely changes in plasma concentration levels of the drug. The clinical implications of changes in plasma concentration are drug specific and relate to the individual drug toxicity signs and symptoms. For example, toxicity signs and symptoms for nitroglycerin include hypotension, worsening angina, ischaemic ECG changes, tachycardia, arrhythmias and others.[10] The toxicity signs and symptoms for transdermal nicotine patch include gastrointestinal symptoms, increased salivation, pallor, weakness and dizziness. In addition, hypertension (at lower doses), hypotension (at higher doses), tachycardia, tachypnoea, headache and other symptoms may occur with toxic nicotine blood concentrations. Educating patients about possible toxicity signs and symptoms is prudent practice for all healthcare providers in order to treat toxicity symptoms as soon as they occur.[10] Table II. General patient information for the appropriate use of a transdermal patch[12] Apply to clean, dry, hairless, non-irritated, intact skin Do not apply to skin where lotion or creams have just been applied Do not apply to the waistline or areas where tight clothing can rub the patch off Keep away from direct heat exposure Some patches contain aluminum and should be removed prior to undergoing an MRI to prevent skin burns Remember to remove the old patch prior to applying a new patch to prevent drug overdose Most patches can be stored at room temperature but some can be stored in the refrigerator and need to be brought to room temperature before use Most patches can be disposed of in the trash but some should be flushed down the toilet (e.g. fentanyl) to prevent accidental exposure or diversion Read specific patch instructions on how to rotate patch application sites Read specific patch instructions for the appropriate application sites Read specific patch instructions for information on what to do if the patch prematurely detaches from the skin Do not cut the patch, especially the reservoir design

Sports Med 2011; 41 (3)

Lenz & Gillespie

182

Healthcare providers, fitness trainers, wellness experts, athletic trainers, coaches and others should be aware of the potential drug-exercise interaction with transdermal patch delivery systems in order to provide the appropriate information to their respective patients, athletes and clients. Patient information regarding the appropriate general use of a transdermal patch is provided in table II. Additional patient information regarding the appropriate use of a transdermal patch for those who exercise and participate in sporting events is provided in table III. Special attention should be paid to a small but distinct group of individuals who may be at a particularly high risk for drug absorption changes resulting from exercise. This group would be individuals who participate in long-duration sporting events, such as marathon running, ultramarathon running or other events in which exercise lasts for long periods of time. These individuals participate in both long-lasting events and long-lasting training sessions, which may make them particularly vulnerable to changes in skin temperature, hydration and blood flow, and increased drug absorption for long periods of time. Enhanced patient education regarding these potential interactions and proper transdermal

patch use can help prevent adverse drug reactions resulting from increased drug toxicity. 5. Conclusions Transdermal patch drug delivery systems are becoming increasingly more popular as a convenient and easy-to-use method for patients to take their medications. Several studies have reported that the absorption of medications delivered via a transdermal patch is increased as a result of exercise. Increasing drug absorption can lead to drug toxicity and untoward adverse reactions. Healthcare providers, coaches, fitness trainers and others should be aware of these potential drug-exercise interactions and provide appropriate patient education to prevent such occurrences. Special precautions should be taken for individuals who wear a transdermal patch and participate in extreme-duration sporting events. Further research should be conducted on medications that use a transdermal patch drug delivery system and that have the potential for unwanted drug toxicity such as hormone replacement medications and drugs used for pain relief (e.g. fentanyl). Acknowledgements

Table III. Exercise-specific patient information for the appropriate use of a transdermal patch Use with caution with first several applications until potential side effects are known Exercise at a lower intensity for the first 1–2 weeks until potential side effects are known Avoid exercising in extreme heat and humidity Exercise during the times of the day when it is cooler and when there is less direct sun exposure Avoid sitting in a sauna immediately after exercising Wear loose fitting clothing that will ‘breathe’ well to dissipate heat Avoid clothing materials made from plastic or rubber that encourage extra sweating Use extreme caution when performing activities that are of a long duration (>60 min) Become familiar with the toxicity signs and symptoms of the drug being administered via the patch Become familiar with the specific product information on what to do if the patch falls off due to increased sweating Tell your doctor or pharmacist if untoward side effects are experienced while wearing the patch

ª 2011 Adis Data Information BV. All rights reserved.

There were no sources of funding or conflicts of interest from either author directly or indirectly relevant to the content of this manuscript.

References 1. Lenz TL, Lenz NJ, Faulkner MA. Potential interactions between exercise and drug therapy. Sports Med 2004; 34 (5): 293-306 2. Mason WD, Kopchak G, Winer N, et al. Effect of exercise on the renal clearance of atenolol. J Pharm Sci 1980; 69: 344-5 3. Van Baak MA, Mooij JM, Schiffers PM. Exercise and the pharmacokinetics of propranolol, verapamil and atenolol. Eur J Clin Pharmacol 1992; 43: 547-50 4. Henry JA, Iliopoulou A, Kaye CM, et al. Changes in plasma concentrations of acebutolol, propranolol, and indomethocin during physical exercise. Life Sci 1981; 28: 1925-9 5. Hurwitz GA, Webb JG, Walle SA, et al. Exercise-induced increments in plasma levels of propranolol and noradrenaline. Br J Clin Pharmacol 1983; 16: 599-608 6. Mooy J, Arends B, Kemenade JV, et al. Influence of prolonged submaximal exercise on the pharmacokinetics

Sports Med 2011; 41 (3)

Exercise and Transdermal Patch Drug Delivery

7.

8.

9.

10.

11. 12.

13.

14.

of verapamil in humans. J Cardiovasc Pharmacol 1986; 8: 940-2 Joreteg T, Jorestrand T. Physical exercise and binding of digoxin to skeletal muscle: effect of muscle activation frequency. Eur J Clin Pharmacol 1984; 27: 567-70 Schlaeffer F, Engelberg I, Kaplanski J, et al. Effect of exercise and environmental heat on theophylline kinetics. Respiration 1984; 45: 438-42 Koivisto VA, Felig P. Effects of leg exercise on insulin absorption in diabetic patients. N Engl J Med 1978; 298: 79-83 Micromedex Healthcare Series [online]. Available from URL: http://www-thromsonhc-com.cuhsl.creighton.edu/ hcs/librarian. [Accessed 2010 Jun 23] Prausnitz MR, Langer R. Transdermal drug delivery. Nat Biotechnol 2008; 26 (11): 1261-8 Tom W-C. Characteristics of transdermal patches. Pharmacist’s Letter. 2008 July; 24 (7): 240711 [online]. Available from URL: http://www.phamacistsletter.com [Accessed 2010 Jun 23] Venkatraman S, Gale R. Skin adhesives and skin adhesion. 1. Transdermal drug delivery systems. Biomaterials 1998; 19: 1119-36 Barkve TF, Langseth-Manrique K, Bradesen JE, et al. Increased uptake of transdermal glyceryl trinitrate during physical exercise and during high ambient temperature. Am Heart J 1986; 112 (3): 537-41

ª 2011 Adis Data Information BV. All rights reserved.

183

15. Weber S, de Luture D, Rey E, et al. The effects of moderate sustained exercise on the pharmacokinetics of nitroglycerine. Br J Clin Pharmacol 1987; 23: 103-5 16. Lefebvre RA, Bogaert MG, Teirlynck O, et al. Influence of exercise on nitroglycerin plasma concentrations after transdermal application. Br J Clin Pharmacol 1990; 30: 292-6 17. Gjesdal K, Klemsdal TO, Rykke EO, et al. Transdermal nitrate therapy: bioavailability during exercise increases transiently after the daily change of patch. Br J Clin Pharmacol 1991; 31: 560-2 18. Health Canada. Canadian adverse drug reaction newsletter. Can Med Assoc 1996; 6 (1) 154: 61-3 19. Klemsdal TO, Gjesdal K, Zahlsen K. Physical exercise increases plasma concentrations of nicotine during treatment with a nicotine patch. Br J Clin Pharmacol 1995; 39: 677-9 20. Bur A, Joukhadar C, Klein N, et al. Effects of exercise on transdermal nicotine release in healthy habitual smokers. Int J Clin Pharmacol Ther 2005; 43 (5): 239-43

Correspondence: Dr Thomas L. Lenz, Associate Professor, Department of Pharmacy Practice, Creighton University, 2500 California Plaza, Omaha, NE 68178, USA. E-mail: [email protected]

Sports Med 2011; 41 (3)

Sports Med 2011; 41 (3): 185-197 0112-1642/11/0003-0185/$49.95/0

REVIEW ARTICLE

ª 2011 Adis Data Information BV. All rights reserved.

A Systematic Review on the Treatment of Acute Ankle Sprain Brace versus Other Functional Treatment Types Ellen Kemler,1 Ingrid van de Port,1 Frank Backx1 and C. Niek van Dijk2 1 Rudolf Magnus Institute of Neuroscience, Department of Rehabilitation, Nursing Science and Sport, University Medical Centre Utrecht, Utrecht, the Netherlands 2 Department of Orthopaedic Surgery, Academic Medical Centre, Amsterdam, the Netherlands

Contents Abstract . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1. Methods . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1.1 Literature Search . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1.2 Data Extraction and Analysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2. Results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.1 Literature Search . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.2 Best Evidence Syntheses. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.2.1 Recurrent Sprains . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.2.2 Residual Complaints (Pain, Swelling and Instability). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.2.3 Functional Outcome . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.2.4 Time to Resumption of Sports, Daily Activities and Work . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3. Discussion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4. Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

Abstract

185 187 187 188 188 188 190 190 190 192 192 193 195

Ankle injuries, especially ankle sprains, are a common problem in sports and medical care. Ankle sprains result in pain and absenteeism from work and/or sports participation, and can lead to physical restrictions such as ankle instability. Nowadays, treatment of ankle injury basically consists of taping the ankle. The purpose of this review is to evaluate the effectiveness of ankle braces as a treatment for acute ankle sprains compared with other types of functional treatments such as ankle tape and elastic bandages. A computerized literature search was conducted using PubMed, EMBASE, CINAHL and the Cochrane Clinical Trial Register. This review includes randomized controlled trials in English, German and Dutch, published between 1990 and April 2009 that compared ankle braces as a treatment for lateral ankle sprains with other functional treatments. The inclusion criteria for this systematic review were (i) individuals (sports participants as well as non-sports participants) with an acute injury of the ankle (acute ankle sprains); (ii) use of an ankle brace as primary treatment for acute ankle sprains; (iii) control interventions including any other type of functional treatment (e.g. Tubigrip, elastic wrap or ankle tape); and (iv) one of the

Kemler et al.

186

following reported outcome measures: re-injuries, symptoms (pain, swelling, instability), functional outcomes and/or time to resumption of sports, daily activities and/or work. Eight studies met all inclusion criteria. Differences in outcome measures, intervention types and patient characteristics precluded pooling of the results, so best evidence syntheses were conducted. A few individual studies reported positive outcomes after treatment with an ankle brace compared with other functional methods, but our best evidence syntheses only demonstrated a better treatment result in terms of functional outcome. Other studies have suggested that ankle brace treatment is a more costeffective method, so the use of braces after acute ankle sprains should be considered. Further research should focus on economic evaluation and on different types of ankle brace, to examine the strengths and weaknesses of ankle braces for the treatment of acute ankle sprains.

The ankle is one of the most frequently traumatized body sites and accounts for 10–30% of all sports injuries.[1] Between 2000 and 2004, 18% of all sports injuries in the Netherlands were ankle injuries.[2] As a rough estimate, one inversion ankle injury occurs per 10 000 people each day, resulting in about 5000 injuries a day in the UK and 23 000 in the US.[3-5] Most ankle injuries occurring in sports involve lateral ankle ligaments, and 77% represent ankle sprains.[1] The severity of acute ankle sprains can vary widely, and can be classified in a number of ways. Grading can be based on anatomical damage, clinical presentation, mechanism of trauma, ‘severity’ of the injury or a combination of these aspects.[6] The most frequently used terms to express severity are mild, moderate and severe, known as grade I, grade II and grade III, respectively.[7] Despite the difficulty of quantifying severity, the consequences of ankle sprains are often clear and can have great impact. Ankle sprains can cause pain and other impairments, resulting in utilization of healthcare resources and absenteeism from work and/or sports. In the Netherlands, Verhagen et al.[8] calculated that the mean total costs (direct and indirect) of one ankle sprain are approximately h360. All ankle sprains in the Netherlands cost about h43.2 million a year, with absence from paid or unpaid work responsible for up to 80% of these costs.[8] In addition to acute restrictions, ankle sprains can lead to chronic physical restrictions such as ankle instability. Chronic ankle instability not ª 2011 Adis Data Information BV. All rights reserved.

only limits physical activity, but can also lead to articular degeneration of the ankle joint and an increased risk of osteoarthritis.[9,10] According to Hubbard and Hicks-Little,[11] up to 30% of patients show objective mechanical laxity and subjective instability up to 1 year after an initial ankle sprain. Another common long-term side effect of ankle sprain is re-injury. Ekstrand and Gilquist[12] and Tropp et al.[13] found that people who have suffered an ankle sprain are more likely to injure the same ankle again. The risk of re-sprain within a period of 3 years after the initial ankle sprain ranges from 3% to 34%.[14] Residual complaints after an ankle sprain range from 6% to 78% after 8 months to 3 years of follow-up.[15-22] In the past, a variety of treatments for ankle sprains have been used, including surgical repair, plaster cast or splint immobilization and functional treatment, consisting of an early mobilization programme frequently combined with the use of an elastic bandage or brace. Kannus and Renstro¨m[23] were among the first to conclude that functional treatment should be the preferred method in cases of complete lateral ankle ligament rupture. Their findings have been corroborated by several other researchers.[24,25] Kerkhoffs et al.[24] assessed the effectiveness of various immobilization methods for acute ankle sprains and compared them with alternative conservative treatments. They found statistically significant differences for six outcome measures (return to sports, return to work, persistent swelling, objective instability, range of motion and patient satisfaction), Sports Med 2011; 41 (3)

Functional Treatment of Ankle Sprains

all in favour of functional treatment compared with cast immobilization. They concluded that functional treatment seems to be a more appropriate approach and should be encouraged. In 2002, Kerkhoffs et al.[26] assessed the effectiveness of various functional treatment strategies for acute lateral ankle ligament injuries in adults. Their findings did not enable them to indicate the most effective treatment, although a lace-up brace or a semi-rigid brace gave better results in terms of reduction of swelling and speed of recovery than bandage alone. The PRICE (Protection, Rest, Ice, Compression, Elevation) treatment protocol is commonly used for acute ankle sprain.[27] The Dutch College of General Practitioners guideline for the treatment of ankle injuries recommends treatment consisting of ICE (Immobilization, Compression and Elevation) during the first week, followed by ankle taping for 6 weeks. Thereafter, sports participants are advised to use an ankle brace while engaging in sports to prevent recurrences.[28] Ankle sprains are very commonly treated with ankle tape. According to the results of previous studies, functional treatment is most effective in acute ankle sprain injuries.[24] Another type of functional therapy might be the use of an ankle brace. It is well known that an ankle brace effectively prevents recurrence of ankle sprains.[29,30] Despite convincing results on prevention, however, braces are rarely used in an earlier stage as a treatment for ankle sprains. Although some studies have investigated the use of braces in the acute stage after injury, none have systematically evaluated whether the use of an ankle brace is a more appropriate treatment for ankle sprains than other forms of protection during functional treatment. The purpose of this review is to evaluate the effectiveness of ankle braces as a treatment method for acute ankle sprains compared with other types of functional treatment (e.g. ankle tape, Tubigrip). 1. Methods 1.1 Literature Search

A computerized literature search was conducted using PubMed, EMBASE, Cumulative ª 2011 Adis Data Information BV. All rights reserved.

187

Index to Nursing and Allied Health Literature (CINAHL) and the Cochrane Clinical Trial Register (CCTR). Randomized controlled trials (RCTs) in English, German and Dutch, published between 1990 and April 2009 that compared ankle braces as a treatment for ankle sprains with other functional treatments were included in this review. Inclusion from 1990 onwards was chosen because functional treatment (tape, bandage or brace) of ankle sprains had been recommended since the early 1990s.[23] Keywords used in this search were ‘ankle brace’, ‘random’, ‘ankle injury’, ‘treatment’, ‘ankle sprain’, ‘ankle trauma’ and ‘inversion ankle injury’. MeSH terms were ‘clinical trials’ and ‘random allocation’, and the MeSH subheading used was ‘therapeutic use’. The methodological filter used in PubMed was therapy, broad sensitive search. In addition, reference lists of included articles were reviewed for potentially valid studies. The complete search strategy is available from the authors. Studies were selected by two reviewers (EK and IP) on the basis of title and abstract. The following criteria for inclusion in this systematic review were used to select randomized or quasi-RCTs: (i) individuals (sports participants as well as non-sports participants) with an acute injury of the ankle; (ii) use of an ankle brace as a primary treatment for acute ankle sprains; (iii) control intervention including any other type of functional treatment (e.g. Tubigrip, elastic wrap or ankle tape); and (iv) one of the following reported outcome measures: re-injuries, residual complaints (pain, swelling, instability), functional outcomes and/or time to resumption of sports, daily activities and/or work. An acute ankle injury was considered an acute ankle sprain, which was defined as a joint injury in which some of the fibres of a supporting ligament are ruptured but the continuity of the ligament remains intact (MeSH). The cause of the injury needed to be acute, which implies a clear onset of injury as a result of trauma (e.g. from tackling, kicking or jumping). Trials aimed at the treatment of, for example, chronic ankle instability or ankle fractures were excluded. The control intervention included functional treatment, which was defined as treatment consisting of Sports Med 2011; 41 (3)

Kemler et al.

188

therapy (supervised or unsupervised) during which the patients conduct functional exercises with the ankle, such as flexion/extension (against resistance) and walking. While conducting the exercises, the patients can wear Tubigrip, elastic wrap or ankle tape, but no ankle brace. Studies comparing treatment with an ankle brace solely with cast immobilization were excluded and no restrictions were used for the follow-up period. A brace was defined as an orthopaedic appliance used to support, align or hold a bodily part (i.e. the ankle) in the correct position.[31] The methodological quality of each study was assessed by two reviewers (EK and IP) using the PEDro scale[32] (table I). The PEDro scale is an 11-item scale designed to rate the methodological quality of RCTs, and is sufficiently reliable for use in systematic reviews.[33] The PEDro scale is used to identify the external (item 1) and internal validity (criteria 2–9), and the amount of statistical information provided to make the results interpretable (criteria 10–11). The maximum score for the PEDro scale is 10 points, since item 1 is not included in the calculation of the total PEDro score. In case of disagreement between the two reviewers, consensus was achieved by discussion. Studies with 4 points or more on the PEDro scale are considered to be of high methodological quality, while those with 3 points or less are considered to be of low quality.[34] Kappa was used to measure the agreement between the two reviewers. When agree-

ment is perfect, Kappa is 1.00. The interpretation of the other values is as follows: (i) <0.20 poor strength of agreement; (ii) 0.21–0.40 fair strength of agreement; (iii) 0.41–0.60 moderate strength of agreement; (iv) 0.61–0.80 good strength of agreement; and (v) 0.81–1.00 very good strength of agreement.[35] 1.2 Data Extraction and Analysis

Data on all relevant outcome measures were extracted by one reviewer (EK). If possible, results of comparable studies were pooled; otherwise, a best evidence synthesis was prepared. The level of evidence was determined using the best evidence synthesis criteria proposed by Steultjens et al.,[36] based on the methodological quality score on the PEDro scale.[34] Steultjens et al.[36] modified the criteria proposed by Van Tulder et al.[37] The results of the best evidence syntheses were categorized into five levels of evidence, depending on the quality of the study and the statistical significance of the results (p < 0.05) [i.e. strong evidence, moderate evidence, limited evidence, indicative findings and no or insufficient evidence (table II)]. 2. Results 2.1 Literature Search

The results of the database search and subsequent assessment of trials identified are summarized in figure 1.

Table I. The PEDro scale[32]a Criteria

Yes

No

1. Eligibility criteria were specified

1

0

2. Subjects were randomly allocated to groups (in a crossover study, subjects were randomly allocated an order in which treatments were received)

1

0

3. Allocation was concealed

1

0

4. The groups were similar at baseline regarding the most important prognostic indicators

1

0

5. There was blinding of all subjects

1

0

6. There was blinding of all therapists who administered the therapy

1

0

7. There was blinding of all assessors who measured at least one key outcome

1

0

8. Measures of at least one key outcome were obtained from >85% of the subjects initially allocated to groups

1

0

9. All subjects for whom outcome measures were available received the treatment or control condition as allocated or, where this was not the case, data for at least one key outcome were analysed by ‘intent to treat’

1

0

10. The results of between-group statistical comparisons are reported for at least one key outcome

1

0

11. The study provides both point measures and measures of variability for at least one key outcome

1

0

a

Total score is calculated using items 2–11 (range 0–10).

ª 2011 Adis Data Information BV. All rights reserved.

Sports Med 2011; 41 (3)

Functional Treatment of Ankle Sprains

189

Table II. Best evidence synthesis[34,36] Evidence level

Definition

Strong evidence

Provided by statistically significant findings in outcome measures in at least two high-quality RCTs, with PEDro scores of at least 4 pointsa

Moderate evidence

Provided by statistically significant findings in outcome measures in at least one high-quality RCT and at least one low-quality RCT (£3 points on PEDro) or one high-quality CCTa

Limited evidence

Provided by statistically significant findings in outcome measures in at least one high-quality RCTa or at least two high-quality CCTsa (in the absence of high-quality RCTs)

Indicative findings

Provided by statistically significant findings in outcome measures in at least one high-quality CCT or low-quality RCTa (in the absence of high-quality RCTs), or two studies of a non-experimental nature with sufficient quality (in the absence of RCTs and CCTs)a

No or insufficient evidence

In the case where the results of eligible studies do not meet the criteria for one of the above-stated levels of evidence or in conflicting (statistically significant positive and statistically significant negative) results among RCTs and CCTs, or where there are no eligible studies

a

If the number of studies showing evidence is <50% of the total number of studies found within the same category of methodological quality and study design (RCT, CCT or non-experimental studies), the classification of ‘no evidence’ is used.

CCT(s) = controlled clinical trial(s); RCT(s) = randomized controlled trial(s).

Eight articles met the inclusion criteria for this review. Differences in outcome measures, intervention types and patient characteristics precluded pooling of the results, and only best evidence analyses were conducted. In total, 1250 patients (both sports participants and non-sports participants) were involved in the included studies, 358 of whom were treated with an ankle brace. One study failed to report

the number of patients in the treatment group.[58] The age of participants ranged from 9 to 61 years, and 63% were male (two studies did not classify sex[58,59]). The severity of the ankle sprains varied between grade I and grade III. Boyce et al.,[60] Karlsson et al.,[61] Leanderson and Wredmark[62] and Dettori et al.[63] focussed on moderate to severe sprains (grade II and III), while Beynnon et al.[58] included only first-time grade I, II and

Search

MEDLINE (n = 439)

Articles excluded after reading titles, due to (n = 585): • no RCT • other subject than ankle sprains Articles excluded after reading abstract, due to (n = 40): • no RCT • prevention instead of treatment • no brace as treatment

EMBASE (n = 26)

CINAHL (n = 48)

Potentially relevant articles identified and screened for retrieval (n = 654) Articles retrieved for abstract screening (n = 69) Full articles retrieved for detailed evaluation (n = 29) RCTs included in the systematic review (n = 8)

CCTR (n = 141)

Articles excluded from systematic review, due to (n = 21): • no functional treatment in control group1 • no brace as treatment2 • review article/no RCT3 • no full text available4 • controls also treated with brace5 • design study6 • second article about same study7

Fig. 1. Total numbers of articles and randomized controlled trials (RCTs) through the different stages of the systematic review. 1 = studies[20,38-42]; 2 = [43-47]; 3 = [48-53]; 4 = [54]; 5 = [55]; 6 = [56]; and 7 = [57]. CCTR = Cochrane Clinical Trial Register; CINAHL = Cumulative Index to Nursing and Allied Health Literature.

ª 2011 Adis Data Information BV. All rights reserved.

Sports Med 2011; 41 (3)

Kemler et al.

190

III sprains. Twellaar et al.[64] made no distinction and included all ankle sprains, while Neumann et al.[59] and Lamb et al.[65] focussed on grade III ankle sprains. Four studies compared an ankle brace (two Aircast, one Air-stirrup and one semi-rigid and specially designed compression pad) with an elastic wrap or bandage.[59-62] Dettori et al.[63] compared an ankle brace with an elastic wrap and a cast, while Twellaar et al.[64] compared a confection brace (Push Med) with adhesive, non-elastic tape. Beynnon et al.[58] investigated the effect of an ankle brace compared with an Aircast brace combined with a wrap and a cast. Lamb et al.[65] compared the Tubigrip with the Aircast brace, a below-knee cast and a Bledsoe boot. Different outcome measures were used (table III). The methodological quality of the eight included studies, as expressed by the PEDro scores varied between 3 and 7 with a median of 5 points. It was not possible to allocate points for items 5 and 6, because blinding of patients and therapists was impossible in the treatment strategies used. Seven trials were scored as high quality; one trial was scored as low quality (table IV). The initial agreement between the two reviewers was 86% (76 of 88 items). The Kappa value for the measure of agreement between the two reviewers on the individual validity items was 0.73. 2.2 Best Evidence Syntheses

Best evidence syntheses were prepared by extracting data for the following outcomes: recurrent sprains, residual complaints (pain, swelling and instability), functional outcome and time to resumption of sports, daily activities and work. 2.2.1 Recurrent Sprains

Four high-quality studies measured the number of re-injuries. Beynnon et al.[58] assessed the total number of re-injuries after 6 months in different treatment groups per injury grade. Neumann et al.[59] assessed the total number of re-injuries after 1 year, while Dettori et al.[63] measured the rate of re-injuries during a 5-week period after injury. Twellaar et al.[64] assessed how many patients had re-injured their ankle between 1.8 and 2.8 years after injury. The follow-up periods used ª 2011 Adis Data Information BV. All rights reserved.

in these studies were different, but none of the four studies found any differences in re-injuries between the groups they studied; therefore, ‘no evidence’ was classified. 2.2.2 Residual Complaints (Pain, Swelling and Instability) Pain

Four high-quality studies measured pain as an outcome variable. After 1.8–2.8 years of followup, Twellaar et al.[64] examined the ankles of patients who had sustained an ankle injury. Pain upon palpation was less frequent in the brace group than in the tape group (20% vs 47%; p < 0.05). Boyce et al.[60] measured the difference in pain scores on a visual analogue scale from 0 to 10, between the initial presentation and day 10. They found no significant difference (p = 0.07). Beynnon et al.[58] measured the period of time from the patient’s onset of injury until they no longer experienced pain during weight bearing exercises. The differences between the groups were not significant. Finally, Dettori et al.[63] classified the pain that patients perceived in different phases of the study from ‘pain at rest’ (4+) to ‘no pain with running’ (0), and found no differences in effect between wraps and braces. Only one of the four studies found significant results, so ‘no evidence’ was classified. Swelling

Four high-quality studies investigated swelling as an outcome variable. Boyce et al.[60] measured the difference in swelling, in millimetres, between the injured and uninjured ankles on day 10, comparing an elastic bandage group with an Aircast brace group. Twellaar et al.[64] and Neumann et al.[59] determined the percentage of patients with swelling within the treatment groups; Twellaar et al. after an average of 2.3 years of follow-up and Neumann et al. after 1 year of follow-up. Dettori et al.[63] measured swelling by volumetry after 5 weeks. None of the studies found statistically significant differences in terms of this outcome measure. Based on these findings, ‘no evidence’ was classified. Instability

Two high-quality studies measured ankle instability as an outcome variable. Twellaar et al.[64] Sports Med 2011; 41 (3)

Study (year)

Sample size (n); sex

Age (y) [range]

Twellaar 116; M 77, F 39 Mean 32 et al.[64] (1993)

Participants

Follow-up

83% SP in the Long term tape group and 1.8–2.8 y 73% SP in the brace group

Neumann 80; NR et al.[59] (1994)

SRO: mean Percentage of 25.2 [16–46]; SP not reported EB: mean 23.2 [11–50]

Dettori 64; M 60, F 4 et al.[63] (1994)

Mean 24.2–25.9

Injury severity

Sports-related injuries

Treatment (n) Control (n)

All ankle injuries

Percentage of sports injuries NR

Confection brace (53)

Adhesive, non- Functional stability; swelling; elastic tape blisters; pain – investigating ankle (63) is painful in brace group: 20% vs 47% in tape group; p < 0.05

SRO (31)

EB (33)

Questionnaire with 88-point scale

Elastic wrap (22); cast (18)

Number of days before return to full duty; ankle swelling; pain

Grade III Percentage of Short term sports injuries (3, 10 d, NR 3, 6 wk, 4 mo); intermediate term (12 mo)

AB (24) All injuries occurred during military activities

Relevant outcomes with significant results

All were military Short term (5 wk) personnel

Moderate and severe injuries

SP + NSP

Short term (24 h, 3–5 d, 2, 4, 10 wk)

Grades II 48% sports and III injuries

ASB (39)

CB (34)

Karlsson’s scoring scale Recording sick leave: ASB 5.3 (0–26 days) vs CB 9.1 (0–21 days); p < 0.05; clinical examination with regard to localized tenderness, degree of swelling and range of plantar flexion-dorsiflexion of the ankle

86; 57 M, 29 F Mean 22 Karlsson [16–38] et al.[61] (1996)

SP

Long term (12–24 mo, mean 18 mo)

Grades II Percentage of and III sports injuries NR

Functional treatment, with specially designed compression pads, weightbearing and range-ofmotion training (45)

Conservative treatment with elastic wrapping, partial weight bearing and crutches (39)

Karlsson’s scoring scale Return to sports activities – functional treatment with compression pads: 9.6 – 4.8 d vs 19.2 – 9.5 days for conventional treatment with elastic wrapping (p < 0.05); reported d of sick leave – functional treatment with compression pads: 5.6 – 4.2 d vs 10.2 – 6.8 d for conventional treatment with elastic wrapping (p < 0.05)

Boyce et al.[60] 50, 12 default, (2005) 3 excluded (1 lateral malleolus,

SP + NSP

Short term (10 d, 1 mo)

Moderate 38% sports or severe injuries

AB (18)

ESB (17)

Karlsson’s scoring scale: (i) after 10 d: ESB mean = 35 vs AB mean = 50; 95% CI 1.7, 27.7; p = 0.028; (ii) after 1 mo: ESB

Leanderson and Wredmark[62] (1995)

73; M 48, F 25 Mean 28 [15–55]

Continued next page

191

Sports Med 2011; 41 (3)

>16 [16–58]

Functional Treatment of Ankle Sprains

ª 2011 Adis Data Information BV. All rights reserved.

Table III. Characteristics of trials included

Kemler et al.

Lamb et al.[65] (2009)

AB = Aircast brace; ASB = Air-stirrup brace; CB = compression bandage; EB = elastic bandage; ESB = elastic support bandage; F = female; FAOS = foot and ankle outcome score; M = male; NR = not reported; NSP = non-sports participants; SP = sports participants; SRO = semi-rigid orthosis.

FAOS including: assessments of Below-knee pain, symptoms, activities of daily cast (119); living, sport and quality of life Bledsoe boot (148); AB (148) Tubigrip (140) Percentage of sports injuries NR Severe ankle sprains SP NR Mean 30

ª 2011 Adis Data Information BV. All rights reserved.

584; M 337, F 247

Short term, intermediate term (6 mo) SP + NSP 1 wrong clinical appointment given and 1 foot injury; ESB M 11, AB M 10

212; NR Beynnon et al.[58] (2006)

[16–61]

Intermediate term (3 and 9 mo)

Grade I: 34% AB (NR) sports injuries; grade II: 39% sports injuries; grade III: 71% sports injuries First-time grade I, grade II and grade III

Sports-related injuries Injury severity Follow-up Participants Age (y) [range] Sample size (n); sex Study (year)

Table III. Contd

Number of d required to return to: no pain during weight bearing, full capability in normal daily activities, full capability at work or school, full capability in usual athletic or recreational physical activity; Karlsson’s scale; re-injuries Elastic wrap (not reported); AB with wrap (NR); cast (NR)

mean = 55 vs AB mean = 68; 95% CI 1.4, 24.8; p = 0.029; pain score; ankle girth difference

Treatment (n) Control (n)

Relevant outcomes with significant results

192

measured mechanical instability and found no significant differences within the treatment groups. Neumann et al.[59] measured subjective instability, but reported no information about significance. We classified ‘no evidence’ for the effect on ankle instability. 2.2.3 Functional Outcome

A total of five studies measured functional outcome, four of high quality and one of low quality. Lamb et al.[65] measured the quality of ankle function using the Foot and Ankle Outcome Score (FAOS). At 3 months, there were significant clinical benefits for the Aircast brace compared with the Tubigrip in terms of the quality of ankle function. No differences were found at 1 and 9 months. Four studies used Karlsson’s scoring scale (or a modified version of it). Boyce et al.[60] used this scale on day 10 and at 1 month after the trauma. In both measurements, the mean score in the patient group treated with an ankle brace was higher (i.e. better) than that in the group treated with an elastic bandage (p = 0.028 on day 10 and p = 0.029 after 1 month). Beynnon et al.[58] used the same scoring scale 6 months after the onset of injury, and found no significant differences between the treatment groups; neither did Karlsson et al.[61] who used the scale after 12–24 months of follow-up. Leanderson and Wredmark[62] applied the scale at 3–5 days, 2, 4 and 10 weeks after the initial injury and did not find any significant differences between the patient group treated with an ankle brace and the group treated with a compression bandage. Since the results of two of the four high-quality studies were significant, the evidence was classified as strong. 2.2.4 Time to Resumption of Sports, Daily Activities and Work Sports

Three high-quality studies used ‘return to sports’ as an outcome measure.[58,59,61] Karlsson et al.[61] found that patients treated with a brace returned to sports activities significantly sooner than patients treated with elastic wrapping (9.6 – 4.8 days vs 19.2 – 9.5 days; p < 0.05). Beynnon et al.[58] and Neumann et al.[59] found no statistical differences between braces and other functional treatments. Sports Med 2011; 41 (3)

Functional Treatment of Ankle Sprains

193

Because the number of studies showing evidence in favour of braces was <50% of the total number of studies in this category, there is no evidence to support the use of braces.

found evidence in favour of braces was <50% of the total number of studies in the same category, so ‘no evidence’ was classified. 3. Discussion

Daily Activity

One high-quality study used ‘return to daily activity’ as an outcome variable, expressed as the number of days it took for patients to fully resume daily activity.[58] Since the study showed no significant results, there is no evidence to show that the treatment of an ankle injury with an ankle brace reduces the number of days necessary to achieve full return to daily activities more quickly than treatment with a bandage or Tubigrip. Work (Sick Leave)

Six studies measured how many days the patients were unable to work. One study was rated as low quality,[62] while five were of high quality.[58,59,61,63,64] The results reported by Leanderson and Wredmark[62] indicated that treatment with an ankle brace significantly reduced the duration of sick leave when compared with treatment with an elastic wrap/compression bandage. They found a mean difference of 4 days (5.3 days [range 0–26 days] vs 9.1 days [range 0–21 days]; p < 0.05). Of the five high-quality studies, only one reported significant results. Karlsson et al.[61] found a difference of approximately 5 days (95% CI 5.6 – 4.2 vs 95% CI 10.2 – 6.8; p < 0.05) in favour of the ankle brace. Although the results of two studies (one high and one low quality) were significant, the total number of studies that

This systematic review examines the available evidence for the effectiveness of ankle braces as a treatment for ankle sprains compared with other functional treatment methods such as Tubigrip, elastic bandages and ankle tape. The validity score of the eight trials that were included ranged from 3 to 7 points. Seven trials were classified as high quality (PEDro score ‡4) and one trial as low quality (PEDro score £3). Heterogeneity of study designs precluded pooling. Strong evidence in favour of treatment with a brace was found for functional outcome, which is a clinically relevant result, because patients who had a good functional outcome might be able to return to their normal functioning levels sooner. The favourable results for functional outcome were measured on two different scales (Karlsson’s scoring scale and the FAOS), with different periods of follow-up. Although the evidence for functional outcome in favour of the ankle brace was classified as strong, the difference between the classifications of strong evidence and no evidence is small and depends on the chosen cut-off point for methodological quality. In total, five studies measured functional outcome, four of high quality and one of low quality. It should be noted that the difference between strong evidence and no evidence

Table IV. Validity assessment of trials included (published between 1990 and April 2009) Study (year)

PEDro items 1

Twellaar et al.

[64]

(1993)

2

3

4

5

6

7

8

9

10

11

Total scorea

Y

1

0

0

0

0

0

0

1

1

1

4

Neumann et al.[59] (1994)

N

1

0

1

0

0

0

0

1

1

1

5 5

Dettori et al.[63] (1994)

Y

1

0

1

0

0

1

1

0

1

0

Leanderson and Wredmark[62] (1995)

N

0

1

0

0

0

0

0

0

1

1

3

Karlsson et al.[61] (1996)

N

0

1

0

0

0

0

1

0

1

1

4

Boyce et al.[60] (2005)

Y

1

1

1

0

0

0

0

1

1

1

6

Beynnon et al.[58] (2006)

Y

1

1

1

0

0

0

0

1

1

1

6

Lamb et al.[65] (2009)

Y

1

1

1

0

0

1

0

1

1

1

7

a

Total score calculated using items 2–11 (range 0–10).

N = no; Y = yes.

ª 2011 Adis Data Information BV. All rights reserved.

Sports Med 2011; 41 (3)

194

was only one point on the PEDro scale. Although the cut-off points are well considered, it is always good to keep in mind the small step from no evidence to strong evidence. Return to work was another important outcome included in this review. In contrast to what was reported by Kerkhoffs et al.,[26] the best evidence synthesis did not show positive results for treatment with a brace. Based on two studies in which elastic bandages were compared with semirigid ankle braces,[61,62] Kerkhoffs et al. concluded that the use of a semi-rigid ankle brace appeared to be associated with a quicker return to work. Due to the use of broader definitions and the inclusion of recent studies our review includes six studies in which ankle braces were compared with other functional treatment methods and sick leave/return to work was used as an outcome variable.[58,59,61-64] Two of the studies,[61,62] which were also included in Kerkhoffs et al., reported a difference in the duration of sick leave of 4–5 days in favour of the ankle brace. However, because the total number of studies reporting evidence was <50%, we classified this as ‘no evidence’. Several of the studies included in this review reported results in favour of ankle braces with regard to pain,[64] time to resumption of sports[61] and sick leave.[61,62] However, the findings for these outcomes were classified as ‘no evidence’ for brace treatment compared with other functional treatments after acute ankle sprain. This might be caused by the heterogeneity of the trials. First, studies in our review included patients with different activity levels, some of them being sports participants and some not. This might have influenced the results, as the outcome of treatment of lateral ankle ligament rupture is significantly affected by the patients’ activity level. For instance, ankle instability and re-injuries are more common in the high-activity groups than in lowactivity groups.[66] Second, we used a broad definition of ankle brace, resulting in the inclusion of articles in which different types of braces with different properties and qualities were used. According to Kerkhoffs et al.,[26] lace-up ankle support appears to reduce swelling more effectively in the short term, compared with semi-rigid ankle support, elastic bandage ª 2011 Adis Data Information BV. All rights reserved.

Kemler et al.

and tape, and rigid braces give more mechanical support than soft braces. Our review did not consider the specific characteristics of the different ankle braces. Significant favourable results were found for the soft brace, the compression pads and for the Aircast brace. A restriction to the Aircast brace would not have changed the results of the review. In contrast, the exclusion of the semi-rigid Aircast ankle brace from our review would have limited the number of studies to two and strong evidence could have been classified for three outcome variables. Furthermore, not every brace might be appropriate for the treatment of ankle sprains.[67] Vaes et al.[67] evaluated the stabilizing effect of external support (nine different ankle braces and taping) in functionally unstable ankles. They used a standard surface electromyogram-controlled stress Roentgen test protocol to measure the talar tilt with and without external support. Two braces had a major influence on the talar tilt, and only these two (Aircast standard brace and Step-in safety brace) were considered to be effective for immediate post-trauma treatment and the prevention of ankle sprain. Despite this conclusion, several studies included in this review[58,61,63,64] reported the opposite or were unable to uphold this conclusion. Although our review examined several outcomes, other endpoints might also be valuable when deciding upon the preferred treatment method. Verhagen et al.[8] calculated that the mean total costs (direct and indirect) of one ankle sprain in the Netherlands to be approximately h360, and that all ankle sprains in the Netherlands together cost about h43.2 million a year. An effective treatment method could reduce the costs of ankle sprains. In a recent study, Cooke et al.[68] evaluated the cost effectiveness of several methods of ankle injury treatment, performing an economic evaluation of the below-knee cast, Aircast brace and Bledsoe boot versus Tubigrip. Cost-utility analysis, comparing incremental costs with the differential impact on health-related quality of life over 9 months, demonstrated that the Aircast brace (d301 per quality-adjusted life-year [QALY]) and below-knee cast (d339 per QALY) were more cost effective than the Bledsoe boot (d2116 per Sports Med 2011; 41 (3)

Functional Treatment of Ankle Sprains

QALY). Cost-effectiveness acceptability curves confirmed that the Bledsoe boot was the least cost effective, and that the Aircast brace and below-knee cast had a comparably higher cost effectiveness; Cooke et al.[68] only compared an ankle brace with Tubigrip. More research is needed to examine the cost effectiveness of ankle braces compared with other functional treatment types such as the widely used ankle tape. Some limitations need to be considered when interpreting the results. We recognize the possibility of publication bias, as we did not systematically search unpublished studies for potentially relevant research. In addition, the literature search was restricted to articles in English, German or Dutch, although no trials were excluded because of language. 4. Conclusions International guidelines for the treatment of ankle sprains are rare. According to the Dutch College of General Practitioners guideline for the treatment of ankle sprains, these should be treated by taping for 6 weeks. Nevertheless, our systematic review found evidence for a better functional outcome when using an ankle brace, and treatment with ankle braces did not show any unfavourable effects. In addition, findings of other studies suggest that this method is more cost effective; therefore, using the brace for treatment of acute ankle sprains should be considered. Research focussing on the effectiveness and cost effectiveness of ankle braces used as a therapeutic method could produce more convincing evidence, allowing the following guidelines to be developed and adapted: 1. In terms of functional outcomes, ankle braces are more effective in the treatment of acute ankle sprains than other types of functional treatment. 2. Compared with other functional treatments, ankle braces are not less effective in the treatment of acute ankle sprains. 3. More research is needed for well defined functional treatment methods. Research should not only focus on functional outcomes, but also on socioeconomic outcomes and on different types of ankle braces. This should provide more ª 2011 Adis Data Information BV. All rights reserved.

195

information on the strengths and weaknesses of ankle braces as a treatment for acute ankle sprains compared with other types of functional treatment. Acknowledgements No sources of funding were used to assist in the preparation of this review. The authors have no conflicts of interest that are directly relevant to the content of this review.

References 1. Fong DT, Hong Y, Chan LK, et al. A systematic review on ankle injury and ankle sprain in sports. Sports Med 2007; 37 (1): 73-94 2. Schmikli SL, Kemler HJ, Backx FJG. Blessureleed in de sport 2000-2004. In: Hildebrandt VH, Ooijendijk WTM, Hopman-Rock M, editors. Trendrapport Bewegen en Gezondheid 2000/2005. Leiden: TNO Kwaliteit van Leven, 2007 3. Brooks SC, Potter BT, Rainley JB. Treatment for partial tears of the lateral ligament of the ankle: a prospective trial. Br Med J 1981; 282: 606-7 4. McCulloch PG, Holden P, Robson DJ, et al. The value of mobilisation and non-steroidal anti-inflammatory analgesia in the management of inversion injuries of the ankle. Br J Clin Pract 1985 Feb; 39 (2): 69-72 5. Ruth CJ. The surgical treatment of injuries of the fibular collateral ligaments of the ankle. J Bone Joint Surg 1961 Mar; 43-A: 229-39 6. Mann G, Nyska M, Constantini N, et al. Mechanics of injury, clinical presentation and staging. In: Nyska M, Mann G, editors. The unstable ankle. Champaign (IL): Human Kinetics, 2002: 54-1 7. Renstro¨m P, Lynch SA. Management of acute ankle sprains. In: Nyska M, Mann G, editors. The unstable ankle. Champaign (IL): Human Kinetics, 2002: 168-78 8. Verhagen E, van Tulder M, van der Beek AJ, et al. An economic evaluation of a proprioceptive balance board training program for the prevention of ankle sprains in volleyball. Br J Sports Med 2005; 39 (2): 111-5 9. Harrington KD. Degenerative arthritis of the ankle secondary to long-standing lateral ligament instability. J Bone Joint Surg Am 1979; 61 (3): 354-61 10. Krips R, Brandsson S, Swensson C, et al. Anatomical reconstruction and Evans tenodesis of the lateral ligaments of the ankle: clinical and radiological findings after follow-up for 15 to 30 years. J Bone Joint Surg Br 2002; 84 (2): 232-6 11. Hubbard TJ, Hicks-Little CA. Ankle ligament healing after an acute ankle sprain: an evidence-based approach. J Athl Train 2008 Sep-Oct; 43 (5): 523-9 12. Ekstrand J, Gilquist J. Soccer injuries and their mechanisms: a prospective study. Med Sci Sports Exerc 1983; 15 (3): 267-70 13. Tropp H, Askling C, Gilquist J. Prevention of ankle sprains. Am J Sports Med 1985; 13 (4): 259-61 14. van Rijn RM, van Os AG, Bernsen RM, et al. What is the clinical course of acute ankle sprains? A systematic literature review. Am J Med 2008 Apr; 121 (4): 324-31.e6

Sports Med 2011; 41 (3)

196

15. Prins JG. Diagnosis and treatment of injury to the lateral ligaments of the ankle. Acta Chir Scand Suppl 1978; 486: 65-137 16. Hedges JR. Mangement of ankle sprains. Ann Emerg Med 1980; 9 (6): 296-302 17. Schaap GR, de Keizer G, Marti RK. Inversion trauma of the ankle. Arch Orthop Trauma Surg 1989; 108: 273-5 18. Cetti R. Conservative treatment of injury to the fibular ligament of the ankle. Br J Sports Med 1982; 16: 47-52 19. Korkala O, Rusanen M, Kyto¨maa J, et al. Treatment of lateral ligament injuries of the ankle: a prospective clinical study [abstract]. Acta Orthop Scand 1986; 57 (6): 579 20. Zwipp Z, Hoffmann R, Thermann H, et al. Rupture of the ankle ligaments. Int Orthop 1991; 15 (1): 245-9 21. van Beek P. Evaluation of ankle injuries using the Cybex II Dynamometer [abstract]. Acta Orthop Scand 1985; 56 (6): 516 22. Brink PRG, de Vette J, Wever J, et al. Treatment of 176 ankle ligament ruptures by taping; results after 2-3 years [abstract]. Acta Orthop Scand 1985; 56 (6): 515 23. Kannus P, Renstro¨m P. Treatment for acute tears of the lateral ligaments of the ankle: operation, cast, or early controlled mobilization. J Bone Joint Surg Am 1991; 73: 305-12 24. Kerkhoffs GMMJ, Rowe BH, Assendelft WJJ, et al. Immobilisation for acute ankle sprain: a systematic review. Arch Orthop Trauma Surg 2001; 121: 462-71 25. Jones MH, Amendola AS. Acute treatment of inversion ankle sprains: immobilization versus functional treatment. Clin Orthop Relat Res 2007 Feb; 455: 169-72 26. Kerkhoffs GM, Struijs PA, Marti RK, et al. Different functional treatment strategies for acute lateral ankle ligament injuries in adults. Cochrane Database Syst Rev 2002; (3): CD002938 27. The Chartered Society of Physiotherapy. PRICE guidelines: guidelines for the management of soft tissue (musculoskeletal) injury with Protection, Rest Ice, Compression and Elevation (PRICE) during the first 72 hours (ACPSM) [online]. Available from URL: http://www.csp.org.uk/di rector/members/libraryandpublications/csppublications.cfm? item_ID=74C876060B1578226A7B4D0CED3866BA [Accessed 2010 Dec 7] 28. NHG (Dutch General Practitioner College). Guideline on ankle sprain [online; in Dutch]. Available from URL: http://nhg.artsennet.nl/kenniscentrum/k_richtlijnen/k_nhg standaarden/Samenvattingskaartje-NHGStandaard/M04_ svk.htm [Accessed 2010 Dec 7] 29. Handoll HHG, Rowe BH, Quinn KM, et al. Interventions for preventing ankle ligament injuries. Cochrane Database Syst Rev 2001; (3): CD000018 30. Verhagen EA, van Mechelen W, de Vente W. The effect of preventive measures on the incidence of ankle sprains. Clin J Sport Med 2000 Oct; 10 (4): 291-6 31. Definition of ankle brace [online]. Available from URL: http://www.thefreedictionary.com/brace [Accessed 2010 Dec 7] 32. Sherrington C, Herbert RD, Maher CG, et al. PEDro: a database of randomized trials and systematic reviews in physiotherapy. Man Ther 2000 Nov; 5 (4): 223-6 33. Maher CG, Sherrington C, Herbert RD, et al. Reliability of the PEDro Scale for rating quality of randomized controlled trials. Phys Ther 2003 Aug; 83 (8): 713-21 34. van Peppen RP, Kwakkel G, Wood-Dauphinee S, et al. The impact of physical therapy on functional outcomes after stroke: what’s the evidence? Clin Rehabil 2004 Dec; 18 (8): 833-62

ª 2011 Adis Data Information BV. All rights reserved.

Kemler et al.

35. Altman DG. Practical statistics for medical research. London: Chapman & Hall, 1999 36. Steultjens EM, Dekker J, Bouter LM, et al. Occupational therapy for stroke patients: a systematic review. Stroke 2003 Mar; 34 (3): 676-87 37. van Tulder MW, Cherkin DC, Berman B, et al. The effectiveness of acupuncture in the management of acute and chronic low back pain: a systematic review within the framework of the Cochrane Collaboration Back Review Group. Spine 1999; 24: 1113-23 38. Dettori JR, Basmania CJ. Early ankle mobilization, part II: a one-year follow-up of acute, lateral ankle sprains (a randomized clinical trial). Mil Med 1994; 159 (1): 20-4 39. Eiff MP, Smith AT, Smith GE. Early mobilization versus immobilization in the treatment of lateral ankle sprains. Am J Sports Med 1994; 22 (1): 83-8 40. Klein J, Rixen D, Albring T, et al. Functional treatment with a pneumatic ankle brace versus cast immobilization for recent rupture of the fibular ligament in the ankle: a randomized clinical trial. Unfallchirurg 1991; 94 (2): 99-104 41. Nyska M, Weisel Y, Halperin N, et al. Controlled mobilization after acute ankle inversion injury. J Sports Traumatol Rel Res 1999; 21 (2); 114-20 42. Regis D, Montanari M, Magnan B, et al. Dynamic orthopaedic brace in the treatment of ankle sprains. Foot Ankle Int 1995; 16 (7): 422-6 43. Avci S, Sayli U. Comparison of the results of short-term rigid and semi-rigid cast immobilization for the treatment of grade 3 inversion injuries of the ankle. Injury 1998; 29 (8): 581-4 44. Johannes EJ, Kaulesar SDMKS, Spruit JLM, et al. Controlled trial of a semi-rigid bandage (‘‘scotch wrap’’) in patients with ankle ligament lesions. Curr Med Res Opin 1993; 13 (3): 154-62 45. Jongen SJM, Pot JH, Dunki Jacobs PB. Treatment of the sprained ankle [in Dutch]. Geneeskd Sport 1992; 25 (3): 98-101 46. Tufft K, Leaman A. A better form of treatment? Comparison of wool and crepe, and elasticated tubular bandages in the treatment of ankle sprains. Prof Nurse 1994; 9 (11): 745-6 47. Watts BL, Armstrong B. A randomised controlled trial to determine the effectiveness of double Tubigrip in grade 1 and 2 (mild to moderate) ankle sprains. Emerg Med J 2001; 18 (1): 46-50 48. Altizer L. Strains and sprains. Orthop Nurs 2003; 22 (6): 404-9 49. Connolly JF. Acute ankle sprains: getting and keeping patients back up on their feet. Consultant 1996 Aug; 36 (8): 1631-43 50. Kerkhoffs GMMJ, Pijnenburg ACM, De Vries JS, et al. Management of acute ankle sprain in athletes. Schweizer Z Med Traumatol 2003; 51 (2): 112-4 51. Loveridge N. Lateral ankle sprains. Emerg Nurse 2002; 10 (2): 29-33 52. MacAuley D. Ankle injuries: same joint, different sports. Med Sci Sports Exerc 1999 Jul; 31 (7 Suppl.): S409-11 53. Meisterling RC, Johnson RJ. Recurrent lateral ankle sprains. Phys Sportsmed 1993; 21 (30): 123-9 54. Soosai NS, Nwachukwu I, Forester A. A prospective randomised trial comparing the Aircast ankle brace with conservative treatment for lateral ligament injuries of the ankle [abstract]. J Bone Joint Surg Br 1997; 79 Suppl. 2: 250 55. Wilkerson GB, Horn-Kingery HM. Treatment of the inversion ankle sprain: comparison of different modes of compression and cryotherapy. J Orthop Sports Phys Ther 1993; 17 (5): 240-6

Sports Med 2011; 41 (3)

Functional Treatment of Ankle Sprains

56. Lamb SE, Nakash RA, Withers EJ, et al. Clinical and cost effectiveness of mechanical support for severe ankle sprains: design of a randomised controlled trial in the emergency department. BMC Musculoskelet Disord 2005; 6: 1 [online]. Available from URL: http://www.biomedcentral.com/ 1471-2474-6-1 [Accessed 2008 Sep 9] 57. Leanderson J, Bergqvist M, Rolf C, et al. Early influence of an ankle sprain on objective measures of ankle joint function: a prospective randomised study of ankle brace treatment. Knee Surg Sports Traumatol Arthrosc 1999; 7 (1): 51-8 58. Beynnon BD, Renstro¨m PA, Haugh L, et al. A prospective, randomized clinical investigation of the treatment of firsttime ankle sprains. Am J Sports Med 2006; 34 (9): 1401-12 59. Neumann K, Wittka¨mper V-I, Muhr G. Functional treatment for acute grade III tears of the lateral ligaments of the ankle with and without a brace: a prospective randomized study. Langenbecks Archiv fur Chirurgie 1994; 379 (Suppl. Kongressbericht): 827-9 60. Boyce SH, Quigley MA, Campbell S. Management of ankle sprains: a randomised controlled trial of the treatment of inversion injuries using an elastic support bandage or an Aircast ankle brace. Br J Sports Med 2005 Feb; 39 (2): 91-6 61. Karlsson J, Eriksson BI, Swa¨rd L. Early functional treatment for acute ligament injuries of the ankle joint. Scand J Med Sci Sports 1996; 6: 341-5 62. Leanderson J, Wredmark T. Treatment of acute ankle sprain. comparison of a semi-rigid ankle brace and compression bandage in 73 patients. Acta Orthop Scand 1995; 66 (6): 529-31 63. Dettori JR, Pearson BD, Basmania CJ, et al. Early ankle mobilization. Part I: the immediate effect on acute, lateral

ª 2011 Adis Data Information BV. All rights reserved.

197

64.

65.

66.

67.

68.

ankle sprains (a randomized clinical trial). Mil Med 1994 Jan; 159 (1): 15-20 Twellaar M, Veldhuizen JW, Verstappen FT. Ankle sprains: comparison of long-term results of functional treatment methods with adhesive tape and bandage (‘‘brace’’) and stability measurement. Unfallchirurg 1993; 96 (9): 477-82 Lamb SE, Marsh JL, Hutton JL, et al., Collaborative Ankle Support Trial (CAST Group). Mechanical supports for acute, severe ankle sprain: a pragmatic, multicentre, randomised controlled trial. Lancet 2009 Feb 14; 373 (9663): 575-81 Haraguchi N, Tokumo A, Okamura R, et al. Influence of activity level on the outcome of treatment of lateral ankle ligament rupture. J Orthop Sci 2009; 14: 391-6 Vaes P, Duquet W, Handelberg F, et al. Objective roentgenologic measurements of the influence of ankle braces on pathologic joint mobility: a comparison of 9 braces. Acta Orthop Belg 1998; 64 (2): 201-9 Cooke MW, Marsh JL, Clark M, et al., CAST trial group. Treatment of severe ankle sprain: a pragmatic randomised controlled trial comparing the clinical effectiveness and cost-effectiveness of three types of mechanical ankle support with tubular bandage. The CAST trial. Health Technol Assess 2009 Feb; 13 (13): iii, ix-x, 1-121

Correspondence: Ellen Kemler, MSc, Rudolf Magnus Institute of Neuroscience, Department of Rehabilitation, Nursing Science and Sport, University Medical Centre Utrecht, Utrecht, PO Box 85500, 3508 GA, the Netherlands. E-mail: [email protected]

Sports Med 2011; 41 (3)

Sports Med 2011; 41 (3): 199-220 0112-1642/11/0003-0199/$49.95/0

REVIEW ARTICLE

ª 2011 Adis Data Information BV. All rights reserved.

Physiology of Small-Sided Games Training in Football A Systematic Review Stephen V. Hill-Haas,1 Brian Dawson,1 Franco M. Impellizzeri2,3 and Aaron J. Coutts4 1 School of Sports Science, Exercise & Health, University of Western Australia, Perth, Western Australia, Australia 2 Neuromuscular Research Laboratory, Schulthess Clinic, FIFA Centre of Excellence, Zurich, Switzerland 3 Research Centre for Sport, Mountain and Health (CSMS) of Rovereto, University of Verona, Verona Italy 4 School of Leisure, Sport & Tourism, University of Technology, Lindfield, New South Wales, Australia

Contents Abstract . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2. Small-Sided Games (SSGs) in Football . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.1 Quantifying Exercise Intensity During SSGs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.2 Time-Motion Measurement in SSGs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.3 Variables Affecting SSG Intensity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.3.1 Pitch Area . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.3.2 Player Number. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.3.3 Concurrent Manipulation of Pitch Area and Player Number . . . . . . . . . . . . . . . . . . . . . . . . 2.3.4 Rule Modifications . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.3.5 Goalkeepers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.3.6 Training Regimen (Including Game Duration and Work : Rest Ratios) . . . . . . . . . . . . . . . . . 2.3.7 Coach Encouragement . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.3.8 Logistics and Planning . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.3.9 Comparisons of SSG Training Intensity with Competitive Match Play . . . . . . . . . . . . . . . . . 3. Studies Comparing SSGs Training with Interval Training . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.1 Acute Physiological Comparisons of SSGs Training with Interval Training . . . . . . . . . . . . . . . . . . . 3.2 Training Studies Comparing SSGs Training with Interval Training . . . . . . . . . . . . . . . . . . . . . . . . . . . 4. Limitations of SSGs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5. Future Research . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6. Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

Abstract

199 200 201 201 202 202 203 203 206 207 208 209 211 212 213 213 213 214 216 217 218

Small-sided games (SSGs) are played on reduced pitch areas, often using modified rules and involving a smaller number of players than traditional football. These games are less structured than traditional fitness training methods but are very popular training drills for players of all ages and levels. At present, there is relatively little information regarding how SSGs can best be used to improve physical capacities and technical or tactical skills in footballers. However, many prescriptive variables controlled by the coach can influence the exercise intensity during SSGs. Coaches usually attempt to change the training stimulus in SSGs through altering the pitch area, player

Hill-Haas et al.

200

number, coach encouragement, training regimen (continuous vs interval training), rules and the use of goalkeepers. In general, it appears that SSG exercise intensity is increased with the concurrent reduction in player number and increase in relative pitch area per player. However, the inverse relationship between the number of players in each SSG and exercise intensity does not apply to the time-motion characteristics. Consistent coach encouragement can also increase training intensity, but most rule changes do not appear to strongly affect exercise intensity. The variation of exercise intensity measures are lower in smaller game formats (e.g. three vs three) and have acceptable reproducibility when the same game is repeated between different training sessions or within the same session. The variation in exercise intensity during SSGs can also be improved with consistent coach encouragement but it is still more variable than traditional generic training methods. Other studies have also shown that SSGs containing fewer players can exceed match intensity and elicit similar intensities to both long- and short-duration highintensity interval running. It also appears that fitness and football-specific performance can be improved equally with SSG and generic training drills. Future research is required to examine the optimal periodization strategies of SSGs training for the long-term development of physiological capacity, technical skill and tactical proficiency.

1. Introduction The main purpose of this review is to provide a summary of the research that has examined the physiological and performance benefits of smallsided games (SSGs) in football. The review is presented in six sections. The first section briefly describes the origins, definition and advantages of SSGs. The second section reviews the use of SSGs in football. The three aspects addressed in the second section include findings from studies that have examined (i) the validity and reliability of quantifiable exercise intensity measures in SSGs; (ii) time-motion analysis of SSGs using Global Positioning System (GPS) technology; and (iii) variables affecting SSGs training intensity. The third section contains two parts. Part A examines studies that describe the acute physiological responses associated with various SSGs and part B examines training studies that compare the effectiveness of using both interval and SSGs training for conditioning. The fourth section describes the limitations of SSGs as a fitness training mode. The final two sections provide suggestions for future research and conclude the review. ª 2011 Adis Data Information BV. All rights reserved.

The articles reviewed here were acquired by searching the electronic databases of AUSPORT, ProQuest 5000, PubMed, SPORTDiscus and Google Scholar. The following keywords were used in various combinations: ‘small-sided soccer games’, ‘small-sided football games’, ‘metabolic conditioning’, ‘soccer-specific conditioning’, ‘football-specific conditioning’, ‘skill-based training’, ‘skill-based conditioning’, ‘soccer training’, ‘football training’ and ‘game-based training’. Due to the focus on football, this reduced the number of articles retrieved and, consequently, no limit to the search period was applied. Electronic database searching was supplemented by examining the bibliographies of relevant articles. This review is justified, given the increasing amount of research conducted into SSGs in football. It represents a useful synthesis of all research into SSGs in football, and helps to identify areas for future research, including the investigation of the technical load and tactical transfer of SSGs to match performance. Finally, it serves to further establish SSGs training as an alternative conditioning method for football players. Sports Med 2011; 41 (3)

Small-Sided Games Training Physiology in Football

2. Small-Sided Games (SSGs) in Football SSGs, also referred to as skill-based conditioning games[1] or game-based training,[2] are modified games played on reduced pitch areas, often using adapted rules and involving a smaller number of players than traditional football games. Formalized SSGs, such as those implemented in football clubs throughout the world and which underpin many junior football development programmes (e.g. Royal Dutch Football Association, Football Federation Australia), appear to have evolved from informal unstructured games of street football. Indeed, many of the world’s top players were introduced to football informally, via street, park or beach football.[3] Although it is still common to observe informal football SSGs being played in the street, park or beach, a structured approach to SSGs training has been adopted in the club setting.[3] SSGs in football are widely considered to offer many practical advantages that have lead to its popularity as a training modality in football at all ages and levels. The primary benefits of SSGs are that they appear to replicate the movement demands, physiological intensity and technical requirements of competitive match play,[4-7] whilst also requiring players to make decisions under pressure and fatigue.[8] SSGs have also been suggested to facilitate the development of technical skills and tactical awareness within the appropriate context of a game.[7,9] Compared with traditional fitness training sessions, SSGs are thought to increase player compliance and motivation, since it is perceived to be sport specific.[6,7] Finally, SSGs are considered to be more time efficient, as physical performance, technical skills and tactical awareness, can be developed concurrently.[6,7] However, the realization of these advantages is dependent on game design. 2.1 Quantifying Exercise Intensity During SSGs

Exercise intensity in SSGs has typically been assessed via heart rate (HR), blood lactate concentration and rating of perceived exertion (RPE). Indeed, HR is the most common measure used for objectively monitoring training intensity ª 2011 Adis Data Information BV. All rights reserved.

201

in many sports,[10] and several studies have shown HR to be a valid indicator of exercise intensity in the mean HR and football.[11,12] For example, . oxygen consumption (VO2) relationship have been reported to be similar during treadmillbased intermittent exercise that reproduced the several demands of a football game.[11] Similarly, . studies have shown that the HR/VO2 relationship established in the laboratory is similar to the . HR/VO2 relationship measured at different intensities during football-specific exercises (five vs five SSGs).[12-14] Collectively, the findings indicate that HR is a valid measure of exercise intensity during football. There are, however, some limitations to using HR to assess exercise intensity during footballspecific activities. For example, it has been suggested that factors inherent in football training, including emotion and the intermittent nature of the activity, may result in HR values that overestimate actual energetic cost of exercise.[15] In contrast, there is also evidence showing that HR monitoring may underestimate the intensity of football drills that have a high anaerobic component, including short-duration SSGs involving few players (e.g. 2-minute bouts; two vs two).[16] Therefore, it seems that other measures of exercise intensity may provide a more appropriate measure of exercise intensity during SSGs. Blood lactate, a by-product of anaerobic glycolysis, has been extensively used as an indicator of exercise intensity in football. The blood lactate concentration has been suggested to represent an overall accumulation of lactate production during football-specific exercise.[17] However, given the intermittent nature of football, blood lactate concentration is a poor indicator of muscle lactate concentration during football match play[17] and, consequently, may be misrepresentative of individual exercise intensities. In contrast to blood lactate concentration, RPE is a simple, non-invasive and inexpensive method of monitoring exercise intensity.[18] Several studies have shown that RPE can be validly used to assess exercise intensity at a specific time during exercise[19] and as a global indicator of overall session intensity (session RPE).[20,21] For example, to validate RPE as a measure of exercise Sports Med 2011; 41 (3)

Hill-Haas et al.

202

intensity during football SSGs, Coutts et al.[19] examined the relationship between RPE with both HR and blood lactate concentration measures. The findings of this study demonstrated that the combination of HR and blood lactate concentration predicted RPE better than HR or blood lactate concentration measures alone. Therefore, it was suggested that RPE may be a more valid marker of global exercise intensity than any physiological measures independently. Similarly, other studies have assessed the validity of the session RPE for assessing exercise intensity in football-specific exercise.[20,21] The session RPE method requires that players provide a single RPE relating to the exercise intensity of the entire session, usually 30 minutes following exercise.[22] However, although several studies have reported that session RPE is a valid indicator of overall perception of effort for intermittent aerobic football-specific exercises (including SSGs) training, it may not be a valid substitute for HR-based methods.[20,21] Nonetheless, due its psychobiological foundations,[18] session RPE measures may be a more valid global measure of exercise intensity during high-intensity intermittent exercise such as SSGs. However, all the methods currently available to assess exercise intensity during SSGs do have limitations. There is no clear evidence to suggest that one particular method is superior to the others. The methodology chosen may depend on what the variable of interest is. Therefore, on the basis of studies examining the validity of HR, RPE and blood lactate concentration during football-specific training, it has been suggested that SSGs training is best monitored via a combination of each of these measures of internal exercise intensity.[19] 2.2 Time-Motion Measurement in SSGs

In addition to physiological measures of exercise intensity during SSGs, recent technological advances now allow for movement characteristics of football players to be collected.[23] This information may be used to design game-related conditioning activities.[23] Specifically, GPS microtechnology is now used by various professional ª 2011 Adis Data Information BV. All rights reserved.

football codes to quantify the movement demands on players during training and games.[23] The validity and reliability of the measures provided by these commercially available (nondifferential) GPS receivers has recently been described.[24-28] In general, the error for total distance travelled (metres/minute) has been reported to be between 3% and 5%.[24] Moreover, the correlations between speed measured by electronic timing gates and values obtained from GPS units, have also been reported to be very high.[25,29] However, there are several limitations associated with this technology, including reduced reliability with increased movement speeds.[24,28] For example, the coefficient of variation (CV) for highintensity running (>14.4 km/h) is reported as 11.2–32.4%, and 11.5–30.4% for very highintensity running (>20.0 km/h).[24] Moreover, lower sampling rates (i.e. 1 vs 5 Hz) may also be a limitation, as this may reduce the devices ability to detect changes in direction at high speed.[24,28] Other limitations, including the number of satellites available from which to collect data, as well as the inability to sample data indoors, should also be considered. However, despite these limitations, the information obtained from these devices, specifically measures of exercise intensity such as total distance, distance covered in wide speed zones (i.e. speed zones that include a wide range of velocities) and peak velocity, may still provide useful data regarding variations in movement demands in the various SSGs. 2.3 Variables Affecting SSG Intensity

The exercise intensity of SSGs can be demonstrated through a player’s movement and/or physiological/perceptual responses. Many prescriptive variables that can be controlled by the coach may influence the exercise intensity during SSGs.[30] These factors include pitch area, player number, coach encouragement, training regimen (continuous or interval, including work : rest manipulations) rule modifications, and the use of goals and/or goalkeepers.[7,31] The following section will review how each of these factors have been manipulated to alter the exercise intensity during football SSGs. Sports Med 2011; 41 (3)

Small-Sided Games Training Physiology in Football

Despite the recent increases in the number of research studies that have investigated the influence of adjusting each of these variables upon the exercise intensity in SSGs, caution should be applied when interpreting the practical suitability of a specific SSG on the basis of a statistical observation. It has been suggested that the small but significant changes in isolated physiological variables between the various SSG designs may have a relatively minor influence on training adaptation.[32] Nonetheless, it is possible that when an alteration in SSG design elicits changes in a combination of physiological variables together (e.g. blood lactate concentration and HR), that a vastly different training response may be elicited. Accordingly, because of these complex interactions, it is important that coaches and scientists carefully interpret changes in the physiological responses to various SSG designs in the context of the global response, rather than simply on the basis of a statistical assessment of single physiological variables. 2.3.1 Pitch Area

The total pitch area, both in absolute and relative terms, can be varied, and this may influence the intensity of SSGs. The relative pitch area per player is defined as the total pitch area divided by the total number of players. Table I is a summary of all the studies that examined the effect of manipulating absolute and relative pitch area (while keeping the number of players constant) on SSG intensity. The majority of studies report an increased HR, RPE and blood lactate concentration response with increased pitch area. For example, Rampinini et al.[32] increased the pitch area by 20% across a variety of SSG formats (three vs three to six vs six, inclusive). Both the percentage of maximum HR (%HRmax) and blood lactate concentration were higher during SSGs played on a large pitch than on a medium-sized or small pitch. RPE was also higher on medium and large pitch sizes compared with small pitches.[32] 2.3.2 Player Number

The number of players on each team in a SSG can also be altered to regulate the intensity of this training mode. Studies that have investigated the ª 2011 Adis Data Information BV. All rights reserved.

203

effect of altering player number on SSGs training intensity have altered player numbers while, at the same time, held many other factors constant, including the pitch area. A summary of all the studies that examined the effect of altering player numbers on SSG intensity is presented in table II. In summary, despite some methodological concerns (very short game duration; differing work : rest ratios), most studies have shown that SSGs containing smaller numbers of players elicit greater HR, blood lactate and perceptual responses.[34,35,37,39] On closer analysis, the results suggest the possible existence of a threshold pitch area. For example, the most pronounced reductions in HR occurred when two versus two was increased to three versus three, and three versus three was increased to four versus four, on a 25 · 20 metre pitch area. In contrast, less pronounced reductions in HR occurred when two versus two was increased to three versus three, and three versus three was increased to four versus four on 20 · 15 metre and 30 · 25 metre pitch areas, respectively.[34] As illustrated in table II, these previous studies only examined the influence of altering the player numbers on teams containing equal numbers of players (e.g. two vs two or three vs three). In training situations, SSGs are often implemented that contain teams of unequal numbers (e.g. four vs three players or six vs five). Reasons for creating an imbalance between opposing teams may include technical development and unavailability of players due to injury. A further variation in player number involves creating temporary ‘overload’ and ‘underload’ situations between opposing teams, via the use of a ‘floater’ player. This neutral player transitions to the team in possession of the ball, to create temporary ‘overload’ and ‘underload’ situations. This SSG game design is typically used to develop defensive or attacking proficiency or to increase the physical load on the ‘floating’ player. The impact of creating fixed and temporary ‘overload’ and ‘underload’ situations (including the use of a ‘floater’) on the physiological, perceptual and time-motion responses in SSGs involving elite youth football players have recently been investigated.[38] The findings from this study Sports Med 2011; 41 (3)

Hill-Haas et al.

204

Table I. Summary of studies examining the effects of pitch dimensions on small-sided game intensity in football players Study

Sample size

Game design

Training prescription

Pitch dimensions (m)

Area per playera (m2)

%HRmax [mean – SD]b

[BLa-] (mmol/L) [mean – SD]

RPE (6–20 AU)c [mean – SD]

Aroso et al.[33]

14

4 vs 4

3 · 6 min/90 s rest

30 · 20 50 · 30

75 188

70.0 – 9.0 -

2.6 – 1.7 › (no value)

13.3 – 0.9 › (no value)

Owen et al.[34]

13

1 vs 1

1 · 3 min/12 min rest

10 · 5 15 · 10 20 · 15 15 · 10 20 · 15 25 · 20 20 · 15 25 · 20 30 · 25 25 · 20 30 · 25 30 · 25 35 · 30 40 · 35

25 75 150 38 75 125 50 83 125 63 94 75 105 140

86.0 88.0 89.0 84.2 87.4 88.1 81.7 81.8 84.8 72.0 78.5 75.7 79.5 80.2

-

-

2 vs 2

3 vs 3

4 vs 4 5 vs 5

Williams and Owen[35]

9

3 vs 3

-

20 · 15 25 · 20 30 · 25

50 83 125

164 – 12d (mean HR) 166 – 9d (mean HR) 171 – 11d (mean HR)

– – –

– – –

Rampinini et al.[32]

20

3 vs 3 (CE) 3 · 4 min/3 min rest

20 · 12 25 · 15 30 · 18 24 · 16 30 · 20 36 · 24 28 · 20 35 · 25 42 · 30 32 · 24 40 · 30 48 · 36

40 63 90 48 75 108 56 88 126 64 100 144

89.5 – 2.9 90.5 – 2.3 90.9 – 2.0 88.7 – 2.0 89.4 – 1.8 89.7 – 1.8 87.8 – 3.6 88.8 – 3.1 88.8 – 2.3 86.4 – 2.0 87.0 – 2.4 86.9 – 2.4

6.0 – 1.8 6.3 – 1.5 6.5 – 1.5 5.3 – 1.9 5.5 – 1.8 6.0 – 1.6 5.2 – 1.4 5.0 – 1.7 5.8 – 1.6 4.5 – 1.5 5.0 – 1.6 4.8 – 1.5

8.1 – 0.6 (CR10) 8.4 – 0.4 (CR10) 8.5 – 0.4 (CR10) 7.6 – 0.5 (CR10) 7.9 – 0.5 (CR10) 8.1 – 0.5 (CR10) 7.2 – 0.9 (CR10) 7.6 – 0.6 (CR10) 7.5 – 0.6 (CR10) 6.8 – 0.6 (CR10) 7.3 – 0.7 (CR10) 7.2 – 0.8 (CR10)

30 · 20 40 · 30 50 · 40

60 120 200

91.0 – 4.0 90.0 – 4.0 89.0 – 2.0

-

-

4 vs 4 (CE)

5 vs 5 (CE)

6 vs 6 (CE)

Kelly and Drust[36]

8

5 vs 5 (CE) 4 · 4 min/2 min rest

a

Total pitch area divided by total number of players.

b

Data for Owen et al.[34] are presented as mean values.

c

RPE is 6–20 AU unless otherwise stated.

d

Age predicted HR values, mean – SD.

AU = arbitrary units; [BLa-] = blood lactate concentration; CE = with coach encouragement; CR10 = category ratio 10 scale; HR = heart rate; %HRmax = percentage of maximum HR; RPE = rating of perceived exertion; › indicates increase; - indicates no data.

were that there were no significant differences between the fixed (four vs three or six vs five) and variable (three vs three + one floater or five vs five + one floater) SSGs in terms of physiological and perceptual responses (see table II). Despite this, either may provide a useful SSGs training variation, or as a technical/tactical training method for defensive, transition and attacking plays. The possibility of fixed and variable SSGs ª 2011 Adis Data Information BV. All rights reserved.

providing a greater technical load needs to be examined by further research. Finally, the use of a floater appears to be more effective in SSGs containing fewer players (e.g. three vs three + one floater), and may be appropriate for either maintaining or developing aerobic fitness.[38] For example, the floater travelled a significantly greater total distance and recorded a greater RPE compared with four-player teams in four- versus Sports Med 2011; 41 (3)

Study

Sample size

Game design

Training prescription

Pitch dimensions (m)

Area per playera (m2)

%HRmax [mean – SD]

[BLa-] (mmol/L) [mean – SD]

RPE (6–20 AU)b [mean – SD]

Aroso et al.[33]

14

2 vs 2 3 vs 3 4 vs 4

3 · 1.5 min/90 s rest 3 · 4 min/90 s rest 3 · 6 min/90 s rest

30 · 20 30 · 20 30 · 20

150 100 75

84.0 – 5.0 87.0 – 3.0 70.0 – 9.0

8.1 – 2.7 4.9 – 2.0 2.6 – 1.7

16.2 – 1.1 14.5 – 1.7 13.3 – 0.9

Owen et al.[34]

13

1 vs 1 2 vs 2 1 vs 1 2 vs 2 3 vs 3 2 vs 2 3 vs 3 4 vs 4 4 vs 4 5 vs 5

1 · 3 min/12 min rest

15 · 10 15 · 10 20 · 15 20 · 15 20 · 15 25 · 20 25 · 20 25 · 20 30 · 25 30 · 25

75 38 150 75 50 125 83 63 94 75

88.0c 84.2c 89.0c 87.4c 81.7d 88.1c 81.8d 72.0e 78.5e 75.7e

-

-

Sampaio et al.[37]

8

2 vs 2 3 vs 3

2 · 1.5 min/90 s rest 2 · 3 min/90 s rest

30 · 20 30 · 20

150 100

83.7 – 1.4 80.8 – 1.7

-

15.5 – 0.6 15.8 – 0.2

Williams and Owen[35]

9

1 vs 1 2 vs 2 3 vs 3 2 vs 2 3 vs 3 4 vs 4 3 vs 3 4 vs 4 5 vs 5

-

20 · 15 20 · 15 20 · 15 25 · 20 25 · 20 25 · 20 30 · 25 30 · 25 30 · 25

150 75 50 125 83 63 125 94 75

183 – 7 (mean HR) 179 – 7 (mean HR) 164 – 12 (mean HR) 180 – 5 (mean HR) 166 – 9 (mean HR) 152 – 14 (mean HR) 171 – 11 (mean HR) 165 – 5 (mean HR) 152 – 6 (mean HR)

-

-

Hill-Haas et al.[38]

12

3 players

24f

37 · 28

148

16

4 players

24f

37 · 28

148

Floater

24

f

37 · 28

148

5 players

24f

47 · 35

149

6 players

24

f

47 · 35

149

Floater

24f

47 · 35

149

82.3 – 3.5 2543 – 187 (TD m) 83.1 – 4.0 2408 – 231 (TD m) 82.7 – 3.0 2668 – 220 (TD m) 82.5 – 5.0 2526 – 302 (TD m) 81.4 – 5.1 2524 – 247 (TD m) 82.5 – 5.6 2610 – 201 (TD m)

2.5 – 0.7 553 – 187 (D m) 2.5 – 0.9 482 – 178 (D m) 2.3 – 0.8 628 – 132 (D m) 2.5 – 1.0 649 – 190 (D m) 2.6 – 1.1 589 – 177 (D m) 2.8 – 0.2 673 – 194 (D m)

16.3 – 1.6 10 – 6 (SP) 14.6 – 1.9 8 – 4 (SP) 16.3 – 1.5 9 – 6 (SP) 15.2 – 1.0 9 – 5 (SP) 14.9 – 0.9 8 – 4 (SP) 16.3 – 1.7 15 – 3 (SP)

8 20

4

Continued next page

205

Sports Med 2011; 41 (3)

24

Small-Sided Games Training Physiology in Football

ª 2011 Adis Data Information BV. All rights reserved.

Table II. Summary of studies examining the effects of player number on small-sided game intensity in football players

Hill-Haas et al.

intensity <11 vs 11 competitive match.

Game duration (min).

Matched team excluding the floater.

f

g

intensity = 11 vs 11 competitive match. d

e

RPE is 6–20 AU unless otherwise stated.

intensity >11 vs 11 competitive match. c

Total pitch area divided by total number of players.

b

a

Underload PN

Overload PN

3 vs 3 and 5 vs 5 6 player and 4 player teams 5 player and 3 player teams Matched PNg

ª 2011 Adis Data Information BV. All rights reserved.

AU = arbitrary units; [BLa-] = blood lactate concentration; D = distance (m): >13.0 km/h; HR = heart rate; %HRmax = percentage of maximum HR; PN = player numbers; RPE = rating of perceived exertion; SP = number of sprints >18.0 km/h; TD = total distance (m); - indicates no data.

15.2 – 1.4 14.7 – 1.5 15.8 – 1.5 2.6 – 1.1 582 – 190 (D m) 2.6 – 1.0 528 – 184 (D m) 2.6 – 1.0 598 – 192 (D m) 82.5 – 4.6 2585 – 204 (TD m) 82.3 – 4.5 2458 – 243 (TD m) 82.3 – 4.0 2535 – 247 (TD m)

Game design Sample size Study

Table II. Contd

Training prescription

Pitch dimensions (m)

Area per playera (m2)

%HRmax [mean – SD]

[BLa-] (mmol/L) [mean – SD]

RPE (6–20 AU)b [mean – SD]

206

three-player games[38] (figure 1). The floater also completed a significantly greater amount of sprints (>18 km/h) compared with five- and six-player teams in six- versus five-player games[38] (see table II). 2.3.3 Concurrent Manipulation of Pitch Area and Player Number

Few studies have systematically examined the influence of the concurrent manipulation of pitch area and player number on exercise intensity in SSGs.[32,40-42] In addition, there are several differences in the design and prescription of the SSGs in the studies that inadvertently manipulated both player number and pitch area, making comparisons between these studies very difficult. Indeed, tables III and IV show that there are subtle differences in the training prescriptions, age and ability of players, intensity measures and sizes in pitch area amongst the studies, all of which may affect the exercise intensity in these SSGs. In general, it appears that a concurrent increase in player number and relative pitch area per player in SSGs elicits lower exercise intensity. For example, Rampinini et al.[32] investigated the effects of concurrently increasing the player number and pitch area on %HRmax, blood lactate concentration and RPE in 20 amateur football players. The main finding of this study was that the exercise intensity during all game formats was decreased when there was an increase in the number of players and more pitch area per player[32] (see table III). Similarly, Jones and Drust[41] also reported a reduction in %HRmax when both player number and pitch area were increased (see table III). One important aspect that has not been considered by studies where both pitch size and player number were altered concurrently was the influence of the relative pitch area per player.[16,32,41-44] In all of these studies, an increase in absolute pitch area and player number also resulted in a greater relative pitch area per player. Therefore, the observed reduction in SSG intensity by several of these studies[32,41,42] may have been due to either the independent effects of increasing the number of players or the inability of the additional players to cover more of the available pitch area. Clearly, more research is required to determine the effect of an increase in player number on Sports Med 2011; 41 (3)

Small-Sided Games Training Physiology in Football

characterized by significantly longer (average and maximal) effort durations and distances for speeds >18 km/h.[45] However, since it is the internal response to training (e.g. HR and RPE) and not the external training load (e.g. distances travelled in speed zones) that determines each players adaptation to a training stimulus,[46] it is recommended that each player’s internal load be monitored to assess how players are coping with different SSGs design (see table IV).

a 3000

TD (m)

2750 2500 2250 2000

b

2.3.4 Rule Modifications

20 RPE (6−20 AU)

207

18 16 14 12 3 Players

4 Players

Floater

Fig. 1. Comparison of (a) total distance (TD; m) and (b) rating of perceived exertion (RPE) [6–20 arbitrary units; AU] with ‘floating’ players and other players in various smaller game formats.[38]

SSG intensity (or vice versa). However, it is important that future studies control for the influence of relative pitch area per player so that an improved understanding of increasing pitch area and player number in SSGs can be obtained. More recently, a study involving youth football players examined the acute physiological and perceptual responses and time-motion characteristics during three variations of SSGs (two vs two, four vs four and six vs six) with a constant ratio of player number to pitch area applied to each SSG variation.[45] The main findings were, as the number of players in the SSG teams decreased, when the relative pitch area per player remained constant, the overall physiological and perceptual responses increased. Notably, the inverse relationship between the number of players in each SSG and exercise intensity did not extend to the time-motion characteristics. In general, the largest game format (six vs six) was associated with a greater range of distances travelled at speeds >18 km/h. In contrast, the four versus four format, compared with the two versus two, was ª 2011 Adis Data Information BV. All rights reserved.

In practice, football coaches quite often modify playing rules in SSGs to achieve greater exercise intensity, or develop specific technical and tactical skills. However, there have only been a few studies that have examined how the modification of rules can influence these variables. Table V provides a summary of studies that have investigated the effects of rule changes on exercise intensity during football SSGs. Two studies[47,48] reported an increase in %HRmax and another reported an increase in blood lactate concentration due to rule changes[33] (table V). Simple rule changes have also been reported to increase the perception of effort[37] (table V), which may be due to the increased cognitive load required of players as a consequence of new rules. To date, the only study to have reported on the influence of rule changes on movement characteristics is by Mallo and Navarro.[48] Compared with normal football rules, these specific rule changes resulted in an increase in total distance travelled (table V) and time spent performing high-intensity running, with less spent time spent stationary.[33,48] Although these simple rule modifications relate to technical aspects of the game, other studies have investigated the influence of providing ‘artificial’ changes.[38] An example of an artificial rule change is the requirement for a player to complete a series of sprints of planned duration during a SSG. Hill-Haas et al.[38] recently examined the acute physiological responses and time-motion characteristics associated with four different rule changes, including the addition of ‘artificial’ rules. The main finding was that changes in SSG playing rules can influence the physiological and time-motion responses, but not Sports Med 2011; 41 (3)

Hill-Haas et al.

208

Table III. Summary of studies examining the effects of concurrent changes in player number and pitch dimensions on small-sided game intensity in football players Study

Sample size; age (y)

Game design

Training prescription

Pitch dimensions (m)

Area per playera (m2)

%HRmax [mean – SD]b

[BLa-] (mmol/L) [mean – SD]

RPE (6–20 AU)c [mean – SD]

Platt et al.[43]

2; 10–12

3 vs 3 5 vs 5

1 · 15 min continuous 1 · 15 min continuous

27 · 18 37 · 27

81 100

88.0d 82.0d

-

-

Little and Williams[16]

28; NR

2 vs 2 3 vs 3 4 vs 4 5 vs 5 6 vs 6 8 vs 8

4 · 2 min/2 min rest 4 · 3.5 min/90 s rest 4 · 4 min/2 min rest 4 · 6 min/90 s rest 3 · 8 min/90 s rest 4 · 8 min/90 s rest

27 · 18 32 · 23 37 · 27 41 · 27 46 · 27 73 · 41

122 123 125 111 104 187

88.9 – 1.2 91.0 – 1.2 90.1 – 1.5 89.3 – 2.5 87.5 – 2.0 87.9 – 1.9

9.6 – 1.0 8.5 – 0.8 9.5 – 1.1 7.9 – 1.7 5.6 – 1.9 5.8 – 2.1

16.3 – 0.9 15.7 – 1.1 15.3 – 0.7 14.3 – 1.5 13.6 – 1.0 14.1 – 1.8

Jones and Drust[41]

8; 7

4 vs 4 8 vs 8

1 · 10 min continuous 1 · 10 min continuous

30 · 25 60 · 40

94 150

83.0 79.0

-

-

Rampinini et al.[32]

20; NR

3 vs 3 (CE) 4 vs 4 (CE) 5 vs 5 (CE) 6 vs 6 (CE)

3 · 4 min/3 min rest

30 · 18 36 · 24 42 · 30 48 · 36

90 108 126 144

90.9 – 2.0 89.7 – 1.8 88.8 – 2.3 86.9 – 2.4

6.5 – 1.5 6.0 – 1.6 5.8 – 1.6 4.8 – 1.5

8.5 – 0.4 (CR10) 8.1 – 0.5 (CR10) 7.5 – 0.6 (CR10) 7.2 – 0.8 (CR10)

a

Total pitch area divided by total number of players.

b

Data for Platt et al.[43] and Jones and Drust[41] are presented as mean values.

c

RPE is 6–20 AU unless otherwise stated.

d

Age predicted heart rate values.

AU = arbitrary units; [BLa-] = blood lactate concentration; CE = coach encouragement; CR10 = category ratio 10 scale; %HRmax = percentage of maximum heart rate; NR = not reported; RPE = rating of perceived exertion; - no data.

perceptual responses, in young elite football players (table V).[38] The artificial rule change that required players to complete extra sprint efforts around the pitch during each SSG at pre-set times, imposed a greater external training load on the players, but did not affect HR, blood lactate concentration or RPE. In contrast, changes in technical rules that were related to a team’s chances of scoring, may have improved player motivation and thereby increased the exercise intensity during the SSGs.[38] Although there have been relatively few studies that have examined the influence of rule modifications on exercise intensity during SSGs, the rule changes that have been investigated are by no means exhaustive. To date, the rule changes that have been investigated have altered either the physiological and/or perceptual responses, as well as the time-motion characteristics of various SSGs. However, this may not be the case for all types of rule changes that could possibly be implemented. Future studies should aim to more systematically classify the types of rules changes that appear to have differential effects on physiological, perceptual and time-motion ª 2011 Adis Data Information BV. All rights reserved.

responses during SSGs. Future studies should examine the effect of common rules modifications on the technical and tactical skills of football players. Factors such as decision making and cognitive load of players should also be assessed (table V). 2.3.5 Goalkeepers

One common rule modification in SSGs is the removal of goalkeepers from the game in an attempt to increase the number of goals scored. Goalkeepers are an integral part of football; however, surprisingly few studies have investigated the use of goalkeepers and their possible effect on SSGs training intensity. Table VI provides a summary of the SSGs studies that investigated the effects of goalkeepers on SSG intensity. Mallo and Navarro[48] reported a significant decrease in %HRmax, total distance and time spent in highintensity running, in three versus three SSGs with goalkeepers. It was suggested that the reduced physiological and time-motion responses were due to increased defensive organization near the goal area, which reduced the tempo of play and subsequently the physiological and time-motion Sports Med 2011; 41 (3)

Small-Sided Games Training Physiology in Football

209

responses.[48] In contrast, Dellal et al.[44] reported a 12% increase in heart rate response in eight versus eight SSGs with goalkeepers. The presence of goalkeepers may have increased the player’s motivation to both attack and defend, thereby increasing the physiological load.[44] At present, the influence of goalkeepers on exercise intensity in football SSGs is not clear. They may have an important role in keeping team structures and formations intact, as well as increasing communication, all of which may influence movement, skill and physiological demands. Future studies are required to determine the influence of goal keepers on the physiological and technical/tactical demands in SSGs. 2.3.6 Training Regimen (Including Game Duration and Work : Rest Ratios)

Similar to interval running, many prescriptive variables can be used in SSGs to alter exercise

intensity. The majority of the studies have used a traditional ‘interval’ training format, whereby several consecutive bouts of SSGs play are interspersed with active or passive rest periods (table VII). The duration of each SSG bout interval, alternating with planned rest periods, is used to determine work : rest ratios. Although most studies examining SSGs have prescribed the SSG bouts using intervals with short rests, some recent studies have used continuous SSG formats of differing duration (e.g. 10–30 minutes). Unfortunately, previous studies have not used consistent work : rest ratios and there is a large variation in the length, duration, and number of work bouts and rest intervals amongst studies (table VII), which makes comparison difficult. For example, a SSG ‘interval’ training prescription consisting of a 1 · 3-minute work bout with a 12-minute rest represents a very low work : rest ratio (1 : 4) and a very short total game duration

Table IV. Summary of studies examining the effects of concurrent changes in player number and pitch dimensions on small-sided game intensity in football players Study

Sample Game size; age design (y)

Training prescription

Pitch Area per %HRR dimensions playera [mean – SD] (m2) (m)

[BLa-] (mmol/L) [mean – SD]

RPE (6–20 AU)b [mean – SD]

Dellal et al.[44]

10; 24–27c

1 vs 1 2 vs 2 4 vs 4 + GK 8 vs 8 + GK 8 vs 8 10 vs 10 + GK

4 · 1.5 min/90 s rest 6 · 2.5 min/2.5 min rest 2 · 4 min/3 min rest 2 · 10 min/5 min rest 4 · 4 min/3 min rest 3 · 20 min/5 min rest

10 · 10 20 · 20 30 · 25 60 · 45 60 · 45 90 · 45

50 100 94 169 169 203

77.6 – 8.6 80.1 – 8.7 77.1 – 10.7 80.3 – 12.5 71.7 – 6.3 75.7 – 7.9

-

-

Hill-Haas et al.[45]

16; 16–18c

2 vs 2

24 min continuous

28 · 21

150

6.7 – 2.6

13.1 – 1.5

4 vs 4

40 · 30

150

6 vs 6

49 · 37

150

89.0 – 4.0 (%HRmax)d 2574 – 16 TD (m) 85.0 – 4.0 (%HRmax)d 2650 – 18 TD (m) 83.0 – 4.0 (%HRmax)d 2590 – 33 TD (m)

25 · 15 40 · 30

63 100

87.6 – 4.8 82.8 – 3.2

-

Katis and Kellis[42] a

34; 3 vs 3 13 – 0.9e 6 vs 6

10 · 4 min/3 min rest

1176 – 8 (D m) 44 – 24 (SP m) 12.2 – 1.8 4.7 – 1.6 1128 – 10 (D m) 65 – 36 (SP m) 10.5 – 1.5 4.1 – 2.0 1142 – 16 (D m) 71 – 36 (SP m) -

Total pitch area divided by total number of players.

b

RPE is 6–20 AU unless otherwise stated.

c

Age range.

d

Age predicted heart rate values.

e

Age presented as mean – SD.

AU = arbitrary units; [BLa-] = blood lactate concentration; D = distance: 13.0–15.9 km/h; GK = including goalkeepers; %HRmax = percentage of maximum heart rate; %HRR = percentage of heart rate reserve; RPE = rating of perceived exertion; SP = number of sprints >18.0 km/h; TD = total distance; - indicates no data.

ª 2011 Adis Data Information BV. All rights reserved.

Sports Med 2011; 41 (3)

210

ª 2011 Adis Data Information BV. All rights reserved.

Table V. Summary of studies examining the effects of rule modifications on small-sided game intensity in football players [BLa-] (mmol/L) [mean – SD]

RPE (6–20 AU) [mean – SD]

TD (m) [mean – SD]

Player-to-player marking Maximum of 3 consecutive touches

› 8.1 – 2.7 4.9 – 2.0

-

-

Free touch

82.0

3.3 – 1.2

-

-

Free touch with pressure

› 91.0

-

-

-

Player-to-player marking Maximum of 2 consecutive touches Player-to-player marking Maximum of 2 consecutive touches

2 2

-

› 17.1 – 0.5 › 16.8 – 0.5

-

2 2

-

› 16.5 – 0.5 › 16.5 – 0.5

-

-

Study

Sample size

Game design

Training prescription

Pitch Rules dimensions (m)

Aroso et al.[33]

14

2 vs 2 3 vs 3

3 · 1.5 min/90 s rest 3 · 4 min/90 s rest

30 · 20

Sassi et al.[47]

9

8 vs 8 + GK 8 vs 8 + GK

4 · 4 min/2.5 min rest 50 · 30

2 vs 2

2 · 1.5 min/90 s rest

3 vs 3

2 · 3 min/90 s rest

Sampaio et al.[37]

8

30 · 20

%HRmax [mean – SD]a

Little and Williams[40]

23

5 vs 5 6 vs 6

5 · 2 min/2 min rest 5 · 2 min/2 min rest

55 · 32 59 · 27

Pressure half switch Pressure half switch

89.9 90.5

-

-

Mallo and Navarro[48]

10

3 vs 3

1 · 5 min/10 min rest

33 · 20

Possession Possession with 2 outside neutral players Normal rules + GK

91.0 2 91.0 2

-

-

747 – 24 749 – 29

88.0 fl

-

-

638 – 34

Hill-Haas et al.[38]

a

b

c

24 23 23 26

3 vs 4 and 24 min continuous 3 vs 3 + 1 floater

37 · 28

Condition a + b Condition a + b + cd Condition a + b + c + de Condition a + b + c + d + ef

83.3 – 3.8 84.8 – 3.8 80.3 – 4.8 83.7 – 4.0

2.8 – 1.0 2.4 – 0.8 2.3 – 1.1 2.8 – 1.1

15.8 – 1.6 15.6 – 2.3 14.8 – 1.2 15.1 – 1.6

2439 – 166 2405 – 201 2450 – 223 2677 – 192

21 22 20 21

5 v 6 and 5 v 5+1 floater

47 · 35

Condition ab + bc Condition a + b + cd Condition a + b + c + de Condition a +b +c + d + ef

81 – 4 83 – 5 83 – 5 80 – 3

2.2 – 1.0 3.2 – 1.2 2.3 – 1.1 2.4 – 0.9

15.3 – 1.1 14.9 – 1.4 14.6 – 0.9 14.9 – 1.1

2471 – 355 2583 – 147 2614 – 178 2639 – 189

24 min continuous

Data for Sassi et al.,[47] Little and Williams[40] and Mallo and Navarro[48] are presented as mean values. Condition a: offside rule in effect (front one-third zone of the pitch).

c

Condition b: kick-in only (ball cannot be thrown in if it leaves the pitch).

d

Condition c: all attacking team players must be in front two zones for a goal to count.

e

Condition d: outside, but along the two lengths of each pitch, two neutral players can move up and down the pitch, but not enter the grid. Before a shot on goal is permitted, the attacking team must pass the ball to either of these players. The ball can also be passed to either player in the defensive half. Each player is only allowed a maximum of one touch on the ball.

f

Condition e: one player from each team (a pair) complete four repetitions of ‘sprint the widths/jog the lengths’ on a 90 s interval (3 vs 4 and 3 vs 3 + 1 games) or three repetitions on a 80 s interval (5 vs 6 and 5 vs 5 + 1 games). TD travelled per player, regardless of game format, would be approximately 440 m.

AU = arbitrary units; [BLa-] = blood lactate concentration; GK = including goalkeepers; %HRmax = percentage of maximum heart rate; RPE = rating of perceived exertion; TD = total distance; › indicates increase; fl indicates decrease; 2 indicates no change; - indicates no data.

Hill-Haas et al.

Sports Med 2011; 41 (3)

b

Small-Sided Games Training Physiology in Football

(3 minutes). Other studies have used different work: rest ratios across various SSGs (table VII).[16] Together, these may confound the physiological and perceptual responses, as well as the timemotion characteristics of the games. A recent study involving youth football players examined the acute physiological and perceptual responses and time-motion characteristics of two different training regimens (continuous and intermittent). These intermittent (4 · 6-minute bouts with 1.5 minutes passive rest) and continuous (24 minutes) regimens were applied to various SSGs including two versus two, four versus four and six versus six.[49] The main finding of this study was that intermittent regimens were characterized by increased distances covered at speeds of >13 km/h. However, paradoxically, the global RPE and %HRmax was significantly higher in continuous regimens. The results of this study demonstrated that both SSG training regimens could be used during a season for match-specific aerobic conditioning, but were unlikely to provide a sufficient stimulus overload for fully . developing maximal oxygen consumption (VO2max).[49] Another study recently investigated the effect of SSG duration, using a 2-, 4- and 6-minute interval format, on both exercise intensity and technical performance during three versus three SSGs.[50] The main findings were that although there was a significant decrease in HR between the 4- and 6-minute

211

game durations and an increase in RPE, the 4-minute bouts appear to provide the optimal physical training stimulus for interval format SSGs.[50] However, the various interval durations did not affect technical performance and, given that the magnitude of changes between each of the different interval bouts was small, football coaches can be confident in using various SSG interval durations to provide an adequate physical and technical training stimulus.[50] In summary, research shows that neither training regimen appears to offer any major advantage over the other, and that both regimens could be used for in-season aerobic fitness maintenance training. 2.3.7 Coach Encouragement

Direct supervision and coaching of exercise sessions have been shown to improve adherence to an exercise programme, increase training intensity and increase performance measures in a variety of training modes.[51,52] In football, active, consistent coach encouragement has also been suggested to have an influence on training intensity.[30,32,37] For example, Rampinini et al.[32] demonstrated that HR, blood lactate concentration and RPE were higher when coaches provided consistent encouragement during SSGs with 20 amateur football players in a variety of SSG formats (three vs three, four vs four, five vs five and six vs six players and on small, medium and large-sized pitches). Similarly,

Table VI. Summary of studies examining the effects of goalkeepers on small-sided game intensity in football players Study Sassi et al.[47]

Sample size 9

Game design

Training prescription

Pitch dimensions (m)

Rules

%HRmaxa [mean – SD]b

[BLa-] (mmol/L) [mean – SD]

Time motion

4 vs 4

4 · 4 min/2.5 min rest

30 · 30

Possession

91.0

6.4 – 2.7

-

fl 88.8

6.2 – 1.4

-

33 · 33

4 vs 4 + GK Mallo and Navarro[48]

10

3 vs 3 + GK

1 · 5 min/10 min rest

33 · 20

Normal rules

88.0 fl

-

fl TD; fl HIR; ›S+W

Dellal et al.[44]

10

8 vs 8

4 · 4 min/3 min rest

60 · 45

-

71.7 – 6.3 (%HRR)

-

-

8 vs 8 + GK

2 · 10 min/5 min rest

60 · 45

-

› 80.3 – 12.5 (%HRR)

-

-

a

%HRmax unless otherwise stated.

b

Data for Sassi et al.[47] and Mallo and Navarro[48] are presented as mean values.

[BLa-] = blood lactate concentration; GK = including goalkeepers; HIR = high-intensity running; %HRmax = percentage of maximum heart rate; %HRR = percentage of heart rate reserve; S + W: standing and walking; TD = total distance; › indicates increase; fl indicates decrease; - indicates no data.

ª 2011 Adis Data Information BV. All rights reserved.

Sports Med 2011; 41 (3)

Hill-Haas et al.

212

Table VII. Summary of different training regimens implemented in small-sided game studies with football players Study Balsom et al.[30]

Sample size 6

Game design

Training prescription

Work : rest ratio

Regimen

3 vs 3

6 · 3 min/2 min rest 15 · 70 s/20 s rest 36 · 30 s/15 s rest 36 · 30 s/30 s rest 1 · 30 min

1.5 : 1 3.5 : 1 2:1 1:1 -

Interval Interval Interval Interval Continuous

Owen et al.[34]

13

1 vs 1 - 5 vs 5

1 · 3 min/12 min rest

1:4

Interval

Aroso et al.[33]

14

2 vs 2 3 vs 3 4 vs 4

3 · 1.5 min/90 s rest 3 · 4 min/90 s rest 3 · 6 min/90 s rest

1:1 2.6 : 1 4:1

Interval Interval Interval

Jones and Drust[41]

-

4 vs 4 and 8 vs 8

1 · 10 min

-

Continuous

Rampinini et al.[32]

20

3 vs 3 - 5 vs 5

3 · 4 min/3 min rest

1.3 : 1

Interval

Kelly and Drust[36]

8

5 vs 5

4 · 4 min/2 min rest

2: 1

Interval

Little and Williams[16]

28

2 vs 2 3 vs 3 4 vs 4 5 vs 5 6 vs 6 8 vs 8

4 · 2 min/2 min rest 4 · 3.5 min/90 s rest 4 · 4 min/2 min rest 4 · 6 min/90 s rest 3 · 8 min/90 s rest 4 · 8 min/90 s rest

1:1 2.3 : 1 2:1 4:1 5.3 : 1 5.3 : 1

Interval Interval Interval Interval Interval Interval

Dellal et al.[44]

10

1 vs 1 2 vs 2 4 vs 4 + GK 8 vs 8 + GK 8vs 8 10 vs 10 + GK

4 · 1.5 min/90 s rest 6 · 2.5 min/2.5 min rest 2 · 4 min/3 min rest 2 · 10 min/5 min rest 4 · 4 min/3 min rest 3 · 20 min/5 min rest

1:1 1:1 1.3 : 1 2:1 1.3 : 1 4:1

Interval Interval Interval Interval Interval Interval

Hill-Haas et al.[49]

16

2 vs 2; 4 v 4; 6 vs 6 2 vs 2; 4 vs 4; 6 vs 6

4 · 6 min/90 s passive rest 1 · 24 min

4:1 -

Interval Continuous

Fanchini et al.[50]

19

3 vs 3

3 · 2 min; 3 · 4 min; 3 · 6 min/4 min rest

1 : 2; 1 : 1; 1.5 : 1

Interval

GK = including goalkeepers; - indicates 1 vs 1, 2 vs 2, 3 vs 3, 4 vs 4 and 5 vs 5 small-sided games were used; - indicates no data.

Sampaio et al.[37] reported a significant increase in RPE (for two vs two and three vs three SSGs) with verbal encouragement, but no significant change in %HRmax. Collectively, these studies support the role of the coach in providing consistent encouragement during SSGs, especially when it is planned that high intensities be achieved. 2.3.8 Logistics and Planning

The logistical considerations associated with organizing SSGs training are also important considerations for coaches, as these have the potential to influence player motivation and exercise intensity. For example, the total number of players available (including goalkeepers) to participate in any session will determine the number of SSG teams that can be formed, as well as the type of games implemented, particularly if the objective is to use evenly balanced teams.[7] In ª 2011 Adis Data Information BV. All rights reserved.

practice, coaches often like to create ‘competitive playing structures’, which typically require all SSG teams in one session to play against each other for an equal number of times. This type of playing structure is thought to increase motivation levels by increasing competition and placing an emphasis on results; however, this has not yet been empirically tested. It is possible that overuse of a competitive playing structure may result in the selection of an inappropriate training regimen and therefore a suboptimal training stimulus. If this occurs frequently, it may compromise longer term training adaptations. Therefore, it is suggested that coaches should select SSGs judiciously. They should also be aware that not all SSG formats will provide sufficient internal stress to provide the desired physiological adaptation. Careful planning and organization of training sessions for SSGs is also important if the approSports Med 2011; 41 (3)

Small-Sided Games Training Physiology in Football

2.3.9 Comparisons of SSG Training Intensity with Competitive Match Play

Several studies have examined how the exercise intensity of various SSGs compares with the exercise intensity of competitive match play.[8,44,55,56] The findings of these studies can also be used to determine if the most intense periods of matches compare with the intensity of various SSGs. For example, Gabbett and Mulvey[8] recruited 13 elite female football players and compared three versus three and five versus five SSGs with (i) domestic football matches against male youth teams; (ii) Australian National Women’s League football matches; and (iii) international women’s football matches. The main finding was that although SSGs simulate the overall movement patterns of domestic, national and international competition, they do not simulate the high-intensity repeated-sprint demands of international competition.[8] In contrast, Allen et al.[55] reported that although total distance was similar, the ratio of high- to low/moderate-intensity work in five versus five SSGs was higher compared with 11 versus 11 games. Similarly, the intensity of two versus two was found to exceed the intensity of State Premier League under 19 matches, while four versus four were similar to, and six versus six were below match intensity (figure 2). Capranica et al.[56] reported that the physiological intensity and movement demands of seven versus seven and 11 versus 11 in prepubescent football players were similar, with HRs exceeding ª 2011 Adis Data Information BV. All rights reserved.

95

Intensity (%HRmax)

priate training stimulus is to be achieved. For example, factors such as planning SSGs according to a prospective training plan designed to meet the physical, technical and tactical requirements of the team, along with the appropriate use of coach encouragement, pitch area, player number, goalkeepers, rule modification and selection of work and rest periods, will help achieve optimal exercise intensity. The variation in individual responses to the various SSG structures within a session and between training sessions should also be considered.[32,53,54] Finally, it is advisable to avoid skill and fitness mismatches between opposing teams in order to avoid compromising training intensity.

213

90 85 80 75 70 2 vs 2

4 vs 4

6 vs 6

Match

Playing format Fig. 2. Box and whisker plot of exercise intensity (percentage of maximum heart rate [%HRmax]) in various small-sided games and matches.[45]

170 beats per minute. In summary, it appears that selected SSG formats containing fewer players can exceed mean match intensity in youth football players. Coaches can use this information for choosing SSGs that are either more intense than match demands to overload the players, or lower than 11 versus 11 match intensity when either technical/tactical requirements or recovery and regeneration is the goal of training. 3. Studies Comparing SSGs Training with Interval Training Despite the widespread use of SSGs in football, there are surprisingly few studies comparing their effectiveness in comparison to traditional forms of fitness training. The previous studies that have been completed can be divided into the following two categories: (i) studies that investigated acute physiological responses of SSGs and compared these with generic (interval) training responses;[30,44,47] and (ii) studies involving the comparison of each training mode on either physiological performance measures and/or direct match performance.[57-59] 3.1 Acute Physiological Comparisons of SSGs Training with Interval Training

Several studies have compared the physiological responses between generic interval training with football-specific SSG training drills. Indeed, Sports Med 2011; 41 (3)

Hill-Haas et al.

214

3.2 Training Studies Comparing SSGs Training with Interval Training

There have been few studies that have examined the efficacy of using SSGs as a conditioning stimulus compared with traditional forms of fitness training. In the first controlled training study to compare both SSGs and generic training, Reilly and White[57] recruited 18 professional youth ª 2011 Adis Data Information BV. All rights reserved.

Tactical training

SSGs

Circuit Interval

100 90 Intensity (%HRmax)

many studies have shown that the exercise intensity achieved during SSGs are similar to generic fitness training drills of similar duration.[30,44,47] For example, Sassi et al.[47] compared the acute physiological responses of two formats of four versus four and eight versus eight SSGs with interval running (4 · 1000 metre repeats, separated by 150 seconds of recovery), using 11 elite professional players from a Spanish first division football club. Although there was no systematic manipulation of pitch area, game format (player number) or rule modifications in this study, the SSG formats elicited a greater %HRmax response compared with the interval running (91% vs 85% HRmax).[47] More recently, Dellal et al.[44] compared the HR response of short-duration (5- to 30-second efforts) high-intensity interval running with a variety of SSG formats, using ten elite footballers from a French first division football club. In contrast to the previous studies, only the two versus two (no goalkeepers) and eight versus eight (including goalkeepers) SSG formats generated similar HR responses compared with the short-duration interval running protocols. The one versus one (no goalkeepers) and four versus four (including goalkeepers) formats generated the lowest HR responses of both the SSGs and interval running.[44] In general, the results of these studies demonstrated that many smaller-format SSGs played on a relatively large pitch area per player, can elicit similar intensities to both longduration interval running[47] and short-duration high-intensity interval protocols.[44] However, it appears that the variability in exercise stimulus is greater in SSGs compared with generic interval training (figure 3), which may be due to the unstructured and stochastic nature of the movement demands in SSGs.

80 70 60 50 0

Fig. 3. Mean (–90% CI) exercise intensity (percentage of maximum heart rate [%HRmax]) in various football training activities. SSGs = small-sided games.

footballers from an English Premier League football club. Using a parallel matched-group design, players were allocated to a SSGs group or an aerobic interval training group (ITG). Players completed the training twice per week, as part of their normal training, over a 6-week period during the competitive season. The SSGs involved five versus five games, played in intervals of 6 · 4 minutes, interspersed with 3-minute active recovery at 50–60% HRmax. The interval running duration was matched with the SSGs, with a target intensity of 85–90% HRmax (active recovery of 3 minutes at 50–60% HRmax). All physiological performance measures, including counter movement jumps, 10–30 metre sprints, 6 · 30 second anaerobic shuttle test, the agility T-test and the multi-stage fitness test, demonstrated similar changes during the study.[57] Based on these results, the authors concluded that both SSGs and interval training are equally effective for maintaining in-season aerobic and anaerobic fitness in elite youth footballers.[57] Unfortunately, the HR responses to each type of training were not reported, making it difficult to determine if both groups received a similar internal training load during the study period. A further limitation of this study was that there was little detail of the periodization and prescription of the SSGs training. For example, the game format was restricted to five versus five for all sessions, and no detail relating to pitch area, rules or coach encouragement was provided. In a comprehensive training study comparing SSGs with generic interval training, Impellizzeri Sports Med 2011; 41 (3)

Small-Sided Games Training Physiology in Football

et al.[58] used a parallel matched-group research design, where 29 youth football players from two junior teams of Italian professional football clubs were randomly allocated to either a SSG or ITG. The 12-week training intervention spanned over 4 weeks of the pre-season and 8 weeks of the competitive season in which the players completed two sessions per week designed to improve aerobic fitness. The interval training comprised a fixed prescription of 4 · 4-minute efforts at a target intensity of 90–95% of HRmax, interspersed by 3 minutes of active recovery at 60–70% of HRmax. The SSGs training involved a mix of SSGs, including three versus three, four versus four and five versus five players. Both the duration and training intensity were matched between the groups. The results demonstrated no difference in mean exercise intensity (%HRmax) or weekly training load (session RPE) between the groups, with the exception of time spent at >95% HRmax, where the SSGs group spent ~30 seconds per session longer in this zone.[58] Fitness test results revealed similar improvements for the ITG and SSG groups for peak oxygen consump. tion (VO2peak) [8% and 7%, respectively], lactate threshold (13% and 11%, respectively) and running economy (3% for both groups) over the 12 weeks of training. Notably, the improvements . in VO2peak for ITG and SSGs for the in-season phase of the study were also very similar to the earlier study of Reilly and White[57] (0.8% and 0.7%, and 0.3% and 0.2%, respectively). Impellizzeri et al.[58] also examined the influence of generic and specific training strategies on physical performance during matches. The results revealed non-significant increases (pre-season training phase only) in low-intensity activity (forwards, backwards and sideways jogging), high-intensity activity (higher speed running and sprinting) and total distance travelled for both the ITG and SSG groups following the 12-week training period. However, when match performance measures for the in-season phase of training were analysed, the magnitude of the increases (for both groups) in low- and high-intensity activity are considerably smaller.[58] Previous training studies comparing SSGs training with interval running have demonstrated ª 2011 Adis Data Information BV. All rights reserved.

215

good research design and high internal validity. However, in the field, there are certain aspects of these studies that rarely occur. For example, it is practically difficult to apply a rigid prescription of interval training that does not have progressive overload when training elite football players. Moreover, in practice, the systematic manipulation of SSGs for the purpose of physical development is problematic, as the technical/tactical training goals of the coach do not always relate to physiological development needs or priorities. Therefore, to examine these issues, Hill-Haas et al.[59] assessed the efficacy of a coach-led SSGs programme and a progressive mixed-methods generic fitness training programme in 25 elite youth football players. Using a parallel matchedgroup research study design, the players were randomly allocated to either SSG or mixed-generic training groups over a 7-week pre-season training period. In contrast to previous research,[58] this study implemented a mixed-generic training programme (consisting mainly of aerobic power training and prolonged intermittent high-intensity interval training), and a SSGs training programme, incorporating a broad range of game formats (i.e. two vs two to seven vs seven).[59] Although the manipulation of the SSGs training variables (such as pitch area and rules) was less systematic than previous studies, a key difference was the planning and implementation of the SSGs training programme by an experienced coach, which increased the external validity of the study. The main finding of this study was that both coach-selected SSGs training and mixed-generic training (comprising short duration, high-intensity intervals of <90 seconds) were effective at significantly improving yo-yo intermittent. recovery test (level 1) performance, but not VO2max.[59] Notably, there were no between-group or traininginduced changes in any other performance measures. In general, the results of these training studies show that SSGs provide similar changes in aerobic fitness and match performance measures, with the majority of changes in fitness/performance observed during pre-season training. The studies also suggest that more effective use of this training mode is still possible. This may be achieved Sports Med 2011; 41 (3)

Hill-Haas et al.

216

4. Limitations of SSGs The first limitation relates to the current research knowledge into the prescriptive variables that affect SSGs intensity. The current volume of systematic research in this area is small and, consequently, definitive conclusions are difficult to form. Despite offering several advantages, there are also a number of suggested limitations that relate to the implementation of SSGs, including (i) the ceiling effect in achieving highexercise intensities for highly fit or skilled players; (ii) the ability to replicate the demands of the most intense periods of match play; (iii) the requirement of a high level of technical and tactical proficiency to achieve appropriate exercise intensity; (iv) the risk of contact injuries during training; and (v) and the availability of enough coaches to control and monitor this type of training. It has. been reported that players with the highest VO2peak elicited the lowest percentage of . VO2peak during SSGs,[60] suggesting that either the technical/tactical constraints of the game or the intermittent nature of the exercise can prevent some players from reaching appropriate training intensities.[61] Therefore, it was suggested that players with a high fitness level and a good skill level will not exercise at sufficient intensity to elicit aerobic fitness adaptations under these training conditions. However, in contrast, we have observed a weak but significant positive correlation between fitness level and exercise intensity during various SSGs (figure 4). These results suggest that players with a high fitness level ª 2011 Adis Data Information BV. All rights reserved.

100 Intensity (%HRmax)

through systematic manipulation of the training variables. However, it is clear that careful selection of SSG formats and training regimens is required to optimize fitness and performance gains. Combined, the evidence suggests that both SSG and interval training drills are suitable for improving fitness and performance in football players. It is most likely that a mixed-methods approach is appropriate for football training; however, the selection of these should be based on the technical, tactical and performance needs of the players.

95 90 85 80 75 1750

2250

2750

3250

MSFT distance (m) Fig. 4. Relationship between player fitness (multi-stage fitness test [MSFT] distance [m]) and exercise intensity (percentage of maximum heart rate [%HRmax]) during various small-sided games, [r = 0.26, p = 0.04].[45] Full line represents the line of best fit and the dashed line represents 95% confidence intervals.

exercise at a higher intensity during SSGs. Therefore, future research is required to elucidate the possible relationships between fitness, skill and exercise intensity during SSGs. Additionally, the intermittent nature of SSGs has been suggested to limit the ability of players to achieve sufficient cardiac load for aerobic fitness adaptations. Indeed, Hoff and Helgerud[61] argue that optimal aerobic adaptations are only possible if cardiac output remains elevated for sustained periods during football training, and that exercise intensities of >90% HRmax are required for improvements in aerobic fitness. Since SSGs are more intermittent than interval running, it has been suggested that the continual re-setting of the muscular venous pump will compromise cardiac output and consequently prevent a sustained high stroke volume being achieved.[61] It has also been reported that SSGs training may not always simulate the high-intensity, repeatedsprint demands of high level competition,[2] and it is not known if they can be used to replicate the most intense periods of the game. However, these potential physiological limitations to SSGs training may be countered by appropriate manipulation of SSGs training variables. Moreover, since SSGs involve a combination of technical/tactical ability, decision making and physical exertion, it seems that concurrent abilities may be required to achieve appropriate exercise intensities. Consequently, it is possible that Sports Med 2011; 41 (3)

Small-Sided Games Training Physiology in Football

less-skilled players may not be able to consistently sustain the technical skill or tactical proficiency to achieve and maintain the required metabolic strain; as such, training may be counterproductive in terms of playing performance.[14] However, this has not been empirically tested and future studies should examine if low technical skill ability limits the exercise intensity of individual players during SSGs. Due to the competitive nature of SSGs in football, there may be an increased risk of contact injuries during training,[7] although rule modifications may help minimize this potential problem. The incidence of injuries in skill-based conditioning games in rugby league have been reported to be lower than that of traditional fitness training.[1] However, to date, there have been no studies that have examined the incidence of injuries during SSGs training in comparison to generic training in football. Other logistic factors involved in the planning of SSGs (e.g. pitch area available, number of staff, number of players available) can also affect the effectiveness of this training mode. These include the ability to control and monitor the intensity of multiple, concurrent SSGs being played on various pitches at any one time. Therefore, a high level of organization and consistent coach encouragement is also needed to maintain player motivation. The use of technology, including real-time HR monitoring of individual players during SSGs, may also promote more effective implementation of SSGs training. In summary, there are several potential limitations to SSGs training in football. Coaches should be aware of these factors, which may reduce the effectiveness of this mode of training for developing both physical attributes and football proficiency. Therefore, for optimal use of SSGs training to improve aerobic fitness, it is suggested that a systematic approach to manipulating SSG prescriptive variables is adopted, with an emphasis on careful control and real-time monitoring 5. Future Research Future research is required to further develop our understanding of the training stimulus proª 2011 Adis Data Information BV. All rights reserved.

217

vided by football-specific SSGs. One important area that requires further investigation is the influence of modifying SSG design variables on the exercise intensity of SSGs training. This systematic review has demonstrated that, with the possible exception of player number, the majority of prescriptive variables have not been investigated thoroughly. Therefore, future research should examine the influences of manipulating selected variables such as pitch area, technical involvements and rule changes. Further research is still required before a complete understanding of how each of the SSG prescriptive variables may influence exercise intensity is gained. Another important area for future research is the influence of different periodization strategies of SSGs training for the development of physiological, technical skill and tactical proficiency. A number of interesting research questions could be posed. For example, are larger SSG formats (e.g. six vs six) more effectively used in early preseason training, while smaller game formats (e.g. two vs two) be used just prior to the competitive season? Is the overall effectiveness of SSGs training improved when implemented as part of a traditional linear periodization approach, or is it better to implement these games using a ‘block periodization model’[62] approach? To date, the training studies comparing the effectiveness of SSGs and interval running suggest that both are equally effective. Consequently, future studies should examine optimal periodization strategies for using both types of training methods for developing football-specific physical qualities. Additionally, although many studies have investigated the technical requirements of SSGs,[8,34,36,39,41-43,48,50,55,56,63,64] research conducted to date has not been very systematic. Future studies should include detailed notational analysis to provide an improved understanding of the technical skill requirements of various SSGs. This may assist coaches to better understand the link between the technical load and exercise intensity of SSGs training. One of the major advantages of SSGs training is thought to be the development of tactical awareness and decision-making capabilities, and the transfer of these to match performance. Future Sports Med 2011; 41 (3)

Hill-Haas et al.

218

research is also needed to understand the nature of the tactical awareness and decision making development provided by different SSG formats. Once established, further research should establish a link between SSGs and the transfer of these skills to match performance. 6. Conclusions Despite the extensive use of SSGs in football, our understanding of their effectiveness as a training tool for developing physical, technical and tactical skills in football players is not complete. Nevertheless, recent research has improved our understanding of some of the variables affecting SSGs intensity. Future studies are required to increase the understanding of the interaction between the technical, tactical and physical demands of SSGs, and how these can be manipulated to improve the training process for football players. However, at present, it seems that exercise intensity in SSGs can be manipulated by altering factors such as player number, numerical balance between teams, rules of play, the use of goalkeepers, pitch area and coach encouragement. It also appears that similar fitness and performance gains can be made with SSGs as is achieved with traditional interval training methods. Acknowledgements In memory of Martyn Crook, the former head coach of the Australian National under 17 and South Australian Sports Institute (SASI) men’s football squads. The authors thank Mr Crook for his coaching expertise and commitment to this project. To all the players, thank you for your time and effort during the SSGs. To Dr Greg Rowsell, thank you for providing valuable feedback on earlier versions of this manuscript. No sources of funding were used to assist in the preparation of this article. The authors have no conflicts of interest that are directly relevant to the content of this article.

References 1. Gabbett T. Skill-based conditioning games as an alternative to traditional conditioning for rugby league players. J Strength Cond Res 2006; 20 (2): 309-15 2. Gabbett T, Jenkins D, Abernethy B. Game-based training for improving skill and physical fitness in team sport athletes. Int J Sports Sci Coach 2009; 4 (2): 273-83

ª 2011 Adis Data Information BV. All rights reserved.

3. Football Federation Australia. Small-sided games handbook [online]. Available from URL: http://www.penrithfc. com/SSG%20Handbook%20for%20clubs%2019Dec07.pdf [Accessed 2010 Dec 10] 4. Gamble P. A skill-based conditioning games approach to metabolic conditioning for elite rugby football players. J Strength Cond Res 2004; 18 (3): 491-7 5. Owen A. Physiological and technical analysis of small-sided conditioned training games within professional football. Wrexham: SAGE Publications, 2003 6. Gregson W, Drust B. The physiology of football drills. Insight 2000; 3 (4): 1-2 7. Little T. Optimizing the use of soccer drills for physiological development. Strength Cond J 2009; 31 (3): 1-8 8. Gabbett T, Mulvey M. Time-motion analysis of small-sided training games and competition in elite women soccer players. J Strength Cond Res 2008; 22 (2): 543-52 9. Allison S, Thorpe R. A comparison of the effectiveness of two approaches to teaching games within physical education: a skills approach versus a games for understanding approach. Br J Phys Ed 1997; 28 (3): 9-13 10. Achten J, Jeukendrup A. Heart rate monitoring-applications and limitations. Sports Med 2003; 33 (7): 517-38 11. Drust B, Reilly T, Cable N. Physiological responses to laboratory-based soccer-specific intermittent and continuous exercise. J Sports Sci 2000; 18 (11): 885-92 12. Esposito F, Impellizzeri FM, Margonato V, et al. Validity of heart rate as an indicator of aerobic demand during soccer activities in amateur soccer players. Eur J Appl Physiol 2004; 93: 167-72 13. Hoff J, Wisløff U, Engen L, et al. Soccer specific aerobic endurance training. Br J Sports Med 2002; 36: 218-21 14. Castagna C, Belardinelli R, Abt G. The oxygen uptake and heart rate response to training with a ball in youth soccer players. J Sports Sci 2004; 22: 532-3 15. Bangsbo J. The physiology of soccer: with special reference to intense intermittent exercise. Acta Physiol Scand 1994; 619: 1-155 16. Little T, Williams A. Measures of exercise intensity during soccer training drills with professional soccer players. J Strength Cond Res 2007; 21 (2): 367-71 17. Krustrup P, Mohr M, Steensberg A, et al. Muscle and blood metabolites during a soccer game: implications for sprint performance. Med Sci Sports Exerc 2006; 38 (6): 1165-7 18. Borg G. Psychophysical basis of perceived exertion. Med Sci Sports Exerc 1982; 14 (5) 377-81 19. Coutts AJ, Rampinini E, Marcora S, et al. Heart rate and blood lactate correlates of perceived exertion during smallsided soccer games. J Sci Med Sport 2009; 12 (1): 79-84 20. Impellizzeri FM, Rampinini E, Coutts AJ, et al. Use of RPEbased training load in soccer. Med Sci Sports Exerc 2004; 36 (6): 1042-7 21. Alexiou H, Coutts AJ. A comparison of methods used for quantifying internal training load in women soccer players. Int J Sports Physiol Perform 2008; 3: 1-12 22. Foster C, Florhaug J, Franklin J, et al. A new approach to monitoring exercise training. J Strength Cond Res 2001; 15 (1): 109-15

Sports Med 2011; 41 (3)

Small-Sided Games Training Physiology in Football

23. Carling C, Bloomfield J, Nelsen L, et al. The role of motion analysis in elite soccer. Sports Med 2008; 38 (10): 839-62 24. Coutts AJ, Duffield R. Validity and reliability of GPS units for measuring movement demands of team sports. J Sci Med Sport 2010; 13 (1): 133-5 25. Macleod H, Morris J, Nevill A, et al. The validity of a nondifferential global positioning system for assessing player movement patterns in field hockey. J Sports Sci 2009; 27 (2): 121-8 26. Townsend A, Worringham C, Stewart I. Assessment of speed and position during human locomotion using nondifferential GPS. Med Sci Sports Exerc 2008; 40: 124-32 27. Edgecomb S, Norton K. Comparison of global positioning and computer-based tracking systems for measuring player movement distance during Australian Football. J Sci Med Sport 2006; 9: 25-32 28. Petersen C, Pyne D, Portus M, et al. Validity and reliability of GPS units to monitor cricket-specific movement patterns. Int J Sports Physiol Perform 2009; 4: 381-93 29. Barbero A´lvarez JC, Coutts AJ, Granda J, et al. The validity and reliability of a Global Positioning Satellite system device to assess speed and repeated sprint ability (RSA) in athletes. J Sci Med Sport 2010; 13 (2): 232-5 30. Balsom P, Lindholm T, Nilsson J, et al. Precision football. Kempele: Polar Electro Oy, 1999 31. Jeffreys I. The use of small-sided games in the metabolic training of high school soccer players. Strength Cond J 2004; 26 (5): 77-8 32. Rampinini E, Impellizzeri FM, Castagna C, et al. Factors influencing physiological responses to small-sided soccer games. J Sports Sci 2007; 25 (6): 659-66 33. Aroso J, Rebelo A, Gomes-Pereira J. Physiological impact of selected game-related exercises [abstract]. J Sports Sci 2004; 22 (6): 522 34. Owen A, Twist C, Ford P. Small-sided games: the physiological and technical effect of altering pitch size and player numbers. Insight FACA J 2004; 7 (2): 50-3 35. Williams K, Owen A. The impact of player numbers on the physiological responses to small sided games [abstract]. J Sports Sci Med 2007; 6 Suppl. 10: 100 36. Kelly D, Drust B. The effect of pitch dimensions on heart rate responses and technical demands of small-sided soccer games in elite players. J Sci Med Sport 2009; 12 (4): 475-9 37. Sampaio J, Garcia G, Macas V, et al. Heart rate and perceptual responses to 2 · 2 and 3 · 3 small-sided youth soccer games. J Sports Sci Med 2007; 6 Suppl. 10: 121-2 38. Hill-Haas S, Coutts AJ, Dawson B, et al. Time-motion characteristics and physiological responses of small-sided games in elite youth players: the influence of player number and rule changes. J Strength Cond Res 2010; 24 (8): 2149-56 39. Duarte R, Batalha N, Folgado H, et al. Effects of exercise duration and number of players in heart rate responses and technical skills during Futsal small-sided games. Open Sports Sci J 2009; 2: 37-41 40. Little T, Williams A. Suitability of soccer training drills for endurance training. J Strength Cond Res 2006; 20 (2): 316-9

ª 2011 Adis Data Information BV. All rights reserved.

219

41. Jones S, Drust B. Physiological and technical demands of 4 v 4 and 8 v 8 games in elite youth soccer players. Kinesiology 2007; 39 (2): 150-6 42. Katis A, Kellis E. Effects of small-sided games on physical conditioning and performance in young soccer players. J Sports Med 2009; 8: 374-80 43. Platt D, Maxwell A, Horn R, et al. Physiological and technical analysis of 3 v 3 and 5 v 5 youth football matches. Insight FACA J 2001; 4 (4): 23-5 44. Dellal A, Chamari K, Pintus A, et al. Heart rate responses during small-sided games and short intermittent running training in elite soccer players: a comparative study. J Strength Cond Res 2008; 22 (5): 1449-57 45. Hill-Haas S, Dawson B, Coutts AJ, et al. Physiological responses and time-motion characteristics of various smallsided soccer games in youth players. J Sports Sci 2009; 27 (1): 1-8 46. Impellizzeri FM, Rampinini E, Marcora SM. Physiological assessment of aerobic training in soccer. J Sports Sci 2005; 23 (6): 583-92 47. Sassi R, Reilly T, Impellizzeri FM. A comparison of smallsided games and interval training in elite professional soccer players [abstract]. J Sports Sci 2004; 22: 562 48. Mallo J, Navarro E. Physical load imposed on soccer players during small-sided training games. J Sports Med Phys Fit 2008; 48 (2): 166-72 49. Hill-Haas S, Rowsell G, Coutts AJ, et al. Acute physiological responses and time-motion characteristics of two smallsided training regimes in youth soccer players. J Strength Cond Res 2008; 22 (6): 1-5 50. Fanchini M, Azzalin A, Castagna C, et al. Effect of bout duration on exercise intensity and technical performance of small-sided games in soccer. J Strength Cond Res. Epub 2010 May 28 51. Coutts AJ, Murphy A, Dascombe B. Effect of direct supervision of a strength coach on measures of muscular strength and power in young rugby league players. J Strength Cond Res 2004; 18 (2): 316-23 52. Mazzetti S, Kraemer W, Volek J, et al. The influence of direct supervision on strength performance. Med Sci Sports Exerc 2000; 32: 1175-84 53. Hill-Haas S, Coutts AJ, Rowsell G, et al. Variability of acute physiological responses and performance profiles of youth soccer players in small-sided games. J Sci Med Sport 2008; 11: 487-90 54. Hill-Haas S, Rowsell G, Coutts AJ, et al. The reproducibility of physiological responses and performance profiles of youth soccer players in small-sided games. Int J Sports Physiol Perform 2008; 3 (3): 393-6 55. Allen J, Butterly R, Welsch M, et al. The physical and physiological value of 5-a-side soccer training to 11-a-side match play. J Hum Movement Stud 1998; 34: 1-11 56. Capranica L, Tessitore A, Guidetti L, et al. Heart rate and match analysis in pre-pubescent soccer players. J Sports Sci 2001; 19: 379-84 57. Reilly T, White C. Small-sided games as an alternative to interval-training for soccer players [abstract]. J Sports Sci 2004; 22 (6): 559 58. Impellizzeri FM, Marcora S, Castagna C, et al. Physiological and performance effects of generic versus specific aerobic

Sports Med 2011; 41 (3)

Hill-Haas et al.

220

training in soccer players. Int J Sports Med 2006; 27 (6): 483-92 59. Hill-Haas S, Coutts AJ, Rowsell G, et al. Generic versus small-sided game training in soccer. Int J Sports Med 2009; 30 (9): 636-42 60. Buchheit M, Laursen P, Kuhnle J, et al. Game-based training in young elite handball players. Int J Sports Med 2009; 30: 251-8 61. Hoff J, Helgerud J. Endurance and strength training for soccer players. Sports Med 2004; 34 (3): 165-80 62. Issurin VB. New horizons for the methodology and physiology of training periodization. Sports Med 2010; 40 (3): 189-206

ª 2011 Adis Data Information BV. All rights reserved.

63. Grant A, Williams M, Johnson S. Technical demands of 7 v 7 and 11 v 11 youth football matches. Insight FACA J 1999; 2 (4): 1-2 64. Grant A, Williams M, Dodd R, et al. Physiological and technical analysis of 11 v 11 and 8 v 8 youth football matches. Insight FACA J 1999; 2 (3): 3-4

Correspondence: Dr Aaron J. Coutts, School of Leisure, Sport & Tourism, University of Technology, Sydney, Kuring-gai Campus, P O Box 222, Lindfield, NSW 2070, Australia. E-mail: [email protected]

Sports Med 2011; 41 (3)

REVIEW ARTICLE

Sports Med 2011; 41 (3): 221-232 0112-1642/11/0003-0221/$49.95/0

ª 2011 Adis Data Information BV. All rights reserved.

Balance Ability and Athletic Performance Con Hrysomallis Institute of Sport, Exercise and Active Living, School of Sport and Exercise Science, Victoria University, Melbourne, Victoria, Australia

Contents Abstract. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1. Static and Dynamic Balance . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2. Balance Ability of Gymnasts Compared with Others . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3. Balance Ability of Various Athletes. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4. Comparison of Balance Ability of Athletes at Different Levels of Competition . . . . . . . . . . . . . . . . . . . 5. Relationship of Balance Ability to Performance Measures. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6. Influence of Balance Training on Sports Performance or Motor Skills . . . . . . . . . . . . . . . . . . . . . . . . . . . 7. Proposed Mechanisms for Enhancement in Performance from Balance Training . . . . . . . . . . . . . . . . 8. Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

Abstract

221 222 223 224 225 227 228 228 230

The relationship between balance ability and sport injury risk has been established in many cases, but the relationship between balance ability and athletic performance is less clear. This review compares the balance ability of athletes from different sports, determines if there is a difference in balance ability of athletes at different levels of competition within the same sport, determines the relationship of balance ability with performance measures and examines the influence of balance training on sport performance or motor skills. Based on the available data from cross-sectional studies, gymnasts tended to have the best balance ability, followed by soccer players, swimmers, active control subjects and then basketball players. Surprisingly, no studies were found that compared the balance ability of rifle shooters with other athletes. There were some sports, such as rifle shooting, soccer and golf, where elite athletes were found to have superior balance ability compared with their less proficient counterparts, but this was not found to be the case for alpine skiing, surfing and judo. Balance ability was shown to be significantly related to rifle shooting accuracy, archery shooting accuracy, ice hockey maximum skating speed and simulated luge start speed, but not for baseball pitching accuracy or snowboarding ranking points. Prospective studies have shown that the addition of a balance training component to the activities of recreationally active subjects or physical education students has resulted in improvements in vertical jump, agility, shuttle run and downhill slalom skiing. A proposed mechanism for the enhancement in motor skills from balance training is an increase in the rate of force development. There are limited data on the influence of balance training on motor skills of elite athletes. When the effectiveness of balance training was compared with resistance training, it was found that resistance training produced superior performance results for jump height and sprint time.

Hrysomallis

222

Balance ability was related to competition level for some sports, with the more proficient athletes displaying greater balance ability. There were significant relationships between balance ability and a number of performance measures. Evidence from prospective studies supports the notion that balance training can be a worthwhile adjunct to the usual training of non-elite athletes to enhance certain motor skills, but not in place of other conditioning such as resistance training. More research is required to determine the influence of balance training on the motor skills of elite athletes.

Balance is the process of maintaining the position of the body’s centre of gravity vertically over the base of support and relies on rapid, continuous feedback from visual, vestibular and somatosensory structures and then executing smooth and coordinated neuromuscular actions.[1] The relationship between balance ability and sport injury risk has been established in many cases,[2] but the relationship between balance ability and athletic performance is less clear. The importance of balance to activities such as gymnastics, rifle shooting and ice hockey may appear apparent, but the relationship to performance in many sports and motor skills has not been fully elucidated. The rationale for inclusion of balance training in an overall conditioning programme can be strengthened if it is also shown to have a positive influence on athletic performance. The aims of this review are to (i) compare the balance ability of athletes from different sports; (ii) determine if there is a difference in the balance ability of athletes at different levels of competition within the same sport; (iii) determine the relationship of balance ability with performance measures; and (iv) examine the influence of balance training on sport performance or motor skills. The review was based on journal articles identified from electronic literature searches using MEDLINE, CINAHL and SportDiscus databases from the years 1970–2009, using the following search terms in various combinations: ‘balance’, ‘postural’, ‘proprioceptive’, ‘ability’, ‘training’, ‘sport’, ‘athlete’ and ‘performance’. 1. Static and Dynamic Balance Static balance is the ability to maintain a base of support with minimal movement. Dynamic balª 2011 Adis Data Information BV. All rights reserved.

ance may be considered as the ability to perform a task while maintaining or regaining a stable position[3] or the ability to maintain or regain balance on an unstable surface[4,5] with minimal extraneous motion. When examining the relationship between balance ability and athletic performance, researchers have used a number of different tests to assess static and dynamic balance. A simple field test for static balance is the timed unipedal stance.[4,6] The most prevalent laboratory test for static balance is monitoring the centre of pressure (CoP) motion for a specified duration as an athlete attempts to stand motionless on a force platform, unipedal or bipedal and with eyes open or shut.[7-9] While it is acknowledged that CoP motion is not identical to centre of gravity motion,[10] minimal CoP motion is indicative of good balance and CoP measured from a force platform is generally considered the gold standard measure of balance.[11] Examples of field tests of dynamic balance include unipedal stance on a wobble board and counting the number of floor contacts in 30 seconds,[12] and the Star Excursion Balance Test (SEBT), which involves stable unipedal stance with maximal targeted reach distance of the free limb in a number of directions.[13,14] Results from the SEBT might also be influenced by strength, flexibility or coordination. Laboratory tests of dynamic balance include the use of a stabilometer, which requires athletes to continuously adjust posture during bipedal stance to maintain an unstable, swinging platform in the horizontal position.[4,15] Another device used to assess dynamic balance is the Biodex Balance System (consisting of an instrumented movable platform, not dissimilar to the motions of a wobble board but with adjustable levels of stability), which measures the degrees of deviation from the horizontal position.[16,17] The force Sports Med 2011; 41 (3)

Balance Ability and Athletic Performance

223

platform has also been incorporated into tests for dynamic balance by monitoring CoP motion for unipedal stance with maximum forward trunk lean[18] or by placing a tilt board on top and monitoring CoP motion.[19] It should be noted that the validity of balance tests, other than those that use a force platform and CoP data, has usually been inferred and has not yet been established by comparing the balance scores with CoP data from a force platform and displaying high correlation.[20] 2. Balance Ability of Gymnasts Compared with Others An athletic population commonly assessed for balance ability is gymnasts (table I), which is not

unexpected, since balance ability is a component of gymnastics. The balance ability of gymnasts has mostly been compared with active control subjects,[4,9,15,21-24] while two studies have compared them with other specific athletes.[13,15] The majority of studies reported some differences in balance ability; the one study that did not[21] had the smallest sample size and might have been underpowered to detect statistical differences. When looking at the data collectively, a number of trends can be identified. Overall, it was found that gymnasts were equal to or outperformed (table I) non-gymnasts. When the balance test duration exceeded 20 seconds, gymnasts performed better than non-gymnasts,[4,9,15,22-24] but not when the test was £20 seconds.[13,21] This result is a little

Table I. Balance ability of gymnasts vs non-gymnasts Study (year)

Athletes and level

Balance test

Significant findings (p < 0.05)

Kioumourtzoglou et al.[4] (1997)

Rythmic gymnasts National 60 F Controls 60 F

Static balance, timed ‘releve’ position. Dynamic balance, stabilometer, bipedal, 90 s, maintaining platform within 10 horizontal

Gymnasts superior static and dynamic balance

Vuillerme et al.[21] (2001)

Gymnasts 6 M Controls 6 M

Static balance, force platform, CoP sway, barefoot, 10 s, bipedal, unipedal, unipedal on foam mat, eyes open, eyes shut

No difference in any test with eyes open (small sample size). Gymnasts superior with no vision and unipedal stance

Aydin et al.[22] (2002)

Gymnasts 20 F Controls 20 F

Unipedal stance for 60 s eyes open then another 60 s with eyes shut on soft surface. Each surface contact with opposite limb counted

Gymnasts superior balance. No difference between limbs within each group

Davlin[15] (2004)

Gymnasts elite 29 M, 28 F Swimmers elite 32 M, 38 F Soccer players elite 30 M, 28 F Controls 31 M, 30 F

Dynamic balance, stabilometer, bipedal, 30 s, maintaining platform within 5 horizontal

Gymnasts superior to all others. Athletes superior to controls No difference between swimmers and soccer. No difference between M and F

Bressel et al.[13] (2007)

Gymnasts college 12 F Soccer players college 11 F Basketball players college 11 F

Static balance, BESS, bipedal, unipedal, tandem on stable and unstable surface, 20 s eyes shut. Dynamic balance, SEBT, results normalized to limb length

No difference between gymnasts and soccer players. Gymnasts superior static balance to basketball players. Soccer players superior dynamic balance to basketball players

Carrick et al.[23] (2007)

Gymnasts elite 156 M/F Controls 80 M/F

Static balance, foam mat on force platform, CoP sway, 25 s, bipedal, eyes shut

Gymnast superior balance

Asseman et al.[9] (2008)

Gymnasts international 13 F Controls 13 F

Static balance, force platform, CoP sway, 30 s, barefoot, unipedal, bipedal, eyes open, eyes shut

Gymnasts superior in unipedal balance with eyes open

Calavalle et al.[24] (2008)

Rhythmic gymnasts elite 15 F Controls 43 F

Static balance, force platform, CoP sway, 60 s barefoot, bipedal, eyes open, eyes shut

Gymnasts had superior balance in lateral direction but inferior in anterior-posterior. Results not normalized despite notable differences in stature and body mass between groups

BESS = balance error scoring system; CoP = centre of pressure; F = female; M = male; SEBT = star excursion balance test.

ª 2011 Adis Data Information BV. All rights reserved.

Sports Med 2011; 41 (3)

Hrysomallis

224

surprising considering that gymnasts do not maintain static postures for much more than 2 seconds during their routines. Gymnasts tended to have superior static unipedal balance,[9,13,22] superior bipedal dynamic balance[4,15] but not static bipedal balance.[9,13,21,24] The ability to maintain balance is likely to be specific to the task and possibly not a general trait. Unipedal balance may be considered difficult and specific to gymnasts; female gymnasts often practice unipedal balance skills on the balance beam, while the floor routine of male gymnasts requires unipedal stability. Bipedal stance may be considered easy and unspecific to gymnasts. There were insufficient data on dynamic unilateral balance to identify any trends. When analysing the comparative studies, it should be noted that gymnasts tend to be shorter and lighter than other athletes and stature and body mass may influence balance ability.[15] Normalizing balance scores relative to height or limb length should be considered when comparing groups with notable differences in stature or body mass[13] but this is not always done.[24] When compared with other specific athletes, gymnasts were found to have superior stabilometer bipedal dynamic balance to soccer players and swimmers.[15] The other study[13] using the Balance Error Scoring System (BESS) and SEBT found no difference in static or dynamic balance when compared with soccer players, but gymnasts had superior static balance to basketball players. The BESS involved three stance positions (bipedal [feet together], unipedal, tandem), stable and unstable surface, holding each position for 20 seconds with hands on hips, eyes shut and then various ‘errors’ were counted: opening eyes, lifting hands off the hips, foot touchdown, lifting forefoot or heel and others.[13] Gymnasts often practice and perform stationary balance and dynamic landings and may develop superior attention focus on cues such as small changes in joint position and acceleration that lead to superior balance.[13] 3. Balance Ability of Various Athletes Although gymnasts and rifle shooters appear to be the most commonly assessed for balance ability, it is the balance ability of soccer players ª 2011 Adis Data Information BV. All rights reserved.

that has been most widely compared with that of other athletes (table II). Soccer players were found to have inferior dynamic bipedal or similar static and dynamic balance to gymnasts.[13,15] They displayed similar dynamic bipedal or superior static unipedal balance to swimmers.[15,29] Compared with basketball players and active control subjects, soccer players had superior static unipedal and dynamic balance ability.[13,14,29] Soccer players frequently support their body mass on one leg when kicking a ball and may be expected to have better unipedal stability than athletes in other sports such as basketball.[29] Basketball players were not shown to have superior balance to any comparison group (table II). They had similar static unipedal balance to swimmers and inferior static and dynamic unipedal balance to soccer players and gymnasts, and inferior dynamic bipedal or similar static bipedal balance to active control subjects.[13,25,29] Swimmers displayed inferior dynamic bipedal balance to gymnasts, similar dynamic bipedal or inferior static unipedal balance to soccer players, similar static unipedal balance to basketball players and control subjects or superior dynamic bipedal balance to control subjects.[15,29] The cross-sectional studies (tables I and II) have found that athletes generally have superior balance ability compared with control subjects; this implies that sport participation improves balance. Based on the available data (table II), gymnasts tended to have the best balance ability followed by soccer players, swimmers, active control subjects and then basketball players. Basketball players rarely engage in unilateral stationary balance. Soccer players often perform dynamic unilateral movements when kicking the ball.[13] Swimmers do not usually practice or perform static or dynamic balance motions and possibly do not provide substantial stimuli to the sensorimotor systems required to enhance balance ability. Surprisingly, no studies were found that compared the balance ability of rifle shooters with other athletes. Rifle shooters were found to have superior static bipedal balance when compared with a control group, and their balance was further enhanced when they wore their competition attire weighing 7–13.5 kg; the stiff and supportive clothing and shoes diminished their body sway.[7] Sports Med 2011; 41 (3)

Balance Ability and Athletic Performance

225

Table II. Comparison of balance ability of athletes in various sports Study (year)

Athletes and level

Balance test

Significant findings (p < 0.05)

Aalto et al.[7] (1990)

Rifle and pistol shooters National 8 M, 2 F Controls 27

Static balance, force plate, CoP sway, 27 s, bipedal, eyes open, eyes shut, with and without competition clothing

Shooters superior balance to control. Rifle shooters superior balance with competitive clothing than without

Kioumourtzoglou et al.[25] (1998)

Basketball players National 13 M Controls 15 M

Dynamic balance, stabilometer, bipedal, 60 s, maintaining platform within 10 horizontal

Basketball players inferior balance but height not reported nor results normalized to height

Perrin et al.[26] (2002)

Judoists elite 17 M Ballet dancers professional 14 F Controls 21 M, 21 F

Static balance, force platform, CoP sway, 20 s, bipedal, eyes open, eyes shut. Dynamic balance, support surface moved – slow rotational oscillations of force platform, 20 s, bipedal, eyes open, eyes shut

Judoists superior to controls in all conditions. Judoists superior static balance with eyes shut than dancers. No difference between M and F controls

Davlin[15] (2004)

Gymnasts elite 29 M, 28 F Swimmers elite 32 M, 38 F Soccer players elite 30 M, 28 F Controls 31 M, 30 F

Dynamic balance, stabilometer, bipedal, 30 s, maintaining platform within 5 horizontal

Gymnasts superior to all others. Athletes superior to controls. No difference between swimmers and soccer players. No difference between M and F

Schmit et al.[27] (2005)

Track runners college 5 M, 5 F Ballet dancers college 5 M, 5 F

Static balance, force platform, with and without foam mat, CoP sway, 30 s, bipedal, barefoot, eyes open, eyes shut

No difference between runners and dancers but sample size was small

Bressel et al.[13] (2007)

Gymnasts college 12 F Soccer players college 11 F Basketball players college 11 F

Static balance, BESS, bipedal, unipedal, tandem on stable and unstable surface, 20 s eyes shut. Dynamic balance, SEBT, results normalized to limb length

No difference between gymnasts and soccer players. Gymnasts superior static to basketball players. Soccer players superior dynamic to basketball players

Gerbino et al.[28] (2007)

Soccer players college 32 F Modern and ballet dancers college 32 F

Static balance, pressure mat with foam mat, CoP sway, 10 s, unipedal, barefoot, eyes open, eyes shut. Dynamic balance, landing from a jump and a side weight shift (cutting)

Soccer players inferior to dancers in 5 of 20 tests, no difference in remaining 15. Ability to stand quietly (sway index) and ability to recover from perturbation (jumps, cutting) mostly differed

Matsuda et al.[29] (2008)

Soccer players non-elite 10 M Basketball players non-elite 10 M Swimmers non-elite 10 M Controls 10 M

Static balance, triangular force platform, CoP sway, 60 s, unipedal

Soccer players were superior to all others. No difference between limbs within each group (basketball players were not taller than other subjects)

Thorpe and Ebersole[14] (2008)

Soccer players college 12 F Controls 12 F

Dynamic balance, SEBT, unipedal stance with maximum targeted reach distance of free limb in anterior, posterior, medial and lateral directions. Results normalized to limb length

Soccer superior in anterior and posterior reach. No difference between limbs within each group

BESS = balance error scoring system; CoP = centre of pressure; F = female; M = male; SEBT = star excursion balance test.

4. Comparison of Balance Ability of Athletes at Different Levels of Competition There are some sports where elite athletes have been shown to possess superior balance ability to their less proficient counterparts (table III). International-level rifle shooters had superior bipedal static balance to national-level shooters who in turn were superior to novice shooters.[30-32] National-level soccer players had superior unipedal ª 2011 Adis Data Information BV. All rights reserved.

and bipedal static and unipedal dynamic balance compared with regional-level players.[5,19] Elite golfers were found to have better unipedal static balance than less proficient golfers;[34] unipedal stability is not automatically associated with golf but it was suggested that it may assist weight shift during the swing. Golfers may also be required to perform the golf swing with an uneven lie of the ball, uphill or downhill lie or a lie that requires one foot in a sand trap and the other on the grass.[34] Superior balance of elite athletes may be the result of Sports Med 2011; 41 (3)

Hrysomallis

226

Table III. Comparison of balance ability of athletes at different levels of competition Study (year)

Athletes and level

Balance test

Significant findings (p < 0.05)

Niinimaa and McAvoy[30] (1983)

Rifle shooters elite 4 M Biathletes experienced 4 M Biathletes rookie 4 M Controls 4 M

Static balance, force platform, bipedal, CoP, at rest, while aiming, 60 s, before and after a bout of 4 min of strenuous exercise (bike riding) to simulate cross-country ski racing

Experienced shooters had superior balance to the less experienced shooters. Balance was better at rest than in the aiming position and was better before exercise

Era et al.[31] (1996)

Rifle shooters International 6 M, 3 F National 8 M Novice 7 M

Static balance, force platform, bipedal, CoP sway while shooting, 1.5 s durations at 7.5 s and 1.5 s before shooting

International level had superior balance to the national level. National level was superior to the novice level

Konttinen et al.[32] (1999)

Rifle shooters International 6 M National 6 M

Static balance, force platform, bipedal, CoP sway while shooting, 6 s before shooting

International level had superior balance to national level

Paillard et al.[8] (2002)

Judoists National and international 11 M Regional 9 M

Static balance, force platform, bipedal, CoP sway, 51.2 s, eyes open, eyes shut

No difference between groups

Noe and Paillard[33] (2005)

Alpine skiers National and international 7 M Regional 7 M

Static balance, force platform, 51.2 s. Dynamic balance, tilt board on force platform, 25.6 s. Both bipedal, CoP sway, barefoot and knees extended, ski boots and knee flexed, eyes open, eyes shut

No difference when tested with ski boots. National and international had inferior barefoot static and dynamic balance to regional skiers

Paillard and Noe[5] (2006)

Soccer players Professional national 15 M Amateur regional 15 M

Static balance, force platform, bipedal, CoP sway, 51.2 s, eyes open, eyes shut

Professional superior balance to amateurs

Paillard et al.[19] (2006)

Soccer players National 15 M Regional 15 M

Static balance, force platform, 51.2 s. Dynamic balance, tilt board on force platform, 25.6 s. Both unipedal, CoP sway, eyes open, eyes shut

National level had superior static and dynamic balance to regional

Sell et al.[34] (2007)

Golfers Handicap <0, 45 M Handicap 0–9, 120 M Handicap 10–20, 92 M

Static balance, force platform, unipedal, 10 s, GRF sway, eyes open, eyes shut

Most proficient golf group had superior balance to other groups

Chapman et al.[35] (2008)

Surfers elite 21 M Intermediate recreational, 20 M

Static balance, balance platform, bipedal, 30 s, sway, head neutral, head back, eyes open, eyes shut

No difference between groups

CoP = centre of pressure; F = female; GRF = ground reaction force; M = male.

repetitive experience that influences motor responses and the athlete’s ability to attend to relevant proprioceptive and visual cues.[13] Training experience might also improve coordination, strength and range of motion that may enhance balance ability.[13] There are other sports where it might be expected that more proficient athletes would display better balance, but this was not found for different competition levels for alpine skiing, surfing and judo.[8,33,35] National- and international-level alpine skiers had similar static and dynamic bipedal balª 2011 Adis Data Information BV. All rights reserved.

ance to regional-level skiers when tested with ski boots but inferior barefoot static and dynamic balance when compared with regional skiers.[33] To explain this unexpected result, it was proposed that elite skiers spend more time in ski boots and possibly do not get as much postural control conditioning of the ankle-foot complex.[33] There was no difference found in the bipedal static balance ability between elite and intermediate recreational surfers.[35] Surfing performance is conducted in a highly unstable and changing environment[35] and a Sports Med 2011; 41 (3)

Balance Ability and Athletic Performance

static balance test is possibly not specific or challenging enough to discern any differences in balance ability; it could be argued that a dynamic test would be more appropriate for surfers. 5. Relationship of Balance Ability to Performance Measures Balance ability has been found to be significantly related to a number of performance measures in a number of sports (table IV). Bipedal static balance while shooting was associated with shooting accuracy for elite and novice rifle shooters.[37,40] Other factors such as rifle stability may be independent of balance and can also influence shooting accuracy.[37] Balance ability was sig-

227

nificantly related to shooting accuracy for junior archers but not senior archers.[36] The senior archers had superior balance ability when compared with junior archers; a high level of stability is a prerequisite to becoming an elite archer and, at this level of expertise, the range of postural sway is small and was not an important discriminating factor for elite senior archers.[36] The dynamic balance of young ice-hockey players displayed a significant relationship with maximum skating speed; balance is required in ice hockey because of the small surface area of the skate blades in contact with the low-friction ice surface.[39] Dynamic unipedal balance as measured by the Biodex Balance System was shown to be associated with speed during simulated luge starts[17] but not with

Table IV. Relationship between balance ability and performance measures Study (year)

Athletes and level

Balance test

Performance measure

Significant relationships (p < 0.05)

Mason and Pelgrim[36] (1986)

Archers national juniors, national seniors

Static balance, force platform, bipedal, CoP sway while shooting arrows, 1 s to shot

Arrow shooting accuracy

Balance ability was associated with shooting accuracy for juniors and less experienced (r = 0.51) but not for seniors or more experienced archers

Ball et al.[37] (2003)

Rifle shooters international 4 M, 2 F

Static balance, force platform, bipedal, CoP sway while shooting, 5, 3, 1 s to shot

Rifle shooting accuracy

Balance ability was associated with performance for four shooters

Marsh et al.[38] (2004)

Baseball pitchers college 16 M

Static balance, force platform, unipedal in the pitching balance point posture, CoP sway, 10 s, eyes open, eyes shut

Pitching accuracy – distance of ball from catcher’s mitt

No association

Behm et al.[39] (2005)

Ice-hockey players high school and junior 30 M

Dynamic balance, timed balance on a wobble board during 30 s

Maximum skating speed

Balance ability was associated with skating speed, particularly for younger players (r = 0.65)

Mononen et al.[40] (2007)

Rifle shooters novice 58 M

Static balance, force platform, bipedal, CoP sway while shooting, 3 s to shot

Rifle shooting accuracy

Balance ability was associated with shooting accuracy (r = 0.291–0.450)

Platzer et al.[17] (2009)

Luge international 13 M

Dynamic balance, Biodex Balance System, unipedal, 30 s

Luge start stimulator – end and maximal speed

Balance ability was associated with end speed (r = 0.590) but not with maximal speed

Platzer et al.[41] (2009)

Snowboarders international 21 M, 16 F

Dynamic balance, Biodex Balance System, unipedal, 30 s

World Cup and International Federation of Skiing points

No association

Wells et al.[6] (2009)

Golfers elite 15 M, 9 F

Static balance, timed unipedal stance

Ball speed and distance, average score, greens in regulation, short game measures, putting accuracy

Balance ability associated with greens in regulation (r = -0.43) and average putt distance after a chip shot (r = 0.50)

CoP = centre of pressure; F = female; M = male; r = correlation coefficient.

ª 2011 Adis Data Information BV. All rights reserved.

Sports Med 2011; 41 (3)

Hrysomallis

228

snowboarders’ ranking points.[41] The static unipedal balance of elite golfers correlated with certain performance measures: greens in regulation and average putt distance after a chip shot; it was proposed that weight shift during the golf swing and standing on uneven ground may require proficient balance.[6] One study investigating the unipedal static balance of college baseball pitchers in the ‘balance point’ posture did not find a significant association with pitching accuracy. It was previously assumed that balance was important for pitching because the action involves a ‘balance point’ during the wind-up where there is unipedal stance as the stride leg reaches the apex of the leg lift.[38] 6. Influence of Balance Training on Sports Performance or Motor Skills Balance training programmes designed to enhance performance might start with exercises on a stable surface and a bipedal stance and then progress to a unipedal stance and an unstable surface (such as foam mat, tilt board, wobble board, inflated rubber disc) with eyes open and shut, and may then incorporate movements such as tilting, rotating, squatting, hopping, jumping, throwing and catching a ball, or resistance exercises while balancing.[42] There have been a number of investigations into the influence of balance training on athletic performance measures (table V). These prospective studies have ranged from 2 to 10 weeks and mostly involved physically active non-elite subjects. It has been found that the addition of a balance training component to the activities of recreationally active subjects or physical education students has resulted in improvements in vertical jump,[12,46] agility,[46] shuttle run[18] and downhill slalom skiing.[16] It is unclear what portion of the improvements is due to the actual balance training stimulus as opposed to just the increased overall volume of physical conditioning brought about by the inclusion of balance training. It has been proposed that improvement with balance could decrease the proportions of muscles allocated to stabilization allowing them to contribute more to the motive force.[12] There are of course activities that would benefit directly from enhanced balance; downhill slalom skiing involves unpredictable surª 2011 Adis Data Information BV. All rights reserved.

faces in addition to the ankle-foot being fixed in the ski boot and unable to make major postural adjustments.[16] The evidence supports the notion that balance training can be a worthwhile adjunct to the usual training of non-elite athletes, but not in place of other conditioning such as resistance training. When the effectiveness of balance training was compared with resistance training, it was found that resistance training produced superior performance results for jump height and sprint time.[43,44] Conditioning programmes for most athletes are multifaceted but it is often unknown what contribution each training component makes to the overall performance. A multifaceted 8-week training programme for recreational golfers that included strength, flexibility and balance training produced significant increases in golf performance measures; a more stable base with greater functional flexibility and strength of the upper body allows for greater upper-body rotational velocity resulting in greater club-head speed.[45] The effectiveness of the programme was not compared with a control group or another conditioning programme that involved just the strength and flexibility training; this would have allowed for an evaluation of the contribution of the balance training component. This is an area for future research. 7. Proposed Mechanisms for Enhancement in Performance from Balance Training The relative contribution of improved motor or sensory function to enhanced performance in a motor task from balance training is unknown. Proprioception is a part of the sensory system that provides information on joint position or detecting joint motion, and is a component of the balance system. Whether proprioception can really be improved by exercise has been questioned and it is speculated that athletes might just become more skilled at focusing on and attending to important sensory cues with training and producing refined motor responses. For example, gymnasts balancing on the beam may learn to pay full attention to ensure they detect all larger body segment acceleration so as to minimize motion and improve performance.[47] Sports Med 2011; 41 (3)

Subjects

Balance training programme and other interventions

Performance measures and balance test Significant findings (p < 0.05)

Comments

Bruhn et al.[43] (2004)

Balance training 6 M, 6 F Strength training 5 M, 6 F Controls 6 M, 4 F

1 h, 2 ·/wk for 4 wk. Balance group: different balancing tasks on wobbly or unsteady surfaces. Strength training group: single repetitions, high intensity

Unipedal isometric MVC and jump height. Only strength training group › Dynamic balance, unipedal, barefoot, MVC, jump height and balance swinging platform (Posturomed) displacement, 40 s

Training status of subjects not reported. Limited details on training programmes

Malliou et al.[16] (2004)

PE students novice skiers balance group 8 M, 7 F Controls 8 M, 7 F

20 min, 4 ·/wk for 2 wk. Indoor, unipedal balance with ski boot on floor, on tilt board, with and without ski poles. Both groups had basic ski lessons for 2 wk

Downhill slalom agility test and snowploughing test. Dynamic balance, unipedal, Biodex Balance System, 20 s

Balance group was better (26%) than control for downhill slalom skiing agility test. Both groups had › balance but no difference between groups

Additional volume of training of the balance group may be partly responsible for improvement

Kean et al.[12] (2006)

Recreationally active wobble board group 11 F Jump landing group 7 F Controls 6 F

20 min, 4 ·/wk for 6 wk. Wobble board: bipedal tilting, squats, ball tosses and unipedal balancing. Jump landing: unipedal, multi-directional, controlled

Vertical jump height, 20 m sprint time. Dynamic balance, wobble board, unipedal, 30 s, number of contacts

Wobble board group › vertical jump (9%) and balance (33%)

Jump landing group used low to moderate heights; training stimulus not high

Yaggie and Campbell[18] (2006)

Recreationally active balance group 17 M/F Controls 19 M/F

20 min, 3 ·/wk for 4 wk. BOSU: unipedal stance, upright, trunk lean, head movement, eyes open, eyes shut

Shuttle run time, timed unipedal balance on BOSU eyes shut, vertical jump and reach test. Static balance, force platform, CoP sway, unipedal, 15 s. Dynamic balance, CoP sway for maximum forward trunk lean

Balance group › shuttle run (6%), static balance CoP sway and timed unipedal balance on BOSU (37%)

Balance training for 4 wk only was possibly insufficient to increase vertical jump

Cressey et al.[44] (2007)

Soccer players college UG 10 M SG 9 M

3 ·/wk for 9 wk; both groups did the same Vertical jump predicted power, 9.1 m and SG › jump power (2.4–3.2%) and resistance training programme but the 36.6 m sprint time, T-test agility time › 36.6 m sprint time more than UG did one of the supplementary UG (1.8% vs 3.9%). Both groups exercises per session (e.g. lunges) on › 9.1 m sprint and agility test inflated rubber discs but no difference between

No control group. Difference between programmes was only one exercise

Lephart et al.[45] (2007)

Recreational golfers multifaceted training group 15 M

3–4 ·/wk for 8 wk. Combined strength, flexibility and balance programme. Elastic resistance for hip, torso and shoulder rotational strengthening. Static stretches for torso rotation, shoulder flexibility and hip flexion/extension. Static squats, unipedal stance on floor and foam mat for balance (1 · 30 s each)

› In multiple strength, flexibility, balance and golf performance measures

No control group. Multifaceted programme, individual components not evaluated

Simek Salaj et al.[46] (2007)

PE students balance group 37 M Controls 38 M

60 min, 3 ·/wk for 10 wk. Tilt and wobble Vertical jump, horizontal jump and agility boards, bipedal, unipedal, static, tilting, eyes open, eye shut, hops, jumps and strength exercises on boards

Balance group › vertical jump (1.2–1.6 cm) and agility

Balance group performed a greater overall training load than the control group

Strength, flexibility, golf performance (club-head speed and total distance). Static balance, force platform, unipedal GRF sway, 10 s, eyes open, eyes shut

BOSU = both sides up balance trainer; CoP = centre of pressure; F = female; GRF = ground reaction force; M = male; MVC = maximal voluntary contraction; PE = physical education; SG = stable training group; UG = unstable training group; › indicates increased.

229

Sports Med 2011; 41 (3)

Study (year)

Balance Ability and Athletic Performance

ª 2011 Adis Data Information BV. All rights reserved.

Table V. Prospective studies on the influence of balance training on performance

Hrysomallis

230

Balance training may lead to task-specific neural adaptations at the spinal and supraspinal levels. It may suppress spinal reflex excitability such as the muscle stretch reflex during postural tasks, which leads to less destabilizing movements[48] and improved balance as required in sports such as gymnastics and rifle shooting. The inhibition of muscle stretch reflexes may enhance agonist-antagonist muscle co-contraction, which increases joint stiffness, stabilizing the joints against perturbations and therefore may improve balance.[49] Taskspecific reduced cortical excitability has also been associated with improved balance from training. It is postulated that balance training promotes a shift in movement control from cortical to subcortical and cerebellar structures.[48] These adaptations help explain the improvement in balance ability from balance training but not the increase in motor skills such as vertical jump. It should be noted that the reduced spinal and supraspinal excitability was task-specific and was demonstrated during the balance tasks. This is not necessarily evident during other movements, so it cannot be assumed that there is reduced neural excitation during various motor skills as it could be counterproductive to force and power production. It was found that balance training increased rectus femoris activation during jump landing. Greater muscle activation might optimize musculotendinous and joint stiffness, which reduces the amortization phase in the stretch-shortening cycle and subsequently improves performance in eccentric-concentric actions such as countermovement jumps.[12] An initial study demonstrated an increase in maximum voluntary isometric contraction (MVIC) force of the knee extensors and flexors of recreationally active subjects after 6 weeks of balance training;[50] however, several subsequent balance training studies have failed to generate any significant increase in strength.[43,51-53] On the weight of the evidence, it appears unlikely that an increase in strength is a significant adaptation to balance training, but what might be likely is an increase in the rate of force development (RFD). Four weeks of balance training was found to increase RFD for MVIC during a multi-joint unipedal leg press action[52] and single-joint ankle plantar flexion action ª 2011 Adis Data Information BV. All rights reserved.

of untrained subjects.[53] An increase in RFD may lead to an increase in power and, subsequently, motor skill performance such as vertical jump. There have been a number of proposed sensory adaptations to the balance training stimuli inherent in many sport activities. As with some other proposed mechanisms, they are based on low-level evidence and not on the findings of any prospective studies. It has been suggested that repetitive experience of expert athletes, such as elite surfers, might enhance balance ability by neurological adaptations that rely less on visual input and more on the other components of postural control such as proprioception.[35] The reduced necessity for visual contribution for postural control may allow more attention to be paid to other sensory input important for balance and sport performance. It has been reported[54] that gymnasts were able to more rapidly re-establish a balance position than non-gymnasts after a period of disturbed proprioceptive information caused by applying vibration to the muscle tendons around the ankle. The authors suggest that the efficiency of the process of integrating and reweighting postural control sensory information is improved by gymnastics training. Another study[55] investigated the influence of disturbing sensory input on the postural control of elite and non-elite soccer players. Sensory input was disturbed by a combination of cooling the subjects’ feet to desensitize plantar cutaneous receptors, electrically stimulating the calf and thigh muscles to disturb myotatic proprioceptive information and bracing the neck to limit information from the cervical vertebral joints. It was found that for both the disturbed and non-disturbed conditions, the elite athletes displayed better static bilateral balance. It was concluded that the elite athletes probably possessed a better knowledge of body axis and verticality. More high-level evidence from prospective studies is required to substantiate many of the proposed mechanisms for enhanced balance ability. 8. Conclusions Cross-sectional studies revealed that gymnasts tended to have the best balance ability, followed by soccer players, swimmers, active control subjects and then basketball players. No studies were found Sports Med 2011; 41 (3)

Balance Ability and Athletic Performance

that compared the balance ability of rifle shooters with other athletes. There were sports such as rifle shooting, soccer and golf where elite athletes were found to have superior balance ability compared with their less proficient counterparts, but this was not found for alpine skiing, surfing and judo. Balance ability was shown to be significantly related to rifle shooting accuracy, archery shooting accuracy, ice hockey maximum skating speed and simulated luge start speed but not for baseball pitching accuracy or snowboarding ranking points. Prospective studies have found that the addition of a balance training component to the activities of recreationally active subjects or physical education students has resulted in improvements in vertical jump, agility, shuttle run and downhill slalom skiing. Balance training may lead to task-specific neural adaptations at the spinal and supraspinal levels. It may suppress spinal reflex excitability, such as the muscle stretch reflex during postural tasks, which leads to less destabilizing movements and improved balance ability. Furthermore, balance training may increase the RFD, which can increase muscular power and subsequent performance of motor skills such as vertical jump. There are limited data on the influence of balance training on motor skills of elite athletes. When the effectiveness of balance training was compared with resistance training, it was found that resistance training produced superior performance results for jump height and sprint time. Balance ability was related to competition level for some sports with the more proficient athletes displaying greater balance ability. There were significant relationships between balance ability and a number of performance measures. Evidence from prospective studies supports the notion that balance training can be a worthwhile adjunct to the usual training of non-elite athletes to enhance certain motor skills, but not in place of other conditioning such as resistance training. More research is required to determine the influence of balance training on the motor skills of elite athletes. Acknowledgements The author has no conflicts of interest that are directly relevant to the content of this review. No sources of funding were used to assist in the preparation of this review.

ª 2011 Adis Data Information BV. All rights reserved.

231

References 1. Nashner LM. Practical biomechanics and physiology of balance. In: Jacobson GP, Newman CW, Kartush JM, editors. Handbook of balance function testing. San Diego (CA): Singular Publishing Group, 1997: 261-79 2. Hrysomallis C. Relationship between balance ability, training and sports injury risk. Sports Med 2007; 37 (6): 547-56 3. Winter DA, Patla AE, Frank JS. Assessment of balance control in humans. Med Prog Technol 1990; 16 (1-2): 31-51 4. Kioumourtzoglou E, Derri V, Mertzanidou O, et al. Experience with perceptual and motor skills in rhythmic gymnasts. Percept Mot Skills 1997; 84 (3): 1363-72 5. Paillard T, Noe F. Effect of expertise and visual contribution on postural control in soccer. Scand J Med Sci Sports 2006; 16 (5): 345-8 6. Wells GD, Elmi M, Thomas S. Physiological correlates of golf. J Strength Cond Res 2009; 23 (3): 741-50 7. Aalto H, Pyykko I, Ilmarinen R, et al. Postural stability in shooters. Otol Rhinol Laryngol 1990; 52 (4): 232-8 8. Paillard T, Costes-Salon C, Lafont C, et al. Are there differences in postural regulation according to the level of competition in judoists? Br J Sports Med 2002; 36 (4): 304-5 9. Asseman F, Caron O, Cremieux J. Are there specific conditions which expertise in gymnastics could have an effect on postural control and performance? Gait Posture 2008; 27 (1): 76-81 10. Winter DA. ABC (anatomy, biomechanics and control) of balance during standing and walking. Waterloo (ON): Waterloo Biomechanics, 1995 11. Clark RA, Bryant AL, Pau Y, et al. Validity and reliability of the Nintendo Wii Balance Board for assessment of standing balance. Gait Posture 2010; 31 (3): 307-10 12. Kean CO, Behm DG, Young WB. Fixed foot balance training increases rectus femoris activation during landing and jump height in recreationally active women. J Sports Sci Med 2006; 5 (1): 138-48 13. Bressel E, Yonker JC, Kras J, et al. Comparison of static and dynamic balance in female collegiate soccer, basketball, and gymnastics athletes. J Athl Train 2007; 42 (1): 42-6 14. Thorpe JL, Ebersole KT. Unilateral balance performance in female collegiate soccer athletes. J Strength Cond Res 2008; 22 (5): 1429-33 15. Davlin CD. Dynamic balance in high level athletes. Percept Mot Skills 2004; 98 (3): 1171-6 16. Malliou P, Amoutzas K, Theodosiou A, et al. Proprioceptive training for learning downhill skiing. Percept Mot Skills 2004; 99 (1): 149-54 17. Platzer H-P, Raschner C, Patterson C. Performancedetermining physiological factors in the luge start. J Sports Sci 2009; 27 (3): 221-6 18. Yaggie JA, Campbell BM. Effects of balance training on selected skills. J Strength Cond Res 2006; 20 (2): 422-8 19. Paillard T, Noe F, Riviere T, et al. Postural performance and strategy in the unipedal stance of soccer players at different levels of competition. J Ath Train 2006; 41 (2): 172-6 20. Riemann BL, Guskiewicz KM, Shields EW. Relationship between clinical and forceplate measures of postural stability. J Sport Rehabil 1999; 8 (2): 71-82

Sports Med 2011; 41 (3)

Hrysomallis

232

21. Vuillerme N, Danion F, Marin L, et al. The effect of expertise in gymnastics on postural control. Neurosci Lett 2001; 303 (2): 83-6 22. Aydin T, Yildiz Y, Yildiz C, et al. Proprioception of the ankle: a comparison between female teenaged gymnasts and controls. Foot Ankle Int 2002; 23 (2): 123-9 23. Carrick FR, Oggero E, Pagnacco G, et al. Posturographic testing and motor learning predictability in gymnasts. Disabil Rehabil 2007; 29 (24): 1881-9 24. Calavalle AR, Sisti D, Rocchi MBL, et al. Postural trials: expertise in rhythmic gymnastics increases control in lateral direction. Eur J Appl Physiol 2008; 104 (4): 643-9 25. Kioumourtzoglou E, Derri V, Tzetzis G, et al. Cognitive perceptual, and motor abilities in skilled basketball performance. Percept Mot Skills 1998; 86 (3): 771-86 26. Perrin P, Deviterne D, Hugel F, et al. Judo, better than dance, develops sensorimotor adaptabilities involved in balance control. Gait Posture 2002; 15 (2): 187-94 27. Schmit JM, Regis DI, Riley MA. Dynamic patterns of postural sway in ballet dancers and track athletes. Exp Brain Res 2005; 163 (3): 370-8 28. Gerbino PG, Griffin ED, Zurakowski D. Comparison of standing balance between female collegiate dancers and soccer players. Gait Posture 2007; 26 (4): 501-7 29. Matsuda S, Demura S, Uchiyama M. Centre of pressure sway characteristics during static one-legged stance of athletes from different sports. J Sports Sci 2008; 26 (7): 775-9 30. Niinimaa V, McAvoy T. Influence of exercise on body sway in standing rifle shooting. Can J Appl Sport Sci 1983; 8 (1): 30-3 31. Era P, Konttinen P, Mehto P, et al. Postural stability and skilled performance: a study on top-level and naive rifle shooters. J Biomech 1996; 29 (3): 301-6 32. Konttinen N, Lyytinen H, Era P. Brain slow potentials and postural sway behaviour during sharpshooting performance. J Mot Behav 1999; 31 (1): 11-20 33. Noe F, Paillard T. Is postural control affected by expertise in alpine skiing? Br J Sports Med 2005; 39: 835-7 34. Sell TC, Tsai Y-S, Smoliga JM, et al. Strength, flexibility, and balance characteristics of highly proficient golfers. J Strength Cond Res 2007; 21 (4): 1166-71 35. Chapman DW, Needham KJ, Allison GT, et al. Effect of experience in a dynamic environment on postural control. Br J Sports Med 2008; 42 (1): 16-21 36. Mason BR, Pelgrim PP. Body stability and performance in archery. Excel 1986; 3 (2): 17-20 37. Ball KA, Best RJ, Wrigley TV. Body sway, aim point fluctuation and performance in rifle shooters: inter- and intraindividual analysis. J Sports Sci 2003; 21 (7): 559-66 38. Marsh DW, Richard LA, Williams LA, et al. The relationship between balance and pitching error in college baseball pitchers. J Strength Cond Res 2004; 18 (3): 441-6 39. Behm DG, Wahl MJ, Button DC, et al. Relationship between hockey skating speed and selected performance measures. J Strength Cond Res 2005; 19 (2): 326-31 40. Mononen K, Konttinen N, Viitasalo J. Relationship between postural balance, rifle stability and shooting accu-

ª 2011 Adis Data Information BV. All rights reserved.

41.

42.

43.

44.

45.

46.

47.

48.

49.

50.

51.

52.

53.

54.

55.

racy among novice rifle shooters. Scand J Med Sci Sports 2007; 17 (2): 180-5 Platzer H-P, Raschner C, Patterson C, et al. Comparison of physical characteristics and performance among elite snowboarders. J Strength Cond Res 2009; 23 (5): 1427-32 Hrysomallis C, Buttifant D, Buckley N. Weight training for Australian football. Melbourne (VIC): Lothian Books, 2006: 105-9 Bruhn S, Kullmann N, Gollhofer A. The effects of a sensorimotor training and a strength training on postural stabilisation, maximum isometric contraction and jump performance. Int J Sports Med 2004; 25 (1): 56-60 Cressey EM, West CA, Tiberio DP, et al. The effects of ten weeks of lower-body unstable surface training on markers of athletic performance. J Strength Cond Res 2007; 21 (2): 561-7 Lephart SM, Smoliga JM, Myers JB, et al. An eight-week golf specific exercise program improves physical characteristics, swing mechanics, and golf performance in recreactional golfers. J Strength Cond Res 2007; 21 (3): 860-9 Simek Salaj S, Milanovic D, Jukic I. The effects of proprioceptive training on jumping and agility performance. Kinesiol 2007; 39 (2): 131-41 Ashton-Miller JA, Wojtys EM, Huston LJ, et al. Can proprioception really be improved by exercises? Knee Surg Sports Traumatol Arthrosc 2001; 9 (3): 128-36 Taube W, Gruber M, Gollhofer A. Spinal and supraspinal adaptations associated with balance training and their functional relevance. Acta Physiol 2008; 193 (2): 101-16 Lloyd D. Rationale for training programs to reduce anterior cruciate ligament injuries in Australian football. J Orthop Sports Phys Ther 2001; 31 (11): 645-54 Heitkamp H-C, Horstmann T, Mayer F, et al. Gain in strength and muscular balance after balance training. Int J Sports Med 2001; 22 (4): 285-90 Holm I, Fosdahl MA, Friis A, et al. Effect of neuromuscular training on proprioception, balance, muscle strength, and lower limb function in female team handball players. Clin J Sport Med 2004; 14 (2): 88-94 Gruber M, Gollhofer A. Impact of sensorimotor training on the rate of force development and neural activation. Eur J Appl Physiol 2004; 92 (1-2): 98-105 Gruber M, Gruber SBH, Taube W, et al. Differential effects of ballistic versus sensorimotor training on rate of force development and neural activation in humans. J Strength Cond Res 2007; 21 (1): 274-82 Vuillerme N, Teasdale N, Nougier V. The effect of expertise in gymnastics on proprioceptive sensory integration in human subjects. Neurosci Lett 2001; 311 (2): 73-6 Paillard T, Bizid R, Dupui P. Do sensorial manipulations affect subjects differently depending on their postural abilities. Br J Sports Med 2007; 41 (7): 435-8

Correspondence: Dr Con Hrysomallis, School of Sport and Exercise Science, Victoria University, PO Box 14488, Melbourne, VIC 8001, Australia. E-mail: [email protected]

Sports Med 2011; 41 (3)

REVIEW ARTICLE

Sports Med 2011; 41 (3): 233-248 0112-1642/11/0003-0233/$49.95/0

ª 2011 Adis Data Information BV. All rights reserved.

L-Arginine as a Potential Ergogenic Aid in Healthy Subjects ´ lvares,1,2 Cla´udia M. Meirelles,1,3 Yagesh N. Bhambhani,4 Vaˆnia M.F. Paschoalin2 Thiago S. A and Paulo S.C. Gomes1 1 Laboratory Crossbridges, Center for Interdisciplinary Research in Health, Department of Physical Education, Universidade Gama Filho, Rio de Janeiro, Brazil 2 Chemistry Institute, Universidade Federal do Rio de Janeiro, Rio de Janeiro, Brazil 3 Escola de Educac¸a˜o Fı´sica do Exe´rcito, Rio de Janeiro, Brazil 4 Department of Occupational Therapy, University of Alberta, Edmonton, Alberta, Canada

Contents Abstract. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1. L-Arginine and Nitric Oxide (NO) Metabolism . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2. Markers of NO Production. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.1 Cyclic Guanosine Monophosphate . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.2 Nitrate and Nitrite . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3. The ‘L-Arginine Paradox’ . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4. Contribution of NO to Exercise-Induced Vasodilation. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5. L-Arginine Supplementation on Exercise Performance. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5.1 Acute Effects . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5.2 Chronic Effects. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6. Potential Side Effects . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7. Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

Abstract

233 235 236 237 237 238 239 240 240 242 244 244

Dietary supplements containing L-arginine, a semi-essential amino acid, are one of the latest ergogenic aids intended to enhance strength, power and muscle recovery associated with both aerobic and resistance exercise. L-arginine is claimed to promote vasodilation by increasing nitric oxide (NO) production in the active muscle during exercise, improving strength, power and muscular recovery through increased substrate utilization and metabolite removal, such as lactate and ammonia. Research on L-arginine has recently tested this hypothesis, under the assumption that it may be the active compound associated with the vasodilator effects of NO. There were only five acute studies retrieved from the literature that evaluated exercise performance after L-arginine supplementation, three of which reported significant improvements. Regarding studies on chronic effects, eight studies were encountered: four reported enhancements in exercise performance, whilst four reports showed no changes. Whether these improvements in exercise performance – regardless of the aerobic or anaerobic nature of the exercise – can be associated with increases in NO production, has yet to be demonstrated in future studies. Low oral doses (£20 g) are well tolerated and clinical side

´ lvares et al. A

234

effects are rare in healthy subjects. In summary, it is still premature to recommend dietary supplements containing L-arginine as an ergogenic aid for healthy physically active subjects.

In recent years, many supplements with ergogenic properties have been developed to optimize gains in muscular strength and hypertrophy during resistance training. Supplements containing L-arginine are the latest ergogenic supplements to become commercially available. The semi-essential amino acid, L-arginine, is the only substrate for endogenous synthesis of nitric oxide (NO). The acute effects of L-arginine supplementation supposedly promote vasodilation due to enhanced NO synthesis in the active muscle during exercise. In animals, L-arginine supplementation has demonstrated positive effects on aerobic exercise performance and skeletal muscle adaptations. Maxwell et al.[1] observed that mice supplemented with L-arginine showed increases in post-exercise urinary nitrate excretion (an indicator of NO production) and aerobic capacity. (measured by maximal oxygen consumption [VO2max]). Long et al.[2] reported increases in myotube density, total nuclei number and nuclear fusion index after L-arginine supplementation. The authors of both studies conclude that enhanced exercise performance and skeletal muscle adaptations might be explained in part by the augmented NO production from L-arginine supplementation. In humans, L-arginine administration has been claimed to promote an increase in blood perfusion in the active muscle,[3] increasing substrates necessary for improving muscular recovery and protein synthesis during and/or after exercise. It also promotes greater removal of metabolites, such as lactate and ammonia,[4] which are related to muscle fatigue during intense physical exercise. In a recent survey, Malinauskas et al.[5] observed that 17% (out of 89 males) and 7% (out of 56 females) of the athletes of the National Collegiate Athletic Association (NCAA) at a southeastern state university in the US were interested in taking supplements for increasing circulation. Among these athletes, 8% and 5% of males and females, respectively, were taking L-arginine. ª 2011 Adis Data Information BV. All rights reserved.

It appears that there is an increasing interest in L-arginine-based supplements, and, therefore, more knowledge about its effects in healthy physically active subjects is needed. In 2007, McConell[6] published a review analysing the effect of both oral and intravenous L-arginine administration on metabolism at rest and during exercise. The author concluded that L-arginine supplementation appears to improve exercise capacity in individuals with cardiovascular disease, but had little impact on aerobic exercise capacity in healthy individuals. Although the author had cited other types of exercise such as resistance and anaerobic power exercise, the conclusion of the review was limited to the effects of L-arginine on aerobic exercise capacity and further suggested that studies were required to elucidate the potential ergogenic effects of L-arginine. Therefore, the aim of this review was to evaluate both the acute and chronic effects of L-arginine supplementation on physical performance during different types of exercise, including aerobic, resistance and anaerobic power exercise in healthy subjects. Furthermore, the proposed underlying mechanism by which L-arginine may function is also addressed. Scientific articles were retrieved based on an extensive search of several databases, including MEDLINE (1966–2010), EMBASE (1974–2010), Cochrane Database of Systematic Reviews (1993–2010), Lilacs (1982–2010), Scielo (1997–2010) and Google Scholar (1980–2010). Computer search engines used the following keywords combined: ‘L-arginine’, ‘supplementation’, ‘exercise’. After using these initial keywords, the search engines were limited to clinical trials, human studies and randomized controlled trials. As a result, 20 articles related to the effects of L-arginine supplementation on metabolism and performance in response to exercise in humans were considered, all of which were full-text articles and published in the English language. However, in order to reach Sports Med 2011; 41 (3)

Ergogenic Effects of L-Arginine

235

Table I. Acute effects of L-arginine supplementation on exercise performance in healthy subjects Study

No. of subjects; sex (sample characteristic); study design

Supplementation

Exercise protocol

Result

Liu et al.[7]

10; M (judo athletes); r, db, co

L-arg (6 g) or PLA 3 d

Cycle ergometer (13 sets at 0.05 kp/kg, 1 min rest at 60 rpm after set 9)

[La], ammonia, nitrate, nitrite, L-citr, MP and Pmax: L-arg = PLA

Stevens et al.[8]

13; M (NR); r, db, co

GAKic or PLA 45, 20 and 0 min before exercise

Isokinetic dynamometer (concentric/eccentric knee extension 35 reps at 90/s (0 min, 5 min, 15 min and 24 h after GAKic or PLA)

PT, TW: GAKic > PLA FI: GAKic < PLA (except after 24 h)

Buford and Koch[9]

10; M (resistance trained); r, db

GAKic or PLA 45, 30 and 10 min before exercise

Cycle ergometer (5 sets of 10 s sprints, 50 s rest intervals

McConell et al.[10]

9; M (endurance trained); r, db, co

L-arg HCl (30 g IV) or PLA after 75 min of exercise

Cycle ergometer (120 min at . 72 – 1% VO2 peak)

MP: GAKic > PLA Pmax and FI: GAKic = PLA . [La], insulin, VO2peak, RPE, FET: L-arg = PLA GCR: L-arg > PLA

Bailey et al.[11]

9; M (trained); r, db, co

L-arg (6 g) or PLA 3 d 1 h before exercise

Cycle ergometer (70–90 rpm)

[La]: L-arg = PLA FET, nitrite: L-arg > PLA . . VO2cost, VO2sc: L-arg < PLA

co = crossover; db = double-blind; FET = fatigue exercise time; FI = fatigue index; GAKic = 2 g glycine + 6 g L-arg HCl + 3.2 g a-ketoisocaproic acid; GCR = glucose clearance rate; IV = intravenous; [La] = lactate concentration; L-arg = L-arginine free form; L-arg HCl = L-arg PT = peak torque; hydrochloride; L-citr = L-citrulline; M = males; MP = mean power; NR = not reported; PLA = placebo; Pmax = peak power; . r = randomized;. reps = repetitions; RPE = rating of perceived exertion; rpm = revolutions per minute; TW = total work; VO2cost = cost of oxygen . consumption; VO2peak = peak oxygen consumption; VO2sc = slow component of oxygen consumption; > or < indicates significant difference between groups (p < 0.05); = indicates no significant differences between groups.

a more objective recommendation on the potential ergogenic effects of L-arginine supplementation, only the studies evaluating exercise performance were considered, which are duly represented in tables I and II of this review. All the studies considered were randomized, double-blind and placebo controlled. References cited on the retrieved articles were also considered in this review. 1. L-Arginine and Nitric Oxide (NO) Metabolism L-arginine is a semi-essential amino acid, which becomes an essential amino acid in special conditions, such as catabolic stress, infant growth, intestinal and kidney dysfunction.[20] L-arginine plays a role in some metabolic pathways. L-arginine is needed to synthesize creatine (Cr) and agmatine.[21] Its conversion into L-ornithine and urea, mediated by arginase, is essential in order to eliminate toxic nitrogen compounds (figure 1). Furthermore, L-arginine is important for the production of NO,[22] a potent vasodilator that acts by elevating the concentration of ª 2011 Adis Data Information BV. All rights reserved.

cyclic guanosine monophosphate (cGMP), resulting in the relaxation of smooth muscle and vasodilation (figure 2). NO is a highly reactive molecule produced endogenously in gas form. The synthesis of NO is dependent upon a family of related enzyme encoded by separate genes called NO synthase (NOS). These enzymes convert L-arginine into NO and L-citrulline in the presence of some cofactors: calmodulin, tetrahydrobiopterine, nicotinamide adenosine dinucleotide phosphate, flavin adenine dinucleotide, nicotinamide adenine dinucleotide and molecular oxygen. There are three isoforms of NOS: two of them expressed constitutively, neuronal NOS (nNOS, or type I) and endothelial NOS (eNOS, or type III) and one, expressed in an inducible way, NOS (iNOS, or type II). Although NO is primarily known for its vasodilatory effects, it is also an important regulatory molecule in many different tissues, including skeletal muscle. Studies have shown that both NOS type I (nNOS) and type III (eNOS) are expressed in skeletal muscle.[23,24] Sports Med 2011; 41 (3)

´ lvares et al. A

236

Table II. Chronic effects of L-arginine supplementation on exercise performance in healthy subjects Study

No. of subjects; sex Supplementation (sample characteristic); study design

Exercise protocol

Campbell et al.[12]

35; M (trained); r, db

AAKG (12 g) or PLA daily; RT (4 ·/wk 3 sets 8–10 reps, 8 wk 70–85% 1RM), 1RM bench-press test, Wingate test. Aerobic activity (3 ·/wk 30 min at 70% of HRmax)

Abel et al.[13]

30; M (trained); r, db

Asp-Arg (14.4 g or 5.0 g) or PLA daily; 4 wk

Cycle ergometer (100 W increased every 3 min by 30 W until exhaustion)

Colombani et al.[14] 14; M (trained); r, db

Asp-Arg (15 g) or PLA daily; 4 wk (1 wk washout)

31 km run

[La], ammonia, TDC: Asp-Arg = PLA Urea: Asp-Arg > PLA

Little et al.[15]

35; M (trained); r, db

Cr + AAKG (0.1 g/kg/d Cr + 0.075 g/kg/d AAKG) or Cr or PLA; 10 d

1RM bench-press test; 3 sets 30 s Wingate cycle tests (2 min rest)

1RM strength: Cr + AAKG = Cr > PLA Pmax: Cr + AAKG > Cr = PLA MP: Cr + AAKG = Cr = PLA

Santos et al.[16]

12; M (untrained); r, db, co

Asp-Arg (3 g) or PLA daily; 15 d

Isokinetic dynamometer (15 reps concentric knee flexion/extension 180/s)

FI: Asp-Arg < PLA FRF (%): Asp-Arg = PLA

Fricke et al.[17]

23; F (PM); r, db

L-arg HCl (18 g) or PLA; 6 mo

Dynamometric grip force and counter-movement jumping on force plate

MIGF (N), PJP (W) and PJF (N): L-arg = PLA PJF/kg: L-arg > PLA

Chen et al.[18]

16; M (cyclists); r, db

L-arg (5.2 g powder form) Cycle ergometer (until exhaustion or PLA; 3 wk at 60% MWR)

AT: L-arg > PLA . [La], VO2max , MP: L-arg = PLA

Camic et al.[19]

50; M (untrained); r, db

L-arg (1.5 g or 3.0 g) or PLA; 4 wk

PWCFT: L-arg > PLA

Cycle ergometer (80 W increasing 30 W each 2 min until exhaustion)

Results

1RM strength, AP: AAKG > PLA FET: AAKG > PLA . . FET, VO2, VCO2, [La]: Asp-Arg = PLA

1RM = one-repetition maximum; AAKG = arginine alpha-ketoglutarate; AP = anaerobic power; Asp-Arg = arginine aspartate; AT = anaerobic threshold; co = crossover; Cr = creatine; db = double-blind; F = females; FET = fatigue exercise time; FI = fatigue index; FRF = fatigue resistance factor; HRmax = maximal heart rate; [La] = lactate concentration; L-arg = L-arginine free form; L-arg HCl = L-arg hydrochloride; M = males; MIGF = maximal isometric grip force; MP = mean power; MWR = maximal work rate; PJF = peak jump force; PJP = peak jump peak; PLA = placebo; PM = postmenopausal; Pmax = peak power; PWCFT = physical . working capacity at the fatigue. threshold; r = randomized; reps = repetitions; RT . = resistance training; TDC = total distance covered; VCO2 = carbon dioxide production; VO2 = oxygen consumption; . VO2max = maximal VO2; > or < indicates greater or lesser significant difference between groups (p < 0.05); = indicates no significant differences between groups; + indicates in association.

Skeletal muscle functions mediated by NO include force and power production,[25,26] vasodilation,[27] protein synthesis,[28,29] activation of satellite cells,[30] mitochondrial biogenesis[31,32] and glucose homeostasis.[33,34] Due to a large amount of information on this topic, the reader should refer to other review articles that specifically address the underlying mechanism of NO on skeletal muscle.[35-37] The most notable function of NO is its effect on regulating vascular tone.[38] However, this function may be compromised by situations that provoke endothelial dysfunction,[39] a condition in which inadequate production of NO has been observed.[40] ª 2011 Adis Data Information BV. All rights reserved.

Many studies in humans have demonstrated the positive effects of L-arginine in modulating vascular tone via increased NO production,[41-46] which may benefit individuals with endothelial dysfunction; however, the positive effects of supplementation on modulating vascular tone in healthy and unhealthy humans are controversial.[47-49] 2. Markers of NO Production Detection of NO in biological samples represents a challenge, since its biological half-life is only a few seconds.[50] The synthesis of NO is not the only way in which the endothelium alters Sports Med 2011; 41 (3)

Ergogenic Effects of L-Arginine

237

CO2 L-arginine

2

Agmatine

3

Urea

4

1 NO

L-ornithine

Creatine L-citrulline

Fig. 1. Overview of the metabolism of L-arginine. (1) Synthesis of nitric oxide (NO) and L-citrulline from L-arginine by NO synthase (NOS); (2) synthesis of L-arginine to L-ornithine and urea by arginase; (3) decarboxylation of L-arginine to agmatine by arginine decarboxylase; (4) synthesis of L-arginine to creatine by L-arginine: glycine amidinotransferase.

vascular tone. The endothelium also triggers vasodilation via prostaglandins and/or endotheliumderived hyperpolarizing factor.[51] Acetylcholine, an endothelium-dependent vasodilator, may alter vascular tone via prostaglandins, NO and/or endothelium-derived hyperpolarizing factor synthesis.[51] Measuring NO is essential to understanding its role in many biological processes, including the L-arginine/NO pathway. Several papers have described techniques to detect NO production, both directly and indirectly. Techniques such as electron paramagnetic resonance[52] and chemiluminescence,[53] as well as electrochemical detection using intravascular probes,[54] have been used to directly quantify NO synthesis in biological models, even though they are expensive and not commonly used. Thus, this review only describes studies utilizing indirect markers of NO production. Quantifying cGMP and nitrate and nitrite in biological fluids are methods commonly used to determine the effects of NO on guanylate cyclase enzyme and on nitrate and nitrite oxidation.[55-59]

concentrations, and thus reducing vascular tone. This pathway is the mechanism by which NO regulates smooth muscle tone, and thus local blood flow. Bo¨ger et al.[55] and Bode-Bo¨ger et al.[56,57] observed significant increases in urinary cGMP concentrations after intravenous L-arginine infusion. Lucotti et al.[58] also observed significant plasma concentrations of cGMP after oral L-arginine supplementation. These data indicate that endogenous NO synthesis increased after L-arginine supplementation via intravenous or oral administration. Levels of cGMP may, however, increase for other reasons besides NO synthesis. Agonists, such as atriopeptin II, released due to the increased plasma volume, may stimulate guanylate cyclase enzyme and, therefore, increase cGMP, triggering increased coronary blood flow, irrespective of NO. 2.2 Nitrate and Nitrite

In cells and blood, oxidation of NO via several metabolic reactions results in the formation of nitrite and nitrate as the two major products.[60] Nitrite is the principal oxidation product of NO synthesis in aqueous solutions (in the absence of biological constituents such as haemoproteins). The further oxidation to nitrate requires the presence of additional oxidizing species such as oxyhaemoproteins.[61] For example, NO is quickly oxidized to nitrite via autoxidation in aqueous solutions such as biological fluids, and may react with superoxide anions to produce peroxynitrites. L-arginine Smooth muscle cells GTP NOS NO

sGC

cGMP

2.1 Cyclic Guanosine Monophosphate

Once released from the endothelial cells, NO quickly spreads to the smooth muscle cells, where it activates the soluble guanylate cyclase to form a second messenger molecule, cGMP, from the breakdown of guanosine triphosphate. The formation of cGMP activates the calcium pump inside smooth muscle cells, reducing intracellular calcium ª 2011 Adis Data Information BV. All rights reserved.

L-citrulline Vasodilation Fig. 2. Mechanism of vasodilation from L-arginine. After synthesis from L-arginine by nitric oxide (NO) synthase (NOS), NO diffuses to smooth muscle cells, in which it stimulates the soluble guanylate cyclase (sGC), resulting in enhanced synthesis of cyclic guanosine monophosphate (cGMP) from guanosine triphosphate (GTP). The increases of cGMP in the smooth muscle cells promote relaxation and, consequently, vasodilation.

Sports Med 2011; 41 (3)

´ lvares et al. A

238

In the presence of haeme groups in proteins such as haemoglobin and myoglobin, NO reacts with oxyhaemoglobin to produce metahaemoglobin and nitrate. Most nitrite and nitrate comes from diet (vegetable products contain the highest levels of nitrate; meat and bean products contain the highest levels of nitrites),[62] mineral water and bacterial synthesis, which may alter the results of the analysis. Thus, endogenous synthesis of NO may not be adequately measured by nitrate and nitrite in plasma and urine when diet is not controlled. This problem may be mitigated by a diet low in nitrate and nitrite, as well as fasting. Dietary nitrate and nitrite excretion takes from 12 hours to 3 days, depending on the prior consumption and renal function.[63] In healthy subjects with a diet low in nitrate and nitrite (210 mmol/day), approximately 50% of urinary nitrate originates from systemic NO synthesis due to L-arginine.[64] After a 12-hour fast, plasma concentrations of nitrate and nitrite appear to reach a steady-state level in healthy subjects with a diet low in nitrate and nitrite.[65] Current methods available for analysing nitrite and nitrate in plasma, serum and urine in experimental and clinical studies include colorimetric and ultraviolet spectrophotometric methods, fluorometric assays, chemiluminescence, highperformance liquid chromatography, capillary electrophoresis, gas chromatography, and gas chromatography/mass spectrometry.[63] In general, nitrite and nitrate are stable metabolites of NO present both in blood and urine, and accessible to quantitative analysis. Therefore, measurement of nitrite and nitrate in various biological fluids, notably plasma or serum and urine appear to be the most suitable, practical and reliable non-invasive method to assess systemic NO synthesis in vivo under basal conditions, as well as upon pharmacological or physical training.[60,66,67] 3. The ‘L-Arginine Paradox’ One of the factors that affect the velocity of a catalyzed reaction by an enzyme is the concentration of the substrate. L-arginine is the only substrate for the NOS, which converts L-arginine into NO and L-citrulline. Pollock et al.[68] reportª 2011 Adis Data Information BV. All rights reserved.

ed that the in vitro Michaelis-Menten constant of endothelial NOS is »3 mmol/L, whereas the L-arginine concentrations in the plasma of both healthy and non-healthy individuals ranges from 40 to 100 mmol/L.[21] The data suggest that physiological concentrations of L-arginine are enough to saturate endothelial NOS, and that supplementary L-arginine does not promote increased enzyme activity – hence the condition known as the ‘L-arginine paradox’. Studies in vivo using L-arginine supplementation have demonstrated improved endothelial function, possibly due to increased NO production. It appears that L-arginine is a limiting factor for NO synthesis in patients at risk for atherosclerosis, but not for healthy individuals. Therefore, L-arginine supplementation may be necessary only for individuals with atherosclerosis risk factors.[41-45] Among the possible explanations for this phenomenon is the presence of high levels of asymmetric dimethylarginine (ADMA), an endogenous NOS inhibitor. Higher concentrations of ADMA were encountered in individuals with atherosclerosis, as well as in individuals with atherosclerosis risk factors, such as hypercholesterolaemia, hypertension, diabetes mellitus, kidney failure, hyperhomocysteinaemia, smoking and aging.[69] Physiological levels of L-arginine and the presence of normal concentrations of ADMA saturate the endothelial NOS enzyme, promoting NO production. In these conditions, L-arginine supplementation does not affect enzyme activity. In contrast, in the presence of elevated plasma concentrations of ADMA the endothelial NOS activity diminishes, resulting in lower physiological levels of NO production. Under these conditions, L-arginine supplementation may re-establish the L-arginine/ADMA ratio in order to activate endothelial NOS.[70] Taken together, these results provide evidence that the endothelial NOS activity could be modulated by the extracellular ADMA and L-arginine levels. In general, the term ‘L-arginine paradox’ refers to specific situations in which L-arginine supplementation appears to stimulate NOS activity, even when endogenous levels are found in a physiological range. Endothelial dysfunction also increases production of reactive oxygen species, mainly superSports Med 2011; 41 (3)

Ergogenic Effects of L-Arginine

oxide anion, which appears to react with NO, producing peroxynitrites that reduce the bioavailability of NO.[71,72] This reaction may also occur immediately following a resistance exercise session, due to the superoxide anion formation during resistance exercise post-ischaemic reperfusion, which results in an imbalance between superoxide anion production and removal.[73] Hudson et al.[74] observed an increase in the plasma concentrations of protein carbonyl, an oxidative stress indicator, after two distinct resistance exercise protocols: one developed for strength and the other for hypertrophy, consisting of 11 sets of three repetitions at 90% of one-repetition maximum (1RM) strength, and four sets of ten repetitions at 75% of 1RM of a squat exercise, respectively. However, Bloomer et al.[75] demonstrated that squatting at 70% of 1RM showed no increase in oxidative stress. Based on contrasting evidence, further studies are needed to evaluate the degree of oxidative stress produced by resistance exercise and its role on NO bioavailability. It is believed that L-arginine supplementation, in addition to restoring systemic NO production, may also reduce superoxide anions released by the endothelium, particularly in hypercholesterolaemia.[76] 4. Contribution of NO to Exercise-Induced Vasodilation In response to acute exercise, numerous phenomena interact to increase blood flow to active muscles, including NO and prostaglandins.[51,77] The production of NO that occurs at the vascular level is directly related to the increase in shear stress. During an exercise session, cardiac output increases and the blood is redistributed to the active muscles. The increased blood flow induced by exercise provokes a rise in shear stress, thus creating a relationship between exercise, increased blood flow and endogenous production of NO.[78,79] There is evidence demonstrating the role of NO in exercise-induced vasodilation by the increased levels of plasma and urinary markers of NO in humans: nitrate, nitrite[67,79-81] and cGMP.[66] Jungersten et al.[79] and Maeda et al.[80,81] observed significant increases in these markers after ª 2011 Adis Data Information BV. All rights reserved.

239

an acute and 2–3 month protocol of incremental cycle ergometer exercise, respectively. However, other authors did not observe any significant changes after an acute treadmill test[82] and cycle ergometer exercise.[83] Bode-Bo¨ger et al.[67] observed increases in urinary nitrate, nitrite and cGMP only during incremental cycle ergometer exercise, when compared 1 hour after exercise. Despite the suitable, practical and reliable non-invasive method to assess changes in systemic NO synthesis in vivo, measuring nitrate and nitrite in plasma and urine require rigorous control. For example, consuming certain foods (vegetable, meat and bean products) may increase endogenous levels of these metabolites, which may bias the measurements.[62] Other techniques have been applied to determine the contribution of NO to vasodilation induced by different exercise protocols. By applying the NOS-inhibiting substance, NG-monomethylL-arginine (L-NMMA), Gilligan et al.,[84] Dyke et al.[85] and Katz et al.[86] observed a significant 7–11%, 20–30% and 10–21% reduction in forearm blood flow during a rhythmic handgrip exercise, respectively. Schrage et al.[51] reported an ~80% reduction in blood flow during a rhythmic handgrip exercise after applying another NOS inhibiting substance, NG-nitro-L-argininemethyl ester (L-NAME). The data suggest that NO contributes to the vasodilation observed during rhythmic handgrip exercise in healthy subjects. However, Radegran and Saltin[87] did not observe any significant changes in blood flow during dynamic knee extension exercises (30–50% of peak power output), but did demonstrate that NO is responsible for approximately 52% of arterial blood flow measured in the femoral region during rest and approximately 34% for the period of post-exercise recovery after L-NMMA infusion. Endo et al.[88] reported significant reductions in forearm blood flow immediately after static handgrip exercise in response to administration of L-NMMA. The contribution of NO to vasodilation may vary depending on the type of exercise. For example, exercise involving large muscle groups greatly increase blood flow and pressure, and may cause greater shear stress on endothelial cells, which is Sports Med 2011; 41 (3)

´ lvares et al. A

240

a stimulus for NO production. Green et al.[89] reported that after L-NMMA administration, blood flow during cycle ergometer exercise reduced significantly more than with rhythmic handgrip exercise. In another study, L-NMMA had no significant effect on blood flow measured during two intensities of wrist flexor exercises (0.2 and 0.4 W).[90] However, other studies showed significant reduction in blood flow during rhythmic handgrip exercise after L-NMMA administration.[84-86] It is important to mention that muscle vasodilation occurring during exercise is the result of a combination of factors besides those attributed to NO, such as prostaglandins, endotheliumderived hyperpolarizing factor, adenosine and bradykinin, among others. Boushel et al.[91] and Kalliokoski et al.[92] inhibited NO and prostaglandin production simultaneously by infusing L-NAME and indomethacin. By using near-infrared spectroscopy and the infusion of indocyanine green as a tracer, the authors observed that both L-NAME and indomethacin reduced blood flow during dynamic knee extension exercise. On the other hand, Schrage et al.[51] demonstrated that NO and prostaglandins act independently in the control of blood flow during exercise. The authors inhibited the NO production with L-NAME and observed a reduction in blood flow of approximately 17%, whereas inhibiting prostaglandins production with ketorolac, the indole moiety of indomethacin, resulted in a 32% reduction in blood flow during dynamic handgrip exercise at 10% of the maximum voluntary contraction. In summary, NO is a potent endogenous vasodilator responsible for increasing blood perfusion via shear stress. It contributes to changes in blood flow during dynamic exercise and postexercise recovery. However, NO is only one of many vasodilator substances produced by the endothelium. 5. L-Arginine Supplementation on Exercise Performance 5.1 Acute Effects

The claim that L-arginine supplementation supposedly modulates NO production and conª 2011 Adis Data Information BV. All rights reserved.

sequently increases blood perfusion to the tissues is of great interest to those who participate in aerobic- and resistance-type exercise. However, the majority of the research regarding L-arginine supplementation has utilized aerobic exercise in order to evaluate its supposed effects on performance. Table I summarizes the results of the studies that evaluated the acute effects of L-arginine supplementation on exercise performance in healthy subjects. Schaefer et al.[4] investigated metabolic changes with 3 g of intravenous L-arginine hydrochloride (HCl) during incremental cycle ergometer exercise. The authors observed a significantly lower increase in plasma lactate concentration and ammonia, besides substantially higher concentrations of L-citrulline (by-product of NO synthesis). This suggests that part of the L-arginine may have been diverted for L-citrulline and NO synthesis during exercise. Theoretically, higher lactate concentration and ammonia concentrations indicate an increase in hydrogen ions and, consequently, intramuscular acidity that reduce both strength and muscular work capacity. If so, L-arginine supplementation may be effective in reducing the aforementioned metabolite concentrations, thereby improving strength and muscle work capacity during exercise. However, Liu et al.[7] did not observe any significant differences in maximum and average anaerobic power during several sets of a cycle ergometer exercise test after orally supplementing ten elite male college judo athletes with 6 g of L-arginine (as free form) or placebo for 3 days. They also did not observe any significant difference in plasma lactate concentration, ammonia, nitrate and nitrite concentrations between groups. Bailey et al.[11] trialled nine healthy recreationally active men with a supplement that contained 6 g of L-arginine (dissolved in 500 mL of water) or placebo 1 hour before a series of moderateand severe-intensity exercise bouts performed on an electronically braked cycle ergometer for 3 days. On day 1 of supplementation, the subjects completed two 6-minute bouts of moderateintensity cycling (at 70–90 revolutions per minute [rpm]); on day 2, they completed one 6-minute bout of moderate-intensity cycling followed by Sports Med 2011; 41 (3)

Ergogenic Effects of L-Arginine

one 6-minute bout of severe-intensity cycling, and on day 3, they completed one 6-minute bout of moderate-intensity cycling followed by one bout of severe-intensity cycling that was continued until task failure, as a measure of exercise tolerance. No significant difference was observed in plasma lactate concentration between L-arginine and placebo groups. There were, however, significant increases observed in plasma nitrite and time to task failure. There was also .a significantly reduced oxygen consumption (VO2) cost of moderate-intensity cycle exercise and reduced . VO2 slow component amplitude observed between groups. It is important to note that this study associated other amino acids besides L-arginine, including L-citruline (quantities not expressed in the study), which have been shown to increase NO production, as measured by plasma concentrations of nitrite[93] and urinary excretion of nitrate and cGMP.[94] Interestingly, the authors did not measure plasma nitrite at baseline; they had just done so 1 hour after supplementation, which is a major methodological limitation, since it is not known whether there were any differences in the samples prior to supplementation. Furthermore, taking into consideration that diet can influence nitrite plasma concentrations, no dietary control to limit the consumption of foods rich in nitrite and nitrate was conducted. The authors’ conclude that the precise mechanisms responsible for improving exercise efficiency and exercise tolerance remain to be elucidated. Upon supplementing 13 subjects orally with a product comprised of L-arginine (6 g) plus glycine (2 g) plus a-ketoisocaproic acid (3.2 g) or 9.46 g sucrose isocaloric control in three equal aliquots at 45, 30 and 10 minutes before exercise, Stevens et al.[8] observed significant increase in peak torque, total work and fatigue index using an isokinetic dynamometer. By using a similar supplement protocol, Buford and Koch[9] observed significant improvement of average power during repeated sets of supra-maximal exercise during cycle ergometry. However, no significant differences in plasma lactate concentration were observed between the groups. The authors did not ª 2011 Adis Data Information BV. All rights reserved.

241

evaluate any underlying mechanism that may explain the observed physical enhancement after supplementation. It is well known that muscle glycogen is an essential fuel source for optimizing performance during moderate- to high-intensity aerobic and anaerobic exercise.[95] Therefore, replenishment of depleted muscle glycogen levels after strenuous exercise is paramount to complete recovery. Muscle glycogen synthesis may be optimized from increased skeletal muscle glucose uptake, which is enhanced by translocation of the GLUT-4 glucose transporter from intracellular vesicles to the plasma membrane in response to insulin.[96] There is evidence that NO may be playing an essential role in the regulation of skeletal muscle glucose uptake during exercise in humans.[97,98] Some studies have examined the effects of increasing endogenous NO production from L-arginine – the only endogenous nitrogencontaining substrate of NOS – on insulin and NO release, which may increase muscle glucose uptake.[10,99-101] Yaspelkis and Ivy[99] supplemented 12 trained subjects with either oral L-arginine HCl (0.08 g/kg of bodyweight) plus carbohydrate (CHO; 1 g/kg of bodyweight) or only CHO at the following intervals: 0, 1, 2 and 3 hours after 2. hours of cycle ergometer exercise at 50–90% of VO2max. They observed that the group supplemented with L-arginine plus CHO had a significantly lower CHO oxidation rate compared with the CHO-only group. They suggested that the lower rate of post-exercise CHO oxidation could increase the availability of glucose for muscular glycogen synthesis during the recovery period. However, the authors’ suggestion cannot be supported since no significant differences in plasma insulin and muscular glycogen concentrations between groups were observed during the recovery period. In a recent study,[100] 12 healthy male judo athletes performed . a single bout of treadmill exercise (at 75% VO2max) during 60 minutes, and then supplemented with oral L-arginine (0.1 g/kg bodyweight of instant powder) or placebo. The authors observed significantly higher concentrations of serum glucose 15 minutes after supplementation and insulin after 30 minutes, when Sports Med 2011; 41 (3)

´ lvares et al. A

242

compared with the placebo group. No differences in the levels of plasma lactate concentration, ammonia, nitrite and nitrate were observed between the two groups. Robinson et al.[101] observed that whole-blood glucose and plasma insulin concentrations after ingesting oral L-arginine (10 g) plus CHO (70 g) were not significantly different from placebo conditions when administered 30 minutes after different exercise protocols (non-exercised, resistance exercise or cycling exercise). McConell et al.[10] submitted nine endurance-trained males to a steady-state cycle . ergometer exercise for 120 minutes at 72 – 1% VO2peak. During the last 60 minutes of exercise, either a placebo or L-arginine HCl (30 g at 0.5 g/min) was administered intravenously. L-arginine had no significant effects on plasma insulin concentration and on cycling exercise performance as measured by mean power output in Watts and total performance time. However, L-arginine infusion significantly increased skeletal muscle glucose clearance compared with placebo. Given that plasma insulin concentration was unaffected by L-arginine infusion, the authors suggested that L-arginine increased NO production, which then increased muscle glucose uptake by skeletal muscle. Matsumoto et al.[102] submitted eight subjects (four males) to a single oral supplement of either a drink containing 2 g of branched chain amino acids (BCAA) and 0.5 g of L-arginine or an isoenergetic placebo at 10 minutes into the first exercise bout. The exercise consisted of three bouts of 20-minute cycling exercise at approximately 126 W, which corresponded to 50% of the maximal work intensity. The authors found that ingestion of BCAA plus L-arginine resulted in a significant suppression of skeletal muscle proteolysis induced by endurance exercise at a moderate intensity compared with the placebo group. The addition of L-arginine to the BCAA supplement in this study was utilized to induce an additional anabolic effect by an increase in insulin level and blood flow, although no difference in either was observed. The L-arginine dosage (500 mg) in the present study was much smaller when compared with other studies that have shown positive results (6 g).[12] This supplementation was probª 2011 Adis Data Information BV. All rights reserved.

ably insufficient to induce an additional metabolic effect via increases in the blood flow and insulin level. Only two studies have analysed the acute effect of L-arginine supplementation on blood flow during resistance exercise, neither of which demonstrated significant changes in blood flow when compared with the control group.[101,103] However, preliminary observations from our laboratory observed significant increases in blood volume – measured by near infrared spectroscopy – during the recovery period of sets of resistance exercise performed 90 minutes after oral L-arginine supplementation (as free form), without simultaneous increases in strength performance. The lack of evidence demonstrates the need to develop acute studies to evaluate the underlying mechanism that may be triggered by L-arginine supplementation in association with exercise – in particular, resistance exercise – such as changes in blood volume and/or flow, muscular oxygenation, NO production and strength performance. 5.2 Chronic Effects

The results of the studies pertaining to the chronic effects of L-arginine supplementation on exercise performance in healthy subjects are summarized in table II. Burtscher et al.[104] submitted 16 trained males to 3 weeks of oral supplementation with either arginine aspartate (3 g/day) or placebo in order to evaluate the effects of prolonged supplementation with L-arginine on metabolic and cardiorespiratory responses to submaximal exercise in healthy subjects. Incremental submaximal cycle ergometer exercise (up to 150 W) was performed before and after the supplementation period. Three weeks of arginine aspartate supplementation resulted in significantly lower plasma lactate concentration, diminished glucose oxidation and reduced ventilation and CO2 production during exercise when compared with the placebo group. Despite having observed some submaximal metabolic and cardiorespiratory improvements, the authors did not evaluate maximum exercise capacity or other physical performance indicators. Another study evaluating only cardiorespiratory response associated to L-arginine supplementation found Sports Med 2011; 41 (3)

Ergogenic Effects of L-Arginine

. no significant difference in VO2max and ventilatory threshold in 18 trained male cyclists after 28 days of oral L-arginine supplementation (12 g [6 g twice daily]).[105] Nevertheless, Abel et al.[13] observed no significant difference in lactate concentration, carbon . dioxide output and VO2 during incremental cycle ergometer exercise after 4 weeks of either high (5.7 g of arginine and 8.7 g of aspartate) or low (2.8 g of arginine and 2.2 g of aspartate) concentrations of oral arginine aspartate supplementation in 30 male endurance-trained athletes. Furthermore, the authors found no improvement in physical performance as measured by time to exhaustion. Colombani et al.[14] also observed no improvement in the time required to run 31 km after 14 days of supplementation with 15 g of oral arginine aspartate in 20 endurance-trained male athletes. They also found no change in lactate concentration and ammonia after the supplementation period. Koppo et al.[106] observed no significant difference in plasma lactate concentration in response to a cycle ergometer test at a frequency of ~70 rpm after 14 days of supplementing seven physically active males with 7.2 g of L-arginine HCl (3 · 3 capsules of 805 mg). No significant difference was observed in urinary nitrite/nitrate (utilized as a nitric oxide production indicator). Chen et al.[18] reported a significant increase in anaerobic threshold in 16 elderly men cyclists after 3 weeks of ingesting 5.2 g of L-arginine (in powder form). However, no significant differences . were observed in plasma lactate concentration, VO2max and power output between L-arginine and placebo groups. Many of the current commercial nutritional supplements that claim to enhance NO levels utilize arginine a-ketoglutarate (AAKG) as the main ‘active ingredient’. a-Ketoglutarate is an important intermediate in the Kreb’s cycle, following isocitrate and prior to succinyl coenzyme A. Campbell et al.[12] reported significant increases in 1RM strength and anaerobic power (Wingate test) after 8 weeks of oral AAKG (6 g of L-arginine and 6 g of a-ketoglutarate) supplementation. Little et al.[15] reported that both Cr (0.1 g/kg/day) and Cr + AAKG (0.075 g/kg/day) supplementation increased the total number of repetitions that could be ª 2011 Adis Data Information BV. All rights reserved.

243

performed over three sets of bench-press exercise compared with placebo. Only Cr + AAKG supplementation induced significant performance improvements in peak power during three repeated Wingate cycling tests. No effect was observed from Cr supplementation alone on repeated Wingate cycle performance. Cr supplementation increases the intramuscular stores of total Cr (i.e. Cr and phosphocreatine [PCr]), leading to an increased capacity to replenish adenosine triphosphate through PCr hydrolysis. PCr is an important energy substrate for repeated resistance exercise bouts.[107] Therefore, increased PCr availability after both Cr and Cr + AAKG supplementation could have enhanced total work capacity. The significant increase in peak power during the Wingate test after Cr + AAKG supplementation might suggest that AAKG improves the ability to generate power on repeated bouts. These results support the work of Campbell et al.[12] who found a significant increase in peak power after supplementing 35 resistance-trained healthy males with 12 g of oral AAKG. Also, Camic et al.[19] observed a significant increase on physical working capacity at the fatigue threshold (the highest power output that can be maintained without neuromuscular evidence of fatigue) in fifty untrained men performing an incremental cycle ergometer test to exhaustion after 4 weeks of 1.5 g or 3.0 g of L-arginine supplementation. Santos et al.[16] observed increased resistance capacity to muscular fatigue evaluated by isokinetic dynamometer (15 repetitions of concentric knee flexion/extension at 180/s) after 15 days of oral supplementation with arginine aspartate (3 g/day). Fricke et al.[17] observed no significant difference in maximal isometric grip force (N), utilizing a hand dynamometer as well as jump height (cm), peak jump power (W) and peak jump force (N) performed on a force plate, after 6 months of L-arginine HCl supplementation (18 g) in postmenopausal women. Peak jump force relative to bodyweight (N/kg) was the only variable that showed a significant increase in the L-arginine group, although this variable is not as important as the other jump variables to assess changes in performance. Sports Med 2011; 41 (3)

´ lvares et al. A

244

Although having observed positive results in 1RM strength,[12] anaerobic power[12,15] and muscular endurance[16] after L-arginine supplementation, the authors of these studies have not evaluated the underlying mechanism that would lead to such effects. Further studies are necessary to identify the physiological mechanism behind strength, power and muscle endurance gains reported. There is still too little scientific evidence to recommend chronic L-arginine supplementation for both aerobic and resistance exercise. 6. Potential Side Effects Studies using high doses of intravenous L-arginine (30 g) have shown side effects in normotensive healthy subjects, such as hypotension with tachycardia,[108] reduced peripheral arterial resistance[65] and an increase in cardiac output.[108] Allergic reactions, including anaphylaxis, may also result from L-arginine supplementation in some individuals,[109] suggesting that such supplementation should be avoided for individuals with allergic tendencies. Hyperkalaemia and hyperphosphataemia have been observed in patients with kidney and liver failure[110,111] and diabetes[112,113] after intravenous L-arginine administration. Evans et al.[114] supplemented healthy subjects with different levels of oral free-form L-arginine (3, 9, 21 or 30 g/day), during 1 week and observed that four of the 12 subjects supplemented with 21 g had diarrhoea, one had nausea and another had nose bleeding. At 30 g/day, nine of ten subjects experienced diarrhoea. Campbell et al.[12] reported no significant clinical side effects by orally supplementing healthy subjects with 12 g of AAKG for 8 weeks. Besides the dosage, it may be that the form of L-arginine supplementation (free-form vs AAKG) caused the observed side effects in the Evans study, compared with the Campbell study. Schulman et al.[115] observed higher mortality in patients supplemented with 9 g of L-arginine for 6 months after myocardial infarction. Therefore, the authors concluded that L-arginine supplementation is not recommended for patients post-infarction. However, Bednarz et al.[116] did not report any serious adverse effects after supplementing 792 patients with myocardial infarcª 2011 Adis Data Information BV. All rights reserved.

tion with 9 g of L-arginine for 30 days. According to the authors, the supplementation was well tolerated, although it showed no benefits. Furthermore, no other study has shown high mortality rates or any other adverse effect as a result of L-arginine supplementation in the dosage as administered by Schulman et al.[115] Sun et al.[117] recently published a metaanalysis with the purpose of analysing the effect of oral L-arginine supplementation on clinical outcomes of patients with acute myocardial infarction. Only two trials (927 participants) were included (Schulman and Bednarz studies, both described above). None of the studies showed a significant difference in event rate between the L-arginine and placebo groups. In an overall pooled estimate, there was a 7% reduction in mortality in the L-arginine treatment group compared with the control group. The authors concluded that oral L-arginine supplementation had no effect on the clinical outcomes of patients with acute myocardial infarction. Shao and Hathcock[118] implemented a methodology for risk assessment – the observed safe level (OSL) – of L-arginine supplementation and concluded that, based on the available published human clinical trial data, there is a strong evidence indicating the absence of adverse effects up to 20 g/day, and these levels are identified as OSL for normal healthy adults. Whereas high doses of both oral and intravenous L-arginine showed adverse effects in specific groups, low oral doses (£20 g) are well tolerated and adverse effects are rare in healthy subjects.[20] However, one should be conservative in recommending L-arginine supplementation until further studies can establish its safety and effectiveness in patients with myocardial infarction, and particularly in non-symptomatic individuals with silent myocardial infarction. 7. Conclusions NO is a potent endogenous vasodilator responsible for increasing blood perfusion via shear stress, and which contributes to changes in blood flow during dynamic exercise and postexercise recovery. L-arginine is a semi-essential Sports Med 2011; 41 (3)

Ergogenic Effects of L-Arginine

amino acid that is the precursor of NO, which has led many to believe that oral supplementation with this amino acid may serve as a NO stimulator. Of the five acute studies retrieved from the literature regarding L-arginine supplementation and exercise performance (table I), three studies reported significant increases in exercise performance: one reported increases in muscular peak torque, total work and reduced muscular fatigue, another study reported increases in anaerobic power and the remaining one reported increases in exercise time to fatigue. Of the eight chronic studies retrieved from the literature that evaluated exercise performance (table II), four showed significant improvements in exercise performance: three studies reported increases in anaerobic power – one of which also demonstrated significant increases in 1RM strength, and one reported a significant reduction in muscular fatigue after L-arginine supplementation. L-arginine supplementation seemed to be safe and well tolerated in the reported studies with healthy subjects, although the dosage used in the studies ranged only from 3 g to 18 g orally. No further dosages have been used in similar groups with the purpose of improving performance. Further studies are required to determine the potential ergogenic aid as well as its side effects. Based on the current information available, it cannot be assumed that the positive results on exercise performance, whether acute and/or chronic, and regardless of the different types of exercise (aerobic or anaerobic) performed, were due to increased NO production via L-arginine supplementation, since none of the reports investigated the underlying mechanisms. There is clearly a need for more studies to verify if L-arginine enhances strength, power performance and muscular recovery associated with increases in NO production in healthy subjects. Acknowledgements Professor Paulo S.C. Gomes is a recipient of a Productivity Research Fellowship from Conselho Nacional de Desenvolvimento Tecnolo´gico (CNPq) from Brazil. Thiago S. A´lvares is supported by a research scholarship from CNPq. The authors have no conflicts of interest that are directly relevant to the content of this review. The authors would like to thank Ricky

ª 2011 Adis Data Information BV. All rights reserved.

245

Toledano for the preparation of the English version of the manuscript.

References 1. Maxwell AJ, Ho HV, Le CQ, et al. L-arginine enhances aerobic exercise capacity in association with augmented nitric oxide production. J Appl Physiol 2001; 90 (3): 933-8 2. Long JH, Lira VA, Soltow QA, et al. Arginine supplementation induces myoblast fusion via augmentation of nitric oxide production. J Muscle Res Cell Motil 2006; 27 (8): 577-84 3. Rector TS, Bank AJ, Mullen KA, et al. Randomized, double-blind, placebo-controlled study of supplemental oral L-arginine in patients with heart failure. Circulation 1996; 93 (12): 2135-41 4. Schaefer A, Piquard F, Geny B, et al. L-arginine reduces exercise-induced increase in plasma lactate and ammonia. Int J Sports Med 2002; 23 (6): 403-7 5. Malinauskas BM, Overton RF, Carraway VG, et al. Supplements of interest for sport-related injury and sources of supplement information among college athletes. Adv Med Sci 2007; 52: 50-4 6. McConell GK. Effects of L-arginine supplementation on exercise metabolism. Curr Opin Clin Nutr Metab Care 2007; 10 (1): 46-51 7. Liu TH, Wu CL, Chiang CW, et al. No effect of short-term arginine supplementation on nitric oxide production, metabolism and performance in intermittent exercise in athletes. J Nutr Biochem 2009; 20 (6): 462-8 8. Stevens BR, Godfrey MD, Kaminski TW, et al. Highintensity dynamic human muscle performance enhanced by a metabolic intervention. Med Sci Sports Exerc 2000; 32 (12): 2102-8 9. Buford BN, Koch AJ. Glycine-arginine-alpha-ketoisocaproic acid improves performance of repeated cycling sprints. Med Sci Sports Exerc 2004; 36 (4): 583-7 10. McConell GK, Huynh NN, Lee-Young RS, et al. L-arginine infusion increases glucose clearance during prolonged exercise in humans. Am J Physiol Endocrinol Metab 2006; 290 (1): E60-6 11. Bailey SJ, Winyard PG, Vanhatalo A, et al. Acute L-arginine supplementation reduces the O2 cost of moderateintensity exercise and enhances high-intensity exercise tolerance. J Appl Physiol 2010; 109 (5): 1394-403 12. Campbell B, Roberts M, Kerksick C, et al. Pharmacokinetics, safety and effects on exercise performance of L-arginine alpha-ketoglutarate in trained adult men. Nutrition 2006; 22 (9): 872-81 13. Abel T, Knechtle B, Perret C, et al. Influence of chronic supplementation of arginine aspartate in endurance athletes on performance and substrate metabolism: a randomized, double-blind, placebo-controlled study. Int J Sports Med 2005; 26 (5): 344-9 14. Colombani PC, Bitzi R, Frey-Rindova P, et al. Chronic arginine aspartate supplementation in runners reduces total plasma amino acid level at rest and during a marathon run. Eur J Nutr 1999; 38 (6): 263-70 15. Little JP, Forbes SC, Candow DG, et al. Creatine, arginine alpha-ketoglutarate, amino acids, and medium-chain

Sports Med 2011; 41 (3)

´ lvares et al. A

246

16.

17.

18.

19.

20. 21. 22. 23.

24.

25.

26.

27.

28.

29.

30.

31.

32. 33.

34.

triglycerides and endurance and performance. Int J Sport Nutr Exerc Metab 2008; 18 (5): 493-508 Santos RS, Pacheco MTT, Martins RABL, et al. Study of the effect of oral administration of L-arginine on muscular performance in healthy volunteers: an isokinetic study. Isok Exerc Sci 2002; 10: 153-8 Fricke O, Baecker N, Heer M, et al. The effect of L-arginine administration on muscle force and power in postmenopausal women. Clin Physiol Funct Imaging 2008; 28 (5): 307-11 Chen S, Kim W, Henning SM, et al. Arginine and antioxidant supplement on performance in elderly male cyclists: a randomized controlled trial. J Int Soc Sports Nutr 2010; 7: 13 Camic CL, Housh TJ, Zuniga JM, et al. Effects of argininebased supplements on the physical working capacity at the fatigue threshold. J Strength Cond Res 2010; 24 (5): 1306-12 Morris Jr S. Arginine: beyond protein. Am J Clin Nutr 2006; 83: 508S-12S Bo¨ger RH, Bode-Bo¨ger S. The clinical pharmacology of L-arginine. Annu Rev Pharmacol Toxicol 2001; 41: 79-99 Moncada S, Higgs A. The L-arginine-nitric oxide pathway. N Engl J Med 1993; 329 (27): 2002-12 Nakane M, Schmidt HH, Pollock JS, et al. Cloned human brain nitric oxide synthase is highly expressed in skeletal muscle. FEBS Lett 1993; 316 (2): 175-80 Frandsen U, Lopez-Figueroa M, Hellsten Y. Localization of nitric oxide synthase in human skeletal muscle. Biochem Biophys Res Commun 1996; 227 (1): 88-93 Morrison RJ, Miller III C, Reid MB. Nitric oxide effects on shortening velocity and power production in the rat diaphragm. J Appl Physiol 1996; 80 (3): 1065-9 Morrison RJ, Miller III C, Reid MB. Nitric oxide effects on force: velocity characteristics of the rat diaphragm. Comp Biochem Physiol 1998; 119 (1): 203-9 Doshi S, Naka K, Payne N, et al. Flow-mediated dilatation following wrist and upper arm occlusion in humans: the contribution of nitric oxide. Clin Sci 2001; 101 (6): 629-35 Smith LW, Smith JD, Criswell DS. Involvement of nitric oxide synthase in skeletal muscle adaptation to chronic overload. J Appl Physiol 2002; 92 (5): 2005-11 Sellman J, DeRuisseau K, Betters J, et al. In vivo inhibition of nitric oxide synthase impairs upregulation of contractile protein mRNA in overloaded plantaris muscle. J Appl Physiol 2006; 100 (1): 258-65 Anderson JE. A role for nitric oxide in muscle repair: nitric oxide–mediated activation of muscle satellite cells. Mol Biol Cell 2000; 11 (5): 1859-74 Nisoli E, Clementi E, Paolucci C, et al. Mitochondrial biogenesis in mammals: the role of endogenous nitric oxide. Science 2003; 299 (5608): 896-9 Nisoli E, Carruba MO. Nitric oxide and mitochondrial biogenesis. J Cell Sci 2006; 119 (Pt 14): 2855-62 Balon TW, Nadler JL. Evidence that nitric oxide increases glucose transport in skeletal muscle. J Appl Physiol 1997; 82 (1): 359-63 McConell GK, Kingwell BA. Does nitric oxide regulate skeletal muscle glucose uptake during exercise? Exerc Sport Sci Rev 2006; 34 (1): 36-41

ª 2011 Adis Data Information BV. All rights reserved.

35. Reid MB. Role of nitric oxide in skeletal muscle: synthesis, distribution and functional importance. Acta Physiol Scand 1998; 162: 401-9 36. Stamler JS, Meissner G. Physiology of nitric oxide in skeletal muscle. Physiol Rev 2001; 81 (1): 209-37 37. Powers SK, Jackson MJ. Exercise-induced oxidative stress: cellular mechanisms and impact on muscle force production. Physiol Rev 2008; 88: 1243-76 38. Arnal J, Dinh-Xuan A, Pueyo M, et al. Endotheliumderived nitric oxide and vascular physiology and pathology. Cell Mol Life Sci 1999; 55 (8-9): 1078-87 39. Maxwell A, Tsao P, Cooke J. Modulation of the nitric oxide synthase pathway in atherosclerosis. Exp Physiol 1998; 83 (5): 573-87 40. Bo¨ger RH, Bode-Bo¨ger SM, Szuda A, et al. Asymmetric dimethylarginine (ADMA): a novel risk factor for endothelial dysfunction. Circulation 1998a; 98 (18): 1842-7 41. Creager M, Gallagher S, Girerd X, et al. L-arginine improves endothelium-dependent vasodilation in hypercholesterolemic humans. J Clin Invest 1992; 90 (4): 1248-53 42. Clarkson P, Adams M, Powe A, et al. Oral L-arginnine improves endothelium-dependent dilation in hypercholesterolemic young adults. J Clin Invest 1996; 97 (8): 1989-94 43. Pieper G, Siebeneich W, Dondlinger L. Short-term oral administration of L-arginine reverses defective endotheliumdependent relaxation and cGMP generation in diabetes. Eur J Pharmacol 1996; 317 (2-3): 317-20 44. Adams M, McCredie R, Jessup W, et al. Oral L-arginine improves endothelium-dependent dilatation and reduces monocyte adhesion to endothelial cells in young men with coronary artery disease. Atherosclerosis 1997; 129 (2): 261-9 45. Lerman A, Burnett Jr J, Higano S, et al. Long-term L-arginine supplementation improves small-vessel coronary endothelial function in humans. Circulation 1998; 97 (21): 2123-8 46. West S, Likos-Krick A, Brown P, et al. Oral L-arginine improves hemodynamic responses to stress and reduce plasma homocysteine in hypercholesterolemic men. J Nutr 2005; 135 (2): 212-7 47. Imaizumi T, Hirooka Y, Masaki H, et al. Effects of L-arginine on forearm vessels and responses to acetylcholine. Hyperthension 1992; 20 (4): 511-7 48. Adams M, Forsyth C, Jessup W, et al. Oral L-arginine inhibits platelet aggregation but does not enhance endotheliumdependent dilation in healthy young men. J Am Coll Cardiol 1995; 26 (4): 1054-61 49. Blum A, Hathaway L, Mincemoyer R, et al. Oral L-arginine in patients with coronary artery disease on medical management. Circulation 2000b; 101 (18): 2160-4 50. Archer S. Measurement of nitric oxide in biological models. FASEB J 1993; 7 (2): 349-53 51. Schrage WG, Dietz NM, Eisenach JH, et al. Agonistdependent variability of contributions of nitric oxide and prostaglandins in human skeletal muscle. J Appl Physiol 2005; 98 (4): 1251-7 52. Xia Y, Zweier JL. Direct measurement of nitric oxide generation from nitric oxide synthase. Proc Natl Acad Sci U S A 1997; 94 (23): 12705-10 53. Laver JR, Stevanin TM, Read RC. Chemiluminescence quantification of NO and its derivatives in liquid samples. Methods Enzymol 2008; 436: 113-27

Sports Med 2011; 41 (3)

Ergogenic Effects of L-Arginine

54. Davies IR, Zhang X. Nitric oxide selective electrodes. Methods Enzymol 2008; 436: 63-95 55. Bo¨ger RH, Bode-Bo¨ger SM, Thiele W, et al. Restoring vascular nitric oxide formation by L-arginine improves the symptoms of intermittent claudication in patients with peripheral arterial occlusive disease. J Am Coll Cardiol 1998b; 32 (5): 1336-44 56. Bode-Bo¨ger SM, Bo¨ger RH, Alfke H, et al. L-arginine induces nitric oxide-dependent vasodilation in patients with critical limb ischemia: a randomized, controlled study. Circulation 1996; 93 (1): 85-90 57. Bode-Bo¨ger SM, Bo¨ger RH, Galland A, et al. L-arginineinduced vasodilation in healthy humans: pharmacokineticpharmacodynamic relationship. Br J Pharmacol 1998; 46 (5): 489-97 58. Lucotti P, Setola E, Monti LD, et al. Beneficial effects of a long-term oral L-arginine treatment added to a hypocaloric diet and exercise training program in obese, insulin-resistant type 2 diabetic patients. Am J Physiol Endocrinol Metab 2006; 291 (5): 906-12 59. Jobgen WS, Jobgen SC, Li H, et al. Analysis of nitrite and nitrate in biological samples using high-performance liquid chromatography. J Chromatogr B 2007; 851 (1-2): 71-82 60. Tsikas D. Methods of quantitative analysis of the nitric oxide metabolites nitrite and nitrate in human biological fluids. Free Radic Res 2005; 39 (8): 797-815 61. Ignarro LJ, Fukuto JM, Griscavage JM, et al. Oxidation of nitric oxide in aqueous solution to nitrite but not nitrate: comparison with enzymatically formed nitric oxide from L-arginine. Proc Natl Acad Sci U S A 1993; 90 (17): 8103-7 62. Griesenbeck JS, Steck MD, Huber Jr JC, et al. Development of estimates of dietary nitrates, nitrites, and nitrosamines for use with the short Willet food frequency questionnaire. Nutr J 2009; 6: 8-16 63. Ellis G, Adatia I, Yazdanpanah M, et al. Nitrite and nitrate analyses: a clinical biochemistry perspective. Clin Biochem 1998; 31 (4): 195-220 64. Castillo L, Beaumier L, Ajami AM, et al. Whole body nitric oxide synthesis in healthy men determined from [15N] arginine-to-[15N]citrulline labeling. Proc Natl Acad Sci U S A 1996; 93 (21): 11460-5 65. Rhodes P, Leone AM, Francis PL, et al. The L-arginine: nitric oxide pathway is the major source of plasma nitrite in fasted humans. Biochem Biophys Res Commun 1995; 209 (2): 590-6 66. Bode-Bo¨ger SM, Bo¨ger RH, Creutzig A, et al. L-arginine infusion decreases peripheral resistance and inhibits platelet aggregation in healthy subjects. Clin Sci 1994; 87 (3): 303-10 67. Bode-Bo¨ger SM, Boger RH, Schroder EP, et al. Exercise increases systemic nitric oxide production in men. J Cardiovasc Risk 1994; 1 (2): 173-8 68. Pollock J, Fo¨rstermann U, Mitchell J, et al. Purification and characterization of particulate endothelium-derived relaxing factor synthase from cultured and native bovine aortic endothelial cells. Proc Natl Acad Sci U S A 1991; 88 (23): 10480-4 69. Cooke PJ. Does ADMA cause endothelial dysfunction? Arterioscler Thromb Vasc Biol 2000; 20 (9): 2032-7

ª 2011 Adis Data Information BV. All rights reserved.

247

70. Bode-Bo¨ger SM, Scalera F, Ignarro LJ. The L-arginine paradox: importance of the L-arginine/asymmetrical dimethylarginine ratio. Pharmacol Ther 2007; 114 (3): 295-306 71. Goumas G, Tentolouris C, Tousoulis D, et al. Therapeutic modification of the L-arginine-eNOS pathway in cardiovascular diseases. Atherosclerosis 2001; 154 (2): 255-67 72. Loscalzo J. What we know and don’t know about L-arginine and NO. Circulation 2000; 101 (18): 2126-9 73. Belviranli M, Go¨kbel H. Acute exercise induced oxidative stress and antioxidant changes. Eur J Gen Med 2006; 3 (3): 126-31 74. Hudson MB, Hosick PA, McCaulley GO, et al. The effect of resistance exercise on humoral markers of oxidative stress. Med Sci Sports Exerc 2008; 40 (3): 542-8 75. Bloomer RJ, Falvo MJ, Fry AC, et al. Oxidative stress response in trained men following repeated squats or sprints. Med Sci Sports Exerc 2006; 38 (8): 1436-42 76. Bo¨ger RH, Bode-Bo¨ger SM, Mu¨gge A, et al. Supplementation of hypercholesterolaemic rabbits with L-arginine reduces the vascular release of superoxide anions and restores NO production. Atherosclerosis 1995; 117 (2): 273-84 77. Maiorana A, O’Driscoll G, Taylor R, et al. Exercise and the nitric oxide vasodilator system. Sports Med 2003; 33 (14): 1013-35 78. Hickner RC, Fisher JS, Ehsani AA, et al. Role of nitric oxide in skeletal muscle blood flow at rest and during dynamic exercise in humans. Am J Physiol 1997; 273 (1 Pt 2): H405-10 79. Jungersten L, Ambring A, Wall B, et al. Both physical fitness and acute exercise regulate nitric oxide formation in healthy humans. J Appl Physiol 1997; 82 (3): 760-4 80. Maeda S, Miyauchi T, Kakiyama T, et al. Effects of exercise training of 8 weeks and detraining on plasma levels of endothelium-derived factors, endothelin-1 and nitric oxide, in healthy young humans. Life Sci 2001; 69 (9): 1005-16 81. Maeda S, Tanabe T, Otsuki T, et al. Moderate regular exercise increases basal production of nitric oxide in elderly women. Hypertens Res 2004; 27 (12): 947-53 82. Poveda JJ, Riestra A, Salas E, et al. Contribution of nitric oxide to exercise-induced changes in healthy volunteers: effects of acute exercise and long-term physical training. Eur J Clin Invest 1997; 27 (11): 967-71 83. Yamamoto K, Kondo T, Kimata A, et al. Lack of effect of aerobic physical exercise on endothelium-derived nitric oxide concentrations in healthy young subjects. Nagoya J Med Sci 2007; 69 (3-4): 167-72 84. Gilligan DM, Panza JA, Kilcoyne CM, et al. Contribution of endothelium-derived nitric oxide to exercise-induced vasodilation. Circulation 1994; 90 (6): 2853-8 85. Dyke CK, Proctor DN, Dietz NM, et al. Role of nitric oxide in exercise hyperaemia during prolonged rhythmic handgripping in humans. J Physiol 1995; 488 (Pt 1): 259-65 86. Katz SD, Krum H, Khan T, et al. Exercise-induced vasodilation in forearm circulation of normal subjects and patients with congestive heart failure: role of endothelium-derived nitric oxide. J Am Coll Cardiol 1996; 28 (3): 585-90 87. Radegran G, Saltin B. Nitric oxide in the regulation of vasomotor tone in human skeletal muscle. Am J Physiol 1999; 6 (Pt 2): 1951-60

Sports Med 2011; 41 (3)

248

88. Endo T, Imaizumi T, Tagawa T, et al. Role of nitric oxide in exercise-induced vasodilation of the forearm. Circulation 1994; 90 (6): 2886-90 89. Green DJ, Bilsborough W, Naylor LH, et al. Comparison of forearm blood flow responses to incremental handgrip and cycle ergometer exercise: relative contribution of nitric oxide. J Physiol 2005; 562 (Pt 2): 617-28 90. Wilson JR, Kapoor S. Contribution of endotheliumderived relaxing factor to exercise-induced vasodilation in humans. J Appl Physiol 1993; 75 (6): 2740-4 91. Boushel R, Langberg H, Gemmer C, et al. Combined inhibition of nitric oxide and prostaglandins reduces human skeletal muscle blood flow during exercise. J Physiol 2002; 543 (2): 691-8 92. Kalliokoski K, Langberg H, Ryberg A, et al. Nitric oxide and prostaglandins influence local skeletal muscle blood flow during exercise in humans: coupling between local substrate uptake and blood flow. Am J Physiol Regul Integr Comp Physiol 2006; 291 (3): 803-9 93. Sureda A, Cordova A, Ferrer MD, et al. Effects of L-citrulline oral supplementation on polymorphonuclear neutrophils oxidative burst and nitric oxide production after exercise. Free Radic Res 2009; 43 (9): 828-35 94. Schwedhelm E, Maas R, Freese R, et al. Pharmacokinetic and pharmacodynamic properties of oral L-citrulline and L-arginine: impact on nitric oxide metabolism. Br J Clin Pharmacol 2008; 65 (1): 51-9 95. Bergstro¨m J, Hultman E. Synthesis of muscle glycogen in man after glucose and fructose infusion. Acta Med Scand 1967; 182 (1): 93-107 96. Hayashi T, Wojtaszewski JF, Goodyear LJ. Exercise regulation of glucose transport in skeletal muscle. Am J Physiol 1997; 273 (6 Pt 1): E1039-51 97. Bradley SJ, Kingwell BA, McConell GK. Nitric oxide synthase inhibition reduces leg glucose uptake but not blood flow during dynamic exercise in humans. Diabetes 1999; 48 (9): 1815-21 98. Kingwell BA, Formosa M, Muhlmann M, et al. Nitric oxide synthase inhibition reduces glucose uptake during exercise in individuals with type 2 diabetes more than in control subjects. Diabetes 2002; 51 (8): 2572-80 99. Yaspelkis 3rd BB, Ivy JL. The effect of a carbohydratearginine supplement on postexercise carbohydrate metabolism. Int J Sport Nutr 1999; 9 (3): 241-50 100. Tsai PH, Tang TK, Juang CL, et al. Effects of arginine supplementation on post-exercise metabolic responses. Chin J Physiol 2009; 52 (3): 136-42 101. Robinson TM, Sewell DA, Greenhaff PL. L-arginine ingestion after rest and exercise: effects on glucose disposal. Med Sci Sports Exerc 2003; 35 (8): 1309-15 102. Matsumoto K, Mizuno M, Mizuno T, et al. Branched-chain amino acids and arginine supplementation attenuates skeletal muscle proteolysis induced by moderate exercise in young individuals. Int J Sports Med 2007; 28 (6): 531-8 103. Fahs CA, Heffernan KS, Fernhall B. Hemodynamic and vascular response to resistance exercise with L-arginine. Med Sci Sports Exerc 2009; 41 (4): 773-9

ª 2011 Adis Data Information BV. All rights reserved.

´ lvares et al. A

104. Burtscher M, Brunner F, Faulhaber M, et al. The prolonged intake of L-arginine-l-aspartate reduces blood lactate accumulation and oxygen consumption during submaximal exercise. J Sports Sci Med 2005; 4: 314-22 . 105. Sunderland KL, Greer F, Morales J. VO2max and ventilatory threshold of trained cyclists are not affected by 28-day L-arginine supplementation. J Strength Cond Res. Epub 2010 Jun 23 106. Koppo K, Taes YE, Pottier A, et .al. Dietary arginine supplementation speeds pulmonary VO2 kinetics during cycle exercise. Med Sci Sports Exerc 2009; 41 (8): 1626-32 107. Lambert CP, Flynn MG. Fatigue during high-intensity intermittent exercise: application to bodybuilding. Sports Med 2002; 32 (8): 511-22 108. Hishikawa K, Nakaki T, Tsuda M, et al. Effect of systemic L-arginine administration on hemodynamics and nitric oxide release in man. Jpn Heart J 1992; 33 (1): 41-8 109. Tiwary CM, Rosenbloom AL, Julius RL. Anaphylactic reaction to arginine infusion [letter]. N Engl J Med 1973; 288 (4): 218 110. Hertz P, Richardson JA. Arginine-induced hyperkalemia in renal failure patients. Arch Intern Med 1972; 130 (5): 778-80 111. Bushinsky DA, Gennari FJ. Life-threatening hyperkalemia induced by arginine. Ann Intern Med 1978; 89 (5 Pt 1): 632-4 112. Massara F, Martelli S, Cagliero E, et al. The hypophosphatemic and hyperkalemic effect of arginine in man. J Endocrinol Invest 1980; 3 (2): 177-80 113. Massara F, Cagliero E, Bisbocci D, et al. The risk of pronounced hyperkalaemia after arginine infusion in the diabetic subject. Diabetes Metab 1981; 7 (3): 149-53 114. Evans RW, Fernstrom JD, Thompson J, et al. Biochemical responses of healthy subjects during dietary supplementation with L-arginine. J Nutr Biochem 2004; 15 (9): 534-9 115. Schulman SP, Becker LC, Kass DA, et al. L-arginine therapy in acute myocardial infarction: the Vascular Interaction With Age in Myocardial Infarction (VINTAGE MI) randomized clinical trial. JAMA 2006; 295 (1): 58-64 116. Bednarz B, Jaxa-Chamiec T, Maciejewski P, et al. Efficacy and safety of oral l-arginine in acute myocardial infarction: results of the multicenter, randomized, double-blind, placebo-controlled ARAMI pilot trial. Kardiol Pol 2005; 62 (5): 421-7 117. Sun T, Zhou WB, Luo XP, et al. Oral L-arginine supplementation in acute myocardial infarction therapy: a metaanalysis of randomized controlled trials. Clin Cardiol 2009; 32 (11): 649-52 118. Shao A, Hathcock JN. Risk assessment for the amino acids taurine, L-glutamine and L-arginine. Regul Toxicol Pharmacol 2008; 50 (3): 376-99

Correspondence: Professor Paulo S.C. Gomes, Laborato´rio Crossbridges, Centro de Pesquisas Interdisciplinares em Sau´de, Universidade Gama Filho, Rua Manoel Vitorino 553, Piedade, Rio de Janeiro, RJ, 20740-900, Brazil. E-mail: [email protected]

Sports Med 2011; 41 (3)

Sports Med 2011; 41 (3): 249-262 0112-1642/11/0003-0249/$49.95/0

RESEARCH REVIEW

ª 2011 Adis Data Information BV. All rights reserved.

Does Physical Activity Impact on Presenteeism and Other Indicators of Workplace Well-Being? Helen E. Brown, Nicholas D. Gilson, Nicola W. Burton and Wendy J. Brown The University of Queensland, School of Human Movement Studies, Brisbane, Queensland, Australia

Abstract

The term ‘presenteeism’ is a relatively new concept in workplace health, and has come to signify being at work despite poor health and performing below par. Presenteeism, which is potentially critical to employers, has been associated with a range of psychosocial outcome measures, such as poor mental health and employee well-being. Physical activity is a potential strategy for reducing presenteeism, and for improving the mental health of employees. This article reviews evidence on the relationships between physical activity and employee well-being and presenteeism in the workplace, and identifies directions for research in an emerging field. Electronic and manual literature searches were used to identify 20 articles that met the inclusion criteria. These included 13 intervention trials (8 randomized controlled trials, 5 comparison trials) and 7 observational studies (3 cohort, 4 cross-sectional). Outcome measures were grouped into ‘workplace well-being’, ‘psychosocial well-being’ and ‘physical well-being’. Studies measured a wide variety of outcomes, with absenteeism being the most commonly assessed. Evidence indicated a positive association between physical activity and psychosocial health in employees, particularly for quality of life and emotional well-being. However, findings were inconclusive as to the role of physical activity in promoting workplace well-being. Only one study reported on presenteeism, with mixed evidence for outcomes. This article indicates that physical activity and employee psychosocial health are positively related, but there is limited evidence of a relationship between physical activity and presenteeism. A standardized definition of presenteeism and an appropriate evaluation tool are key research priorities if the complex relationships between physical activity and workplace well-being are to be better understood.

1. Background Presenteeism is a relatively new concept in workplace health. Originally coined by Professor Cary Cooper, a psychologist specializing in organ-

izational management at Manchester University in the UK, the term has come to signify being at work ‘on the job’, but performing below par, because of illness or medical conditions.[1] Chapman[2] has described presenteeism as the

Brown et al.

250

measurable extent to which physical or psychosocial symptoms, conditions and diseases adversely affect the work productivity of individuals who choose to remain at work. Conceptualizations of presenteeism indicate that it is not simply the opposite of absenteeism but, rather, a reduced ability to work productively.[3] It has been suggested that the impact of presenteeism is reflected in costs associated with reduced work output, errors on the job and failure to meet company standards.[4] A recent policy article[5] reported that presenteeism losses were between 1.9[6] and 5.1[7] times more than the costs incurred for absenteeism. One study, which examined the financial burden of ten common health conditions, found that presenteeism-related costs were greater than direct health costs in most cases, and that presenteeism accounted for 18–60% of all expenses for each of the ten conditions.[7] The WHO[3] recognizes emotional well-being as an important psychosocial marker of health, and well-being may play a pivotal role in presenteeism-associated productivity outcomes and employee-employer relations. Specific psychosocial conditions associated with presenteeism include anxiety, chronic fatigue, depression, nervousness, panic attacks and low energy levels.[2] Some of these conditions are reported to be among the most frequent causes of occupational disability whilst at work.[8] Statistics on work-related health and safety in the UK, for example, indicate that 13.8 million work days were ‘lost’ in 2006/7 due to stress, anxiety and depression.[9] These data highlight the economic impact of employee well-being and raise questions on how to reduce presenteeism and promote productivity in the workplace. Encouraging employees to be physically active may be a useful strategy, particularly considering evidence that suggests exercise interventions may be cost effective.[10-12] Regular physical activity has been found to improve mental health[13] and protect against depression, anxiety and stress.[14] It has also been found to reduce symptoms of fatigue[15] and somatization,[16] promote coping,[17] enhance mood[18] and increase quality of life (QOL)[19] and life satisfaction.[18] Yet, whilst these associations are well accepted, links between physical activity and emª 2011 Adis Data Information BV. All rights reserved.

ployee presenteeism remain unclear and often anecdotal. Given this, and the established link between mental health and productivity,[4] it seems worthwhile to explore this emerging area. This article examines the impact physical activity has on employee well-being and presenteeism. It provides a review of current evidence, identifying issues and recommendations for further research. 2. Methods 2.1 Search Methods

A search of PsycInfo, PubMed, Science Direct, Web of Science, MEDLINE and the Cochrane Library was conducted in November 2009. Keywords reflected the study variables (e.g. physical activity, exercise, sport) and outcome measures (e.g. presenteeism, productivity, job satisfaction, emotional well-being; see figure 1 for full search details). Bibliographies from included studies and additional review articles were additionally screened for relevant references. 2.2 Inclusion and Exclusion Criteria

All articles that included some form of physical activity (e.g. exercise, sport) as a study variable, at least one of the outcome measures listed in figure 1, and were conducted with employees or in a workplace setting, were initially selected. Both intervention and observational studies were included. Articles were excluded if they did not report on associations between physical activity and employee outcomes, involved a clinical or treatment population, were not available in hard copy or full text, or were not written in English. Review articles and discussion articles were retained and screened for further references. 2.3 Information Extraction and Analysis

Information about study location and design, participants, setting, physical activity, outcome measures and results was extracted from each article independently by two authors. Studies were categorized according to their design as interventions or observational studies. The outcome measures reported in all studies were reSports Med 2011; 41 (3)

Physical Activity and Workplace Presenteeism

251

(physical activity OR walking OR cycling OR exercise OR sport OR sitting OR sedentary OR active travel OR lifestyle activity OR structured exercise OR fitness) Number of records identified through database searching 20 448

AND

Number of records after duplicates were removed 20 068

AND

(employer OR employee OR worker OR manager OR colleague OR worksite OR office OR work OR workplace)

Number of records excluded (titles not relevant) 19 968

Number of records screened 380

Number of full-text articles assessed for eligibility 97

Number of studies included in qualitative synthesis 20

(presenteeism OR job satisfaction OR engagement OR emotional well-being OR psychosocial wellbeing OR productivity OR psychosocial outcomes OR depression OR anxiety)

Number of articles excluded (clinical population, full text not available, not in English) 283

Number of full-text articles excluded Discussion papers (22) Systematic review or metaanalysis (15) Not in workplace (8) No physical activity details (19) No psychosocial outcomes (9) Measurement paper (2) Not employees (2)

Fig. 1. Eligibility screening identification.

viewed and categorized as measures of ‘workplace well-being’ (including absenteeism, presenteeism and productivity), ‘psychosocial well-being’ (including depression, stress and emotional wellbeing) or ‘physical well-being’ (including physical QOL and general health). 3. Results A summary of the search process is shown in figure 1. In the first stage, the majority of articles were excluded because article titles were not relevant (e.g. see Flannery[20]). In the next stage, 283 articles were excluded because the full text version was not available in English, or included ª 2011 Adis Data Information BV. All rights reserved.

a clinical population only (e.g. see Klemetti et al.[21]). Of the remaining 97 articles, 20 were included in this review. 3.1 Study Design and Participant Characteristics

Of these 20 articles, 13 were intervention studies, with eight randomized controlled trials (RCTs)[22-29] and five comparison intervention trials[30-34] (see table I). There were seven observational studies, of which three were cohort studies[35-37] and four were cross-sectional studies[38-41] (see table II). Participants were predominantly female and aged between 30–45 years. Only two intervention Sports Med 2011; 41 (3)

Study (year)

252

ª 2011 Adis Data Information BV. All rights reserved.

Table I. Summary table of intervention studies (randomized controlled trials [RCT(s)] and comparison intervention trials) No. of participants (% female); age; workplace setting

Intervention; description; delivery; duration

Assessment period, construct (measure)

Results

Atlantis et al.[22] (2004)

Sydney, NSW, Australia; RCT, intervention vs control

n = 44 (55%); 30 – 6.8 y (intervention), 33 – 8.3 y (control) active <2 ·/wk; one casino

Supervised aerobic exercise; 20 min + weights 3 ·/wk offsite fitness facility, 5 · behaviour modification seminars (included 60 min/mo 1 on 1 counselling); unknown, e-mail reminders; 24 wk

Pre-/post-intervention; depression (DASS); anxiety (DASS); stress (DASS); HR-QOL (SF-36)

fl Depression (p = 0.048); fl stress (p = 0.036); fl bodily pain (p = 0.005); › vitality (p < 0.001); › general health (p = 0.009); › physical functioning (p = 0.004); 2 anxiety

Block et al.[23] (2008)

California, USA; RCT, intervention vs wait-list control

n = 549 (74%); 44 y (19–65); one healthcare organization

Web-based PA; nutritional feedback and support, goal setting; electronic; 16 wk

Pre-/post-intervention; HR-QOL (SF-8); presenteeism (difficulties concentrating because of back pain, depression/ anxiety, accomplishing tasks during the past 4 wk); mental health (SF-8); health status (SF-8)

› QOL (p = 0.02); › mental health (p = 0.02); › health status (p = 0.001); 2 presenteeism (difficulties concentrating because of back pain) [p = 0.054]; 2 presenteeism (difficulties concentrating because of depression/anxiety) [p = 0.06]; fl presenteeism (difficulties concentrating accomplishing tasks) [p = 0.02]

Brand et al.[24] (2006)

Stuttgart, Germany; RCT, intervention vs control

n = 109 (28%); [most common = 36–45 y]; three business organizations (both blue and white collar)

Free onsite group and individual exercise session; 2 ·/wk (26 sessions, 13 supervised); fitness coach and self-directed; 13 wk

Pre-/post-intervention; HR-QOL (WHOQOL-BREF); job satisfaction (FLZ life satisfaction questionnaire)

› Global QOL (p = 0.001); › physical QOL (p = 0.002); › psychosocial QOL (p = 0.003); 2 social relationships; 2 job satisfaction

Hinman et al.[25] (1997)

Texas, USA; RCT, intervention vs control

n = 50 (100%); 37.7 y; one School of Nursing and one allied Health Sciences Department

Stretching/relaxation exercises; 15 min 2 ·/d; computer directed; 8 wk

Pre-/post-intervention Job stress (PSQ)

2 Stress

RCTs

Continued next page

Brown et al.

Sports Med 2011; 41 (3)

Study location and design

Study location and design

No. of participants (% female); age; workplace setting

Intervention; description; delivery; duration

Assessment period, construct (measure)

Results

Nurminen et al.[26] (2002)

Helsinki, Finland; RCT, intervention group vs control

n = 217 (100%); 38.9 y; one laundry company

1 · 30 individual exercise prescription; 60 min group exercise session/wk; physiotherapist; 8 mo (26 sessions)

Pre-/post-intervention 3, 8, 12, 15-mo follow-up; work ability and impairment absenteeism (personnel records); well-being (stress, life satisfaction) job satisfaction

Intervention effect (all timepoints): 2 work ability; 2 absenteeism; 2 job satisfaction; 2 well-being

Proper et al.[27] (2004)

Netherlands; RCT, intervention vs control

n = 264 (68%); 43.8 y – 8.1 (intervention), 43.7 – 9.3 (control); three municipal services centres

PA and nutrition counselling sessions; 7 · 20 min; physiotherapist; 9 mo

Pre-/post-intervention; absenteeism (personnel records)

2 Absenteeism

Serxner et al.[28] (2001)

California, USA; RCT, intervention vs control

n = 450 (60.5%); 42.5 y; variety: telecommunications employees on shortterm disability (+ families)

On-site health management programme (e.g. weight loss, smoking cessation, physical therapy nutrition, weight management); fitness centre, cost reimbursement for fitness and wellness programmes, telephone counselling and support, self-care book; local third-party; 3 y

Pre-/post-intervention 2-y follow-up; absenteeism (personnel records)

fl Absenteeism (p < 0.05)

Tveito and Eriksen[29] (2009)

Bergen, Norway; RCT, intervention vs control

n = 29 (100%); 45 y; one nursing home

Aerobic exercise; 1 h 3 ·/wk, 15 · 1 h/wk; health and lifestyle information and stress management training, examination of workplace, aerobics instructor with healthcare education; 9 mo

Intervention effect: Pre-/post-intervention 2 absenteeism (post and 1-y follow-up (absenteeism only); follow-up); 2 QOL; 2 coping; absenteeism (personnel records); 2 job stress; 2 subjective QOL (SF-36); coping (IMOCF); job health complaints stress (PD-DCM); subjective health complaints (SHCI); subjective effects Subjective experiences of (health work environment, work intervention: › health; › work situation, fitness, pain, stress, situation; › fitness; › health health knowledge) maintenance; fl stress; fl muscle pain; 2 work environment Continued next page

253

Sports Med 2011; 41 (3)

Study (year)

Physical Activity and Workplace Presenteeism

ª 2011 Adis Data Information BV. All rights reserved.

Table I. Contd

Study (year)

254

ª 2011 Adis Data Information BV. All rights reserved.

Table I. Contd Study location and design

No. of participants (% female); age; workplace setting

Intervention; description; delivery; duration

Assessment period, construct (measure)

Results

Comparison intervention trials (no control group) New York, USA; RT, aerobic vs non-aerobic stretching exercise

n = 43 (88%); 32.05 – 10.68 y; two treatment facilities for children and adults with disabilities

Classes; 30 min 3 ·/wk; unknown; 8 wk

Pre-/post-intervention; weekly pre-/post-exercise session (state anxiety); anxiety (STAI); job satisfaction (JDI); absenteeism (personnel records)

Intervention effect: fl state anxiety in aerobic (p = 0.018); fl trait anxiety in aerobic (p = 0.018); 2 job satisfaction; 2 absenteeism Collapsed groups: fl state anxiety post-exercise (p = 0.005); fl trait anxiety (p < 0.001)

Eriksen et al.[31] (2002)

Bergen, Norway; RT, PE vs IHP vs SM

n = 860 (61%); 38.1 y (35.8–40.6); 29 post offices and two postal terminals

2 h/wk PE: 2 · 1 h/wk aerobic exercise; unknown; IHP: 1 h/wk PE + 1 h/wk information about stress, coping, nutrition; instructors; SM; cognitivebehavioural training; unknown; 12 wk

Pre-/post-intervention 1-y follow-up; Job stress (Cooper Job Stress Questionnaire); absenteeism sick leave (self-report); subjective health complaints (SHCI); subjective effects (health work environment, work situation, fitness, pain, stress, health knowledge)

Intervention effect: 2 job stress; 2 absenteeism; 2 subjective health complaints PE group only – subjective effects: › health; › fitness; fl muscle pain IHP group only – subjective effects: › health, › work situation, › fitness; › health maintenance; fl stress; fl pain; 2 work environment

Galinsky et al.[32] (2007)

Ohio, USA; RT, stretch breaks vs breaks without stretching

n = 51 (92%); 36 y; one IRS service centre

Stretch breaks vs breaks without stretching; 2 · 15 min/d for 4 wk, (2 · 15) + (4 · 5) min/d for 4 wk; self-directed with timer reminders; 8 wk

Pre-/post-intervention; physical/musculoskeletal discomfort (Feeling State Questionnaire); feeling state (POMS); work performance (data entry tracking)

2 Physical discomfort; 2 energy; 2 cheerfulness; 2 tension; 2 fatigue; 2 work performance

Goetzel et al.[33] (2009)

New York, USA; RT, moderate intensity health promotion vs high intensity health promotion

n = 7799 (72.1%); 44.2 y; ten education, healthcare, transportation or utility firms

Health-risk education programmes; moderate intensity, moderate + tailored individual behaviour change counselling, high intensity; unknown; 2 y

Pre-/post-intervention 2-y follow-up; stress (75-item HRA); emotional health (75-item HRA)

Intervention effect: 2 stress; 2 emotional health Moderate-intensity group: 2 stress; 2 poor emotional health High-intensity group: fl stress (p = 0.001); 2 poor emotional health Continued next page

Brown et al.

Sports Med 2011; 41 (3)

Altchiler and Motta[30] (1994)

ª 2011 Adis Data Information BV. All rights reserved.

studies had specific physical activity inclusion criteria. One trial stipulated that participants should not currently engage in more than two aerobic or weight training sessions of 20 minutes duration per week,[22] whilst another only included those participants who were involved in ‘physically demanding’ work.[26] A range of workplace settings was used, including corporate environments, medical or healthcare facilities and the public service sector. Studies were predominantly conducted in either the US[23,25,28,30,32,33,38,39] or Europe,[24,26,27,29,31,34-37,40,41] with only one Australian study.[22] 3.2 Outcomes Measured

BREF; › indicates increase; fl indicates decrease; 2 indicates no change; < indicates less than.

SHCI = Subjective Health Complaints Inventory; SM = stress management; STAI = State Trait Anxiety Inventory; WHOQOL-BREF = World Health Organization Quality of Life -

255

education; POMS = profile of mood states; PSQ = Personal Strain Questionnaire; QOL = quality of life; RCT(s) = randomized controlled trial(s); RT = randomized trial; SF = short form;

Descriptive Index; MBI-GS = Maslach Burnout Inventory-General Survey; PA = physical activity; PD-DCM = Psychological Demands – Demand Control Model; PE = physical

4DSQ = Four Dimensional Symptom Questionnaire; DASS = Depression Anxiety Stress Scale; FLZ = Fragen zur Lebenszufriedenheit Life Satisfaction Scale; HRA = Health Risk

Appraisal; HR-QOL = Health-related quality of life; IHP = Integrated Health Programme; IMOCF = Instrumental Mastery Coping Factor; IRS = Inland Revenue Service; JDI = Job

2 burnout; 2 fatigue Physical and relaxation exercise fatigue (fatigue checklist) re-shaping; 4 · 60 min sessions; psychologists; 8 wk company cognitive intervention

fl anxiety; fl burnout; fl fatigue

2 distress; 2 depression; 2 anxiety; 2 somatization; 4 · 60 min sessions; physical therapists; 8 wk and cognitive 18–63 y; one telecommunications Netherlands, RT, physical and relaxation exercise vs et al.[34] (2005)

only: fl distress; fl depression;

Intervention effect: Pre-/post-intervention;

psychosocial complaints (4DSQ); burnout (MBI-GS);

Physical training sessions; n = 90 (<10%); Amsterdam, the Van Rhenen

Table I. Contd

Assessment period, construct (measure)

Results Intervention; description; delivery; duration No. of participants (% female); age; workplace setting Study location and design Study (year)

Physical Activity and Workplace Presenteeism

Outcome measures used in each study (grouped as workplace well-being, psychosocial well-being or physical well-being) are shown in table III. More than 21 different constructs were assessed. Sixteen studies included at least one measure of workplace well-being. Absenteeism was the most frequently used outcome measure in this category, and indeed across all articles, as it was assessed in 10 of the 20 studies.[26-31,36-39] It was measured using objective indicators, such as company personnel records and sick leave data, and included a range of data such as frequency and overall duration of absenteeism, and shortand long-term illness. Measures of job satisfaction and enthusiasm,[24,26,30,40,41] as well as productivity and work performance[26,32,35,39,41] were also commonly assessed workplace wellbeing indicators (each measured in five different studies). Notably, only one study measured and reported on presenteeism;[23] two additional articles introduced presenteeism as a study variable, but did not explicitly report on the construct in the results.[33,38] Fifteen studies assessed at least one indicator of psychosocial well-being. Emotional well-being (e.g. vitality, energy, cheerfulness, tension, nervousness, relaxation, self-esteem) was measured in almost half of the studies,[22,32,33,35,38,40,41] a measure of stress was included in five studies.[22,25,26,33,34] Eight studies included a measure of physical well-being. The most commonly included constructs Sports Med 2011; 41 (3)

Study (year)

Study location and design

256

ª 2011 Adis Data Information BV. All rights reserved.

Table II. Observational studies (prospective cohort trials and cross-sectional studies) No. of participants (% female); age; workplace setting

Description of PA

Construct (measure)

Results

Prospective cohort trials Tuomi et al.[35] (2004)

Helsinki, Finland; 2 y cohort

n = 1389 (39.74%); 43.9 y, (most common = 45–54 y); 91 metal and retail industry companies

Leisure time physical exercise

Work ability (WAI); organizational commitment (questionnaire); mental well-being (questionnaire)

Higher levels of PA: › work ability (p < 0.05); 2 organizational commitment; 2 mental well-being

van den Heuvel et al.[36] (2005)

Netherlands; 4 y cohort

n = 1228 (NS); 18–59 y; variety: employees were pre-selected from a large cohort study (SMASH) including industrial, administrative and service roles

‘Physically demanding’ sport in the past 12 mo

Absenteeism (personnel records)

All participants: fl total duration of sick leave among active (p < 0.0005); 2 frequency total sick leave Sedentary workers: fl total duration of sick leave among active (p < 0.05); 2 frequency total sick leave Less sedentary workers: 2 total duration of sick leave; 2 frequency total sick leave

Laaksonen et al.[37] (2009)

Helsinki, Finland; cohort study (mean 3.9 y)

n = 6934 (78.8%); 40–60 y; civil services

PA behaviours with similar intensity to walking, vigorous walking to jogging, jogging, and running. MET values calculated

Absenteeism (personnel records)

Lowest PA level: › absenteeism

Cross-sectional studies Illinois, USA; cross-sectional

n = 999 (67.7%); 35.1 y; one financial services company (customer service and call centre)

Fitness centre participation

Work limitations (WLQ); absenteeism (short-term disability claims records from the personnel department)

Fitness centre participants: fl difficulty with time management (p = 0.02); fl difficulty with physical work (p = 0.04); fl difficulty with output demand (p = 0.05); fl difficulty with overwork impairment (p = 0.05); fl claims incidence (p = 0.03); fl average days absenteeism (p = 0.018); 2 difficulty with mental/interpersonal issues

Pronk et al.[39] (2004)

Minnesota, USA; crosssectional

n = 683 (60.8%); 46.4 – 0.4 y; three healthcare organizations and one airline

Leisure time PA behaviours (Godin PAQ) differentiating moderate and vigorous activity

Work performance (HRA); absenteeism (self-report)

Higher levels of moderate activity: › work quality (p = 0.0017); › overall performance (p = 0.0047); 2 absenteeism; 2 work quantity; 2 co-worker relations; 2 work effort Continued next page

Brown et al.

Sports Med 2011; 41 (3)

Burton et al.[38] (2005)

Study (year)

Study location and design

No. of participants (% female); age; workplace setting

Description of PA

Construct (measure)

Results Higher levels of vigorous activity: › overall performance (p = 0.0039); 2 absenteeism; 2 co-worker relations; 2 work quality; 2 work quantity; 2 work effort

Aberystwyth, Wales; crosssectional

n = 312 (34.6%); 34.11 – 8.07 y; one information technology company

PA (BHPAQ): at work (sitting, standing etc.), exercise/sport and leisure time PA (including walking or cycling)

Job affect (JAS); physical self-worth (ASPP); physical satisfaction (PSS); job satisfaction (JSQ); life satisfaction (SWLS); self-esteem (ASPP)

Higher levels of exercise: › physical self-worth (p < 0.001); › physical satisfaction (p < 0.001); › enthusiasm at work (p < 0.001); › life satisfaction (p = 0.02); 2 job satisfaction; 2 nervousness; 2 fatigue; 2 relaxation; 2 self-esteem Higher levels of PA: › physical self-worth (p < 0.001); › physical satisfaction (p < 0.001); › enthusiasm at work (p < 0.001); › relaxed at work (p = 0.01); › self-esteem (p = 0.005); 2 job satisfaction; 2 nervousness; 2 fatigue; 2 relaxation; 2 life satisfaction

Coulson et al.[41] (2008)

Bristol, UK; cross-sectional

n = 201 (67%); 38.2 – 23.8 y; one public and two private office-based companies, employees who exercised regularly

Regular exercise on site (72% cardiovascular)

Work performance and engagement (WLQ); affect post-exercise (PANAS) mood (diary); satisfaction with achievements

Exercise day (vs non-exercise day): › satisfaction with day’s achievement (p < 0.01); › positive affect during day (p < 0.01); › tranquility during day (p < 0.01); › managing time demands (p < 0.01); › managing mental/ interpersonal demands (p < 0.01); › managing output demands (p < 0.01); fl negative affect during day (p < 0.01); fl work limitations (p < 0.01); 2 day’s workload estimate

ASPP = Adult Self-Perception Profile; BHPAQ = Baecke’s Habitual Physical Activity Questionnaire; HRA = Health Risk Assessment; JAS = Job Affect Scale; JSQ = Job Satisfaction Questionnaire; MET = Metabolic Equivalent of Task; NS = not specified; PA = physical activity; PANAS = Physical Activity Affect Scale; PAQ = Physical Activivity Questionnaire; PSS = Physical Satisfaction Scale; SMASH = Study on Musculoskeletal Disorders, Absenteeism, Stress and Health; SWLS = Satisfaction With Life Scale; WAI = Work Ability Index; WLQ = Work Limitations Questionnaire; › indicates increase; fl indicates decrease; 2 indicates no change.

257

Sports Med 2011; 41 (3)

ThogersenNtoumani et al.[40] (2005)

Physical Activity and Workplace Presenteeism

ª 2011 Adis Data Information BV. All rights reserved.

Table II. Contd

Brown et al.

258

were physical QOL and physical functioning, which were measured in four studies,[22-24,29] and subjective health complaints or somatization[24,29,34] and fatigue,[32,34,40] which were each measured in three studies (see table III). 3.3 Evidence from Intervention Studies

Summaries of the eight RCTs[22-29] are provided in the first section of table I. A wide range of intervention strategies was used to promote physical activity, from internet health promotion schemes to onsite exercise sessions. All studies showed favourable intervention effects across a variety of employee outcomes. Internet-based health promotion schemes[23] and onsite exercise sessions[24] demonstrated a positive intervention effect on QOL (psychosocial and physical). The same internet-based intervention[23] and a study combining aerobic exercise sessions with behaviour modification classes[22] improved general health. These exercise classes[22] also impacted on emotional well-being. Only one of the RCTs that reported on absenteeism showed a favourable intervention effect.[28] However, this intervention was a health management programme targeting a range of behaviours in addition to physical activity, including weight loss, smoking cessation and nutrition. Results of the other intervention trials that compared two intervention strategies with no true control group are summarized in the second section of table I. Two studies showed improvements in anxiety and burnout. One found that aerobic exercise reduced anxiety more than nonaerobic exercise,[30] and the other showed that physical and relaxation exercises reduced burnout and anxiety more than a cognitive intervention.[34] However, the impact of this physical and relaxation intervention,[34] and an additional one that used stretch breaks for employees,[32] showed little evidence of a reduction in fatigue. 3.4 Evidence from Observational Studies

The results of the prospective cohort studies are shown in the first section of table II. A variety of physical activity behaviours were studied, inª 2011 Adis Data Information BV. All rights reserved.

cluding incidental physical activity (e.g. increasing stair use), walking and jogging, leisure time physical exercise and physically demanding sport. The follow-up period was 2–4 years. There were favourable associations between physically demanding sport (participated in during the previous 12 months)[36] and physical activity such as walking or jogging[37] with objectively measured absenteeism (e.g. company sick leave data). Productivity was measured in only one cohort study,[35] where there was a positive association with leisure time physical exercise. The same study[35] identified no association between leisure time physical exercise and organizational commitment or emotional well-being. Physical wellbeing, including fatigue, subjective health complaints and somatization, was not measured in any of the cohort studies. The cross-sectional studies are summarized in the second section of table II. There was a positive association between leisure time physical exercise and productivity,[39] and between regular onsite exercise and productivity.[41] This latter study found that the same employees performed better at work on self-selected exercise versus non-exercise days, with improvements in performance outcomes linked to acute changes in mood.[41] Another study, which measured incidental physical activity at work (e.g. standing at desks), exercise and sport and leisure time physical activity, found positive associations between activity and physical self-worth, job satisfaction and emotional well-being.[41] One study demonstrated that fitness centre participation was inversely associated with job limitations, such as difficulties with time management and output demand.[38] Another study[39] showed no association between leisure time physical exercise and social relations. 4. Discussion 4.1 Physical Activity, Presenteeism and Related Constructs

The aim of this review was to examine the impact of physical activity on employee well-being and presenteeism. Results indicated that many Sports Med 2011; 41 (3)

Physical Activity and Workplace Presenteeism

259

Table III. Outcome measures used in each study Constructs

RCT references [22]

[23]

[24]

[25]

Comparison trial references [26]

[27]

[28]

[29]

[30]

[31]

[32]

[33]

[34]

Cohort references

Cross-sectional study references

[35]

[38]

[36]

[37]

[39]

[40]

[41]

O

O

Totalsa

Workplace well-being O

Job satisfaction/ enthusiasm

O

O O

Organizational commitment O

Absenteeism

O

O

O

O

O

1 O

O

O

O

10

O

Presenteeism

5

1 O

Productivityb

O O

Job stress

O

O

O

O

O

5 3

O

Work limitations

1

Psychosocial well-being Anxiety

O

Depression

O

Stress/distress/burnout

O

O O O

Psychosocial QOL

O

O

3

O

2

O

4

O

3

O

Social relations Emotional/mental well-beingc

O

O

O O

O

O

O

O

3 O O

Life satisfaction O

Coping

O

7 1 1

Physical well-being Pain/discomfort

O

Physical QOL/functioning

O

O

O

General health

O

O

O

Subjective health complaintsd

3 4

O O

3

O

O O

Fatigue Physical worth/satisfaction a

O

O

O

3 O

3

O

1

Total number of studies reporting on outcome measure.

b

Performance, ability, effort.

c

Vitality, energy, cheerfulness, tension, nervousness, relaxation, self-esteem.

d

Includes somatization.

QOL = quality of life; RCT = randomized controlled trial.

different types of physical activity had a positive impact on employee well-being, across a wide range of outcome measures. These findings are encouraging, particularly when set against the context of the recent economic collapse and the need to stimulate effective recovery through costeffective strategies that promote a healthy, valued and productive workforce. Two-thirds of the 20 studies were intervention trials, which provide a good base for building the ª 2011 Adis Data Information BV. All rights reserved.

evidence in this emerging area of interest. Across intervention strategies,[22-24] positive effects were shown for QOL, general health and emotional well-being. Aerobic exercise[30] and exercise and relaxation in combination,[34] were shown to reduce anxiety and burnout in employees. However, evidence from intervention trials indicated no association between physical activity and workplace well-being, particularly absenteeism, which was only reduced with a multi-strategy Sports Med 2011; 41 (3)

Brown et al.

260

intervention[28] including weight loss, smoking cessation, increasing incidental physical activity and improving nutritional intake. The results of this study, although promising, do not indicate whether it was an increase in physical activity or positive changes in other factors that reduced absenteeism. With this in mind, evaluation of single compared with multi-component interventions, is an important consideration for future studies. The majority of the intervention studies tended to focus on either psychosocial well-being or workplace indicators of performance as outcome measures and not on both. Despite the established links between physical activity and health,[13-19] and between health and productivity,[4] as yet we can only hypothesize about causal relationships between physical activity and productivity. Further studies are needed to expand the evidence base. Some other outcome measures, for example, coping,[29] organizational commitment,[35] physical satisfaction,[40] were only reported by one study each, so that evidence is insufficient to draw comprehensive conclusions about their relationships with physical activity. Productivity was only measured in one cohort study,[35] and this showed a positive association with leisure time physical exercise. In the other cohort studies, ‘physically demanding sport’[36] and physical activity[37] were shown to be inversely associated with absenteeism. It was notable that most studies measured absenteeism[26-31,36-39] rather than productivity. This was perhaps to be expected, given that absenteeism data are easily and objectively assessed using workplace personnel records. Productivity on the other hand is arguably more complex. It can be measured as a ratio-based concept, where inputs such as time spent working are compared with outputs, such as number of tasks completed. It also incorporates perceived performance indicators such as work ability and commitment,[33] which tend to be difficult to objectify. 4.2 Conceptualization and Measurement of Presenteeism

The initial objective of this review was to examine the impact of physical activity on preª 2011 Adis Data Information BV. All rights reserved.

senteeism; however, only one study[23] reported data on presenteeism. Two further studies referred to presenteeism,[33,38] but their definitions of presenteeism differed and did not concur with the conceptualization provided in the former study.[23] Block et al.[23] measured presenteeism using a self-administered questionnaire, and reported on the number of hours in a typical work day that back pain or depression/anxiety interfered with accomplishing work tasks, as well as difficulty with concentrating at work because of physical and emotional problems. Goetzel et al.[33] conceptualized presenteeism as healthrelated productivity loss, subsumed within a specially designed 75-item Health Risk Assessment (which also measured risk behaviours and healthcare utilization). The third study, by Burton et al.,[38] used the terms presenteeism and productivity interchangeably and measured this construct with the Work Limitations Questionnaire,[42] an inventory that rates the frequency of difficulty performing specific job demands without specifying if this is attributable to poor health or well-being. 4.3 Recommendations and Study Limitations

An important finding of this review was the wide variation in how studies conceptualized and assessed presenteeism. A standard definition and evaluation tool are therefore key research priorities if workplace trials are to explore the complex relationships between physical activity and presenteeism. Given these issues, a limitation of our review is that keywords used in the initial search may not have yielded all published articles of interest. However, the comprehensive search that was conducted did enable us to explore the relationships between physical activity and those interpretations of presenteeism and workplace well-being that have been published in recent years. As a result, it is interesting to speculate on how the well documented links between physical activity and mental health,[13-19] and between mental health and employee productivity,[4] might be combined to improve our understanding of the temporal relationship between physical activity and presenteeism. Sports Med 2011; 41 (3)

Physical Activity and Workplace Presenteeism

The studies we reviewed included different types of physical activities and demonstrated different types of associations dependent on the measures used. The emerging message is that physical activity is beneficial, particularly for emotional wellbeing; however, given the current level of evidence, it is difficult to identify the frequency, intensity, time and type of physical activity that is most effective. Better quantification of activity prescription is therefore an additional research priority. The use of combined physical activity measures such as accelerometers and self-report will play an important role in this process, with quality data providing useful insights for researchers, exercise professionals and human resource personnel involved in workplace practice. Our review identified a number of other important priorities. Future studies need to account for pre-existing presenteeism and the influence this has on recruitment bias. Disengaged employees, for example, are a key target group, yet these individuals may be less inclined to participate in workplace physical activity initiatives and may therefore be under represented in study samples. The majority of participants in those studies reviewed were women. We suggest therefore that more research with male participants is conducted, as there may be sex differences in experiences of presenteeism. Finally, closer examination of the wide range of occupational roles, geographical locations and participants’ exercise history (and indeed, type of exercise – be it leisure-time recreation, incidental physical activity or organized sport) are needed. This will better inform the applicability of interventions to different workplace contexts, while controlling for the impact typical activity patterns have on participant recruitment and study outcomes. 5. Conclusions The articles reviewed in this study found positive relationships between physical activity and a range of indicators of employee psychosocial well-being. There was, however, only limited evidence of a relationship between physical activity and presenteeism and other indicators of workplace well-being. We suggest this is because the ª 2011 Adis Data Information BV. All rights reserved.

261

concept of presenteeism has not yet been widely considered, agreed upon or formally evaluated by physical activity researchers. Considering these issues, there is clearly a need for well designed studies with robust and valid measures of both physical activity and presenteeism, if dose-response and causal relationships are to be examined. Acknowledgements Nicola Burton was supported by a Heart Foundation Research Fellowship (PH08B3905) and an NHRMC Program Grant (Owen, Bauman and Brown 569940). The authors have no conflicts of interest that are directly relevant to the content of this article.

References 1. Cooper C, Dewe P. Well-being: absenteeism, presenteeism, costs and challenges. Occup Med (Lond) 2008 Dec; 58 (8): 522-4 2. Chapman LS. Presenteeism and its role in worksite health promotion. Am J Health Promot 2005 Mar-Apr; 19 (4): 1-8 3. Hemp P. Presenteeism: at work-but out of it. Harvard Bus Rev 2004 Oct; 82 (10): 49-58 4. Schultz AB, Edington DW. Employee health and presenteeism: a systematic review. J Occup Rehab 2007 Sep; 17 (3): 547-79 5. The Sainsbury Centre for Mental Health. Mental health at work: developing the business case [policy paper 8; online]. Available from URL: http://www.centreformentalhealth. org.uk/pdfs/mental_health_at_work.pdf [Accessed 2011 Jan 31] 6. Hilton M. Getting upstream of psycho-social disability in the workforce: who are we not seeing and at what cost? Workshop presentation at MRC conference on employment and mental health: absence from work due to mild and moderate mental ill health; 2007 Jan 15-16; Manchester. Cited in Sainsbury Centre for Mental Health. Mental health at work: developing the business case [policy paper 8; online]. Available from URL: http://www.cen treformentalhealth.org.uk/pdfs/mental_health_at_work.pdf [Accessed 2011 Jan 31] 7. Goetzel RZ, Long SR, Ozminkowski RJ, et al. Health, absence, disability, and presenteeism cost estimates of certain physical and mental health conditions affecting US employers. J Occup Environ Med 2004 Apr; 46 (4): 398-412 8. Wang PS, Simon G, Kessler RC. The economic burden of depression and the cost-effectiveness of treatment. Int J Methods Psychiatr Res 2003; 12 (1): 22-33 9. HSC National Statistics. Health and safety statistics 2006/7 [online]. Available from URL: http://www.hse.gov.uk/ statistics/overall/hssh0607.pdf [Accessed 2011 Jan 31] 10. Hagberg LA, Lindholm L. Cost-effectiveness of healthcarebased interventions aimed at improving physical activity. Scand J Public Health 2006; 34 (6): 641-53 11. Roux L, Pratt M, Tengs TO, et al. Cost effectiveness of community-based physical activity interventions. Am J Prev Med 2008 Dec; 35 (6): 578-88

Sports Med 2011; 41 (3)

Brown et al.

262

12. Hagberg LA, Lindholm L. Is promotion of physical activity a wise use of societal resources? Issues of cost-effectiveness and equity in health. Scand J Med Sci Sports 2005 Oct; 15 (5): 304-12 13. Paluska SA, Schwenk TL. Physical activity and mental health: current concepts. Sports Med 2000 Mar; 29 (3): 167-80 14. Strohle A. Physical activity, exercise, depression and anxiety disorders. J Neural Transm 2009 Jun; 116 (6): 777-84 15. Penedo FJ, Dahn JR. Exercise and well-being: a review of mental and physical health benefits associated with physical activity. Curr Opin Psychiatry 2005 Mar; 18 (2): 189-93 16. Martinsen EW. Physical activity in the prevention and treatment of anxiety and depression. Nord J Psychiatry 2008; 62 Suppl. 47: 25-9 17. Taylor CB, Sallis JF, Needle R. The relation of physical activity and exercise to mental health. Public Health Rep 1985 Mar-Apr; 100 (2): 195-202 18. Fox KR. The influence of physical activity on mental wellbeing. Public Health Nutr 1999 Sep; 2 (3A): 411-8 19. Brown DW, Balluz LS, Heath GW, et al. Associations between recommended levels of physical activity and healthrelated quality of life: findings from the 2001 Behavioral Risk Factor Surveillance System (BRFSS) survey. Prev Med 2003 Nov; 37 (5): 520-8 20. Flannery RB. Violence in the workplace, 1970-1995: a review of the literature. Aggression Violent Behav 1996 Spring; 1 (1): 57-68 21. Klemetti M, Santavirta N, Sarvima¨ki A, et al. Tension neck and evaluation of a physical training course among office workers in a bank corporation. J Adv Nurs 1997; 26: 962-7 22. Atlantis E, Chow CM, Kirby A, et al. An effective exercisebased intervention for improving mental health and quality of life measures: a randomized controlled trial. Prev Med 2004; 39 (2): 424-34 23. Block G, Sternfeld B, Block CH, et al. Development of alive! (A lifestyle intervention via email), and its effect on healthrelated quality of life, presenteeism, and other behavioral outcomes: randomized controlled trial. J Med Internet Res 2008; 10 (4): E43 24. Brand R, Schlicht W, Grossman K, et al. Effects of a physical exercise intervention on employees’perceptions quality of life: a randomized controlled trial. Soz Praventivmed 2006; 51 (1): 14-23 25. Hinman M, Ezzo L, Hunt D, et al. Computerized exercise program does not affect stress levels of asymptomatic VDT users. J Occup Rehab 1997; 7 (1): 45-51 26. Nurminen E, Malmivaara A, Ilmarinen J, et al. Effectiveness of a worksite exercise program with respect to perceived work ability and sick leaves among women with physical work. Scand J Work Environ Health 2002; 28 (2): 85-93 27. Proper KI, de Bruyne MC, Hildebrandt VH, et al. Costs, benefits and effectiveness of worksite physical activity counseling from the employer’s perspective. Scand J Work Environ Health 2004; 30 (1): 36-46 28. Serxner S, Gold D, Anderson D, et al. The impact of a worksite health promotion program on short-term

ª 2011 Adis Data Information BV. All rights reserved.

29.

30.

31.

32.

33.

34.

35.

36.

37.

38.

39.

40.

41.

42.

disability usage. J Occup Environ Med 2001 Jan; 43 (1): 25-9 Tveito TH, Eriksen HR. Integrated health programme: a workplace randomized controlled trial. J Adv Nurs 2009 Jan; 65 (1): 110-9 Altchiler L, Motta R. Effects of aerobic and nonaerobic exercise on anxiety, absenteeism, and job-satisfaction. J Clin Psychol 1994 Nov; 50 (6): 829-40 Eriksen HR, Ihlebaek C, Mikkelsen A, et al. Improving subjective health at the worksite: a randomized controlled trial of stress management training, physical exercise and an integrated health programme. Occup Med Oxf 2002 Oct; 52 (7): 383-91 Galinsky T, Swanson N, Sauter S, et al. Supplementary breaks and stretching exercises for data entry operators: a follow-up field study. Am J Indust Med 2007; 50 (7): 519-27 Goetzel RZ, Roemer EC, Short ME, et al. Health improvement from a worksite health promotion private-public partnership. J Occup Environ Med 2009 Mar; 51 (3): 296-304 Van Rhenen W, Blonk RW, van der Klink JJ, et al. The effect of a cognitive and a physical stress-reducing programme on psychological complaints. Int Arch Occup Environ Health 2005; 78: 139-48 Tuomi K, Vanhala S, Nykyri E, et al. Organizational practices, work demands and the well-being of employees: a follow-up study in the metal industry and retail trade. Occup Med Oxf 2004 Mar; 54 (2): 115-21 van den Heuvel SG, Boshuizen HC, Hildebrandt VH, et al. Effect of sporting activity on absenteeism in a working population. Br J Sports Med 2005 Mar; 39 (3): E15 Laaksonen M, Piha K, Martikainen P, et al. Health-related behaviours and sickness absence from work. Occup Environ Med 2009 Dec; 66 (12): 840-7 Burton WN, McCalister KT, Chen CY, et al. The association of health status, worksite fitness center participation, and two measures of productivity. J Occup Environ Med 2005 Apr; 47 (4): 343-51 Pronk NP, Martinson B, Kessler RC, et al. The association between work performance and physical activity, cardiorespiratory fitness, and obesity. J Occup Environ Med 2004 Jan; 46 (1): 19-25 Thogersen-Ntoumani C, Fox KR, Ntoumanis N. Relationships between exercise and three components of mental well-being in corporate employees. Psychol Sport Exerc 2005 Nov; 6 (6): 609-27 Coulson JC, McKenna J, Field M. Exercising at work and self-reported work performance. Int J Workplace Health Manage 2008; 1 (3): 176-97 Lerner D, Amick 3rd BC, Rogers WH, et al. The work limitations questionnaire. Med Care 2001 Jan; 39 (1): 72-85

Correspondence: Helen Elizabeth Brown, The University of Queensland, School of Human Movement Studies, Brisbane, QLD 4072, Australia. E-mail: [email protected]

Sports Med 2011; 41 (3)