Sports Med 2009; 39 (4): 257-278 0112-1642/09/0004-0257/$49.95/0
REVIEW ARTICLE
ª 2009 Adis Data Information BV. All rights reserved.
Muscle Fatigue in Males and Females during Multiple-Sprint Exercise Franc¸ois Billaut1 and David Bishop2 1 Department of Kinesiology, University of Lethbridge, Lethbridge, Alberta, Canada 2 Team Sport Research Group, Facolta` di Scienze Motorie, Universita` degli Studi di Verona, Verona, Italy
Contents Abstract. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1. General Physiological Sex Differences . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1.1 Morphology and Body Composition . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1.2 Endocrine Status . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1.3 Enzyme Activities . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1.4 Substrate Utilization. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1.5 Muscle Fibre Properties . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1.6 Neural Activation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1.7 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2. Sex Differences in Sprint-Induced Muscle Fatigue . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.1 Single-Sprint Exercise . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.1.1 Determinants of Fatigue . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.1.2 Fatigue during Single-Sprint Exercise in Males and Females . . . . . . . . . . . . . . . . . . . . . . . . . 2.2 Multiple-Sprint Exercise. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.2.1 Determinants of Fatigue . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.2.2 Fatigue during Multiple-Sprint Exercise in Males and Females. . . . . . . . . . . . . . . . . . . . . . . . 3. Sex Differences in Physiological Responses to Sprint Exercise Reanalysed . . . . . . . . . . . . . . . . . . . . . . . 3.1 Methodological Concerns . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.2 Influence of Total Mechanical Work . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4. Summary and Future Directions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Abstract
257 258 259 259 260 261 261 261 262 263 263 263 265 267 267 269 269 269 271 272
Females have often been reported to have a greater muscle fatigue resistance than males, especially during exercise at low-to-moderate intensities. Differences in muscle mass, muscle metabolism and voluntary activation patterns have been the primary explanations for the differences in performance and physiological responses to exercise between sexes. However, while ample data are available for isometric contractions, dynamic activity is a less studied mode of exercise, and there is even less information regarding multiple-sprint exercise (MSE). This is surprising given that MSE places unique demands on metabolic processes in the muscle where energy supply oscillates between fuelling contractile activity and restoring homeostasis. As such, MSE provides a rich area for future applied research. This review examines the limited data available concerning the physiological responses of males and females to sprint exercise, and discusses the methodological confounds arising from non-appropriate comparison methods. Based on original findings,
Billaut & Bishop
258
we highlight that sex differences in the absolute mechanical work performed during a given task might explain a significant part of the differences in physiological responses of males and females to sprint exercise. We therefore suggest that future studies using male and female subjects to answer basic physiological questions use mechanical work as a covariate.
Human skeletal muscle fatigue can be defined as a transient, exercise-induced reduction in the maximal force capacity of the muscle.[1,2] Several mechanisms have been proposed that contribute concurrently to the fatigue exhibited by a muscle or muscle group following exercise, and the classic approach used to identify the cause of muscle fatigue has been to distinguish between ‘central’ and ‘peripheral’ mechanisms. Typically, peripheral skeletal muscle fatigue involves processes occurring at or distal to the neuromuscular junction, in the presence of unchanged or increasing central motor output.[3-6] On the other hand, central fatigue is due to failure at a site within the CNS.[2,7,8] Studies applying an electrical stimulus to peripheral nerves and/or a magnetic stimulus to the motor cortex have demonstrated that both ‘central’ and ‘peripheral’ mechanisms are involved during fatiguing contractions, and a number of good scientific reviews on this topic are available to the reader.[1,2] It has also been demonstrated that human skeletal muscle fatigue is influenced by the biological sex of the individual.[9-11] Studies on the physiological function of females have mostly concentrated on isolated muscle exercise (e.g. isometric and isokinetic contractions of a single muscle group). This work has provided tremendous advances in our understanding of possible mechanisms for these specific tasks. In the last decade, however, the popularity of team and court sports, which require the athlete to sprint intermittently over the course of the game, has increased.[12-14] Consequently, many authors have explored the physiology of multiple-sprint exercise (MSE). This particular pattern of activity involves repeated bouts of short . duration (£8 seconds), high-intensity (>300% VO2max) exercise, separated by short rest duration (£30 seconds). Such short rest periods have been shown to negatively affect subsequent sprint performance.[15,16] Current ª 2009 Adis Data Information BV. All rights reserved.
knowledge of MSE physiology is based largely on the responses of young adult males, and this is somewhat surprising since in most countries team and court games are also popular sports among females. More importantly, there is now strong evidence that the mechanisms underlying force decline are highly task specific.[1,2] This means that muscle fatigue can be induced by a combination of processes contributing in different ways to the decline in force, according to the details of the task (intensity, duration, mode of contraction, muscle, etc.). As such, one cannot rely on data arising from isometric contraction research to explain sex differences in performance and muscle fatigue during MSE. Such whole-body tasks (e.g. running and cycling) need to be further explored to achieve greater understanding of female physiology. This paper begins with an updated review of general physiological sex differences that could potentially contribute to the sex differences observed during MSE. Then we discuss the relative importance of these factors in the fatigue processes during sprint exercise. Finally, we summarize the limited number of studies that have investigated the physiological responses to MSE in females versus males, and evaluate the evidence for and against the existence of a sexrelated difference in the manifestation of skeletal muscle fatigue in response to MSE. In particular, we question whether males and females display different degrees of fatigue during MSE. 1. General Physiological Sex Differences It is well accepted that males possess greater absolute muscle strength and produce greater power output scores than their female counterparts in a variety of muscles and in a variety of exercise conditions.[17-30] Research has also fairly well documented the influence of sex on muscle Sports Med 2009; 39 (4)
Sex and Sprint Fatiguability
fatigue, typically reporting that a female’s muscle is capable of longer endurance times (i.e. greater resistance to fatigue) and faster recovery (i.e. ability to restore force or power output) than a male’s muscle.[23,29,31-35] This has been observed with the use of various fatiguing, isometric protocols at low to moderate intensities, and during sprint exercise, where females have been observed to maintain the initial absolute power output for a longer time than males.[19,36,37] Several mechanisms may explain the effect of sex on performance and fatiguability. Evidence supporting body composition (e.g. lean and fat mass), muscle metabolism (e.g. hormonal regulation, enzymatic activities and substrate utilization), muscular characteristics (e.g. typology) and motor unit discharge rate as factors accounting for the sex differences are discussed below. 1.1 Morphology and Body Composition
Differences in morphology and body composition are the most visible and obvious differences when one first compares the two sexes. On average, and at the same age, males are taller (+11 cm), heavier (+13 kg) and have a greater lean mass (+18 kg) and lower fat mass (-5 kg) than their female counterparts.[28,38-41] These parameters have therefore often been suggested to explain sex differences in performance.[10,38,42,43] The primary intrinsic determinants of maximum voluntary strength include cross-sectional area (CSA) of the muscle or muscle groups, specific tension (force per unit CSA, which may be affected by the fibre type distribution and the amount of non-contractile tissue present in the muscle), and possible anatomical differences in the mechanical advantage of a muscle acting across a joint.[44-47] Males, having greater segment length and muscle mass, develop higher absolute muscle force and power output than females. For example, a significant correlation (r = 0.91; p < 0.05) between total mechanical work (during repeated cycle sprints) and body mass (BM) has been reported in athletes.[48] Thus, it is not surprising to observe smaller sex differences when indices of performance are expressed as a ratio to body mass, an index of lower limb ª 2009 Adis Data Information BV. All rights reserved.
259
volume, or CSA of the thigh.[27,30,41,49-52] Performances must then be scaled for body size differences to permit meaningful comparisons between males and females. It is important, however, to point out that some studies indicate that muscle mass is not the only factor accounting for the sex difference in fatigue. In fact, while differences in performance and fatiguability are reduced, they often persist when the two sexes exercise at the same percentage of initial performance, or when the data are expressed relative to BM, lean BM (LBM), lean volume (LV) of the active limb, and when subjects are matched for strength.[16,30,37,43,50,51,53-59] For example, Fulco and colleagues[55] have shown that the fatigue rate of the adductor pollicis muscle during intermittent, isometric, submaximal contractions was still »2-fold slower in females than in males matched for strength. When comparing the performances of males and females during two consecutive 8-second sprints on a cycle ergometer, it has also been observed that males remained more powerful than females when data were expressed relative to LBM (+17%) and lower limb LV (+16%).[16,20] Thus, even though body dimensions explain the major discrepancies between the sexes in performances, sex differences still persist when body dimensions are appropriately controlled. Accordingly, physiological factors (as opposed to muscle mass quantity) must also contribute to the sex difference in performance. 1.2 Endocrine Status
The secretion of sex hormones is another difference between males and females. Androgens (e.g. testosterone) increase protein synthesis and lead to muscle hypertrophy.[60,61] The higher androgen concentration found in males is therefore likely to contribute to some sex differences (e.g. muscle mass, selective hypertrophy of type II fibres). On the other hand, estrogens (i.e. estrone, estradiol and estriol) increase growth hormone (GH) concentration, which is known to stimulate lipolysis and to reduce glycogenolytic activity by reducing plasma adrenaline (epinephrine) secretion.[62,63] However, although the higher estrogen Sports Med 2009; 39 (4)
Billaut & Bishop
260
concentration in females does indeed increase GH release at rest in young and adult females compared with males of the same age,[64-67] exercise seems to evoke a similar incremental GH response in both sexes.[67-69] Sex also appears to affect the sympathetic responses to supramaximal exercise,[56,70,71] lowering plasma catecholamine levels (and subsequently blood lactate) in females during exercise at the same relative intensity compared with males with similar fitness levels. While some have attributed these differences to a direct inhibitory effect of estradiol on the sympathetic nervous system,[72] Sandoval and Matt[68] concluded that these differences were most likely due to differences in the absolute workload performed by males and females. Less absolute work would lead to less lactate production and glycogen use, and thus less glucose would be taken up by the muscle for refuelling glycogen stores after exercise in females.[68] This study therefore demonstrates indirectly the importance of matching subjects for total work performed before attempting to draw any conclusion regarding sex differences. Researchers continue to debate whether the different phases of the menstrual cycle affect athletic performance and fatiguability,[73] and possibly modify the magnitude of sex-related differences.[74,75] Some authors have reported higher voluntary muscle strength and total work in a short-term, all-out performance during the luteal phase.[76-78] There is also evidence of enhanced blood lactate removal at high intensities,[79,80] greater O2 consumption during recovery from repeated cycle sprints,[76] and increased excess post-exercise O2 consumption after prolonged exercise[81] during the luteal phase relative to the follicular phase. While these metabolic changes could improve multiple-sprint performance by enhancing recovery between sprints,[76] such findings need to be balanced by the many studies that suggest that hormonal fluctuations throughout the cycle do not contribute to sex difference in performance and fatiguability.[82-85] In summary, the multiple contrasting findings clearly demonstrate that there is currently no consensus about the impact of the monthly hormonal fluctuations upon sex differences observed in performance and muscle fatigue. ª 2009 Adis Data Information BV. All rights reserved.
1.3 Enzyme Activities
In vitro measurements of muscle enzyme activities are related to whole-body energy metabolism,[86] and can provide an insight into the relative contribution of the different energy production pathways in males and females.[70] Very few studies have analysed the impact of sex on the activity of enzymes involved in the phosphagen (or alactic) energy systems. However, activities of myosine adenosine triphosphatase (ATPase) and creatine phosphokinase have been found to be higher in males than females,[87,88] suggesting a greater potential phosphagen energy provision in males to support performance. The maximal activities of glycolytic and glycogenolytic enzymes (glycogen phosphorylase, phosphofructokinase [PFK] and lactate dehydrogenase) are also lower in females.[19,42,62,87-92] Interestingly, Jaworowski et al.[42] reported that the higher maximal activities of glycolytic enzymes in males were related to the height, mass, muscle CSA and relative area of type II fibres, which were all significantly larger in males than in females. As males participate more in intense activities than females, particularly during childhood,[93-96] the greater glycolytic enzyme activities reported in males may then be caused by daily activity patterns rather than intrinsic physiological sex differences. An increased muscle mass associated with a greater muscle recruitment is likely to contribute to the greater anaerobic potential in males.[36,38,55,62,97] Assessment of oxidative enzyme activities of the Kreb’s cycle has been contradictory; higher values for males[87,90,92] or no sex differences[19,42,91,98] have been reported for various enzymes. For enzymes involved in lipid oxidation, no sex differences have been reported.[19,42,92] Again, however, when sex differences have been reported it is difficult to establish whether these are attributable to training or actual sex differences. Evidence against the existence of sex differences is provided by the observation that oxidative enzyme activities and mitochondrial volume density increase have been found to adapt in a similar manner (maximal activity and Sports Med 2009; 39 (4)
Sex and Sprint Fatiguability
magnitude) following endurance exercise training in the two sexes.[98-100] 1.4 Substrate Utilization
Sex differences in muscle metabolism during exercise have been investigated quite thoroughly.[19,62,63,101,102] Overall, muscle characteristics account for an energetic balance in which the aerobic contribution is more important in females during prolonged sprints[19,39,101] and during submaximal, isometric contractions.[62,97] For example, Hill and Smith[39] estimated the aerobic contribution to total work during a 30second cycle sprint to be 25% in females and 20% in males. According to Fulco et al.[55] and Russ and Kent-Braun,[103] this predisposition to oxidative phosphorylation would allow a faster ATP resynthesis during recovery. On the contrary, a greater reliance on the anaerobic glycolytic pathway in males would induce a greater fatiguability and a slower recovery.[23,97] The reduced reliance on glycolytic energy in females may be related to other factors in addition to the lower maximal activation velocity of glycolytic enzymes and the greater reliance on fat metabolism. For example, the lower increase in plasma catecholamine concentration during sprints in females may also influence the stimulation of glycolytic enzymes.[54,56,104,105] Therefore, a reduced maximal activation velocity of glycolytic enzymes, a reduced stimulation of glycolytic pathways, and greater reliance on fat oxidation are all likely to contribute to muscle glycogen sparing and the lower blood lactate concentration reported in females after a 30-second sprint on a cycle ergometer.[54,56,106] While a lack of sex difference in ATP or phosphocreatine (PCr) reduction has been reported in type I and type II fibres after a 30-second Wingate test,[54] one study has demonstrated a smaller ATP reduction in females than in males in type II fibres among three Wingate tests separated by 20 minutes of recovery.[36] The authors explained these differences via a faster ATP resynthesis in females via a greater inosine monophosphate (IMP) reamination during the recovery phases.[36] On another occasion, the ª 2009 Adis Data Information BV. All rights reserved.
261
same protocol (three 30-second Wingate tests interspaced with 20 minutes of recovery) resulted in a similar ATP and PCr content decrease and alactic ATP turnover rate in males and females.[107] Males, compared with females, have also been found to exhibit a greater decrease in ATP and PCr concentrations after MSE consisting of 5 · 6-second sprints every 30 seconds.[108] This lack of consistency for exercise-induced metabolism changes is likely to be related to biopsy timing and the protocols used rather than actual sex differences. 1.5 Muscle Fibre Properties
Based on the histochemical staining properties for the myofibrillar myosin ATPase, different fibre categories can be distinguished in human skeletal muscle. It is generally accepted that untrained females have smaller fibre CSA in all fibre types than untrained males in the muscles of the upper and lower limbs, as do female athletes and bodybuilders compared with their male counterparts.[19,29,38,42,54,62,91,92,98,109-112] Studies in which fibre percentage has been estimated are less consistent. While several authors failed to observe any differences in the mean proportion of type I, IIA and IIX fibres (% number) between males and females from similar sport specialties and fitness level,[19,36,38,42,98,101,110,111,113] others have reported a greater distribution of type I fibres and lower distribution of type IIX fibres in females.[29,91,92,109] Such discrepancies may be related to sampling bias in subject selection and/or problems with the accurate estimation of fibre percentage.[29] Nonetheless, fibre size and property differences between sexes may have an impact on the peak power output and on the subsequent ability to maintain power output. 1.6 Neural Activation
Neuromuscular activation patterns are increasingly described in males and females during maximal tasks, but it is difficult to form precise conclusions about a typical trend in neuromuscular responses to exercise, as results are highly dependent upon the nature of the task.[103] Further difficulty arises from the controversial Sports Med 2009; 39 (4)
Billaut & Bishop
262
use of electromyogram (EMG) recordings. Indeed, while useful information (e.g. the net motor unit activity) can be extracted from an appropriately recorded surface EMG, there remains a complex mismatch between the spinal cord output and both EMG amplitude and frequency parameters, which limits the interpretation of EMG data when recorded alone.[114,115] Nonetheless, the maximum voluntary EMG has been found to decrease only in males, after performing 20 repetitions of a maximal squatlift with a load of 100% of the one repetition maximum.[116] This suggests an attenuation of skeletal muscle recruitment after a strenuous heavy-resistance exercise in males compared with females. Failure in voluntary activation during maximal tasks has also been directly examined between sexes. For example, Russ and KentBraun[103] observed a greater neural activation deficit (i.e. reduction in voluntary activation as assessed from a supramaximal train of stimuli superimposed onto a maximal voluntary contraction [MVC] without any concomitant change in the compound muscle action potential nor muscle twitch characteristics) in males than in females during maximal, intermittent, isometric contractions of the dorsiflexor muscles (5-second contraction, 5-second rest conducted to exhaustion). More recently, transcranial magnetic stimulation of the motor cortex to examine the contribution of supraspinal fatigue in performance decrement[8,117,118] has been applied to males and females during a maximal task (six 22-second MVC of the elbow flexors, separated by 10 seconds).[119] The authors concluded that the greater muscle fatigue (i.e. torque reduction) in males than in females was not explained by a difference in supraspinal fatigue but rather involved mechanisms located within the muscles. A greater peripheral fatigue (as assessed via M-wave amplitude alterations with no concomitant reduction in EMG amplitude) for males than females has also been reported during intermittent MVCs of the adductor pollicis muscle (5-second contraction, 2-second rest conducted for 3 minutes).[53] Even though males were stronger than females in this study (mean MVC ª 2009 Adis Data Information BV. All rights reserved.
force: males 10.0 – 2.7 kg vs females 6.6 – 1.1 kg; p < 0.05), the fatigue index did not show a significant sex difference (males 45% vs females 38%; p>0.05), which suggests a similar amount of fatigue in both sexes. These data indicated that males were more susceptible to transmission failure at the neuromuscular junction and/or decreases in muscle membrane excitability.[53] Interestingly, during a submaximal contraction of the lumbar musculature sustained to exhaustion, a faster compression of the median frequency has been reported in males than in females,[31] which also suggests higher fibre conductibility impairments[114,120-122] in males compared with females. Overall, the clear distinction of neuromuscular activation patterns in males versus females in exercise physiology is not an easy task, as they are underpinned by the task characteristics. Therefore, it is likely that the central motor output to locomotor muscles will differ during MSE, as well as the ‘central’ and/or ‘peripheral’ nature of fatigue mechanisms. Nonetheless, such differences in neuromuscular physiology between the sexes could contribute to the difference in fatigue resistance observed during repeated sprints.
1.7 Summary
Morphological, metabolic and neuromuscular properties of the muscle tissue are different between males and females, and predominately explain the differences in strength, power output and fatigue resistance between the sexes. Historically, body composition has been the principal factor used to account for sex differences in performances, but research has demonstrated that enzyme activities, substrate use and central motor output also contribute. Muscle fatigue in males and females during sprint exercise has not been studied extensively in the literature, and as a consequence, the influence of the abovediscussed factors is not well understood. The final sections of this review focus on male versus female performance and muscle fatiguability during sprint exercise, with particular emphasis on analysing the appropriateness of current methods used to compare the two sexes. Sports Med 2009; 39 (4)
Sex and Sprint Fatiguability
The introduction and section 1 of this article have highlighted the importance of sprinting and repeated sprinting to team-sport performance, and the origins of sex differences in terms of general performance and fatiguability. An understanding of sprint metabolism is, at this point, necessary to comprehend the demands placed on team-sport athletes during competition, and the reader is referred to the detailed reviews of Glaister[123] and Spencer and colleagues.[124] Although studies have shown that females have a greater muscular endurance than males during isometric exercises,[31,55,57,103,119] there is a lack of information on the sex difference in sprint-induced fatigue, considering the amount of information documenting strength and power output differences.[19,28,29,33,35,51,101,125-127] After brief synopses of the determinants of fatigue during sprint exercise, this section focuses on the sex difference in fatigue during single and multiple sprints of varied duration. 2.1 Single-Sprint Exercise 2.1.1 Determinants of Fatigue
Many studies have examined the participation of the energy-producing systems during maximal sprinting exercise of varying duration (10-, 20and 30-second sprints). During such sprints, peak power is quickly reached in 2–3 seconds (before peak pedal speed), and thereafter power declines.[20,128-131] This implies a very high energy need from the very beginning of the sprint. Overall, during a single, short-duration sprint (£6 seconds), the rate of ATP utilization is extremely high, with a mean value of ~15 mmol/kg/sec dry muscle (d.m.).[132] Approximately 50% of the ATP is supplied by the degradation of PCr, while intramuscular ATP stores, anaerobic glycolysis and aerobic energy provide the remainder.[106,132-134] As the sprint duration increases, energy system contribution is modified, and one notes a progressively greater participation of anaerobic glycolysis and oxidative metabolism (figure 1).[128,135-138] For all sprint durations, it has been demonstrated that ATP depletion is minimal and is unlikely to constitute a ª 2009 Adis Data Information BV. All rights reserved.
limiting factor of performance.[130,132,133,137,140] However, such intensities result in a severe reduction in intramuscular PCr concentration. Indeed, PCr depletion after 6 seconds of sprinting has been reported to be around 35–55% of resting values.[132,133,140,141] The study of muscle metabolic responses to 10 and 20 seconds of cycle ergometer sprinting demonstrated that PCr was reduced by about 55% after 10 seconds and about 73% after 20 seconds.[135,136,142] Following a 30-second sprint, the depletion is even greater (i.e. up to 80%).[130,133,135,143-146] Maximal sprinting activity thus requires considerable contribution of PCr to provide energy, and it is likely that the ability to sustain sprint exercise will be affected by PCr availability in the working muscles. This is supported by the direct relationship (r2 = 0.74; p < 0.05) between the percentage recovery of PCr following a recovery period and the subsequent recovery of performance, expressed as percentage of mean power output (figure 2).[143] These data were later confirmed by high correlations between %PCr resynthesis and the percentage recovery of mean power (r = 0.84; p < 0.05) and mean pedalling speed (r = 0.91; p < 0.05) during the initial 10 seconds of a second 30-second sprint.[135] From muscle lactate concentrations, it has been estimated that >40% of total anaerobic Anaerobic glycolysis PCr ATP Oxidation
120 100
3 7
80 ATP (%)
2. Sex Differences in Sprint-Induced Muscle Fatigue
263
60
50
13 7
20
40
25
18
50
55
25
5
2
40 20
40
40
0 6
10 20 Sprint duration (sec)
30
Fig. 1. Schematic illustration of relative energy system contribution to ATP resynthesis (in percentage of total energy) during sprint of varying duration (adapted from Bogdanis et al.,[128,135,136] Gaitanos et al.,[132] Medbø and Tabata[139] and Spriet et al.[138]). PCr = phosphocreatine.
Sports Med 2009; 39 (4)
Billaut & Bishop
264
Percentage of MPO recovery
100 96 92 88 84
r2 = 0.74
80 0 0 60
70 80 90 Percentage of PCr resynthesis
100
Fig. 2. Relationship between the percentage of phosphocreatine (PCr) resynthesized during the 3-minute recovery period and the mean power output (MPO) achieved during the 6-second sprint (relative to resting value) performed 3 minutes after a 30-second cycle sprint (reproduced from Bogdanis et al.[143] with permission).
energy during a single 6-second sprint bout is provided via anaerobic glycolysis.[132,133] The subsequent decline in muscle glycogen that occurs during repeated, maximal sprints could theoretically contribute to impaired performance via a reduction in substrate and subsequent glycolytic flux. However, the decrease in muscle glycogen has been reported to be only ~30% during a 30-second cycle or treadmill sprint.[130,147-149] Therefore, glycogen stores do not represent a limiting factor in this type of activity. The most significant source of anaerobic ATP during intense activities lasting at least 10–20 seconds is from glycolysis. Glycogenolysis-glycolysis has been associated with the accumulation of lactate and hydrogen (H+) ions. Thus, high levels of power output have been associated with a decrease in blood and muscle pH on several occasions. For example, a muscle pH of 6.73 units was estimated after a maximal 30-second sprint on a non-motorized treadmill.[130] This would have, in vitro, inhibited PFK and glycogen phosphorylase activities, the key regulatory or rate-limiting enzymes in this pathway.[150-153] This would lead to a reduced rate of ATP production, which might set an important limitation to muscle performance.[128,135,151,152,154-156] A more important consequence of the decrease in pH may be in affecting the muscle contractile mechanism itself, by decreasing the energy availª 2009 Adis Data Information BV. All rights reserved.
able for contraction per ATP hydrolysed.[152,154] In fact, acidosis interferes with the effectiveness of calcium (Ca2+) activation at many sites in the excitation-contraction process.[150,154,157,158] Finally, a decline in muscle pH may contribute to the occurrence of central fatigue. Indeed, the typical association between pH and EMG[7] is consistent with the role of pH in feedback to the CNS and a subsequent alteration in central motor drive during the development of fatigue. These metabolic perturbations have been found to act on nerve terminations of group III and IV afferents, inducing a reflex inhibition of the central drive.[3,90,159-161] Thus, H+ accumulation may contribute to fatigue during sprint exercise. Sprint exercise also results in other important ionic perturbations that may contribute to fatigue during sprint exercise. In particular, sprint exercise changes the extracellular potassium (K+) ion concentration ([K+]) far beyond the narrow limits seen in resting subjects. It has been suggested by some[150,162-166] that subsequent alterations in sarcolemma excitability induce muscle fatigue by preventing cell activation. For example, Medbø and Sejersted[165] reported a >200% increase in plasma [K+] after a 1-minute running sprint on a motor-driven treadmill (10.5% inclination). In muscles contracting at high workloads, inorganic phosphate (Pi) also accumulates because PCr is broken down to creatine and Pi. The [Pi] increases substantially in the myoplasm during intense exercise and affects both the myofibrillar proteins and activation processes.[167-169] Although not as extensively studied, changes in skeletal muscle recruitment may contribute to performance decrement during maximal sprinting exercise. The only two studies to have examined neuromuscular fatigue (via EMG recording) during a single sprint are those of Vandewalle et al.[170] and Hunter et al.[171] The first study observed a parallel decline in power output and integrated EMG during a 45-second cycle sprint, and suggested a progressive attenuation of spatial and/or temporal recruitment of motor units during this type of exercise. On the other hand, the EMG amplitude has been shown to remain unchanged (whereas power output declined) during a 30-second cycle sprint.[171] Sports Med 2009; 39 (4)
Sex and Sprint Fatiguability
The authors completed the data with an analysis of the frequency power spectrum, and demonstrated a shift of the mean power frequency towards lower values (-14.7%; p < 0.05). According to several authors,[1,114,121,172] this might be caused by an accumulation of metabolites and a consequent decrease in muscle pH, and/or some form of neural control through reflex regulation of muscle force to prevent muscle damage.[173-175] Clearly, more studies using electrically evoked stimulation will need to be conducted to clarify whether voluntary drive parallels the power output decline observed during maximal sprint, to ascertain if a failure of excitation is present under these conditions. 2.1.2 Fatigue during Single-Sprint Exercise in Males and Females
The few studies providing information on performances and fatiguability during sprints in males and females are shown in table I. During a 30-second, supramaximal cycling exercise (such as the Wingate anaerobic test), males develop greater absolute power output levels than females (on average, peak power and mean power are 40% and 30% greater, respectively). These discrepancies are reduced, but overall remain significant, when results are expressed relative to BM, LBM or leg LV,[37,39,51,56,58,106,178] meaning that other factors, likely related to the capacity to sustain a high ATP resynthesis rate, should account for the sexual dimorphism in prolonged sprints.[51] As previously discussed (section 1), the greater potential for anaerobic metabolism in males (due to greater glycolytic enzyme activities) and the larger muscle fibre CSA (associated with greater concentration of male sex steroids) are likely to explain the greater absolute scores in males during such sprints. With respect to fatigue, however, it is usually accepted that, compared with males, females are capable of maintaining their peak power output for a longer time within sprints. For example, Froese and Houston[37] reported a greater fatigue index (decrement in absolute power output) in males than in females over the course of a 30-second cycle sprint. Such observations may be related to the greater reliance of females on ª 2009 Adis Data Information BV. All rights reserved.
265
aerobic metabolism[55,100,103,179,180] associated in turn with reduced muscle H+ ion accumulation and reduced ionic disturbances.[97,181] Additionally, the greater ability of females to maintain motor unit activation at exhaustion[103,116] may contribute to a better maintenance of power output. However, when expressed per unit of BM or leg volume, the sex difference in muscle fatigue disappears; although males exhibited a greater absolute power decrease than females (-433 W vs -315 W, respectively; p < 0.05) during a 30-second cycle sprint, these discrepancies disappeared when values were related to the peak power developed during the sprint (mean fatigue index: males 47% vs females 48%; p > 0.05).[19] Hill and Smith[39] and Weber et al.[52] confirmed these data, showing a similar relative power output decline in males (mean 48%) and females (mean 52%). Thus, these results suggest that rather than sex differences, differences in fatiguability during 30-second sprints may actually be related to the greater initial power output of males. When looking at the data obtained during shorter sprints (<10 seconds), leg peak power output (PPO) reached during a cycling forcevelocity test was greater in boys than in girls at the same age (14–17.5 years old).[27] Moreover, for the same leg length, the optimal pedalling frequency was higher in boys than in girls, with no sex difference observed for the optimal force.[27] Better performances (+12% to +22%) have also been reported in males during 30 m and 36.5 m track sprints (»4–6 seconds),[28,51] 5-second treadmill sprints,[59] and 8-second cycle sprints.[16,20] These results are likely to reflect the influence of androgens on qualitative muscular factors (i.e. type II muscle fibres, glycolytic ability) and then on male mechanical scores. The observed sex differences of the optimal and maximal pedalling frequency during cycle ergometer sprint[16,27] may be related to differences in proportion and/or recruitment of fast-twitch fibres. In males (in contrast to females), it has been suggested that a selective hypertrophy of type II fibres[42,182] may occur in response to greater circulating testosterone levels.[183] However, in contrast to results for the longer sprints, the sex discrepancy for shortsprint performance remained present when data Sports Med 2009; 39 (4)
Billaut & Bishop
266
Table I. Sex differences in mechanical, metabolic and hormonal responses to sprint exercise Study
Exercise mode
Protocol
Principal significant observations
Jacobs et al.[106] (1983)
Cycle
30 s Wingate
MPO (abs. & rel. to BM): M > F D [Lac-] after sprint: M > F
Murphy et al.[176] (1986)
Cycle
30 s Wingate
PPO & MPO (abs. & rel. to BM, LBM): M > F
Froese and Houston[37] (1987)
Cycle
30 s Wingate
PPO & TW (abs. & rel. to BM, LLV): M > F Fatigue index (abs.): M > F; (rel. to BM): NS
Hill and Smith[39] (1993)
Cycle
30 s Wingate
PPO, MPO, TW (abs. & rel. to BM): M > F Fatigue index: NS Anaerobic work (abs. & rel. to BM): M > F Aerobic proportion to TW: F > M
Gratas-Delamarche et al.[56] (1994)
Cycle
30 s Wingate
PPO & MPO (abs. & rel. to BM, LBM): M > F D [Lac-] after sprint: M > F D [adrenaline] after sprint: M > F
Esbjo¨rnsson-Liljedahl et al.[54] (1999)
Cycle
30 s Wingate
PPO & MPO (abs. & rel. to BM): M > F MPO (rel. to LBM): M > F D [ATP] & [PCr] after sprint in type I & II: NS D [glycogen] after sprint in type I: M > F D [Lac-] after sprint in type I: M > F
Vincent et al.[58] (2004)
Cycle
30 s Wingate
PPO & MPO (abs. & rel. to BM, LBM): M > F D [glucose] after sprint: F > M D [insulin] after sprint: F > M
Weber et al.[52] (2006)
Cycle
30 s Wingate
PPO & MPO (abs. & rel. to BM): M > F PPO & MPO (rel. to LBM and LLV): NS Fatigue index: NS
Perez-Gomez et al.[51] (2008)
Cycle
30 s Wingate
PPO & MPO (abs.): M > F PPO (rel. to LLV): NS MPO (rel. to LLV): M > F
Esbjo¨rnsson-Liljedahl et al.[19] (1993)
Cycle
3 · 30 s Wingate (20 min)
PPO & MPO (abs. & rel. to BM): M > F Fatigue index (abs.): M > F; (rel. to PPO): NS Total LDH activity: M > F
Bodin et al.[107] (1994)
Cycle
3 · 30 s Wingate (20 min)
D [PCr] & [ATP] after sprints: NS Alactic ATP turnover rate: M = F
Esbjo¨rnsson-Liljedahl et al.[36] (2002)
Cycle
3 · 30 s Wingate (20 min)
PPO & MPO (abs.): M > F; (rel.): NR PPO decrease from sp1 to sp3: M > F MPO decrease from sp1 to sp3: M = F D [ATP] & [IMP] after sprint in type II: M > F D [glycogen] after sprint in type I: M > F
Jacobs et al.[106] (1983)
Cycle
10 s sprint
MPO (abs. & rel. to BM): M > F D [Lac-] after sprint: M > F
Winter et al.[30] (1991)
Cycle
4 · 8 s sprint (5 min)
PPO (abs.): M > F; (rel. to LLV): NS PPO (rel. to LLV-analysis of covariance): M > F
Martin et al.[27] (2004)
Cycle
2 · 5–8 s sprint (3 min)
Vopt (rel. to LL-analysis of covariance): M > F (young adults: 14–17.5 y of age)
Dore´ et al.[177] (2005)
Cycle
3 · 5–8 s sprint (4 min)
PPO (rel. to LLV-analysis of covariance): M > F (young adults: 16–20 y of age)
Mayhew and Salm[28] (1990)
Run-track
36.5 m
Sprint time: F > M
abs. = absolute value; BM = body mass; F = females; IMP = inosine monophosphate; Lac- = lactate; LBM = lean body mass; LDH = lactate dehydrogenase; LL = leg length; LLV = lean leg volume; M = males; min = minutes; MPO = mean power output; NR = not reported; NS = not significant; PCr = phosphocreatine; PPO = peak power output; rel. = relative value; run-track = over-ground running; s = seconds; sp1 = sprint 1; TW = total work; Vopt = optimal velocity (i.e. velocity to reach peak power output); [y y] indicates concentration; D indicates delta changes from rest to the end of the exercise.
ª 2009 Adis Data Information BV. All rights reserved.
Sports Med 2009; 39 (4)
Sex and Sprint Fatiguability
were expressed per unit of BM (20.5%), LBM (17%), leg LV (12%) and thigh LV (16%).[16,20,59] Comparisons of within-sprint fatigue patterns during short-duration sprints (<10 seconds) are difficult to find in the literature. We are aware of only one study that has directly investigated sex differences in fatigue pattern within a brief sprint.[16] Subjects in this study were only matched on the basis of their physical activity (males 11.8 – 6.5 vs females 10 – 4.2 h/week; p > 0.05). The authors demonstrated that during a single, all-out, 8-second cycle sprint bout against an optimal force (males 86 – 12 N vs females 53 – 8 N; p < 0.05; optimal forces corresponded to »10% of body mass for each sex), females had a greater decrement in relative power output than males (-31% vs -19% of PPO, respectively; p < 0.05). In another study conducted on ten intermittent, 6-second sprints on a non-motorized treadmill,[104] the two sexes seemed to display similar power decrement within the first sprint of the series (males -32%, females -27%), but unfortunately statistical analysis was not performed on these data. In conclusion, more data clearly need to be collected from females to better define metabolic and neuromuscular changes, and to examine fatigue patterns during sprint exercise, especially during brief sprint exercise (<10 seconds). In light of the available results, one may imagine that males and females would exhibit similar relative performance decrements when repeated sprints are involved. 2.2 Multiple-Sprint Exercise 2.2.1 Determinants of Fatigue
There are very few data on the relative energy system contributions during MSE involving consecutive, all-out sprints of short duration. During brief periods of maximal work, ATP provision is maintained through the integration of various metabolic processes. However, as work bouts are repeated, the metabolic response to subsequent work bouts will be affected by the previous exercise and the duration of the intervening rest periods. Due to the complexity of physiological processes that regulate this type of ª 2009 Adis Data Information BV. All rights reserved.
267
activity, research shows that MSE places considerable demands on both aerobic and anaerobic pathways, although the relative contribution from each of these sources is still an issue of controversy.[48,123,132,184,185] Muscular fatigue that develops during MSE is associated with signs of energy deficiency, i.e. increased concentrations of IMP, inosine, hypoxanthine and uric acid.[15,48,148,186-189] Since energy provision during MSE is maintained predominantly by anaerobic sources (PCr degradation and anaerobic glycolysis), deficiencies in energy provision are likely to be associated with limitations in anaerobic metabolism.[123,124] In particular, close relationships (0.84 < r < 0.86; p < 0.05) have been reported between PCr resynthesis and the recovery of power output in different sprinting conditions,[135,143] suggesting that the ability to reproduce high power outputs is directly related to the resynthesis of PCr. This is supported by studies showing that occlusion during recovery (and hence the prevention of PCr resynthesis) impairs the recovery of power output, while creatine supplementation improves repeated-sprint performance.[190-193] Consequently, some of the decrease in power output during MSE can probably be attributed to the decrease in the absolute contribution of PCr to the total ATP production from sprint one to sprint ten (44.3 – 4.7 vs 25.3 – 9.7 mmol/kg d.m., respectively).[132] In addition, a large decrease in the contribution of anaerobic glycogenolysis (11-fold reduction) and glycolysis (8-fold reduction) to energy supply has been reported from the first to tenth sprint (10 · 6-second sprints, 30-second recovery),[132] which is also likely to contribute to the appearance of fatigue during MSE. The accumulation of metabolites has also been demonstrated to correlate with fatigue during MSE. In particular, the accumulation of H+ (acidosis) may impair performance through effects on the contractile machinery and its potential role in glycolytic inhibition (through negative effects on glycolytic enzymes). This is supported by studies demonstrating a correlation between repeated-sprint ability and both muscle buffer capacity and changes in blood pH.[48,184,194] Greater improvements in repeated-sprint ability Sports Med 2009; 39 (4)
268
following training have also been reported in subjects with greater improvements in muscle buffer capacity[195] and the sodium-hydrogen exchanger[196] – a ubiquitously expressed integral membrane protein that mediates the exchange of one extracellular sodium ion with one intracellular proton, which plays a central role in the regulation of intracellular pH in most cells. In addition, Bishop et al.[197] have reported a significant reduction in fatigue during 5 · 6-second sprints (24-second recovery) following sodium bicarbonate (NaHCO3) administration. In contrast, Gaitanos et al.[198] indicated no effect of NaHCO3 ingestion on performance scores and fatigue throughout ten 6-second sprints (30-second recovery). This discrepancy may be related to different exercise protocols used (cycling vs running) or the large variability that has often been observed in performance improvement in response to alkalosis.[199] Further investigations are clearly required to fully establish the role, if any, of H+ accumulation on the development of fatigue during MSE. Recent applied physiological findings[150,167,200] have revealed that K+ and Pi accumulation may also have a significant role in muscle fatigue. However, even though the negative effects of a rise in interstitial [K+] and intracellular [Pi] have been studied during highintensity exercise,[165,201,202] there is no study to the authors’ knowledge investigating such ionic aspects during MSE. An interesting observation, however, was provided by Mohr et al.[166] who investigated K+ kinetics during three repeated, intense, one-legged knee extensions with 10-minute recovery (exercise protocol not specific to the activity patterns of field-based team sports). They found that, when intense exercise was repeated, the rate of K+ accumulation in the initial phase of exercise was lowered and [K+] at exhaustion decreased, suggesting that it is not the accumulation of K+ in the muscle interstitium per se that depresses performance when exercise is repeated.[166] Further research is required to investigate the consequences of K+ and Pi accumulation during the development of fatigue when short bouts of exercise are repeated over a long period of time. ª 2009 Adis Data Information BV. All rights reserved.
Billaut & Bishop
Finally, neural adjustments have been linked to fatigue occurrence during MSE. However, very few studies are available on this topic, and the uncertainty regarding the extent, if any, to which muscle recruitment impairs MSE performance is reflected in the contrasting results of investigations into the EMG signal. Observation of steady levels of EMG signal amplitude (assessed through integrated EMG or root mean square) in prime mover muscles during and after MSE[203-205] suggests that despite mechanical performance becoming progressively impaired, the neural system still recruits motor unit pools at their highest firing rate. On the other hand, proof of neural adjustments (i.e. reduction in the central neural drive to active musculature and power spectrum frequency, and inter-muscle coordination pattern changes) gathered from several studies demonstrate the progressive inability of the brain to maintain the initial pattern of motor unit activation throughout repeated sprint bouts.[204,206-209] Once again, further research is required to clarify the neural adjustments that occur during MSE. In summary, while fatigue during MSE is likely to be the result of a spectrum of events, research supports a predominantly anaerobic ATP provision during work periods and an exclusively aerobic process of recovery.[15,48,128,132,135,186,210-215] Moreover, that both PCr resynthesis and H+ removal are oxygendependent processes suggests that a high level of aerobic fitness may convey an enhanced ability to resist fatigue during this type of work.[123,216] This is further supported by studies showing that endurance-trained athletes are better able to resist fatigue during MSE than their sprinttrained counterparts,[198,211] and by the correlation between work decrement during MSE and the peak oxygen consumption obtained during a graded exercise test (r = -0.62; p < 0.05).[184] Therefore, if there is a sex difference in repeatedsprint ability, it is likely to be associated with sex differences in the aerobic contribution to repeated sprints, the ability to breakdown and resynthesise PCr, buffer H+ and/or the ability to maintain an optimal muscle recruitment pattern as sprints are repeated. Sports Med 2009; 39 (4)
Sex and Sprint Fatiguability
2.2.2 Fatigue during Multiple-Sprint Exercise in Males and Females
Sex differences in fatiguability during MSE (i.e. the ability to recover from one sprint to the subsequent one) have been poorly examined. To the best of our knowledge, the first study to investigate sex differences during MSE was conducted in 1990, with a protocol involving a 6-second sprint on a non-motorized treadmill repeated ten times with 30 seconds of recovery between each sprint.[104] The authors demonstrated significant differences between the sexes; males had a greater peak power (+25%) and total work (+25%) than females. Once again (see section 1), the higher absolute scores in males versus females during MSE may be due to the typical sex difference in growth factors, anaerobic metabolism and the area occupied by type II muscle fibres. Despite the greater initial sprint performance of the males (in the study by Brooks and colleagues[104] above), the fatigue index (based on work done) calculated from sprint 1 to sprint 10 was not significantly different between the sexes. One could imagine that females were less fit than males in this study, but not enough data were provided on the subjects’ physical characteristics to support this assumption. However, another study using a similar protocol (10 · 5-second cycle sprints, 10 seconds of recovery) indicated that mean power decrement (absolute but not relative to leg volume) from the first to the tenth repetition was greater in teenage boys than girls (43.8 – 7.5% vs 33.9 – 7.9%, respectively; p < 0.05).[59] In addition, males were found to perform better (both absolute and relative work) than females during five 6-second cycle sprints every 30 seconds, but experienced a greater work decrement than females (13.7 – 5.1% vs 11.0 – 2.8%, respectively; p < 0.05).[108] Thus, despite the limited research, it appears that males experience greater absolute decrements in performance during MSE. This may be related to the greater involvement of anaerobic glycolysis, due to the greater initial power output in males than in females, and hence subsequent inhibition of muscle glycolysis and contractile mechanisms during later sprints (see sections 1 and 2.2.1). The sex difference in the depletion rate of high-energy phosphate stores and the reduction in central drive may also ª 2009 Adis Data Information BV. All rights reserved.
269
contribute to the difference in fatigue resistance in MSE, but has not yet been investigated. Finally, a greater aerobic contribution to energy supply in females would be beneficial to PCr resynthesis during recovery periods and to the maintenance of high ATP resynthesis rates during the final sprints. Esbjo¨rnsson-Liljedahl and colleagues[19,36,54] have also demonstrated significantly higher peak (+30%) and mean (+28%) power output and greater fatigue in males versus females during a repeated-sprint protocol consisting of repeated 30-second cycle sprints interspersed with 20 minutes of rest. Indeed, during three 30-second Wingate tests separated with 20 minutes of rest, a decline in power output among the three sprints was reported in males (8%; p < 0.05) but not in females (4%; p = NS). The results of this study suggest, therefore, that females have a greater ability to restore power between prolonged sprints separated by long recovery periods. 3. Sex Differences in Physiological Responses to Sprint Exercise Reanalysed When comparing skeletal muscle fatiguability between males and females, there are some methodological confounds that may affect the interpretation of the results. For example, difficulties arise when attempting to match the sexes for absolute power output, relative power output and training background. This has contributed to the inability to definitively establish sex differences in muscle metabolism, the degree of fatigue development, and the rate of impairment of possible contributing mechanisms. The following section therefore demonstrates how this methodological issue affects sprint exercise, and proposes a different interpretation of the current sprint literature. 3.1 Methodological Concerns
Current research suggests that it is important to consider differences in absolute performance (and training background) when investigating the sex difference in metabolism, performance and fatigue. This appears particularly important during MSE, where a correlation between the initial Sports Med 2009; 39 (4)
Billaut & Bishop
270
Males Females
a Absolute changes (mmol/kg d.m.)
140 120 100 80
* *
60 40 20
I
I
G
ly
co
ge
La
c
n
fib
fib
re
re
PC r
AT P
0
than age-matched girls from the beginning of the series. A deeper examination of the studies conducted by Billaut et al.[206] and Gaitanos et al.[132] also shows that the higher the performance reached in the first sprint during a series of ten all-out bouts, the greater the decline in power output across the ten sprints. Interestingly, Gaitanos and coworkers[132] additionally reported a correlation between the total work done over the first five sprints and the increase in blood lactate concentration (r = 0.88; p < 0.05), and a correlation between blood lactate concentration and power output decrement (r = 0.82; p < 0.05). Thus, the greater fatiguability reported in males during sprint tasks may be related Males Females
a b
120 Absolute changes (mmol/kg d.m.)
Relative changes (mmol/kg d.m./W)
0.20
0.15
0.10
0.05
100 80 60 *
*
40 20 * 0
0
Cr
Lac
Cr
Lac
re
b
La
c
fib
fib
Fig. 3. (a) Absolute and (b) relative (relative to mean power output) changes in muscle metabolism during a maximal 30-second sprint on a cycle ergometer in males and females (adapted from Esbjo¨rnsson-Liljedahl et al.,[54]). d.m. = dry muscle; Lac = lactic; PCr = phosphocreatinine. * significant sex difference (p < 0.05) as reported by the authors.
mechanical score and performance decrement over subsequent sprints has consistently been reported. Indeed, the percentage decrement in mechanical output during MSE has been reported to be positively correlated with initial sprint performance (0.57 < r < 0.89; p < 0.05).[48,59,211] Accordingly, Yanagiya et al.[59] explained the greater fatiguability in teenage boys during an MSE test (10 · 6 seconds, 30-second recovery) via the observation that boys developed higher power levels ª 2009 Adis Data Information BV. All rights reserved.
6 Relative changes (mmol/kg d.m./kJ)
n G
ly co ge
PCr
I
I re
PC r
AT P
ATP
5 4 3 *
*
2 1 0 ATP
PCr
Fig. 4. (a) Absolute and (b) relative (per kJ of total mechanical work) changes in muscle metabolism of males and females during multiple-sprint exercise. When expressed per kJ of total mechanical work, the greatest metabolite changes appear in females (adapted from Bishop et al.[108]). Cr = creatinine; d.m. = dry muscle; Lac = lactic; PCr = phosphocreatinine. * significant sex difference (p < 0.05) as reported by the authors.
Sports Med 2009; 39 (4)
Sex and Sprint Fatiguability
to the greater initial sprint performance rather than actual sex differences. 3.2 Influence of Total Mechanical Work
A deeper examination of the results obtained during repeated supramaximal sprints reveals that many of the differences in metabolism and performance between males and females are likely to be due to sex differences in the absolute work performed during the actual task. For example, the greater the work performed during a sprint, or a series of sprints, the greater the changes in muscle substrates and hormonal responses.[58,108] Thus, it would be logical to observe greater metabolic and hormonal disturbances in males, and hence greater performance decrements, as they typically perform more work during a given exercise. It is then conceivable that this ‘methodological confound’ might contribute to reported sex differences in skeletal muscle metabolism (i.e. substrate depletion and hormonal responses) and subsequent performance reported during sprint exercises. In type I fibres, the exercise-induced glycogen reduction during a 30-second cycle sprint has been reported to be 42% smaller in females than in males, and this was associated with a 22% smaller increase in blood lactate concentration in females.[54] However, a closer look at the results of Esbjo¨rnsson-Liljedahl et al.[54] shows that, when expressed relative to mean power output (MPO), the impact of sex on glycogen changes is strongly reduced, and relative changes in muscle lactate concentration, actually become higher in females than in males (figure 3). Interestingly, the same observations can be made from the data collected in another study by the same group.[36] The authors reported that three 30-second cycle sprints with 20 minutes of recovery induced a smaller reduction of ATP (absolute values) in females than in males in type II muscle fibres (48% vs 62%, respectively; p < 0.05). However, once again, males developed higher power output levels throughout the sprint bouts, and a closer examination of those data demonstrates that sex differences in ATP concentration changes after sprints are nullified when expressed ª 2009 Adis Data Information BV. All rights reserved.
271
relative to MPO. Thus, the higher absolute changes for ATP after exercise in males seem almost completely reduced when values integrate power output level as a covariate. These results are consistent with those of Gaitanos et al.,[132] who reported a strong correlation between work done during a 6-second sprint and changes in metabolites. These analyses highlight the role of the work performed during exercise to account for the reported sex differences in performance, muscle metabolism and hormonal responses during repeated-sprint exercise. The greater perturbations observed in males might come from the fact that males actually perform significantly more mechanical work than females during a given sprint. This in turn is likely to lead to greater decrements in performance, as Gaitanos and colleagues[132] have shown that performance decrement is related to anaerobic metabolism during the first sprint. Furthermore, the validity of comparing powerful with less powerful subjects (males and females, respectively, in this review) has frequently been questioned, as a greater initial sprint performance is positively correlated with a greater performance decrement.[48,59,132] This raises the possibility that the often-observed sex differences in fatiguability may actually be due to inappropriate comparison methods rather than actual sex differences. Rather, it is likely that if body dimensions and initial maximal performance are satisfactorily covaried, the changes in muscle metabolism and fatigue associated with sprint activity will largely depend on the absolute mechanical work performed by subjects. Such a methodological confound has previously been highlighted by our group in abstract form.[108] We investigated the sex difference in muscle metabolism during a MSE consisting of five all-out sprints lasting 6 seconds repeated every 30 seconds. Both absolute (kJ) and relative (J/kg) work values were greater in males than in females. As expected, the work decrement (%) over five sprints was greater in males (males 13.7 – 5.1%, females 11.0 – 2.8%; p < 0.05). The sprints were accompanied by greater absolute changes in ATP, PCr, creatine and lactate concentrations in males than in Sports Med 2009; 39 (4)
Billaut & Bishop
272
females (figure 4). However, when expressed in relative terms (i.e. per kJ of work), only sex differences in PCr and creatine persisted – but were inverted. Sex differences in muscle metabolism then appeared to be largely due to differences in the absolute work performed by males and females.[108] Thus, these results suggest that unless males and females are matched for total work (or total work is used as a covariate), it is very difficult to compare decrements in performance during repeated sprint exercise, as differences in hormonal and metabolic responses are likely to affect the development of fatigue. 4. Summary and Future Directions The investigation of skeletal muscle fatigue in males and females must be made with appropriate comparison methods. Studies of the fatiguability and metabolic and hormonal responses of males and females during MSE have, in our opinion, not been optimally designed to control for possible covariates. Rather than taking into account only the strength or power output capacity of the rested muscle, studies dedicated to understanding sex differences should incorporate the total mechanical work done by each sex during exercise as a covariate. Furthermore, studies are warranted where males and females are matched for both initial power and activity levels. Although it is tempting to propose that males are more susceptible to fatigue than females for a given sprint, we must emphasize the need for more basic research comparing exercise tolerance between the sexes. In addition, greater investigation of the influence of initial force on the mechanisms of fatigue in males versus females is needed. It is encouraging to see a number of studies using advanced techniques to analyse muscle fatigue aetiology under a variety of conditions. It is probable, however, that our uncertainty surrounding the understanding of the sex difference in muscle fatigue stems from scientists from diverse specialist fields having taken a local approach, in an attempt to solve the underlying cause of fatigue. Multiple-sprint activity requires bouts of all-out intensity to be repeated several times with incomplete recovery. The potential ª 2009 Adis Data Information BV. All rights reserved.
sex difference in voluntary activation of active muscles has not been sufficiently examined during sprint activity; the greater impairment of central drive in males during discrete tasks (e.g. contraction of dorsiflexor and elbow flexor muscles) may also be found during whole-body sprinting. Furthermore, the use of transcranial magnetic stimulation would be a powerful complement for investigating the contribution of supraspinal fatigue to task failure during repeated sprints in males and females. 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.8pt?>
References 1. Enoka RM, Stuart DG. Neurobiology of muscle fatigue. J Appl Physiol 1992; 72 (5): 1631-48 2. Gandevia S. Spinal and supraspinal factors in human muscle fatigue. Physiol Rev 2001; 81: 1725-89 3. Bigland-Ritchie B, Furbush F, Woods JJ. Fatigue of intermittent submaximal voluntary contractions: central and peripheral factors. J Appl Physiol 1986; 61 (2): 421-9 4. Fuglevand A, Zackowski K, Huey K, et al. Impairment of neuromuscular propagation during human fatiguing contractions at submaximal forces. J Physiol 1993; 460: 549-72 5. Gibson H, Edwards RHT. Muscular exercise and fatigue. Sports Med 1985; 2: 120-32 6. Merton P. Voluntary strength and fatigue. J Physiol 1954; 123: 553-64 7. Kent-Braun JA. Central and peripheral contributions to muscle fatigue in humans during sustained maximal effort. Eur J Appl Physiol 1999; 80: 57-63 8. Taylor JL, Allen GM, Butler JE, et al. Supraspinal fatigue during intermittent maximal voluntary contractions of the human elbow flexors. J Appl Physiol 2000; 89: 305-11 9. Drinkwater BL. Women and exercise: physiological aspects. Exerc Sport Sci Rev 1984; 12: 21-51 10. Hicks AL, Kent-Braun J, Ditor DS. Sex differences in human skeletal muscle fatigue. Exerc Sport Sci Rev 2001; 29 (3): 109-12 11. Shephard RJ. Exercise and training in women: part I. Influence of gender on exercise and training responses. Can J Appl Physiol 2000; 25 (1): 19-34 12. Bangsbo J, Norregaard L, Thorso F. Activity profile of competition soccer. Can J Sport Sci 1991; 16 (2): 110-6 13. Mendez-Villanueva A, Fernandez-Fernandez J, Bishop D, et al. Activity patterns, blood lactate concentrations and ratings of perceived exertion during a professional singles tennis tournament. Br J Sports Med 2007; 41 (5): 296-300
Sports Med 2009; 39 (4)
Sex and Sprint Fatiguability
14. Spencer M, Lawrence S, Rechichi C, et al. Time-motion analysis of elite field hockey, with special reference to repeated-sprint activity. J Sports Sci 2004; 22: 843-50 15. Balsom PD, Seger JY, Sjodin B, et al. Maximal-intensity intermittent exercise: effect of recovery duration. Int J Sports Med 1992; 13 (7): 528-33 16. Billaut F, Giacomoni M, Falgairette G. Maximal intermittent cycling exercise: effects of recovery duration and gender. J Appl Physiol 2003; 95: 1632-7 17. Batterham AM, Birch KM. Allometry of anaerobic performance: a gender comparison. Can J Appl Physiol 1996; 21: 45-62 18. Cramer JT, Housh TH, Weir JP, et al. Power output, mechanomyographic, and electromyographic responses to maximal concentric, isokinetic muscle actions in men and women. J Strength Cond Res 2002; 16: 399-408 19. Esbjo¨rnsson-Liljedahl M, Sylve´n C, Holm I, et al. Fast twitch fibres may predict anaerobic performance in both females and males. Int J Sports Med 1993; 14: 257-63 20. Falgairette G, Billaut F, Giacomoni M, et al. Effect of inertia on performance and fatigue pattern during repeated cycle sprints in males and females. Int J Sports Med 2004; 25: 235-40 21. Falkel JE, Sawka MN, Levine L, et al. Upper to lower body muscular strength and endurance ratios for women and men. Ergonomics 1985; 28 (12): 1661-70 22. Green S. Measurement of anaerobic work capacities in humans. Sports Med 1995; 19 (1): 32-42 23. Hunter SK, Enoka RM. Sex differences in the fatigability of arm muscles depends on absolute force during isometric contractions. J Appl Physiol 2001; 91: 2686-94 24. Kanehisa H, Ikegawa S, Fukunaga T. Comparison of muscle cross-sectional area and strength between untrained women and men. Eur J Appl Physiol Occup Physiol 1994; 68 (2): 148-54 25. Krivickas LS, Suh D, Wilkins J, et al. Age- and genderrelated differences in maximum shortening velocity of skeletal muscle fibers. Am J Phys Med Rehabil 2001; 80 (6): 447-55 26. Laubach LL. Comparative muscular strength of men and women: a review of the literature. Aviat Space Environ Med 1976; 47 (5): 534-42 27. Martin RJ, Dore E, Twisk J, et al. Longitudinal changes of maximal short-term peak power in girls and boys during growth. Med Sci Sports Exerc 2004; 36 (3): 498-503 28. Mayhew J, Salm P. Gender differences in anaerobic power tests. Eur J Appl Physiol 1990; 60: 133-8 29. Miller AEJ, MacDougall JD, Tarnopolsky MA, et al. Gender differences in strength and muscle fibre characteristics. Eur J Appl Physiol 1993; 66: 254-62 30. Winter EM, Brookes FBC, Hamley EJ. Maximal exercise performance and lean leg volume in men and women. J Sports Sci 1991; 9: 3-13 31. Clark BC, Manini TM, The´ DJ, et al. Gender differences in skeletal muscle fatigability are related to contraction type and EMG spectral compression. J Appl Physiol 2003; 94: 2263-72 32. Linnamo V, Hakkinen K, Komi PV. Neuromuscular fatigue and recovery in maximal compared to explosive
ª 2009 Adis Data Information BV. All rights reserved.
273
33.
34.
35.
36.
37.
38.
39.
40.
41.
42.
43.
44.
45. 46.
47.
48.
49.
strength loading. Eur J Appl Physiol Occup Physiol 1998; 77 (1-2): 176-81 Maughan RJ, Harmon M, Leiper JB, et al. Endurance capacity of untrained males and females in isometric and dynamic muscular contractions. Eur J Appl Physiol 1986; 55: 395-400 Semmler JG, Kutzscher DV, Enoka RM. Gender differences in the fatigability of the human skeletal muscle. J Neurophysiol 1999; 82: 3590-93 West W, Hicks AL, Clements L, et al. The relationship between voluntary electromyogram, endurance time and intensity of effort in isometric handgrip exercise. Eur J Appl Physiol 1995; 71: 301-5 Esbjo¨rnsson-Liljedahl M, Bodin K, Jansson E. Smaller muscle ATP reduction in women than in men by repeated bouts of sprints exercise. J Appl Physiol 2002; 93: 1075-83 Froese E, Houston M. Performance during the Wingate anaerobic test and muscle morphology in males and females. Int J Sports Med 1987; 8: 35-9 Glenmark B, Hedberg G, Jansson E. Changes in muscle fibre type from adolescence to adulthood in women and men. Acta Physiol Scand 1992; 146 (2): 251-9 Hill DW, Smith JC. Gender difference in anaerobic capacity: role of aerobic contribution. Br J Sports Med 1993; 27 (1): 45-8 Kumagai K, Abe T, Brechue WF, et al. Sprint performance is related to muscle fascicle length in male 100-m sprinters. J Appl Physiol 2000; 88 (3): 811-6 Maughan RJ, Watson JS, Weir J. Strength and crosssectional area of human skeletal muscle. J Physiol 1983; 338: 37-49 Jaworowski A, Porter MM, Holmback AM, et al. Enzyme activities in the tibialis anterior muscle of young moderately active men and women: relationship with body composition, muscle cross-sectional area and fibre type composition. Acta Physiol Scand 2002; 176 (3): 215-25 Wright A, Marino F, Kay D, et al. Influence of lean body mass on performance differences of male and female distance runners in warm, humid environments. Am J Phys Anthropol 2002; 118: 285-91 Hakkinen K, Keskinen KL. Muscle cross-sectional area and voluntary force production characteristics in elite strength- and endurance-trained athletes and sprinters. Eur J Appl Physiol Occup Physiol 1989; 59 (3): 215-20 Maughan RJ. The limits of human athletic performance. Ann Transplant 2005; 10 (4): 52-4 Seynnes OR, de Boer M, Narici MV. Early skeletal muscle hypertrophy and architectural changes in response to high-intensity resistance training. J Appl Physiol 2007; 102 (1): 368-73 Tesch PA. Skeletal muscle adaptations consequent to longterm heavy resistance exercise. Med Sci Sports Exerc 1988; 20 (5 Suppl.): S132-4 Bishop D, Lawrence S, Spencer M. Predictors of repeatedsprints ability in elite females hockey players. J Sci Med Sport 2003; 6 (2): 199-209 Maud PJ, Schultz BB. Gender comparisons in anaerobic power and anaerobic capacity test. Br J Sports Med 1986; 20: 51-4
Sports Med 2009; 39 (4)
274
50. Nindl BC, Mahar MT, Harman EA, et al. Lower and upper body anaerobic performance in male and female adolescent athletes. Med Sci Sports Exerc 1995; 27 (2): 235-41 51. Perez-Gomez J, Rodriguez GV, Ara I, et al. Role of muscle mass on sprint performance: gender differences? Eur J Appl Physiol 2008; 102 (6): 685-94 52. Weber CL, Chia M, Inbar O. Gender differences in anaerobic power of the arms and legs: a scaling issue. Med Sci Sports Exerc 2006; 38: 129-37 53. Ditor DS, Kent-Braun J. The effect of age and gender on the relative fatigability of the human adductor pollicis muscle. Can J Physiol Pharmacol 2000; 78: 781-90 54. Esbjo¨rnsson-Liljedahl M, Sundberg CJ, Norman B, et al. Metabolic response in type I and type II muscle fibers during a 30-s cycle sprint in men and women. J Appl Physiol 1999; 87: 1326-32 55. Fulco CS, Rock PB, Muza SR, et al. Slower fatigue and faster recovery of the adductor pollicis muscle in women matched for strength with men. Acta Physiol Scand 1999; 167: 233-9 56. Gratas-Delamarche A, Le Cam R, Delamarche P, et al. Lactate and catecholamine responses in male and female sprinters during a Wingate test. Eur J Appl Physiol 1994; 68: 362-6 57. Hunter SK, Critchlow A, Shin I-S, et al. Men are more fatigable than strength-matched women when performing intermittent submaximal contractions. J Appl Physiol 2004; 96: 2125-32 58. Vincent S, Berthon P, Zouhal H, et al. Plasma glucose, insulin and catecholamine responses to a Wingate test in physically active women and men. Eur J Appl Physiol 2004; 91 (1): 15-21 59. Yanagiya T, Kanehisa H, Kouzaki M, et al. Effect of gender on mechanical power output during repeated bouts of maximal running in trained teenagers. Int J Sports Med 2003; 24: 304-10 60. Ahtiainen JP, Pakarinen A, Alen M, et al. Muscle hypertrophy, hormonal adaptations and strength development during strength training in strength-trained and untrained men. Eur J Appl Physiol 2003; 89 (6): 555-63 61. Sinha-Hikim I, Cornford M, Gaytan H, et al. Effects of testosterone supplementation on skeletal muscle fiber hypertrophy and satellite cells in community-dwelling older men. J Clin Endocrinol Metab 2006; 91 (8): 3024-33 62. Nygaard E. Skeletal muscle fibre characteristics in young women. Acta Physiol Scand 1981; 112 (3): 299-304 63. Ruby B, Robergs R, Waters D, et al. Effects of estradiol on substrate turnover during exercise in amenorrheic females. Med Sci Sports Exerc 1997; 29 (9): 1160-9 64. Giustina A, Veldhuis JD. Pathophysiology of the neuroregulation of growth hormone secretion in experimental animals and the human. Endocr Rev 1998; 19 (6): 717-97 65. Pincus SM, Gevers EF, Robinson IC, et al. Females secrete growth hormone with more process irregularity than males in both humans and rats. Am J Physiol 1996; 270 (1 Pt 1): E107-15 66. Veldhuis JD. The neuroendocrine regulation and implications of pulsatile GH secretion: gender effects. Endocrinology 1995; 5: 198-213
ª 2009 Adis Data Information BV. All rights reserved.
Billaut & Bishop
67. Wideman L, Weltman JY, Shah N, et al. Effects of gender on exercise-induced growth hormone release. J Appl Physiol 1999; 87 (3): 1154-62 68. Sandoval DA, Matt KS. Gender differences in the endocrine and metabolic responses to hypoxic exercise. J Appl Physiol 2002; 92 (2): 504-12 69. Thompson DL, Weltman JY, Rogol AD, et al. Cholinergic and opioid involvement in release of growth hormone during exercise and recovery. J Appl Physiol 1993; 75 (2): 870-8 70. Tarnopolsky LJ, MacDougall JD, Atkinson SA, et al. Gender differences in substrate for endurance exercise. J Appl Physiol 1990; 68 (1): 302-8 71. Weber CL, Schneider DA. Maximal accumulated oxygen deficit expressed relative to the active muscle mass for cycling in untrained male and female subjects. Eur J Appl Physiol 2000; 82 (4): 255-61 72. Dar DE, Zinder O. Short term effect of steroids on catecholamine secretion from bovine adrenal medulla chromaffin cells. Neuropharmacology 1997; 36 (11-12): 1783-8 73. Lebrun CM, Rumball JS. Relationship between athletic performance and menstrual cycle. Curr Women’s Health Rep 2001; 1: 232-40 74. Glenmark B. Skeletal muscle fibre types, physical performance, physical activity and attitude to physical activity in women and men: a follow-up from age 16 to 27. Acta Physiol Scand Suppl 1994; 623: 1-47 75. Glenmark B, Nilsson M, Gao H, et al. Difference in skeletal muscle function in males versus females: role of estrogen receptor-b. Am J Physiol Endocrinol Metab 2004; 287: E1125-31 76. Middleton LE, Wenger HA. Effects of menstrual phase on performance and recovery in intense intermittent activity. Eur J Appl Physiol 2006; 96: 53-8 77. Phillips SK, Sanderson AG, Birch K, et al. Changes in maximal voluntary force of human adductor pollicis muscle during the menstrual cycle. J Physiol 1996; 496 (Pt 2): 551-7 78. Sarwar R, Niclos BB, Rutherford OM. Changes in muscle strength, relaxation rate and fatiguability during the human menstrual cycle. J Physiol 1996; 493 (Pt 1): 267-72 79. Jurkowski JE, Jones NL, Toews CJ, et al. Effects of menstrual cycle on blood lactate, O2 delivery, and performance during exercise. J Appl Physiol 1981; 51 (6): 1493-9 80. McCracken M, Ainsworth B, Hackney AC. Effects of the menstrual cycle phase on the blood lactate responses to exercise. Eur J Appl Physiol Occup Physiol 1994; 69 (2): 174-5 81. Matsuo T, Saitoh S, Suzuki M. Effects of the menstrual cycle on excess postexercise oxygen consumption in healthy young women. Metabolism 1999; 48 (3): 275-7 82. DiBrezzo R, Fort IL, Brown B. Relationships among strength, endurance, weight and body fat during three phases of the menstrual cycle. J Sports Med Phys Fitness 1991; 31 (1): 89-94 83. Gur H. Concentric and eccentric isokinetic measurements in knee muscles during the menstrual cycle: a special
Sports Med 2009; 39 (4)
Sex and Sprint Fatiguability
84.
85.
86.
87.
88.
89.
90.
91.
92.
93.
94. 95.
96.
97.
98.
99.
100.
reference to reciprocal moment ratios. Arch Phys Med Rehabil 1997; 78 (5): 501-5 Janse de Jonge XA, Boot CR, Thom JM, et al. The influence of menstrual cycle phase on skeletal muscle contractile characteristics in humans. J Physiol 2001; 530 (Pt 1): 161-6 Lebrun CM, McKenzie DC, Prior JC, et al. Effects of menstrual cycle phase on athletic performance. Med Sci Sports Exerc 1995; 27 (3): 437-44 Blomstrand E, Radegran G, Saltin B. Maximum rate of oxygen uptake by human skeletal muscle in relation to maximal activities of enzymes in the Krebs cycle. J Physiol 1997; 501 (Pt 2): 455-60 Borges O, Essen-Gustavsson B. Enzyme activities in type I and II muscle fibres of human skeletal muscle in relation to age and torque development. Acta Physiol Scand 1989; 136 (1): 29-36 Komi PV, Karlsson J. Skeletal muscle fiber types, enzyme activities and physical performance in young males and females. Acta Physiol Scand 1978; 103: 212-8 Gauthier JM, Theriault R, Theriault G, et al. Electrical stimulation-induced changes in skeletal muscle enzymes of men and women. Med Sci Sports Exerc 1992; 24 (11): 1252-6 Green HJ, Fraser IG, Ranney DA. Male and female differences in enzyme activities of energy metabolism in vastus lateralis muscle. J Neurol Sci 1984; 65: 323-31 Simoneau JA, Bouchard C. Human variation in skeletal muscle fiber-type proportion and enzyme activities. Am J Physiol 1989; 257 (4 Pt 1): E567-72 Simoneau JA, Lortie G, Boulay MR, et al. Skeletal muscle histochemical and biochemical characteristics in sedentary male and female subjects. Can J Physiol Pharmacol 1985; 63 (1): 30-5 Dovey SM, Reeder AI, Chalmers DJ. Continuity and change in sporting and leisure time physical activities during adolescence. Br J Sports Med 1998; 32 (1): 53-7 Engstrom LM. Physical activity of children and youth. Acta Paediatr Scand Suppl 1980; 283: 101-5 Santos MP, Gomes H, Mota J. Physical activity and sedentary behaviors in adolescents. Ann Behav Med 2005; 30 (1): 21-4 Telama R, Yang X. Decline of physical activity from youth to young adulthood in Finland. Med Sci Sports Exerc 2000; 32: 1617-22 Kent-Braun JA, Ng AV, Doyle JW, et al. Human skeletal muscle responses vary with age and gender during fatigue due to incremental isometric exercise. J Appl Physiol 2002; 93: 1813-23 Carter SL, Rennie CD, Hamilton SJ, et al. Changes in skeletal muscle in males and females following endurance training. Can J Physiol Pharmacol 2001; 79 (5): 386-92 Hoppeler H, Howald H, Conley K, et al. Endurance training in humans: aerobic capacity and structure of skeletal muscle. J Appl Physiol 1985; 59 (2): 320-7 McKenzie S, Phillips SM, Carter SL, et al. Endurance exercise training attenuates leucine oxidation and BCOAD activation during exercise in humans. Am J Physiol Endocrinol Metab 2000; 278 (4): E580-7
ª 2009 Adis Data Information BV. All rights reserved.
275
101. Esbjo¨rnsson-Liljedahl M, Holm I, Christer S, et al. Different responses of skeletal muscle following sprint training in men and women. Eur J Appl Physiol 1996; 74: 375-83 102. Harber V, Petersen S, Chilibeck P. Thyroid hormone concentrations and muscle metabolism in amenorrheic and eumenorrheic athletes. Can J Appl Physiol 1998; 23 (3): 293-306 103. Russ DW, Kent-Braun JA. Sex differences in human skeletal muscle fatigue are eliminated under ischemic conditions. J Appl Physiol 2003; 94: 2414-22 104. Brooks S, Nevill ME, Meleagros L, et al. The hormonal responses to repetitive brief maximal exercise in humans. Eur J Appl Physiol 1990; 60: 144-8 105. Nevill ME, Holmyard DJ, Hall GM, et al. Growth hormone responses to treadmill sprinting in sprint- and endurance-trained athletes. Eur J Appl Physiol Occup Physiol 1996; 72 (5-6): 460-7 106. Jacobs I, Tesch P, Bar-Or O, et al. Lactate in human skeletal muscle after 10 and 30 s of supramaximal exercise. J Appl Physiol 1983; 55 (2): 365-7 107. Bodin K, Esbjo¨rnsson-Liljedahl M, Jansson E. Alactic ATP turnover rate during a 30-s cycle sprint in females and males [abstract]. Clin Sci 1994; 87 Suppl.: 205 108. Bishop D, Edge J, Dawson B, et al. Gender differences in muscle metabolism during repeated-sprint exercise [abstract]. International Biochemistry of Exercise Conference; 2003 Jul 13-16; Maastricht 109. Brooke MH, Engel WK. The histographic analysis of human muscle biopsies with regard to fiber types, 1: adult male and female. Neurology 1969; 19 (3): 221-33 110. Gerdle B, Karlsson S, Crenshaw AG, et al. The relationships between EMG and muscle morphology throughout sustained static knee extension at two submaximal force levels. Acta Physiol Scand 1997; 160 (4): 341-51 111. Ruby B, Robergs R. Gender differences in substrate utilisation during exercise. Sports Med 1994; 17 (6): 393-410 112. Sale DG, MacDougall JD, Alway SE, et al. Voluntary strength and muscle characteristics in untrained men and women and male bodybuilders. J Appl Physiol 1987; 62 (5): 1786-93 113. Gerdle B, Karlsson S, Crenshaw AG, et al. The influences of muscle fibre proportions and areas upon EMG during maximal dynamic knee extensions. Eur J Appl Physiol 2000; 81 (1-2): 2-10 114. De Luca CJ. The use of surface electromyography in biomechanics. J Appl Biomechanics 1997; 13: 135-63 115. Farina D, Merletti R, Enoka RM. The extraction of neural strategies from the surface EMG. J Appl Physiol 2004; 96: 1486-95 116. Hakkinen K. Neuromuscular fatigue and recovery in male and female athletes during heavy resistance exercise. Int J Sports Med 1993; 14: 53-9 117. Gandevia SC, Allen GM, Butler JE, et al. Supraspinal factors in human muscle fatigue: evidence for suboptimal output from the motor cortex. J Physiol 1996; 490 (Pt 2): 529-36 118. Todd G, Taylor JL, Gandevia SC. Measurement of voluntary activation of fresh and fatigued human muscles
Sports Med 2009; 39 (4)
Billaut & Bishop
276
119.
120.
121. 122.
123.
124.
125.
126.
127.
128.
129.
130.
131.
132.
133.
134.
135.
using transcranial magnetic stimulation. J Physiol 2003; 551 (Pt 2): 661-71 Hunter SK, Butler JE, Todd G, et al. Supraspinal fatigue does not explain the sex difference in muscle fatigue of maximal contractions. J Appl Physiol 2006; 101 (4): 1036-44 Gerdle B, Fugl-Meyer AR. Is the mean power frequency shift of the EMG a selective indicator of fatigue of the fast twitch motor units? Acta Physiol Scand 1992; 145 (2): 129-38 Juel C. Muscle action potential propagation velocity changes during activity. Muscle Nerve 1988; 11: 714-9 Kupa E, Roy S, Kandarian S, et al. Effects of muscle fiber type and size on EMG median frequency and conduction velocity. J Appl Physiol 1995; 79: 23-32 Glaister M. Multiple sprint work: physiological responses, mechanisms of fatigue and the influence of aerobic fitness. Sports Med 2005; 35: 757-77 Spencer M, Bishop D, Dawson B, et al. Physiological and metabolic responses of repeated-sprint activities: specific to field-based team sports. Sports Med 2005; 35 (12): 1025-44 Bilodeau M, Schindler-Ivens S, Williams DM, et al. EMG frequency content changes with increasing force and during fatigue in the quadriceps femoris muscle of men and women. J Electromyogr Kinesiol 2003; 13: 83-92 Evetovich T, Housh T, Johnson G, et al. Gender comparisons of the mechanomyographic responses to maximal concentric and eccentric isokinetic muscle actions. Med Sci Sports Exerc 1998; 30: 1697-702 Hunter SK, Ryan DL, Ortega JD, et al. Task differences with the same load torque alter the endurance time of submaximal fatiguing contractions in humans. J Neurophysiol 2002; 88: 3087-96 Bogdanis GC, Nevill ME, Lakomy HK, et al. Power output and muscle metabolism during and following recovery from 10 and 20 s of maximal sprint exercise in humans. Acta Physiol Scand 1998; 163 (3): 261-72 Bogdanis GC, Nevill ME, Lakomy HK, et al. Effects of active recovery on power output during repeated maximal sprint cycling. Eur J Appl Physiol Occup Physiol 1996; 74 (5): 461-9 Cheetham ME, Boobis LH, Brooks S, et al. Human muscle metabolism during sprint running. J Appl Physiol 1986; 61 (1): 54-60 Lakomy H. Measurement of work and power output using friction-loaded cycle ergometer. Ergonomics 1986; 29 (4): 509-17 Gaitanos GC, Williams C, Boobis LH, et al. Human muscle metabolism during intermittent maximal exercise. J Appl Physiol 1993; 75 (2): 712-9 Boobis L, Williams C, Wootton S. Human muscle metabolism during brief maximal exercise [abstract]. J Physiol (Lond) 1982; 338 (21P): 22P Parolin ML, Chesley A, Matsos MP, et al. Regulation of skeletal muscle glycogen phosphorylase and PDH during maximal intermittent exercise. Am J Physiol 1999; 277 (5 Pt 1): E890-900 Bogdanis GC, Nevill ME, Boobis LH, et al. Contribution of phosphocreatine and aerobic metabolism to energy
ª 2009 Adis Data Information BV. All rights reserved.
136.
137.
138.
139.
140.
141.
142.
143.
144.
145.
146.
147.
148.
149.
150. 151.
152.
supply during repeated sprint exercise. J Appl Physiol 1996; 80 (3): 876-84 Bogdanis GC, Nevill ME, Boobis LH, et al. Recovery of power output and muscle metabolism after 10s and 20s of maximal sprint exercise in man. Clin Sci 1994; 87 Suppl. 1: 121-2 Boobis LH. Metabolic aspects of fatigue during sprinting. In: MacLeod D, Maughan RJ, Nimmo MA, et al., editors. Exercise: benefits, limitations and adaptations. London: E & FN Spon ed., 1987: 116-40 Spriet L, So¨derlund K, Bergstro¨m M, et al. Anaerobic energy release in skeletal muscle during electrical stimulation in men. J Appl Physiol 1987; 62 (2): 611-5 Medbø JI, Tabata I. Anaerobic energy release in working muscle during 30 s to 3 min of exhausting bicycling. J Appl Physiol 1993; 75 (4): 1654-60 Dawson B, Goodman C, Lawrence S, et al. Muscle phosphocreatine repletion following single and repeated short sprint efforts. Scand J Med Sci Sports 1997; 7: 206-13 Parra J, Cadefau J, Rodas G, et al. The distribution of rest periods affects performance and adaptations of energy metabolism induced by high-intensity training in human muscle. Acta Physiol Scand 2000; 169: 157-65 Bogdanis GC, Nevill ME, Lakomy HKA, et al. Muscle metabolism during repeated sprint exercise in man. J Physiol 1994; 475: 25P-6P Bogdanis GC, Nevill ME, Boobis LH, et al. Recovery of power output and muscle metabolites following 30 s of maximal sprint cycling in man. J Physiol 1995; 482 (Pt 2): 467-80 Casey A, Constantin-Teodosiu D, Howell S, et al. Metabolic response of type I and II muscle fibers during repeated bouts of maximal exercise in humans. Am J Physiol 1996; 271 (1 Pt 1): E38-43 Hirvonen J, Rehunen S, Rusko H, et al. Breakdown of high-energy phosphate compounds and lactate accumulation during short supramaximal exercise. Eur J Appl Physiol Occup Physiol 1987; 56 (3): 253-9 Karatzaferi C, de Haan A, Ferguson RA, et al. Phosphocreatine and ATP content in human single muscle fibres before and after maximum dynamic exercise. Pflu¨gers Arch-Eur J Physiol 2001; 442: 467-74 Bangsbo¨ J, Graham TE, Kiens B, et al. Elevated muscle glycogen and anaerobic energy production during exhaustive exercise in man. J Physiol 1992; 451: 205-27 Hargreaves M, McKenna MJ, Jenkins DG, et al. Muscle metabolites and performance during high-intensity, intermittent exercise. J Appl Physiol 1998; 84 (5): 1687-91 Nevill ME, Boobis LH, Brooks S, et al. Effect of training on muscle metabolism during treadmill sprinting. J Appl Physiol 1989; 67 (6): 2376-82 Fitts R. Cellular mechanisms of muscle fatigue. Physiol Rev 1994; 74 (1): 49-94 Harris R, Sahlin K, Hultman E. Phosphagen and lactate contents of m. quadriceps femoris of man after exercise. J Appl Physiol 1977; 43 (5): 852-7 Spriet L, Lindinger M, McKelvie R, et al. Muscle glycogenolysis and H+ concentration during maximal intermittent cycling. J Appl Physiol 1989; 66 (1): 8-13
Sports Med 2009; 39 (4)
Sex and Sprint Fatiguability
153. Spriet L, So¨derlund K, Bergstro¨m M, et al. Skeletal muscle glycogenolysis, glycolysis, and pH during electrical stimulation in men. J Appl Physiol 1987; 62 (2): 616-21 154. Allen DG, Westerblad H, La¨nnergren J. The role of intracellular acidosis in muscle fatigue. Adv Exp Med Biol 1995; 384: 57-68 155. Harris R, Edwards R, Hultman E, et al. The time course of phosphorylcreatine resynthesis during recovery of the quadriceps muscle in man. Pflu¨gers Arch 1976; 367: 137-42 156. Mercier J, Mercier B, Prefaut C. Blood lactate increases during the force velocity exercise test. Int J Sports Med 1991; 12 (1): 17-20 157. Bergstro¨m M, Hultman E. Relaxation and force during fatigue and recovery of human quadriceps muscle: relations to metabolite changes. Eur J Appl Physiol 1991; 418: 153-60 158. MacIntosh BR, Allen DG. Contractile changes and mechanisms of muscle fatigue. In: Nigg BM, MacIntosh BR, Mester J, editors. Biomechanics and biology of movement. Champaign (IL): Human Kinetics, 2000: 365-83 159. Rotto DM, Kaufman MP. Effect of metabolic products of muscular contraction on discharge of group III and IV afferents. J Appl Physiol 1988; 64 (6): 2306-13 160. Sacco P, Newberry R, McFadden L, et al. Depression of human electromyographic activity by fatigue of a synergistic muscle. Muscle Nerve 1997; 20: 710-7 161. Sinoway LI, Hill JM, Pickar JG, et al. Effects of contraction and lactic acid on the discharge of group III muscle afferents in cats. J Neurophysiol 1993; 69 (4): 1053-9 162. Juel C, Pilegaard H, Nielsen JJ, et al. Interstitial K(+) in human skeletal muscle during and after dynamic graded exercise determined by microdialysis. Am J Physiol Regul Integr Comp Physiol 2000; 278 (2): R400-6 163. Lindinger MI, Heigenhauser GJ. The roles of ion fluxes in skeletal muscle fatigue. Can J Physiol Pharmacol 1991; 69 (2): 246-53 164. Lindinger MI, Heigenhauser GJ, McKelvie RS, et al. Blood ion regulation during repeated maximal exercise and recovery in humans. Am J Physiol 1992; 262 (1 Pt 2): R126-36 165. Medbo JI, Sejersted OM. Plasma potassium changes with high intensity exercise. J Physiol 1990; 421: 105-22 166. Mohr M, Nordsborg N, Nielsen JJ, et al. Potassium kinetics in human muscle interstitium during repeated intense exercise in relation to fatigue. Pflugers Arch 2004; 448 (4): 452-6 167. Allen DG, Westerblad H. Role of phosphate and calcium stores in muscle fatigue. J Physiol 2001; 536: 657-65 168. Steele D, Duke A. Metabolic factors contributing to altered Ca2+ regulation in skeletal muscle fatigue. Acta Physiol Scand 2003; 179: 39-48 169. Westerblad H, Lee JA, La¨nnergren J, et al. Cellular mechanisms of fatigue in skeletal muscle. Am J Physiol 1991; 261: C195-209 170. Vandewalle H, Maton B, Le Bozec S, et al. An electromyographic study of an all-out exercise on a cycle ergometer. Arch Int Physiol Biochim Biophys 1991; 99 (1): 89-93 171. Hunter AM, St Clair Gibson A, Lambert MI, et al. Effects of supramaximal exercise on the electromyographic signal. Br J Sports Med 2003; 37: 296-9
ª 2009 Adis Data Information BV. All rights reserved.
277
172. Taylor AD, Bronks R, Smith P, et al. Myoelectric evidence of peripheral muscle fatigue during exercise in severe hypoxia: some references to m. vastus lateralis myosin heavy chain composition. Eur J Appl Physiol Occup Physiol 1997; 75 (2): 151-9 173. Linnamo V, Bottas R, Komi PV. Force and EMG power spectrum during and after eccentric and concentric fatigue. J Electromyogr Kinesiol 2000; 10: 293-300 174. Linssen W, Jacobs M, Stegeman D, et al. Muscle fatigue in McArdle’s disease: muscle fibre conduction velocity and surface EMG frequency spectrum during ischaemic exercise. Brain 1990; 113: 1779-93 175. St Clair Gibson A, Lambert MI, Noakes TD. Neural control of force output during maximal and submaximal exercise. Sports Med 2001; 31: 637-50 176. Murphy M, Patton J, Frederick F. Comparative anaerobic power of men and women. Aviat Space Environ Med 1986; 57: 636-41 177. Dore´ E, Martin R, Ratel S, et al. Gender differences in peak muscle performance during growth. Int J Sports Med 2005; 26: 274-80 178. Esbjo¨rnsson-Liljedahl M, Jansson E. Sex difference in plasma ammonia but not in muscle inosine monophosphate accumulation following sprint exercise in humans. Eur J Appl Physiol Occup Physiol 1999; 79 (5): 404-8 179. Tarnopolsky MA. Gender differences in substrate metabolism during endurance exercise. Can J Appl Physiol 2000; 25 (4): 312-27 180. Tarnopolsky MA. Gender differences in metabolism, nutrition and supplements. J Sci Med Sport 2000; 3 (3): 287-98 181. Russ DW, Lanza IR, Rothman D, et al. Sex differences in glycolysis during brief, intense isometric contractions. Muscle Nerve 2005; 32: 647-55 182. Glenmark B, Hedberg G, Kaijser L, et al. Muscle strength from adolescence to adulthood: relationship to muscle fibre types. Eur J Appl Physiol Occup Physiol 1994; 68 (1): 9-19 183. Krotkiewski M, Kral JG, Karlsson J. Effects of castration and testosterone substitution on body composition and muscle metabolism in rats. Acta Physiol Scand 1980; 109 (3): 233-7 184. Bishop D, Edge J, Goodman C. Muscle buffer capacity and aerobic fitness are associated with repeated-sprint ability in women. Eur J Appl Physiol 2004; 92: 540-7 185. Mayhew SR, Wenger HA. Time-motion analysis of professional soccer. J Hum Mov Stud 1985; 11 (1): 49-52 186. Balsom PD, Seger JY, Sjodin B, et al. Physiological responses to maximal intensity intermittent exercise. Eur J Appl Physiol Occup Physiol 1992; 65 (2): 144-9 187. Spencer M, Bishop D, Lawrence S. Longitudinal assessment of the effects of field-hockey training on repeated sprint ability. J Sci Med Sport 2004; 7: 323-34 188. Stathis CG, Zhao S, Carey MF, et al. Purine loss after repeated sprint bouts in humans. J Appl Physiol 1999; 87: 2037-42 189. Sjodin B, Hellsten Westing Y. Changes in plasma concentration of hypoxanthine and uric acid in man with short-distance running at various intensities. Int J Sports Med 1990; 11 (6): 493-5
Sports Med 2009; 39 (4)
278
190. Harris R, Hultman E, Kaijser L, et al. The effect of circulatory occlusion on isometric exercise capacity and energy metabolism of the quadriceps muscle in man. Scand J Clin Lab Invest 1975; 35: 87-95 191. Trump ME, Heigenhauser GJ, Putman CT, et al. Importance of muscle phosphocreatine during intermittent maximal cycling. J Appl Physiol 1996; 80 (5): 1574-80 192. Wiroth JB, Bermon S, Andrei S, et al. Effects of oral creatine supplementation on maximal pedalling performance in older adults. Eur J Appl Physiol 2001; 84 (6): 533-9 193. Yquel RJ, Arsac LM, Thiaudiere E, et al. Effect of creatine supplementation on phosphocreatine resynthesis, inorganic phosphate accumulation and pH during intermittent maximal exercise. J Sports Sci 2002; 20 (5): 427-37 194. Bishop D, Edge J. Determinants of repeated-sprint ability in females matched for single-sprint performance. Eur J Appl Physiol 2006; 97 (4): 373-9 195. Edge J, Bishop D, Hill-Haas S, et al. Comparison of muscle buffer capacity and repeated-sprint ability of untrained, endurance-trained and team-sport athletes. Eur J Appl Physiol 2006; 96 (3): 225-34 196. Mohr M, Krustrup P, Nielsen JJ, et al. Effect of two different intense training regimens on skeletal muscle ion transport proteins and fatigue development. Am J Physiol Regul Integr Comp Physiol 2007; 292 (4): R1594-602 197. Bishop D, Edge J, Davis C, et al. Induced metabolic alkalosis affects muscle metabolism and repeated-sprint ability. Med Sci Sports Exerc 2004; 36 (5): 807-13 198. Gaitanos G, Nevill M, Brooks S, et al. Repeated bouts of sprint running after induced alkalosis. J Sports Sci 1991; 9: 355-70 199. Street D, Nielsen JJ, Bangsbo J, et al. Metabolic alkalosis reduces exercise-induced acidosis and potassium accumulation in human skeletal muscle interstitium. J Physiol 2005; 566 (Pt 2): 481-9 200. Sejersted OM, Sjogaard G. Dynamics and consequences of potassium shifts in skeletal muscle and heart during exercise. Physiol Rev 2000; 80 (4): 1411-81 201. McKenna MJ, Heigenhauser GJ, McKelvie RS, et al. Sprint training enhances ionic regulation during intense exercise in men. J Physiol 1997; 501 (Pt 3): 687-702 202. Nielsen JJ, Mohr M, Klarskov C, et al. Effects of highintensity intermittent training on potassium kinetics and performance in human skeletal muscle. J Physiol 2004; 554 (Pt 3): 857-70 203. Billaut F, Basset FA. Effect of different recovery patterns on repeated-sprint ability and neuromuscular responses. J Sports Sci 2007; 25 (8): 905-13
ª 2009 Adis Data Information BV. All rights reserved.
Billaut & Bishop
204. Billaut F, Basset FA, Falgairette G. Muscle coordination changes during intermittent cycling sprints. Neurosci Lett 2005; 380: 265-9 205. Hautier C, Arsac L, Deghdegh K, et al. Influence of fatigue on EMG/force ratio and cocontraction in cycling. Med Sci Sports Exerc 2000; 32 (4): 839-43 206. Billaut F, Basset FA, Giacomoni M, et al. Effect of highintensity intermittent cycling sprints on neuromuscular activity. Int J Sports Med 2006; 27: 25-30 207. Drust B, Rasmussen P, Mohr M, et al. Elevations in core and muscle temperature impairs repeated sprint performance. Acta Physiol Scand 2005; 183: 181-90 208. Mendez-Villanueva A, Hamer P, Bishop D. Physical fitness and performance: fatigue responses during repeated sprints matched for initial mechanical output. Med Sci Sports Exerc 2007; 39 (12): 2219-25 209. Racinais S, Bishop D, Denis R, et al. Muscle deoxygenation and neural drive to the muscle during repeated sprint cycling. Med Sci Sports Exerc 2007; 39 (2): 268-74 210. Balsom PD, Gaitanos GC, Soderlund K, et al. Highintensity exercise and muscle glycogen availability in humans. Acta Physiol Scand 1999; 165 (4): 337-45 211. Bishop D, Spencer M. Determinants of repeated-sprint ability in well-trained team-sport athletes and endurancetrained athletes. J Sports Med Phys Fitness 2004; 44: 1-7 212. Blonc S, Casas H, Duche´ P, et al. Effect of recovery duration on the force-velocity relationship. Int J Sports Med 1998; 19: 272-6 213. Chamari K, Ahmaidi S, Fabre C, et al. Pulmonary gas exchange and ventilatory responses to brief intense intermittent exercise in young trained and untrained adults. Eur J Appl Physiol Occup Physiol 1995; 70 (5): 442-50 214. Hamilton A, Nevill M, Brooks S, et al. Physiological responses to maximal intermittent exercise: differences between endurance-trained runners and game players. J Sports Sci 1991; 9: 371-82 215. Wadley G, Le Rossignol P. The relationship between repeated sprint ability and the aerobic and anaerobic energy systems. J Sci Med Sport 1998; 1 (2): 100-10 216. Tomlin DL, Wenger HA. The relationship between aerobic fitness and recovery from high intensity intermittent exercise. Sports Med 2001; 31 (1): 1-11
Correspondence: Dr Franc¸ois Billaut, Department of Kinesiology, University of Lethbridge, 4401 University Drive, Lethbridge, AB, TIK 3M4, Canada. E-mail:
[email protected]
Sports Med 2009; 39 (4)
Sports Med 2009; 39 (4): 279-294 0112-1642/09/0004-0279/$49.95/0
REVIEW ARTICLE
ª 2009 Adis Data Information BV. All rights reserved.
Physical Activity as a Predictor of Adolescent Body Fatness A Systematic Review Felipe Fossati Reichert,1,2 Ana Maria Baptista Menezes,1 Jonathan C.K. Wells,3 Samuel Carvalho Dumith1 and Pedro Curi Hallal1 1 Postgraduate Program in Epidemiology, Federal University of Pelotas, Pelotas, Brazil 2 Centre of Physical Education and Sports, Department of Physical Education, State University of Londrina, Londrina, Brazil 3 Childhood Nutrition Research Centre, Institute of Child Health, London, UK
Contents Abstract. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1. Methods. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2. Results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.1 Measuring the Exposure: Physical Activity. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.1.1 Experimental and Quasi-Experimental Studies . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.1.2 Observational Studies . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.2 Measuring the Outcome: Adiposity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.2.1 Experimental and Quasi-Experimental Studies . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.2.2 Observational Studies . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.3 Effects of Physical Activity on Adiposity Measures . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.3.1 Experimental and Quasi-Experimental Studies . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.3.2 Observational Studies . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.4 Prevention versus Treatment Effects. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.4.1 Experimental and Quasi-Experimental Studies . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.4.2 Observational Studies . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.5 Measuring Potential Confounders . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3. Discussion. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4. Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Abstract
279 281 282 286 286 287 287 287 287 287 287 288 288 288 288 288 289 292
Adolescent obesity has increased dramatically in several countries in recent decades; however, the contribution of physical activity level to adolescent adiposity requires clarification. This article investigates the effect of physical activity on subsequent levels of adiposity in adolescence. The methodological aspects of the studies included in this article, particularly in terms of measurement accuracy for both exposure (physical activity) and outcome (adiposity) variables, are also evaluated. Systematic searches of the literature were undertaken using online databases, including PubMed/MEDLINE, examination of citations and contacting of authors. The online databases were searched from their earliest records until 2007. Only longitudinal studies with 50 or more adolescents were included. Two independent reviewers assessed
Reichert et al.
280
the quality of the studies using the Downs and Black checklist. Thirteen observational, five experimental and six quasi-experimental studies (without a control group) were identified. Almost all studies were carried out in highincome settings and showed protective effects of physical activity for both prevention and treatment of adolescent obesity. However, experimental studies undertaken with obese adolescents at baseline usually combined physical activity with dietary changes, making it difficult to assess the effect of physical activity itself on the treatment of obesity. Physical activity estimated from questionnaires and body mass index (BMI) were the most frequently used measures. Despite the feasibility of using these approaches in epidemiological studies, significant limitations are evident. Questionnaires are subjective and adolescents may not report physical activity level accurately. Furthermore, BMI is not an accurate measure of fatness for adolescents, as it is also associated with lean mass, hence bias may arise from its longitudinal association with physical activity level. Despite the majority of studies reviewed showing protective effects of physical activity on adiposity, particularly in individuals who are obese at baseline, the current literature on this issue is sparse and several methodological drawbacks are evident. The main limitations relate to a lack of validity in the measurements of both physical activity and body composition. Further studies are needed in order to generate evidence-based recommendations for the quantity and quality of adolescent physical activity required to prevent or treat adolescent obesity.
Adolescent obesity has increased dramatically in several countries in recent decades.[1] Currently, high rates of excessive bodyweight are observed not only in developed societies but also in low- and middle-income countries. For example, in Brazil, the prevalence of overweight more than tripled in boys aged 10–19 years (from 2.6% to 11.8%) and more than doubled in girls (from 5.8% to 15.3%) from 1975 to 1997.[2] Obesity in adolescence is associated with a broad range of adverse health effects. First, adolescent obesity has been shown to track strongly into adulthood, and the health consequences of adulthood obesity are well established.[3] More recently, studies have demonstrated that some diseases, including elevated blood pressure, type 2 diabetes mellitus, asthma and sleep disorders, are more frequent among obese than non-obese adolescents.[4-7] Obesity is also associated with psychological disorders.[8] Thus, studies investigating the factors that play a role in the prevention or treatment of adolescent obesity are warranted. The rapid change in the prevalence of obesity worldwide suggests that factors other than genes play an important role in the aetiology of obesity, ª 2009 Adis Data Information BV. All rights reserved.
although an interaction between these factors may occur. Fat accumulation is ultimately the result of chronic positive energy balance (energy intake > energy expenditure). In this context, physical activity, which accounts for a large part of total energy expenditure, is predicted to influence adiposity. Several cross-sectional studies have addressed this issue; however, their results are conflicting.[9-12] Such lack of consistency may arise because the association between physical activity and adiposity is highly susceptible to reverse causality (adolescents may change their physical activity level depending on the degree of adiposity). Theoretically, longitudinal studies might also be affected by reverse causality, but this bias is much less likely in randomized controlled trials. Thus, longitudinal studies are more appropriate to investigate this issue. Nonetheless, longitudinal studies performed during the period of adolescence (age 10–19 years) are rare and very heterogeneous concerning many aspects that may likewise lead to conflicting findings. Few systematic reviews on the association between physical activity and adiposity in youth Sports Med 2009; 39 (4)
Adolescent Physical Activity and Body Fatness
have been carried out and, overall, their results indicate that increased physical activity may play a role in both prevention and treatment of obesity. Not only did some reviews include only experimental studies[13,14] and one included only observational studies,[15] none of them focused exclusively on adolescents. Physical activity can be subdivided into sleep, sedentary behaviours and motion behaviours. We carried out a systematic review of the literature with the primary aim of investigating the effect of physical activity, specifically the motion aspect, on subsequent levels of adiposity in adolescents. Our review included both observational and experimental studies. We also evaluated methodological aspects of the studies included in the review, particularly in terms of measurement accuracy of both exposure (physical activity) and outcome (adiposity) variables. This issue was not addressed in any previous review, but is critical for interpreting the results of a collation of studies using heterogeneous methods. 1. Methods In July 2007, the following electronic databases were searched from the earliest record: MEDLINE, SportDiscus, SCIELO, BioMed Central and PsycINFO. The reference lists from identified articles were searched manually. The first author of the articles included was contacted and questioned about other published or unpublished data. The following keywords were used: ‘abdominal fat’, ‘abdominal obesity’, ‘adiposity’, ‘body composition’, ‘body fat’, ‘body fat distribution’, ‘body mass index’, ‘central adiposity’, ‘central fat’, ‘central fatness’, ‘central obesity’, ‘centrally distributed fat’, ‘centrally distributed obesity’, ‘fat’, ‘fat patterning’, ‘fatness’, ‘metabolic syndrome’, ‘metabolic syndrome x’, ‘obese’, ‘obesity’, ‘syndrome x’, ‘truncal fat’, ‘truncal obesity’, ‘trunk adiposity’, ‘trunk fat’ and ‘trunk obesity’. These keywords were combined with ‘exercise’, ‘inactivity’, ‘motor activity’, ‘physical activity’, ‘physical exercise’ or ‘sports’. Given that the aim of the present article was to assess the role of physical activity during adolesª 2009 Adis Data Information BV. All rights reserved.
281
cence on subsequent levels of adiposity, crosssectional studies were excluded because of their inability to establish temporality. The following keywords were thus used: ‘clinical trial’, ‘cohort’, ‘experimental’, ‘experimental design’, ‘follow-up’, ‘intervention’, ‘intervention studies’, ‘longitudinal’, ‘panel’, ‘prospective’ and ‘trial’. The keywords ‘adolescence’, ‘adolescents’, ‘teenager’ and ‘youth’ were also used in the literature search. Studies were considered if the outcome variable (adiposity) was collected when subjects were aged 10–19 years, even if the exposure (physical activity) had been measured before 10 years of age. This strategy was adopted since too few studies were carried out exclusively during adolescence. However, studies were only considered if most of the ages of the participants were within the adolescence period. Studies with fewer than 50 subjects in the sample were not considered because of the low statistical power associated with small samples. The search resulted initially in 571 articles. After reading the titles and abstracts, 43 articles were selected, and after reading the full texts, 19 articles were included in the review. A further five articles were selected through the reference lists of these articles, and contact with experts in the area. Two independent reviewers assessed the quality of the studies using the Downs and Black checklist.[16] In case of eventual differences between the two reviewers, the articles were re-assessed until both referees agreed with the evaluation. Because the original Downs and Black checklist is applicable to experimental designs, a modified version of the scale[17] was used to assess the quality of observational studies (table I). Therefore, questions 8, 13, 23 and 24 of this instrument were not considered for observational studies. All questions were coded as 0 (representing poor quality) or 1, with the exception of question 5, which was coded with 0, 1 or 2. Furthermore, question 27, originally coded from 0 to 5, was dichotomized into 0 or 1 (code 1 was attributed to studies that mentioned a statistical power ‡80%). The final scale ranged from 0 (poorest quality) to 24 (best quality) points for observational studies or 28 points for experimental studies. Sports Med 2009; 39 (4)
Reichert et al.
282
Table I. Quality of the studies, as defined by the Downs and Black modified scale[16] Criteria
No. of articles adequate
inadequate
1. Is the hypothesis/aim/objective of the study clearly described?
24
2. Are the main outcomes to be measured clearly described in the Introduction or Methods section?
24
0
3. Are the characteristics of the patients included in the study clearly described?
24
0
4. Are the interventions of interest clearly described?
24
0
6
18
6. Are the main findings of the study clearly described?
24
0
7. Does the study provide estimates of the random variability in the data for the main outcomes?
24
0
9. Have the characteristics of patients lost to follow-up been described?
11
13
10. Have actual probability values been reported (e.g. 0.035 rather than <0.05) for the main outcomes except where the probability value is <0.001?
17
7
11. Were the subjects asked to participate in the study representative of the entire population from which they were recruited?
8
16
10
14
14. Was an attempt made to blind study subjects to the intervention they have received?b
4
20
15. Was an attempt made to blind those measuring the main outcomes of the intervention?b
4
20
5. Are the distributions of principal confounders in each group of subjects to be compared clearly described?
0
8. Have all important adverse events that may be a consequence of the intervention been reported?a
12. Were those subjects who were prepared to participate representative of the entire population from which they were recruited? 13. Were the staff, places and facilities where the patients were treated representative of the treatment the majority of patients receive?a
16. If any of the results of the study were based on ‘data dredging’, was this made clear?
22
2
17. In trials and cohort studies, do the analyses adjust for different lengths of follow-up of patients or, in case-control studies, is the time period between the intervention and outcome the same for cases and controls?
20
4
18. Were the statistical tests used to assess the main outcomes appropriate?
21
3
19. Was compliance with the intervention/s reliable?b
11
13
20. Were the main outcome measures used accurate (valid and reliable)?
5
19
21. Were the patients in different intervention groups (trials and cohort studies) or were the cases and controls (case-control studies) recruited from the same population?
21
3
22. Were study subjects in different intervention groups (trials and cohort studies) or were the cases and controls (case-control studies) recruited over the same period of time?
19
5
18
6
23. Were study subjects randomized to intervention groups?a 24. Was the randomized intervention assignment concealed from both patients and healthcare staff until recruitment was complete and irrevocable?a 25. Was there adequate adjustment for confounding in the analyses from which the main findings were drawn? 26. Were losses of patients to follow-up taken into account?
9
15
27. Did the study have sufficient power to detect a clinically important effect where the probability value for a difference being due to chance is <5%?
3
21
a
These questions are specific for experimental studies and, therefore, were not rated in the table.
b
Observational studies were coded as 0 (‘unable to determine’).
2. Results Table II presents a description and summary of the studies included in the review. Studies are presented according to the design and in alphabetical order of the first author’s name. ª 2009 Adis Data Information BV. All rights reserved.
A total of five experimental,[18-22] six quasiexperimental (without control group),[23-28] and 13 observational (cohort) studies[29-41] were found and fulfilled the inclusion criteria. Only five studies exclusively included adolescents (individuals aged 10–19 years) and all but one[35] study were Sports Med 2009; 39 (4)
Study (country)
Sample size
Follow-up duration
Weight status of the sample
Definition of the exposure
Definition of the outcome
Main results
Eliakim et al.[18] (Israel)
177 individuals aged 6–16 y
3 mo
Obese
Dietary, behaviour and exercise intervention
BMI
BMI decreased in 74% of the individuals (from 26.1 – 0.3 to 25.4 – 0.3 kg/m2)
Gortmaker et al.[22] (US)
1295 individuals
21 mo
General population
School-based intervention aimed to change PA and dietary habits
Combination of BMI and triciptal skinfold
Intervention was successful among girls, but not boys
Gutin et al.[19] (US)
80 adolescents aged 13–16 y
8 mo
Obese
Lifestyle intervention plus either moderate or vigorous PA
Total (DEXA) and visceral adiposity (magnetic resonance)
High- and moderateintensity PA were similarly effective in reducing visceral and total-body adiposity
McMurray et al.[20] (US)
1140 adolescents (630 females) aged 11–14 y
8 wk
Normal
Aerobic exercise programme and education in schools
Sum of skinfold thickness
Exercise group had smaller gains in skinfold thickness than control group
Savoye et al.[21] (US)
209 individuals aged 8–16 y
1y
Overweight
Weight management family-based programme including exercise, nutrition and behaviour modification
Change in weight, BMI and body fat estimated from Tanita, TBF 300
Individuals in the intervention group presented better indicators in all outcomes
Dao et al.[23] (France)
55 individuals aged 9–17 y
6–12 mo
Obese
Dietary and physical activity programme
Total and regional body composition determined by DEXA
Body fat decreased in both sexes and steepest declines were observed in the trunk
Reinehr et al.[24] (Germany)
75 individuals aged 7–15 y
1y
Obese
Intervention consisted of physical exercise, nutritional course and behaviour therapy for participants and their parents
BMI
Participation in exercise groups was associated with a decrease in SD scores of BMI
Experimental studies
Adolescent Physical Activity and Body Fatness
ª 2009 Adis Data Information BV. All rights reserved.
Table II. Characteristics of the studies included in the review according to design, first author’s name, country, sample size, follow-up duration, weight status of the sample, definition of exposure and outcome, and main results
Quasi-experimental studies
283
Sports Med 2009; 39 (4)
Continued next page
284
ª 2009 Adis Data Information BV. All rights reserved.
Table II. Contd Sample size
Follow-up duration
Weight status of the sample
Definition of the exposure
Definition of the outcome
Main results
Reinehr et al.[25] (Germany)
170 individuals (mean age 10.5 y)
1y
Obese
1 y of physical exercises and 3 mo of nutrition education
SD score in BMI in the third year after intervention
66% of individuals presented a reduction of SD score in BMI 3 y after the end of intervention
Sothern et al.[27] (US)
87 individuals (39 boys) aged 7–17 y
1y
Obese
Intervention consisting of dietary restrictions, moderate-intensity PA and behaviour modification sessions
Percentage of body fat
All individuals presented better body composition indicators
Sothern et al.[26] (US)
56 individuals aged 7–17 y
1y
Obese
Dietary restrictions and physical activities
Variation in bodyweight and percentage of body fat, estimated by skinfold thickness
Individuals had a significant decline in all outcomes after 10 wk of the programme, which was maintained at 1-y follow-up
Wong et al.[28] (Singapore)
112 adolescents
2y
Obese
Weight control programme
BMI
Long-term weight loss was associated with increased PA after intervention
Berkey et al.[29] (US)
6149 girls and 4620 boys aged 9–14 y at baseline
1y
General population
Self-reported PA estimated by questionnaire
1-y change in BMI
PA was inversely associated with large increases in BMI in girls only
Berkey et al.[30] (US)
6767 girls and 5120 boys aged 10–15 y
1y
General population
Self-reported PA estimated by questionnaire
1-y change in BMI
Elgar et al.[31] (UK)
355 adolescents (mean age 12.3 y at baseline)
4y
General population
Self-reported PA estimated by the Health Behaviour of School-aged Children Questionnaire
BMI
PA effects were sexdependent and stronger in overweight than normal weight adolescents. The effect of PA on BMI was stronger than the effect of sedentary activities Baseline PAL was associated with BMI change, but no BMI at follow-up
Observational studies
Sports Med 2009; 39 (4)
Continued next page
Reichert et al.
Study (country)
Sample size
Follow-up duration
Weight status of the sample
Definition of the exposure
Definition of the outcome
Main results
Heelan et al.[32] (US)
320 individuals aged 10.2 – 0.7 y
6 mo
General population
Active commuting to and from school estimated from the self-administered PA checklist
BMI and body fat determined by skinfold thickness
Significant positive correlation between active commuting and overweight was observed
Kettaneh et al.[33] (France)
436 individuals aged 8–18 y
2y
General population
Self-reported and parent-assisted PA estimated from the Kriska’s modifiable activity questionnaire
BMI, sum of skinfold thickness and waist circumference
A decline in PAL during the follow-up period was associated with lower adiposity in girls, but not boys
Kimm et al.[34] (US)
2379 girls followed up from age 9/10–18/19 y
9y
General population
Self-reported PA estimated from the Habitual Activity Questionnaires
BMI and sum of skinfold thickness
Active girls had smaller gains in both outcomes than inactive ones
Mo-suwan et al.[35] (Thailand)
2252 schoolchildren aged 5–16 y at baseline
5y
General population
Exercise level compared with other individuals of the same age, as reported by the parents
BMI
Lower level of exercise was associated with increases in BMI
Mundt et al.[36] (Canada)
208 individuals aged 8–19 y
Maximum of 7 y; median of 5 y
General population
Self-reported PA estimated from the PA Questionnaire for Children or Adolescents
Fat mass determined by DEXA
PAL was negatively associated with fat mass accumulation in boys, but not girls
Must et al.[37] (US)
173 premenarcheal girls aged 8–12 y followed up until 4 y postmenarche
7.5 y (average)
General population
Self-reported PA estimated from questionnaire
Percentage of body fat estimated from bioelectrical impedance and BMI
PA was negatively associated with percentage of body fat only among those who had at least one parent overweight
O’Loughlin et al.[38] (Canada)
2951 schoolchildren aged 9–12 y at baseline
1 or 2 y
General population
Self-reported PA estimated from questionnaire
BMI
Highest decile of change in BMI was more frequent among those with low levels of physical activity Continued next page
285
Sports Med 2009; 39 (4)
Study (country)
Adolescent Physical Activity and Body Fatness
ª 2009 Adis Data Information BV. All rights reserved.
Table II. Contd
Reichert et al.
Higher levels of physical education classes were associated with lower gains in adiposity in boys
No association between past year PAL and BMI increases was observed
2727 students aged 11–12 y at baseline Wardle et al.[41] (UK)
ª 2009 Adis Data Information BV. All rights reserved.
In the 24 studies revised here, physical activity was both measured and promoted in the intervention studies in a variety of different ways. BMI = body mass index; DEXA = dual-energy x-ray absorptiometry; PA = physical activity; PAL = PA level; SD = standard deviation.
BMI and waist circumference Self-reported and teacher-reported number of physical education classes per wk General population
496 girls aged 11–15 y Stice et al.[40] (US)
5y
BMI Self-reported PA estimated from the Past Year Leisure Physical Activity Scale General population
1083 students from the fourth grade of seven elementary schools Rosenberg et al.[39] (US)
4y
Active commuting to school over 2 y was not associated with BMI change or overweight status BMI and skinfold thickness
Sample size
2y
General population
Self-reported active commuting estimated from questionnaire
Main results
undertaken in high-income countries. Sample sizes ranged from 55[23] to 11 887,[30] and follow-up duration was as short as 3 months[18] and as long as 9 years.[34] Table I describes the methodological quality of the articles according to the Downs and Black checklist. Figure 1 presents potential confounders considered by the studies reviewed. 2.1 Measuring the Exposure: Physical Activity
Study (country)
Table II. Contd
Follow-up duration
Weight status of the sample
Definition of the exposure
Definition of the outcome
286
2.1.1 Experimental and Quasi-Experimental Studies
All experimental and quasi-experimental studies included components other than a physical activity programme in the intervention (table II). For example, Gutin et al.[19] assigned participants to one of the following three groups: (i) 1 hour of lifestyle education every 2 weeks (which served as a control condition because it was offered to all groups); (ii) lifestyle intervention plus moderate physical . activities (55–60% of peak oxygen uptake [VO2peak]); and (iii) lifestyle intervention plus high-intensity physical activities . (75–80% VO2peak). Energy expenditure for each session was held at approximately 250 kcal/ session; therefore, the training duration was related to physical activity intensity. However, individuals of the high-intensity group performed the exercises at a heart rate (154 beats/min) significantly lower than that prescribed (167 beats/min), and the attendance rate was low for both physical training groups (~54%). Savoye et al.[21] randomly assigned overweight individuals to either a control group (receiving traditional clinical weight management counselling) or a 12-month weight management group, which received a family-based programme involving exercise, nutrition and behaviour modification. The exercises consisted of aerobic activities performed at 65–85% of the age-adjusted maximal heart rate for twice a week (50 minutes each session) during the first 6 months and 100 minutes per month for the last 6 months. The intervention proposed by Reinehr et al.[25] was based on 1 year of physical exercise, nutrition Sports Med 2009; 39 (4)
Adolescent Physical Activity and Body Fatness
287
2.2.1 Experimental and Quasi-Experimental Studies
Height Parental obesity-related variables Socioeconomic status Baseline value of the outcome Sexual maturationrelated variables Race/ethnicity Nutrition-related variables Age 0
2
4 6 8 No. of studies
10
12
Fig. 1. Potential confounders considered in the observational studies reviewed.
education and behaviour therapy for children and parents separately (first 3 months) and individual psychological care of the child and its family. Twenty of the 75 participants dropped out during the study, but training session attendance among those who remained in the study was >90%. Sothern et al.[26] also proposed an intervention with three main branches: nutrition, physical exercises and behaviour modification. The exercise was of moderate. intensity (45–55% of maximal oxygen uptake [VO2max]) and the frequency and duration of the sessions varied according to the degree of obesity, although no specific information for these variables was provided. Ninetythree percent of individuals completed the acute phase (10–20 weeks) of intervention and 62.5% completed the 1-year programme. Mean attendance during the acute phase was 91% with 57% for the remaining phase. 2.1.2 Observational Studies
All observational studies estimated physical activity level by questionnaires (table II). Most of the questionnaires were self-reported, with the remainder completed by parents[33,35] or teachers.[41] 2.2 Measuring the Outcome: Adiposity
Adiposity was also measured in a variety of different ways in the 24 different studies. ª 2009 Adis Data Information BV. All rights reserved.
Most of the studies included more than one method to estimate adiposity. Body mass index (BMI) or change in this variable during the follow-up period was the most frequently used outcome.[18,21,24,25,28] One study defined obesity based on both BMI and triciptal skinfold thickness.[22] Percentage of body fat derived by either skinfold thickness[26,27] or bioelectrical impedance[21] was another outcome frequently used in the studies. Dual-energy x-ray absorptiometry (DEXA) was used in two studies[19,23] and MRI in one study.[19] 2.2.2 Observational Studies
Observational and experimental/quasi-experimental studies used similar methods to estimate adiposity. Virtually all observational studies used BMI as a measure of adiposity (table II), with six of the 13 studies using only this method. Skinfold thickness measurements and bioelectrical impedance as well as circumferences were also used in some studies. One study[36] used DEXA. 2.3 Effects of Physical Activity on Adiposity Measures 2.3.1 Experimental and Quasi-Experimental Studies
All studies showed favourable effects of the intervention on adiposity level of adolescents. However, it should be highlighted that all interventions included other exposures besides physical activity (e.g. dietary change). Thus, it is difficult to assess the independent impact of physical activity on adiposity. One school-based intervention decreased obesity only among girls,[22] while the remaining studies found comparable results of the intervention among both sexes. One experimental study measured total body composition by DEXA and visceral adipose tissue and subcutaneous abdominal tissue by MRI.[19] Based on efficacy analyses adjusted for potential confounders and baseline values, the authors showed a significant decline in the following outcomes after 8 months of physical training: visceral adipose tissue (-42.0 – 9.3 cm3), subcutaneous abdominal adipose tissue (-69.7 – 55.9 cm3), Sports Med 2009; 39 (4)
Reichert et al.
288
and percentage of body fat (-3.57 – 0.80%). The efficacy analyses included only those participants who attended at least 40% of intervention sessions. However, the intent-to-treat analyses (i.e. effectiveness analysis) showed no association of the intervention with either of the outcomes. Furthermore, the authors highlighted that there was no evidence that the high-intensity physical training was more efficacious than the moderate-intensity training, although no analyses for this association were shown. 2.3.2 Observational Studies
All but two studies[39,40] showed significant inverse associations of physical activity with body composition or BMI, although the magnitude of the association was markedly different between studies. Some studies suggest that the effect of physical activity on these outcomes depended on sex,[29,36] ethnicity[34] and baseline BMI.[30] For example, Berkey et al.[29] showed that active girls had smaller gains in BMI (-0.0284; p = 0.046) over a 1-year period. Among boys, this difference was not significant, despite similarity in the strength and direction of the effect (coefficient = -0.0261; p = 0.094). In contrast, Mundt et al.[36] showed that physical activity was associated with reduced increments in fat mass (measured by DEXA) in boys but not girls. 2.4 Prevention versus Treatment Effects 2.4.1 Experimental and Quasi-Experimental Studies
The evidence that physical activity prevents adolescence obesity is very limited, since only two experimental studies on normal weight individuals were located[20,22] and one of them found no effect of the intervention on boys.[22] The remaining experimental and quasi-experimental studies were carried out with either overweight or obese individuals at baseline and showed positive effects of the intervention.[18,19,21,23-28] Overall, the results were consistent regardless of the method used to estimate adiposity. However, McMurray et al.[20] found no differences in the BMI changes among four groups studied (educational intervention, exercise intervention, educational plus exercise ª 2009 Adis Data Information BV. All rights reserved.
interventions, and control group). In contrast, the sum of skinfolds increased less in the exercise intervention groups than in the control and education-only groups (p < 0.001). It is very difficult to draw confident conclusions regarding the actual effects of physical activity from these studies because the interventions also focused on other aspects, for example diet. Therefore, although there is some evidence that physical activity is important in the prevention and treatment of adolescence obesity, the real impact of physical activity, as well as the type, frequency and duration that is most beneficial, remains unknown. 2.4.2 Observational Studies
Observational studies demonstrated that physical activity might play a role in the prevention of fat accumulation in normal-weight subjects. For example, Kimm et al.[34] demonstrated a clear dose-response effect of physical activity practice on skinfold thicknesses: girls who were more active from ages 9–19 years had smaller gains in skinfold thicknesses throughout adolescence. However, although most studies estimated physical activity from questionnaires, there are numerous differences between these instruments. Some of them estimated only leisure-time physical activities, while others measured all-domain activities. They also used a variety of cut-off points and units of measures (metabolic equivalents, minutes of physical activity and kilocalories spent per week, etc.). Another pitfall of questionnaires concerns their subjectivity. Therefore, the relative importance of type, frequency and duration of physical activities for preventing or treating obesity is also unknown. 2.5 Measuring Potential Confounders
All observational studies adjusted their analyses for potential confounders. Figure 1 shows the factors most frequently considered as confounders and the number of studies that included them in analyses. The figure does not show sex as a potential confounder, but most of the studies performed analyses stratified by this variable. Confounders not included in the figure, but Sports Med 2009; 39 (4)
Adolescent Physical Activity and Body Fatness
considered by at least one study, were smoking habits,[34] hours of television viewing, number of parents,[31] compensatory behaviours (i.e. vomiting for weight-control purposes, laxative abuse and diuretic abuse), depressive symptoms[40] and school.[38] Table I describes the methodological quality of the articles, as defined by the Downs and Black modified scale. The average quality score of the observational studies was 16.4/24 (standard deviation [SD] 2.1; median 17; range 12–19) and of the experimental studies was 17.2/28 (SD = 2.7; median = 18.0; range 14–21). 3. Discussion This review focused on the role of physical activity (or lack of activity) on subsequent levels of adiposity with the latter, at least, being investigated during adolescence. Despite most of the articles reviewed showing protective effects of physical activity against adiposity, several limitations are evident in the literature. Although the longitudinal relationship between these variables is of most importance, few studies on this subject (particularly experimental studies) could be located. Furthermore, there are virtually no relevant studies from low- or middle-income countries. This result was expected, given the fact that longitudinal studies are very expensive and time consuming. It is plausible that the physiological effects of a specific physical activity will have similar effects on body composition regardless of the population and setting. However, different activities are practised in low- and middle-income country populations compared with those of high-income countries (e.g. leisure-time vs travel or subsistence activity).[42-45] Therefore it is important to assess the effectiveness of different activity patterns on changes in body composition. The only study undertaken outside high-income countries from Thailand[35] showed similar results (i.e. an increase in BMI was associated with lower levels of exercise); however, many populations and settings still remain unexplored. The best study design for testing the hypothesis that physical activity prevents excessive fat accumulation is randomized field trials. ª 2009 Adis Data Information BV. All rights reserved.
289
However, virtually all available data regarding the prevention of fat accumulation through increased physical activity derive from observational studies, and several biases are therefore of concern. For example, high rates of loss of follow-up and refusal to participate are observed in the reviewed literature, which may lead to an overestimation of the actual effects of physical activity. All but two[20,22] of the experimental (randomized intervention) and quasi-experimental studies reviewed were carried out in individuals overweight/obese at baseline. Although all of these studies suggested favourable effects of the intervention on adiposity levels, such studies are very heterogeneous regarding both exposure and outcome, and their results must be interpreted with caution. First, none of the studies verified the effect of physical activity itself on adiposity levels. Interventions usually consisted of a combination of factors such as change in dietary habits, behaviour and physical activity levels.[18,21,23,25,27,28] Thus, the results are likely to be a consequence of an interaction between these variables. Secondly, there were noticeable differences in terms of baseline body composition, duration of the intervention, and number of subjects included in the study, which all may affect the results. An important methodological aspect to be considered is the decision of adjusting or not for the baseline value of the outcome. One should consider that physical activity practice may have different consequences on later body composition depending on the current nutritional status of the individual. In order to address this issue, one of the possibilities is to adjust the association between physical activity and later body composition by baseline values of the outcome variable. An alternative approach is to test interactions between the variables. Both approaches were rarely used in the reviewed studies. Regardless of the design of the study, a further important limitation relates to the measurement of both physical activity and body composition. Despite the recognized importance of these variables for patterns of morbidity and mortality worldwide, accurate determination of physical activity and body composition in large-scale Sports Med 2009; 39 (4)
290
studies remains a challenge. A non-systematic error in the measurement of any of these variables would underestimate the effect of physical activity; however, the effect of a systematic error is unknown. The optimum technique for assessing fat mass is the multi-component model, although deuterium dilution is also considered to have high accuracy.[46] MRI is also considered a gold standard for adipose tissue quantity and distribution, though it should be noted that adipose tissue is not equal to fat mass, hence reducing comparability across studies. DEXA is often described as an accurate objective technique; however, bias is systematically associated with factors such as sex, body size and obesity status.[47,48] In this context, we found that BMI was the most frequently used outcome measure (18 out of 24 studies reviewed). It has been shown that for a given value of BMI, a wide range of fatness is observed in children.[49] The limitation of BMI is much more evident if the aim is to investigate its association with physical activity. Physical activity can increase lean mass as well as decrease fat.[50] However, BMI does not distinguish between these two compartments and, therefore, either stability or even increase in this index may actually correspond to favourable changes in body composition. In fact, the lack of association between physical activity and BMI in boys observed in some studies[29,33] may be related to this issue. Nonetheless, one should note that studies that used BMI and another method to estimate adiposity (i.e. bioelectrical impedance, skinfold thickness or circumferences) usually found similar effects of physical activity regardless of the outcome measurement. For example, Savoye et al.[21] showed an effect of comparable magnitude (p < 0.001) of physical activity on BMI and percentage of body fat derived from a body fat analyser (Tanita, TBF 300). Skinfold thickness was another frequently used outcome measure. This approach at least measures one component of adiposity directly, but may not reflect deeper fat depots,[51] which are most strongly associated with health outcomes.[52] Furthermore, published equations have been shown to be inaccurate. However, the maª 2009 Adis Data Information BV. All rights reserved.
Reichert et al.
jority of adolescent fat mass is subcutaneous rather than intra-abdominal, hence raw skinfold data provide valuable information about energy stores. Raw data expressed as standard deviations are frequently based on reference data that may not be appropriate for contemporary populations;[53,54] however, ranking within the population is likely to be accurate. Some studies, with noticeably lower numbers of participants, have used more sophisticated methods such as DEXA. In addition to the limitations in accuracy discussed above, their results are frequently converted into percentage of body fat. However, percentage of body fat has been criticized because it is both statistically and conceptually problematical.[54-56] For example, percentage fat is the inverse of percentage lean mass, hence the actual body component associated with other variables remains unclear. Proposed alternatives, such as adjusting body fat to body size (as kg/m2, for example), have not yet been associated with physical activity in longitudinal studies; however, in cross-sectional studies, both fat mass and lean mass adjusted for height have been associated with physical activity level.[57] Similar lack of validity is also evident in the measurement of physical activity. The gold standards for estimating physical activity in children and adolescents have been claimed to be direct observation, double-labelled water or indirect calorimetry.[58] None of the studies used any of these methods. Instead, most studies estimated physical activity by questionnaires. Although questionnaires have some advantages over other methods, they are very subjective. Collecting valid data through questionnaires depends on the reliability of the interviewee in reporting accurately the practice of physical activities over a determined period of time. This is particularly problematic with children and adolescents, given their low ability to record their activities. Furthermore, physical activities in these ages are generally characterized by irregular bouts of short duration and varied intensity activities, making it even more difficult to obtain accurate data. An alternative to overcome these shortcomings would be the application of more Sports Med 2009; 39 (4)
Adolescent Physical Activity and Body Fatness
objective measures of physical activity, such as motion sensors to determine physical activity level. Such techniques have been shown to be valid and reliable in young populations, with no evidence of high reactivity.[59] However, our review did not identify any studies using these approaches. Possible reasons for this finding are the high costs associated with the use of such devices in large-scale studies, or the limited feasibility of assessing physical activity over long time periods using these instruments. The costs and the storage capacity of recent accelerometer models have changed favourably and it is possible that in the near future these devices will be used in epidemiological studies. Experimental and quasi-experimental studies usually focused on aerobic activities lasting at least 30 minutes and performed at least twice a week. Gutin et al.[19] verified the effect of different intensities on body composition and concluded that either high- or moderate-intensity activities impacted on body composition similarly. However, these authors highlighted important limitations such as the low rate of attendance at the exercise sessions, and the inability to perform the exercises in the target heart rate zone, which might have compromised the results. In this context, based on our review, the current recommendation of 60 min/day of physical activities practice on most days of the week to prevent/treat adolescent obesity[60] lacks evidence. In fact, this recommendation has been criticized for the same reason in other studies.[61] Some methodological aspects of our review should be highlighted. Several studies on the association between physical activity and body composition were cross-sectional, and thus were not included in this review. Although crosssectional studies are valuable for addressing various research questions, they are unable to establish temporality of the association between the exposure and outcome under study (physical activity and adiposity levels), which was the objective of our review. Therefore, our review only included prospective studies, since they are less likely to be affected by reverse causation than cross-sectional studies. Studies with fewer than 50 participants were excluded due to the lack of ª 2009 Adis Data Information BV. All rights reserved.
291
statistical power associated with such a low number of individuals. Furthermore, we were interested in the effectiveness (i.e. intent-to-treat analysis) rather than efficacy of physical activity, whereas smaller studies usually carry out only efficacy analyses. However, it should be noted that small-sized studies usually allow for more sophisticated measures of both physical activity and adiposity, and their results can represent valuable pilot data requiring confirmation in larger samples. The likelihood of publication bias in our review must be considered. If studies with inconclusive results or indicating an unexpected association (i.e. unfavourable effects of physical activity on adiposity) were not located but do exist, then the beneficial effects of physical activity on adiposity would be overestimated. In order to decrease the likelihood of this bias, several strategies of the literature search were adopted and the authors of the articles included were asked about other studies (either published or unpublished) on the same topic. Finally, we assessed the methodological quality of the studies using a modified version of the Downs and Black Scale.[16] Such a scale has been used in other reviews.[17,62-64] Overall, results from studies with higher scores were similar to those with lower scores. Although it is important to have an indicator of the quality of the studies included in the review, some limitations of the Downs and Black Scale itself must be pointed out. Firstly, some criteria (e.g. numbers 14, 15 and 19) are not applicable to observational studies and were coded as 0 (thus, rated as ‘inadequate’). Secondly, the absence of some items of the scale in an article might be a reflection of factors other than solely poorer methodological quality. For example, it is reasonable to believe that authors from studies with small sample sizes are under greater editorial pressure to discuss the power of analyses than authors from larger studies. In fact, several large studies were rated as inadequate in criteria number 20 because sample size calculations were not presented. Some items of the scale are vague and difficult to evaluate. For example, regarding criteria number 19, there is no widely acceptable rate of desired compliance. Likewise, some studies did Sports Med 2009; 39 (4)
Reichert et al.
292
not clearly indicate the compliance rate with the intervention. Therefore, we decided to rate all studies as adequate for this item, as this is also recommended in the original article reporting the scale, being relevant whenever misclassification error is likely to bias the association to the null.[16]
4.
5.
6.
4. Conclusions This article focused on the longitudinal association between physical activity and adiposity in adolescence. Most studies showed protective effects of physical activity against adiposity, mainly in individuals who were obese at baseline. Nonetheless, few studies, in particular experimental ones, are available and several methodological drawbacks are evident. The main limitations relate to a lack of validity in the measurement of both physical activity and body composition. Thus, based on the current available data, we conclude that the literature offers only limited support for a causal link between physical activity and adiposity in adolescence. Further studies are needed in order to generate evidence-based recommendations for the quantity and quality of adolescent physical activity to prevent and treat adolescent obesity. Acknowledgements Felipe F. Reichert thanks the National Council of Technological and Scientific Development (CNPq) for supporting him with an academic scholarship during the writing of the manuscript. No other sources of funding were used to assist in the preparation of this review. The authors do not have any conflicts of interest that are relevant to the contents of this manuscript.
References 1. Wang Y, Monteiro C, Popkin BM. Trends of obesity and underweight in older children and adolescents in the United States, Brazil, China, and Russia. Am J Clin Nutr 2002 Jun; 75 (6): 971-7 2. da Veiga GV, da Cunha AS, Sichieri R. Trends in overweight among adolescents living in the poorest and richest regions of Brazil. Am J Public Health 2004 Sep; 94 (9): 1544-8 3. Wyatt SB, Winters KP, Dubbert PM. Overweight and obesity: prevalence, consequences, and causes of a growing
ª 2009 Adis Data Information BV. All rights reserved.
7.
8.
9.
10.
11.
12.
13.
14.
15.
16.
17.
18.
public health problem. Am J Med Sci 2006 Apr; 331 (4): 166-74 Thomsen SF, Ulrik CS, Kyvik KO, et al. Association between obesity and asthma in a twin cohort. Allergy 2007 Oct; 62 (10): 1199-204 Smith Jr SC. Multiple risk factors for cardiovascular disease and diabetes mellitus. Am J Med 2007 Mar; 120 (3 Suppl. 1): S3-11 Ko GT, Chan JC, Chan AW, et al. Association between sleeping hours, working hours and obesity in Hong Kong Chinese: the ‘better health for better Hong Kong’ health promotion campaign. Int J Obes 2007 Feb; 31 (2): 254-60 Gold DR, Damokosh AI, Dockery DW, et al. Body-mass index as a predictor of incident asthma in a prospective cohort of children. Pediatr Pulmonol 2003 Dec; 36 (6): 514-21 Scott KM, Bruffaerts R, Simon GE, et al. Obesity and mental disorders in the general population: results from the world mental health surveys. Int J Obes 2008 Jan; 32 (1): 192-200 Janssen I, Katzmarzyk PT, Boyce WF, et al. Overweight and obesity in Canadian adolescents and their associations with dietary habits and physical activity patterns. J Adolesc Health 2004 Nov; 35 (5): 360-7 Levin S, Lowry R, Brown DR, et al. Physical activity and body mass index among US adolescents: youth risk behavior survey, 1999. Arch Pediatr Adolesc Med 2003 Aug; 157 (8): 816-20 Ekelund U, Neovius M, Linne Y, et al. Associations between physical activity and fat mass in adolescents: the Stockholm Weight Development Study. Am J Clin Nutr 2005 Feb; 81 (2): 355-60 Gazzaniga JM, Burns TL. Relationship between diet composition and body fatness, with adjustment for resting energy expenditure and physical activity, in preadolescent children. Am J Clin Nutr 1993 Jul; 58 (1): 21-8 Connelly JB, Duaso MJ, Butler G. A systematic review of controlled trials of interventions to prevent childhood obesity and overweight: a realistic synthesis of the evidence. Public Health 2007 Jul; 121 (7): 510-7 Doak CM, Visscher TL, Renders CM, et al. The prevention of overweight and obesity in children and adolescents: a review of interventions and programmes. Obes Rev 2006 Feb; 7 (1): 111-36 Must A, Tybor DJ. Physical activity and sedentary behavior: a review of longitudinal studies of weight and adiposity in youth. Int J Obes 2005 Sep; 29 Suppl. 2: S84-96 Downs SH, Black N. The feasibility of creating a checklist for the assessment of the methodological quality both of randomised and non-randomised studies of health care interventions. J Epidemiol Commun Health 1998 Jun; 52 (6): 377-84 Monteiro PO, Victora CG. Rapid growth in infancy and childhood and obesity in later life: a systematic review. Obes Rev 2005 May; 6 (2): 143-54 Eliakim A, Kaven G, Berger I, et al. The effect of a combined intervention on body mass index and fitness in obese children and adolescents: a clinical experience. Eur J Pediatr 2002 Aug; 161 (8): 449-54
Sports Med 2009; 39 (4)
Adolescent Physical Activity and Body Fatness
19. Gutin B, Barbeau P, Owens S, et al. Effects of exercise intensity on cardiovascular fitness, total body composition, and visceral adiposity of obese adolescents. Am J Clin Nutr 2002 May; 75 (5): 818-26 20. McMurray RG, Harrell JS, Bangdiwala SI, et al. A schoolbased intervention can reduce body fat and blood pressure in young adolescents. J Adolesc Health 2002 Aug; 31 (2): 125-32 21. Savoye M, Shaw M, Dziura J, et al. Effects of a weight management program on body composition and metabolic parameters in overweight children: a randomized controlled trial. JAMA 2007 Jun 27; 297 (24): 2697-704 22. Gortmaker SL, Peterson K, Wiecha J, et al. Reducing obesity via a school-based interdisciplinary intervention among youth: Planet Health. Arch Pediatr Adolesc Med 1999 Apr; 153 (4): 409-18 23. Dao HH, Frelut ML, Oberlin F, et al. Effects of a multidisciplinary weight loss intervention on body composition in obese adolescents. Int J Obes Relat Metab Disord 2004 Feb; 28 (2): 290-9 24. Reinehr T, Brylak K, Alexy U, et al. Predictors to success in outpatient training in obese children and adolescents. Int J Obes Relat Metab Disord 2003 Sep; 27 (9): 1087-92 25. Reinehr T, Temmesfeld M, Kersting M, et al. Four-year follow-up of children and adolescents participating in an obesity intervention program. Int J Obesity (2005) 2007 Jul; 31 (7): 1074-7 26. Sothern, Udall Jr JN, Suskind RM, et al. Weight loss and growth velocity in obese children after very low calorie diet, exercise, and behavior modification. Acta Paediatr 2000 Sep; 89 (9): 1036-43 27. Sothern MS, von Almen TK, Schumacher HD, et al. A multidisciplinary approach to the treatment of childhood obesity. Del Med J 1999 Jun; 71 (6): 255-61 28. Wong ML, Koh D, Lee MH, et al. Two-year follow-up of a behavioural weight control programme for adolescents in Singapore: predictors of long-term weight loss. Ann Acad Med Singapore 1997 Mar; 26 (2): 147-53 29. Berkey CS, Rockett HR, Field AE, et al. Activity, dietary intake, and weight changes in a longitudinal study of preadolescent and adolescent boys and girls. Pediatrics 2000 Apr; 105 (4): E56. doi:10.1542/peds.105.4.e56 30. Berkey CS, Rockett HR, Gillman MW, et al. One-year changes in activity and in inactivity among 10- to 15-year-old boys and girls: relationship to change in body mass index. Pediatrics 2003 Apr; 111 (4 Pt 1): 836-43 31. Elgar FJ, Roberts C, Moore L, et al. Sedentary behaviour, physical activity and weight problems in adolescents in Wales. Public Health 2005 Jun; 119 (6): 518-24 32. Heelan KA, Donnelly JE, Jacobsen DJ, et al. Active commuting to and from school and BMI in elementary school children: preliminary data. Child Care Health Dev 2005 May; 31 (3): 341-9 33. Kettaneh A, Oppert JM, Heude B, et al. Changes in physical activity explain paradoxical relationship between baseline physical activity and adiposity changes in adolescent girls: the FLVS II study. Int J Obes 2005 Jun; 29 (6): 586-93 34. Kimm SY, Glynn NW, Obarzanek E, et al. Relation between the changes in physical activity and body-mass index
ª 2009 Adis Data Information BV. All rights reserved.
293
35.
36.
37.
38.
39.
40.
41.
42.
43.
44.
45.
46.
47.
48.
49.
during adolescence: a multicentre longitudinal study. Lancet 2005 Jul 23-29; 366 (9482): 301-7 Mo-suwan L, Tongkumchum P, Puetpaiboon A. Determinants of overweight tracking from childhood to adolescence: a 5 y follow-up study of Hat Yai schoolchildren. Int J Obes Relat Metab Disord 2000 Dec; 24 (12): 1642-7 Mundt CA, Baxter-Jones AD, Whiting SJ, et al. Relationships of activity and sugar drink intake on fat mass development in youths. Med Sci Sports Exerc 2006 Jul; 38 (7): 1245-54 Must A, Bandini LG, Tybor DJ, et al. Activity, inactivity, and screen time in relation to weight and fatness over adolescence in girls. Obesity (Silver Spring) 2007 Jul; 15 (7): 1774-81 O’Loughlin J, Gray-Donald K, Paradis G, et al. Oneand two-year predictors of excess weight gain among elementary schoolchildren in multiethnic, low-income, inner-city neighborhoods. Am J Epidemiol 2000 Oct 15; 152 (8): 739-46 Rosenberg DE, Sallis JF, Conway TL, et al. Active transportation to school over 2 years in relation to weight status and physical activity. Obesity (Silver Spring) 2006 Oct; 14 (10): 1771-6 Stice E, Presnell K, Shaw H, et al. Psychological and behavioral risk factors for obesity onset in adolescent girls: a prospective study. J Consult Clin Psychol 2005 Apr; 73 (2): 195-202 Wardle J, Brodersen NH, Boniface D. School-based physical activity and changes in adiposity. Int J Obes 2007 Sep; 31 (9): 1464-8 Rafferty AP, Reeves MJ, McGee HB, et al. Physical activity patterns among walkers and compliance with public health recommendations. Med Sci Sports Exerc 2002 Aug; 34 (8): 1255-61 Hallal PC, Azevedo MR, Reichert FF, et al. Who, when, and how much? Epidemiology of walking in a middleincome country. Am J Prev Med 2005 Feb; 28 (2): 156-61 Salles-Costa R, Heilborn ML, Werneck GL, et al. Gender and leisure-time physical activity [in Portuguese]. Cad Saude Publica 2003; 19 Suppl. 2: S325-33 Gal DL, Santos AC, Barros H. Leisure-time versus full-day energy expenditure: a cross-sectional study of sedentarism in a Portuguese urban population. BMC Public Health 2005 Feb 15; 5: 16. doi:10.1186/1471-2458-5-16 Wells JC, Fuller NJ, Dewit O, et al. Four-component model of body composition in children: density and hydration of fat-free mass and comparison with simpler models. Am J Clin Nutr 1999 May; 69 (5): 904-12 Williams JE, Wells JC, Wilson CM, et al. Evaluation of lunar prodigy dual-energy x-ray absorptiometry for assessing body composition in healthy persons and patients by comparison with the criterion 4-component model. Am J Clin Nutr 2006 May; 83 (5): 1047-54 Wong WW, Hergenroeder AC, Stuff JE, et al. Evaluating body fat in girls and female adolescents: advantages and disadvantages of dual-energy x-ray absorptiometry. Am J Clin Nutr 2002 Aug; 76 (2): 384-9 Wells JC. A Hattori chart analysis of body mass index in infants and children. Int J Obes Relat Metab Disord 2000 Mar; 24 (3): 325-9
Sports Med 2009; 39 (4)
294
50. Torun B, Viteri FE. Influence of exercise on linear growth. Eur J Clin Nutr 1994 Feb; 48 Suppl. 1: S186-S189 51. Lara Fernandez A, Escolar Castellon JL, Aguilar Cuevas R, et al. Obesity and distribution of body fat: correlation between anthropometric and tomographic data on areas at the abdominal level [in Spanish]. Rev Clin Esp 1996 Jul; 196 (7): 437-45 52. Rosin BL. The progression of cardiovascular risk to cardiovascular disease. Rev Cardiovasc Med 2007; 8 Suppl. 4: S3-8 53. Wells JC, Coward WA, Cole TJ, et al. The contribution of fat and fat-free tissue to body mass index in contemporary children and the reference child. Int J Obes Relat Metab Disord 2002 Oct; 26 (10): 1323-8 54. Wells JC. A critique of the expression of paediatric body composition data. Arch Dis Child 2001 Jul; 85 (1): 67-72 55. Wells JC, Victora CG. Indices of whole-body and central adiposity for evaluating the metabolic load of obesity. Int J Obes 2005 May; 29 (5): 483-9 56. Rennie KL, Wells JC, McCaffrey TA, et al. The effect of physical activity on body fatness in children and adolescents. Proc Nutr Soc 2006 Nov; 65 (4): 393-402 57. Ness AR, Leary SD, Mattocks C, et al. Objectively measured physical activity and fat mass in a large cohort of children. Plos Med 2007 Mar; 4 (3): e97. doi:10.1371/journal. pmed.0040097 58. Sirard JR, Pate RR. Physical activity assessment in children and adolescents. Sports Med 2001; 31 (6): 439-54
ª 2009 Adis Data Information BV. All rights reserved.
Reichert et al.
59. Rowlands AV. Accelerometer assessment of physical activity in children: an update. Pediatr Exerc Sci 2007; 19: 252-66 60. Biddle S, Cavill N, Sallis J. Young and active? Young people and health-enhancing physical activity: evidence and implications. London: Health Education Authority, 1998 61. Twisk JW. Physical activity guidelines for children and adolescents: a critical review. Sports Med 2001; 31 (8): 617-27 62. Buscemi N, Vandermeer B, Hooton N, et al. Efficacy and safety of exogenous melatonin for secondary sleep disorders and sleep disorders accompanying sleep restriction: meta-analysis. BMJ (Clin Res Ed) 2006 Feb 18; 332 (7538): 385-93 63. Machado M, Bajcar J, Guzzo GC, et al. Sensitivity of patient outcomes to pharmacist interventions: part I. Systematic review and meta-analysis in diabetes management. Ann Pharmacother 2007 Oct; 41 (10): 1569-82 64. Teasell R, Bayona N, Marshall S, et al. A systematic review of the rehabilitation of moderate to severe acquired brain injuries. Brain Inj 2007 Feb; 21 (2): 107-12
Correspondence: Dr Felipe Fossati Reichert, Department of Physical Education, State University of Londrina, Rodovia Celso Garcia Cid, km 380, CEP: 86051-990, Londrina, PR, Brazil. E-mail:
[email protected]
Sports Med 2009; 39 (4)
REVIEW ARTICLE
Sports Med 2009; 39 (4): 295-312 0112-1642/09/0004-0295/$49.95/0
ª 2009 Adis Data Information BV. All rights reserved.
The Respiratory Health of Swimmers Vale´rie Bougault,1 Julie Turmel,1 Benoıˆt Levesque2 and Louis-Philippe Boulet1 1 Centre de Recherche de l’Institut Universitaire de Cardiologie et de Pneumologie de Que´bec, Que´bec, Canada 2 Institut National de Sante´ Publique du Que´bec, Direction des Risques Biologiques, Environnementaux et Occupationnels, Que´bec, Canada
Contents Abstract. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1. Respiratory Symptoms . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2. Prevalence of Respiratory Disorders in Swimmers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.1 Prevalence of Atopy and Rhinitis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.2 Asthma and Airway Hyper-Responsiveness (AHR) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3. Respiratory Health in Swimmers versus Other Athletes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4. Determinants and Mechanisms of Development of Respiratory Disorders in Swimmers . . . . . . . . . . . 4.1 Nasal Disorders . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.2 Mechanisms of AHR and Asthma in Swimmers. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.2.1 Airway Inflammation and Epithelial Damage . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.2.2 Airway Remodelling . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.2.3 Accidental Exposure to Chlorine: Reactive Airway Dysfunction Syndrome . . . . . . . . . . . . 5. Diagnosis and Management of Respiratory Disorders in Swimmers . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6. Areas for Future Research . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7. Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Abstract
295 296 298 298 299 299 301 302 302 302 304 304 306 308 309
Regular physical activity is recognized as an effective health promotion measure. Among various activities, swimming is preferred by a large portion of the population. Although swimming is generally beneficial to a person’s overall health, recent data suggest that it may also sometimes have detrimental effects on the respiratory system. Chemicals resulting from the interaction between chlorine and organic matter may be irritating to the respiratory tract and induce upper and lower respiratory symptoms, particularly in children, lifeguards and high-level swimmers. The prevalence of atopy, rhinitis, asthma and airway hyper-responsiveness is increased in elite swimmers compared with the general population. This may be related to the airway epithelial damage and increased nasal and lung permeability caused by the exposure to chlorine subproducts in indoor swimming pools, in association with airway inflammatory and remodelling processes. Currently, the recommended management of swimmers’ respiratory disorders is similar to that of the general population, apart from the specific rules for the use of medications in elite athletes. Further studies are needed to better understand the mechanisms related to the development or worsening of respiratory disorders in recreational or competitive swimmers, to determine how we can optimize treatment and possibly help prevent the development of asthma.
Bougault et al.
296
In the general population, swimming is ranked among the preferred physical activities, after walking and cycling,[1,2] and is also now frequently being introduced in some schools’ sports curriculae, especially for young children. This activity is considered beneficial for health and particularly suitable for asthma patients since a humid and warm environment is less ‘asthmogenic’.[3-5] The limitations imposed by bodyweight on physical activity are lessened in water, so this sport is often suggested for obese subjects, pregnant women, the elderly and those with injuries or handicap. Finally, swimming lessons may also be motivated by the wish to help prevent drowning. Although swimming is considered to contribute to general health, potential respiratory problems associated with this sport have been described, mostly with respect to the effects of environmental conditions in which swimming is performed. Indeed, swimming pools are mostly disinfected with chlorine or its derivatives, known to be highly efficient to control viruses and bacterial growth, but not without adverse effects.[6] Chemical by-products are released in swimming pools resulting from interactions between chlorine and organic matter. They include trihalomethanes, mostly represented by chloroform, and halocetic acids. Chlorine reacts with ammonia, brought to the pool water by users and originating from sweat, urine, soap residues, cosmetics and suntan oil.[6] Halocetic acids may produce eye and skin irritation, whereas chlorine gas and chloramines, particularly nitrogen trichloride, are mainly irritants to the respiratory system. Therefore, whilst the warm and humid air of indoor swimming pools theoretically constitutes a beneficial environment for asthmatic subjects, research has shown that athletes who regularly use chlorinated swimming pools for prolonged periods of time may have a higher risk of developing respiratory health problems than the general population.[7-10] After reviewing the prevalence of respiratory symptoms and disorders among swimmers compared with the general population and other athletes, this article will discuss the determinants and mechanisms by ª 2009 Adis Data Information BV. All rights reserved.
which an indoor pool environment may induce respiratory problems. Finally, this article concludes with a review of the diagnosis and management of respiratory disorders in swimmers and provides perspectives for future research and prevention. For this review, publications available until April 2008 in the peer-reviewed literature (mostly PubMed) using keywords such as ‘chlorine’, ‘airway hyperresponsiveness’, ‘asthma’ or ‘rhinitis’ in combination with ‘athletes’ or ‘swimmers’ were analyzed. Various governmental and sport organization reports were also analyzed.
1. Respiratory Symptoms A high incidence of upper and lower airway respiratory symptoms has been reported in competitive athletes and schoolchildren attending swimming pools.[8,11] These are summarized in table I. Upper respiratory symptoms may be quite troublesome for athletes, particularly during periods of increased allergen exposure (e.g. during the pollen season). Several studies have shown that nasal symptoms were highly prevalent in swimmers regularly attending pools, with 25–74% of swimmers complaining of chronic rhinitis symptoms.[7,15] The main upper respiratory symptoms reported are sneezing, itching, rhinorrhoea, nasal obstruction and symptoms associated with sinusitis.[7,9,15] Swimmers also frequently report headache, ocular symptoms and throat pain.[9,14] Both upper and lower airway respiratory symptoms may have a deleterious effect on swimmers’ performance, unless minimal.[16] Zwick et al.[8] reported that 79% of the swimmers training 27–37 hours per week presented various upper and lower respiratory symptoms compared with 21% of controls. Helenius et al.[12] observed swimming-induced lower airway respiratory symptoms in 57% of the 42 Finnish National Team swimmers. The most frequently reported symptoms were cough and asthma-like symptoms such as breathlessness, wheezing and chest tightness.[9,10,14,13] Respiratory symptoms are not only frequently reported by elite swimmers, but Sports Med 2009; 39 (4)
Study
n
Sporta
Lower airway symptoms (%)a
Upper airway symptoms (%)a
cough wheezing breathless- chest ness tightness
chest congestion
sneezing/ rhino- nasal itching rrhoea obstruction
sinusitis
40
40
throat irritation
eye headache irritation
Swimmers Deitmer and Scheffler[7] Zwick et al.[8]
20 14
Momas et al.[11]
210
24
Potts[9]
44
738
36
26
39
25
Langdeau et al.[10]
25
24
’
20
-
Helenius et al.[12]
16
’
63–81
-
Turcotte et al.[13]
95
10
10
33
8
15
305
26
9
24
NA
9
Levesque et al.[14] Gelardi et al.[15]
44
23
40
30
43
7
29
62
11
16
19
27
36
36
45 40
The Respiratory Health of Swimmers
ª 2009 Adis Data Information BV. All rights reserved.
Table I. Prevalence of respiratory tract symptoms obtained by questionnaire
12 30 ’
74
-
10
15
0
19
37
0
7
29
Healthy non-swimmers Deitmer and Scheffler[7]
20
Zwick et al.[8]
16
Potts[9]
16
Langdeau et al.[10]
50
21 12
’
28
-
20
25 Dry airb 20
’
12
-
36
25 Cold airc
76
’
48
-
48
25 Mixed aird
32
’
16
-
36
25
Other athletes Langdeau et al.[10]
297
Sports Med 2009; 39 (4)
Continued next page
Bougault et al.
298
Triathlon. d
Long-distance running and mountain biking. b
c Biathlon, cross-country skiing and speed skating.
Data are expressed as the percentage of subjects developing symptoms. a
9 6 14 499 Soccer Levesque et al.[14]
ª 2009 Adis Data Information BV. All rights reserved.
n = number of subjects included in the study and responding to the questionnaire; NA = data not available; ’- indicates bringing together of several symptoms.
4 9
23 2 37 0 9 43 Mixed air
16 5 32 10 176 Cold air
5
14 4 40 6 384 Dry air
7
Upper airway symptoms (%)a
sneezing/ rhino- nasal itching rrhoea obstruction cough wheezing breathless- chest ness tightness
Turcotte et al.[13]
Table I. Contd
Sporta n Study
chest congestion Lower airway symptoms (%)a
sinusitis
10
throat irritation
4
19
eye headache irritation
also by children regularly attending swimming pools with their school.[11] 2. Prevalence of Respiratory Disorders in Swimmers 2.1 Prevalence of Atopy and Rhinitis
Although nasal symptoms such as sneezing are the most common complaints in swimmers, there is still a paucity of data available with regard to rhinitis in swimmers.[9] In this respect, swimmers seem to be a specific population of athletes associated with a high prevalence of upper respiratory illnesses.[17,18] Indeed, elite swimmers are more susceptible to rhinitis, whether seasonal or not.[8,18] Moreover, a large number of competitive swimmers (50–65%) are sensitized to various allergens, among which are seasonal allergens, as documented by allergy skin-prick tests, compared with a control group (29–36%).[8,12,19,20] However, despite a significantly higher prevalence of allergy in swimmers, Gelardi et al.[15] showed in a recent study of 40 swimmers training 90–240 minutes three to five times a week that rhinitis was mostly non-allergic. In their study, only 16 swimmers had a typical allergic rhinitis.[15] The remaining 24 swimmers presented with non-allergic rhinitis, including infectious rhinosinusitis, non-allergic rhinitis with eosinophilia and, more frequently, neutrophilic rhinitis, suggesting that chlorine may play a role in the development of this nasal disorder and not only seasonal allergen exposure. In addition, Passali et al.[17] observed an altered nasal mucociliary transport in 35 elite swimmers compared with other athletes, which may be due to the effect of chlorine sub-products and may not only contribute to nasal problems such as rhinosinusitis and otitis but also to asthma. Rhinitis in athletes seems under-treated. In this regard, during the summer Sydney 2000 Olympic Games only half of all athletes with allergic rhinitis reported taking an antiallergic drug.[18] Finally, not only can untreated rhinitis interfere with current activities, but this condition, highly prevalent in swimmers, may also contribute to lower airway respiratory symptoms and dysfunction.[21] Sports Med 2009; 39 (4)
The Respiratory Health of Swimmers
2.2 Asthma and Airway Hyper-Responsiveness (AHR)
Symptoms compatible with exercise-induced bronchoconstriction (EIB), describing a transient airway narrowing following intense exercise, are commonly reported by athletes, many of whom have no physician-diagnosed asthma. The key characteristics of bronchial asthma include variable airway obstruction, inflammation, remodelling and airway hyper-responsiveness (AHR). AHR is defined as an abnormal susceptibility to airway narrowing following exposure to a wide range of bronchoconstrictor stimuli.[22] Interestingly, Potts[9] reported an increased prevalence of AHR in swimmers compared with other athletes, without any difference in exercise-induced bronchoconstriction compared with the general population. The increased airway responsiveness to histamine observed in athletes was frequently associated with atopy, as in non-athletes.[23] Asthma is generally diagnosed clinically on the basis of symptoms of wheezing, dyspnoea, phlegm production and cough, associated with objective evidence of variable airway obstruction[24] or of AHR to a pharmacological agent such as methacholine. However, most studies have reported either respiratory symptoms or mostly AHR separately with only a few showing their comparative prevalence in a given athletes’ population. Helenius et al.[12,20] reported that 23–31% of competitive swimmers had asthma, as defined by the presence of an AHR (histamine provocative dose of inhaled drug producing a 15% decrease in forced expiratory volume in 1 second [PD15FEV1] £1.6 mg) associated with at least one exerciseinduced bronchial symptom monthly during the last year. In the studies available, swimmers had a previous physician’s diagnosis of asthma in up to 21% of athletes compared with up to 9% in healthy non-swimmer controls.[9,10,12,14,19,20,25] An asthma diagnosis should always be confirmed by objective means,[26] and in this regard Dickinson et al.[27] observed that 21% of athletes with a previous physician-diagnosed asthma failed to produce a positive test for asthma according to the International Olympic Committee (IOC) criteria. ª 2009 Adis Data Information BV. All rights reserved.
299
The prevalence of AHR in the general population varies from 4% to 35%.[28] It has been found in 36–79% of indoor competitive swimmers and in 14% of outdoor or sea swimmers.[8-10,12,19,20,29] Langdeau et al.[10] found a mean methacholine provocative concentration of inhaled drug producing a 20% decrease in FEV1 (PC20) of 7.3 – 5.5 mg/mL in 25 swimmers versus 35.4 – 56.9 mg/mL in 50 control subjects. In a population of 22 swimmers, Boulet et al.[30] found AHR (defined as PC20 <16 mg/mL) in 12, with a mean PC20 of 2.26 mg/mL compared with 36.5 mg/mL in swimmers without AHR and with a measurable PC20. The prevalence of asthma and AHR in swimmers is summarized in table II. A discrepancy has often been observed between the prevalence of asthma symptoms and AHR in swimmers. In a population of 25 swimmers, Langdeau et al.[10] observed a prevalence of 20% of symptomatic asthma while it was surprisingly 76% for AHR (PC20 £16 mg/mL). This report may suggest that symptom recognition is impaired in swimmers; it may not inevitably reflect the exact prevalence of asthma or AHR in the population, as assessed by a physician.[9,10,26,32] Athletes may consider that exercise-related symptoms are a normal effect of high-intensity training and not the consequence of airway dysfunction, and thus may not report or deny these symptoms.[9,18,33] Perception of sensitory stimuli may possibly be altered in the airways of athletes (damaged sensitory receptors?) following chlorine exposure, although this remains to be investigated. Moreover, coughing may be explained by an increased cough reflex to environmental stimuli possibly mediated through neurogenic mechanisms, in the presence or absence of airway narrowing.[34] 3. Respiratory Health in Swimmers versus Other Athletes Elite endurance athletes are more commonly affected by respiratory symptoms,[35] rhinitis,[17] asthma[9] and AHR[10] than other athletes or recreational competitive athletes.[21,35-37] The increased airway ailments seem to be partly related to the duration and intensity of exercise.[36] Sports Med 2009; 39 (4)
300
ª 2009 Adis Data Information BV. All rights reserved.
Table II. Prevalence of asthma and bronchial hyper-responsiveness in swimmers Study
Swimmers n age (y) [mean or range (SD)]
Method (response to)
AHR (%) S C
Chosen AHR threshold
Atopy (%)
Asthma diagnosis S C
Training level
Years of swimming [mean]
19
12–15
Questionnaire
NA
NA
NA
NA
38
6–9
2–3 h/d
NA
Zwick et al.[8]
14
11–24
Methacholine
78.6
35.7
PD20 = 3150 (650) mg
62
NA
NA
27–37 h/wk
1–6
Potts[9]
35
14–22
Methacholine
60 34
12.5 0
PC20 £16 mg/mL PC20 £8 mg/mL
NA
NA
NA
Competitive swimmers
NA
Helenius et al.[20]
42
18.6 (2.7)
Histamine
35.7
11.1
PD15 £1.6 mg
50
19
2
1780 – 710 km previous year
8.5
Langdeau et al.[10]
25
20.6 (1.5)
Methacholine
76 60
28 18
PC20 <16 mg/mL PC20 <8 mg/mL
84
8
4
19.1 h/wk
10.7
Helenius et al.[12]
42
23.9 (2.6)
Histamine
36
12
PD15 £1.6 mg
50
26
4
Finnish National team. Controls are former swimmers
12
7
23.3 (7.7)
Methacholine
14
NA
PC20 £8 mg/mL
NA
NA
NA
Outdoor 32 – 15 km/wk
11.3
22
20–22
Methacholine
55
NA
PC20 £16 mg/mL
83
NA
NA
NA. At least 4.5 h/wk
NA
305
8–22
Questionnaire
NA
NA
NA
NA
15.4
NA
Competitive swimmers
>1
Bonsignore et al.[25]
Sports Med 2009; 39 (4)
Boulet et al.[30]
Levesque et al.[14]
AHR = airway hyper-responsiveness; C = control subjects; n = number of swimmers; NA = data not available; PD20 or PC20 = dose or concentration of inhaled methacholine causing a 20% decrease of the forced expiratory volume in 1 second (FEV1); PD15 = dose of inhaled histamine causing a 15% decrease of the FEV1; S = swimmers.
Bougault et al.
Carlsen et al.[31]
The Respiratory Health of Swimmers
Among elite summer athletes, swimmers seem to develop rhinitis and allergies more frequently.[17,38] Despite conflicting results, swimmers seem more at risk to develop asthma than control subjects.[9,10,20,37] This may also be true for other sports.[39] Helenius et al.[20] reported a 25-fold increase in the relative risk of asthma (defined as increased AHR together with at least one exercise-induced symptom) in atopic speed and power athletes, a 42-fold increase in atopic longdistance runners, and as much as a 97-fold increase in atopic swimmers compared with nonatopic controls. In 2004, swimmers had the highest prevalence of exercise-induced asthma at the summer Olympics, 44% of swimmers of the Great Britain team responding to the criteria for asthma diagnosis according to the IOC guidelines.[27] Winter cold air athletes are also frequently affected by respiratory symptoms, mainly cough, in association with a prevalence rate of AHR of up to 75% in elite cross-country skiers.[40] However, Langdeau et al.,[10] who classified a cohort of 100 athletes into four groups according to their sport environment (dry air, cold air, mixed air and swimmers) and 50 control subjects, found that swimmers had more frequent or more marked AHR than coldair athletes. The 25 swimmers had the lowest mean methacholine PC20 (7.3 – 5.5 mg/mL) [controls mean PC20 = 35.4 – 56.9 mg/mL, mean of all athletes: 16.9 – 16 mg/mL], including coldair athletes (15.8 – 16 mg/mL). The proportion of swimmers with methacholine <8 mg/mL was 60% compared with 20% in cold- and dryair athletes and 32% in triathletes.[10] As previously described, swimmers seem to report less cough than cold-air athletes,[10] but more than soccer players.[14] If athletes, other than swimmers, frequently practise swimming as a recreational activity or for training, this may have been misleading. This may have caused an underestimation of the exposure to chlorinated products in certain previous studies. However, in previous studies conducted in our team, cold-air athletes were not attending swimming pools (unpublished observations), suggesting minimal influence on airway function and symptoms. ª 2009 Adis Data Information BV. All rights reserved.
301
4. Determinants and Mechanisms of Development of Respiratory Disorders in Swimmers Chlorine derivatives present in the ambient air of indoor pool environments may irritate the airway, and are increasingly blamed for the occurrence of respiratory symptoms in pool attendees. Among recreational swimmers, infants and young children are especially exposed to chlorine derivatives and thus may be more subject to the harmful effects of these substances.[41-43] Occupational asthma to indoor pool contaminants contained in the pool atmosphere has been reported in lifeguards.[44-47] As athletes spend many hours per week training while sustaining high levels of ventilation, they can be markedly affected by the constituents and characteristics of inhaled air.[19,20,48] Hence, swimmers training for 30 hours per week are 20 times more exposed to chlorine compounds than lifeguards and about 100 times more than recreational swimmers.[20,49] The marked penetration of ambient air pollutants from indoor or outdoor environments in proximal, but also possibly peripheral, airways may contribute to the development of respiratory tract disorders, and reduce exercise performance.[50-53] Chlorine is currently the main product used to disinfect the water of indoor swimming pools worldwide.[43,54] When added to water in its liquid form, at the pH found in swimming pools (7.2–7.8), chlorine reacts to form hypochlorous (HOCl) and hydrochloric acids (HCl), which are non-volatile, potentially harmful compounds.[6,9,55,56] Chlorine gas may provoke acute damage to the respiratory tract through the generation of free oxygen radicals in the upper as well as the lower respiratory tract.[56,57] HCl and HOCl may also cause cellular injury, particularly to epithelial mucosa of the ocular conjunctivae and the upper respiratory tract.[6,56] When reacting with nitrogenous and carbon contaminants brought by swimmers, chlorine generates a number of volatile chemical sub-products that are known irritants, sensitizing agents and possible carcinogens.[6,44,55,58-60] Exposure to these chlorinated compounds may cause an oxidative Sports Med 2009; 39 (4)
Bougault et al.
302
stress in swimmers that could partly explain their toxicity on the respiratory tract.[61] Among chloramines, nitrogen trichloride is highly volatile and responsible for the chlorine smell in pools[43,60] and for the ocular and respiratory symptoms felt by swimming-pool attendees.[44,58] 4.1 Nasal Disorders
The mechanisms involved in nasal disorders in response to chlorine derivative inhalation remain unclear. The response to these products seems different according to the atopic history of the swimmers. Indeed, Shusterman et al.[62,63] have shown that subjects with seasonal allergic rhinitis had more marked nasal congestion after chlorine gas inhalation than non-rhinitic subjects, with increased nasal resistance. In healthy subjects without AHR or allergic rhinitis, short-term laboratory exposure to chlorine concentrations of 0.1–0.5 ppm did not induce an inflammatory response in the nasal epithelium.[64] Neither mast cell degranulation nor central or peripheral neurogenic reflexes seem to be responsible for chlorine-induced nasal airflow obstruction.[63] Because chlorine gas is not often used as a disinfectant in swimming pools, except in the case of chlorine accidents, these observations remain to be confirmed in the presence of chlorine subproducts used in swimming pools. Gelardi et al.[15] showed that the majority of cases of rhinitis in competitive swimmers were non-allergic, with prevalent nasal obstruction and neutrophilia. Exercise itself is recognized as a potential cause of rhinitis,[65] causing a rhinorrhoea more than nasal congestion; the combination of chlorine by-product exposure and exercise may thus act in synergy to induce rhinitis in swimmers. Further studies are required to characterize populations at risk of developing rhinitis and to clarify the mechanisms of action of chlorine on the nasal function. Regarding swimmers and allergy, lung permeability may increase in subjects who use swimming pools, in response to chlorinated products, probably reflecting airway epithelial damage. This may also occur in the nasal mucosa and facilitate the penetration of aeroallergens, ª 2009 Adis Data Information BV. All rights reserved.
increasing the risk of allergic manifestations as observed in swimmers.[43,66] In a recent excellent review, this relationship between atopy and swimming pool attendance was highlighted.[43] The author suggested that atopic subjects could be more at risk to develop atopic manifestations, as chlorine-related irritants in swimming pools could act as adjuvants not only in regard to sensitization to allergens, but may also contribute to the clinical expression. 4.2 Mechanisms of AHR and Asthma in Swimmers
Asthmatic individuals may have chosen swimming as a sport based upon a suggestion from their physician or peers, as it is frequently considered less ‘asthmogenic’ than others. However, in almost 80% of swimmers with a diagnosis of asthma, this condition likely appeared after the beginning of the swimming career, suggesting that they did not start swimming because of their asthma condition.[12] This seems to indicate that the indoor swimming environment may be involved in the development of airway dysfunction.[23,43,48,67] It has then been suggested that the increasing rate of asthma among children in the European population over a few years was partly due to the introduction of swimming in the school programmes in industrialized countries.[43,68] The ‘united airways disease’ hypothesis, suggesting that upper and lower airway diseases are both manifestations of a single inflammatory process within the respiratory tract,[69,70] may partly explain the high prevalence of asthma and AHR in swimmers with regard to the high prevalence of rhinitis in this population. 4.2.1 Airway Inflammation and Epithelial Damage
Carbonnelle et al.[71] and Bernard et al.[41] studied the acute effects of a swimming session on the lung epithelium in recreational swimmers. They observed transient epithelium damage due to chlorinated swimming pool attendance in recreational swimmers. They compared serum levels of lung proteins such as CC16 released in the airways and alveolar surfactant-associated serum proteins (SP-A and SP-B) before and after Sports Med 2009; 39 (4)
The Respiratory Health of Swimmers
2 hours spent in a chlorinated swimming pool.[72] These proteins are used as markers of lung epithelial barrier integrity in a variety of acute or chronic lung disorders. The serum concentration of CC16 and surfactant proteins reflects an increase in the permeability of the alveolarcapillary barrier.[72] Indeed, surfactant proteins are normally secreted by airway or epithelial cells and their release in the blood can only be explained by assuming that they leak from the lung into the bloodstream.[72] After a 2-hour exposure to a chlorinated pool environment, an increase of blood SP-A and SP-B levels was reported, even after an hour without exercising, excluding a major role of exercise on the loss of pulmonary epithelium integrity. No significant variation of CC16 was noted in children or adults at the end of the second hour. These data indicate firstly that 2 hours spent in a swimming pool environment is sufficient to significantly alter lung permeability. Secondly, deep lung seems preferentially affected by such exposure with regard to changes in CC16 compared with SP-A and SP-B. This evidence of epithelial damage may be attributed to the presence of chlorine subproducts, as no such SP-A and SP-B changes have been observed after an exercise in swimming pools disinfected by a copper/silver process.[71] We should, however, note that the subjects developed neither respiratory symptoms nor a reduced pulmonary function in that last study. Cumulated school pool attendance was positively correlated to the serum concentration of pneumoproteins, airway inflammation and asthma indicators (SP-A, SP-B and CC16).[41,42,71] The clinical significance of these effects on the lung epithelium of recreational swimmers exposed long term to chlorination products in indoor swimming pools are uncertain, but deserve further investigation, as they seem to persist and increase with the number of hours spent in the swimming-pool environment. That is also true for very young children attending pools as baby swimmers or with the highest pool attendance, who present epithelial damage, similar to that observed in current smokers.[41] Therefore, as a result of a prolonged exposure to chlorine, a reduction in the level of CC16 in blood serum siª 2009 Adis Data Information BV. All rights reserved.
303
milar to the one observed in smokers or in subjects exposed to chemical smoke was observed at rest in children who attended chlorinated swimming pools as infants compared with other children.[43,73-75] Levesque et al.[14] reported a positive correlation between the chloramine levels in the air or water of the swimming pools and upper and lower respiratory symptoms in 72 competitive swimmers to whom a questionnaire on respiratory symptoms was administered. Neutrophils were also more abundant in the induced sputum of elite swimmers at rest compared with healthy controls.[19,25] Helenius et al.[19] studied induced sputum in 29 elite swimmers who competed for a mean of 9.1 years versus 19 healthy controls. Contrary to the observation made by previous authors on recreational swimmers, they found a significantly increased eosinophil cell count in swimmers. Moreover, 21% of them had an eosinophilia, with a sputum differential eosinophil count over 4% compared with none in controls. During a 5-year follow-up, the eosinophilia increased in active swimmers.[12] Swimmers with EIB symptoms presented higher eosinophil counts (7.6%) compared with swimmers without symptoms (0.7%).[12,19] Neutrophils and eosinophils were more activated in swimmers than in control subjects, as indicated by eosinophil peroxidase and human neutrophil lipocalin levels.[19] These results suggest that exposure to chlorine derivatives, when repeated for many years, may contribute to the persistence of airway inflammation at rest and probably AHR in swimmers. Chloramine-induced changes to the lung epithelium may become permanent in the case of a long-term exposure, for example in elite swimmers. Eosinophilic airway inflammation has been reported after long-term exposure to the pool environment in elite swimmers, but not in recreational swimming-pool attendees. However, other inflammatory cells (table III) remain to be studied in recreational and elite swimmers to better understand the inflammatory mechanisms involved, although epithelial damage seems to precede this process as suggested by our recent study.[81] Further studies are required to inSports Med 2009; 39 (4)
Bougault et al.
304
vestigate whether epithelial damage is reversible as well as the time-course of recovery of pneumoproteins (SP-A and SP-B) and alveolo-capillary barrier permeability. 4.2.2 Airway Remodelling
Inflammatory cell recruitment in the respiratory tract may be consecutive to airway epithelial damage, and especially bronchial epithelial cell (BEC) changes.[82] The high ventilation sustained during exercise itself may affect airway epithelium in changing the viscosity and tonicity of airway surface liquid.[53] BEC numbers and apoptosis seem to be increased after intense exercise, and a correlation was found between this change and ventilation rate sustained during exercise.[78,83] Moreover, after 45 days of a low- to moderate-intensity physical programme in mice training 5 days a week on a motorized rotor, Chimenti et al.[82] found a 2-fold increase in BEC apoptosis, a 4-fold reduction in ciliated epithelial cell counts, a 5-fold increase in proliferating cell counts and a 56% increase of epithelial thickness in bronchiolar epithelium compared with sedentary mice. Therefore, epithelial damage may appear consecutively to the intense hyperventilation through the dehydration of the mucosa and an induced shear stress on the airway wall, followed by a repair process.[78,82] However, as other authors found no change in the BEC level in sputum at rest and after a running or swimming race,[25,76] further studies are needed. Basic mechanisms of chlorinated product toxicity are related to the high solubility of chlorine in water and, at the physiological pH found in the bronchi, to the formation of hydrochloric and hypochlorous acids within epithelial tissues. These acid compounds are known to be highly irritating to the mucosa. Moreover, repeated cleavage of chlorinated products in contact with respiratory tract water provokes a release of oxygen free radicals. Functional and pathological changes in the airways resulting from chlorine by-product exposure are often considered to be mainly caused by such resulting oxidative stress.[84] ª 2009 Adis Data Information BV. All rights reserved.
The effects of long-term chlorine by-product exposure on a swimmer’s airways structure in the context of an indoor swimming pool are currently unknown. However, high-level acute exposure to chlorine has been described in experimental studies[84-86] or in the case of accidental exposure.[87-94] Airway epithelial damage and desquamation are the main consequences of an acute chlorine exposure, with an alteration in the morphology of the cells and replacement of cuboidal epithelial with flat cells.[84-86] Increased subepithelial fibrosis has been observed, following subepithelial haemorrhage and inflammatory infiltrates, after accidental high-level short-term chlorine exposure.[85,90] Repeated low- to moderate-level exposure to chlorine may therefore induce an airway remodelling process though the activation of fibrogenic cytokines in elite swimmers. To our knowledge, Karjalainen et al.[40] and Sue-Chu et al.[95] are the sole authors who examined bronchial biopsies of athletes. They observed in elite cross-country skiers an inflammatory infiltrate and an increased subepithelial tenascin, as well as lymphoid aggregates.[40,95] These changes could be due to the combination of endurance training and repeated cold dry air exposure, but the time-course and mechanisms by which these changes occur are unknown. 4.2.3 Accidental Exposure to Chlorine: Reactive Airway Dysfunction Syndrome
Two main types of occupational asthma have been described, the ‘classic form’ with a latency period (months to years) between initiation of exposure and the development of symptoms, and another without such a latency period called ‘irritant-induced asthma’, initially described as the reactive airway dysfunction syndrome (RADS) after a single exposure to high levels of toxic irritant substances, or also after repeated subacute exposures.[96] Indeed, RADS has been described as resulting from exposure to high levels of toxic substances such as chlorine, sulphur dioxide, ammonia or other substances with highly irritant properties inducing acute and often persistent airway damage.[87-94] Acute accidental release of high concentrations of chlorine in swimming pools may cause respiratory Sports Med 2009; 39 (4)
The Respiratory Health of Swimmers
305
Table III. Airway inflammatory cells in athletes and markers of activation of inflammatory cells in swimmers’ airways Reference
Subjects (n)
NO
Methods
Condition
TCC
BEC
Eos
Mac
Neu
Lym
Swimmers Helenius et al.[19]
29 Finnish national team swimmers
Sputum
Resta
2
›
›
2
fl
2
Bonsignore et al.[25]
Six non-elite outdoor swimmers
Sputum
Resta
2
2
›
2
fl
2
2
After sea race compared with rest
2
›
2
›
fl
2
fl
After outdoor swimming-pool race compared with rest
2
2
2
2
2
2
›
›
fl
2
› fl
2
Other sports athletes Karjalainen et al.[40]
40 cross-country skiers
Biopsy
12 mild asthma subjects Bonsignore et al.[76]
Five amateur long-distance runners
Sputum ›
Rest during the peak of their training programmea
›
Compared with skiers
›
Resta
2
›
›
2
After a marathon compared with rest
2
=
›
2
fl
2
Lumme et al.[77]
68 national level ice-hockey players
Sputum
Resta
2
›
›
2
fl
2
Morici et al.[78]
Nine rowers
Sputum
After an exercise test
2
2
2
2
2
›
Sue-Chu et al.[79]
29 elite cross-country skiers with EIA symptoms
Biopsy BAL
Before and after treadmill exercise at -15C
2
2
2
2
fl
2
Verges et al.[80]
29 cross-country skiers and ten triathletes
Sputum
Resta
National or international level Subgroup of AHR positive
›
2
›
2
2
Subgroup of AHR negative
2
2
2
2
2
2
EPO ›
HNL ›
NE
HI
ECP
L-S
After sea race compared with rest
2
2
2
fl
After outdoor swimming-pool race compared with rest
2
2
2
fl
Markers of activation of inflammatory cells and histamine level in swimmers airways Helenius et al.[20]
29 Finnish national team swimmers
Sputum
Resta
Bonsignore et al.[25]
Six non-elite outdoor swimmers
Sputum
Resta
a
Rest data are compared with control subjects.
AHR = airway hyper-responsiveness; BAL = bronchoalveolar lavage; BEC = bronchial epithelial cell count; ECP = eosinophil cationic protein; EIA = exercise-induced asthma; Eos = eosinophils; EPO = eosinophil peroxidase; HI = histamine; HNL = human neutrophil lipocalin; L-S = L-selectin; Lym = lymphocytes; Mac = macrophages; n = number of subjects; NE = neutrophil elastase; Neu = neutrophils; NO = nitric oxide; TCC = total cell count; fl indicates decreased counts; › indicates increased counts; 2 indicates no difference.
ª 2009 Adis Data Information BV. All rights reserved.
Sports Med 2009; 39 (4)
306
injuries, although very rarely death of the victims.[59,88-93] In general, cough, breathlessness and wheezing are the main respiratory symptoms in patients exposed to an accidental high-level exposure to chlorine. In contrast to a low level exposure to common irritants in non-asthmatic workers, in the case of RADS, expiratory flows may sometimes be reduced by >50%,[93] and patients develop transient AHR of variable intensity according to the inhaled dose.[59,92-94] Subjects with rhinitis, atopy, AHR or a chronic respiratory disease and those engaged in a physical exercise at the time of the accident report more severe respiratory distress than others, especially over a concentration of 1 ppm.[59,92,94,97] Reduced lung function generally recovers within approximately 15 days depending on the inhaled dose of chlorine.[59,91,94,97] However, the restitution of functional integrity does not necessarily mean histological integrity, as many days after expiratory flows and partly AHR have recovered, patients may still show epithelial desquamation and an inflammatory infiltrate, mainly neutrophilic.[87,89,91] Deschamps et al.[89] reported the case of an atopic patient without a diagnosis of asthma before the accidental inhalation of chlorine in the industry, but still with asthma 2 years after exposure; bronchial biopsies revealed in this patient a marked epithelial damage with a slight inflammatory lymphocytic process. The authors suggested that epithelial destruction could impair the epithelial release of bronchodilator substances and contribute to persistent asthma.[89] The model of acute intense exposure may help to better understand what happens in the case of repetitive lower-dose chlorine by-products exposure, as in regular swimmers. Epithelial cell damage is evidenced by an increase in CC16 in the blood serum, as reported in a group of 18 children, victims of an accidental highchlorine exposure during their swimming lesson.[91] It is, however, conceivable that low-level long-term exposures could cause a process of the same nature as RADS in some individuals, especially in those regularly attending swimming pools, although at a much lower grade.[98] However, this remains to be confirmed. ª 2009 Adis Data Information BV. All rights reserved.
Bougault et al.
5. Diagnosis and Management of Respiratory Disorders in Swimmers The diagnosis of rhinitis and asthma is initially based on the clinical features of these diseases. For asthma, the demonstration of a variable airway obstruction from measurements of expiratory flow response to a bronchodilator, spontaneous variations of airway obstruction, or measurement of airway responsiveness (e.g. methacholine, isocapnic hyperventilation) is essential to distinguish this entity from non-specific symptoms attributable to another condition. Currently, the management of asthma in swimmers is similar to asthma in other populations.[99] It is mostly based on the use of rapid-acting bronchodilators for intercurrent symptoms or prevention of EIB, associated with an inhaled corticosteroid if asthma symptoms are regularly experienced, and with added long-acting inhaled bronchodilators if asthma control is not achieved by a low dose of inhaled corticosteroid and, in some cases, additional medications such as leukotriene antagonists.[99] The specific response to the various asthma medications in swimmers needs to be studied and reports in various sports have demonstrated a reduced response to some of these agents in highlevel athletes,[100,101] possibly in relation to the different (neutrophilic) type of airway inflammation or other changes in the asthma phenotype. Asthma medications in competitive athletes are regulated and the IOC and World Anti-Doping Agency (WADA) regulations should be reviewed before using these medications.[102] In this population, long-term intermittent chlorine exposure may possibly impair the respiratory function and affect athletic performance, as observed in the presence of other strong pollutants.[50-53,103] As they may be particularly affected by the adverse effects of untreated rhinitis, elite swimmers should take the necessary precautions to minimize the impact of upper respiratory conditions on their ability to perform.[18,104] This holds especially true during spring and summer for pollen-sensitive swimmers. Allergy testing allows identification of sensitization to common airborne allergens and should be performed in athletes Sports Med 2009; 39 (4)
The Respiratory Health of Swimmers
presenting upper or lower respiratory symptoms. Rhinitis should be recognized in order to reduce its impact on their daily life and performance.[104] Medications used to treat rhinitis have been summarized in a review from the Allergic Rhinitis and its Impact on Asthma (ARIA) group.[38,105] It has been shown that the symptoms, the quality of life, and the performance of athletes with allergic rhinoconjunctivitis were improved with appropriate medication and environmental measures.[104] Current guidelines emphasize that an asthma diagnosis should be confirmed by objective means, in the presence of suggestive symptoms, to identify alternative aetiologies of the symptoms, assess the severity of the condition, and offer appropriate interventions to reduce the untoward effects of asthma and help prevent its worsening. It may be important to regularly assess swimmers’ respiratory health to identify such problems at an early stage. However, the increasing proportion of athletes taking asthma medication, particularly b2-agonists, led the IOC and WADA to institute guidelines to confirm asthma and document AHR.[106] The sole report of exercise-induced respiratory symptoms cannot justify asthma medication anymore, for previously described reasons (see section 2.2). Therefore, athletes using asthma medications must now be authorized beforehand by the proper sport authorities after a formal report by a physician. According to IOC and WADA, at least one authorized bronchial provocation testing must be positive to allow the use of regulated drugs for asthma, and the results must be provided if the swimmer performs at an international level. Different bronchial challenges are presently available to confirm the presence of variable airway obstruction such as eucapnic voluntary hyperpnoea, exercise test, hypertonic saline, methacholine and, recently, a mannitol challenge.[103] Guidelines for these challenges have previously been described,[107,108] as well as the positivity threshold for each test, which allows international athletes to use approved asthma medications, as summarized in table IV. For the Turin Olympic Games in 2006, the IOC recommended the following: a methacholine challenge was considered as positive in nonª 2009 Adis Data Information BV. All rights reserved.
307
steroid-treated athletes if the PD20 was as low as 2 mmol or 400 mg or less (PC20 £4 mg/mL).[108] The criteria remained similar for the 2008 Beijing Olympic Games. In the general population, AHR has often been defined as a methacholine PC20 £16 mg/mL,[107] although there is a ‘grey zone’ of AHR between 4 and 16 mg/mL. However, if negative to methacholine, they may show a positive response to the other tests.[32] Respiratory symptoms in subjects with a PC20 between 4 and 16 mg/mL may be due to asthma as previously suggested.[109] The IOC and WADA guidelines are regularly updated.[102,106] According to the WADA guidelines, recurrent symptoms of bronchial obstruction are a diagnostic prerequisite for asthma or EIA in athletes, and laboratory tests alone are not sufficient for the diagnosis.[102,106] However, athletes, and especially swimmers, may not recognize, may ignore or may not feel symptoms associated with Table IV. International Olympic Committee and World Anti-Doping Agency positivity criteria according to the test used Challenge
The test is positive if:
Bronchial reversibility
An increase of ‡12% and of at least 200 mL of FEV1 from the prechallenge value is observed after the inhalation of an authorized b2-agonist
Exercise test (field or laboratory)
A fall of at least 10% of FEV1 from the pre-challenge value is observed within 30 min after the end of the test. A 10% FEV1 fall during 5 consecutive minutes is consistent with a hyperventilationinduced bronchoconstriction
Eucapnic voluntary hyperpnoea
Methacholine challenge
In Olympic Games in Beijing 2008: (i) for subjects not taking corticosteroids: PC20 £4 mg/mL or PD20 £2 mmoL (or 400 mg); (ii) for subjects taking inhaled corticosteroids for at least 3 mo: PC20 £16 mg/mL or PD20 £8 mmoL (or 1600 mg)
Hypertonic aerosol challenges (saline solution or mannitol)
A fall of ‡15% in FEV1 from the prechallenge value is observed after inhalation of 22.5 mL of saline solution concentrated at 4.5% or 635 mg of dry powder mannitol
FEV1 = forced expiratory volume in 1 second; PC20 = concentration of inhaled methacholine that causes a 20% decrease of the FEV1; PD20 = cumulative doses of inhaled methacholine that cause a 20% decrease of the FEV1.
Sports Med 2009; 39 (4)
Bougault et al.
308
bronchoconstriction and accordingly not consult a physician for a diagnosis or intervention.[10,13,26,33,110] Therefore, in elite swimmers, in addition to confirming the presence of AHR, particular attention should be paid to documenting relevant symptoms. To prevent the consequences of exposure to chlorine derivates, it would be preferable to ensure that ambient levels of these components are kept at a minimum. In a review on chlorination products, Bernard[43] reported mean levels of nitrogen trichloride in the atmosphere of public indoor pools between 0.3 and 0.5 mg/m3, sometimes reaching 2 mg/m3. In the study of Levesque et al.,[14] the mean concentrations of chloramines in the air of seven swimming pools varied from 0.26 to 0.41 mg/m3. Using the same technique, Hery et al.[61] documented chloramine concentrations varying from <0.5 to 1.25 mg/m3 in 13 swimming pools. Massin et al.[44] also reported mean chloramine concentrations in the air of 46 swimming pools between 0.24 mg/m3 and 0.46 mg/m3 in 17 community centres. Elite swimmers, often training 2 hours twice daily and sometimes >30 hours per week, sustaining a high level of ventilation and inhaling immediately above the water surface, are a highly exposed population to chlorine derivatives. Because chloramine levels in the air above swimming-pool water is influenced by ventilation and the pool water chemistry, a proper aeration and reduction of human protein load in the water may reduce respiratory health problems of swimmers. Chloramine accumulations in the air above the water may be removed by efficient ventilation, which would increase turnover and remove concentrated chloramines. An efficient measure to reduce chloramine production would also be a modification of swimmers’ behaviour; for example, encouraging showering before entering any pool, wearing a swim cap and facilitating frequent bathroom breaks for swimmers, particularly children, since this could significantly reduce the amount of urine and other nitrogenous waste contaminating the water, which leads to the accumulation of chloramines in the swimming pool’s ambient air. Certain preventative measures are easy to implement and ª 2009 Adis Data Information BV. All rights reserved.
may contribute to a better respiratory health of swimmers and bathers. Alternative disinfection processes such as chlorine dioxide, brominebased disinfectants, ozone or UV use may also replace chlorine, but further studies on their effects on the respiratory health should be performed. Bromine-based disinfectants are volatile liquids with toxic fumes that irritate the skin, eyes and respiratory tract, while ozone use, the most powerful oxidizing and disinfecting agent available, is often followed by a deozonation process necessitating chlorine or bromine use and is more expensive.[55] Ozone is also a strong respiratory irritant.[55] The risks posed by brominated byproducts and the use of a disinfectant as an oxidizing agent have yet to be elucidated. Other disinfectant systems may be used and limit the production of chlorinated disinfection byproducts, especially in smaller and domestic pools, as hydrogen peroxide is associated with copper/silver ions or iodine.[55] Logically, the air of a copper/silver pool has been shown to contain no detectable trichloramines. Epithelial damage observed in swimmers attending chlorinated pools was not found when copper/silver disinfection was used.[71] However, copper/silver disinfection cannot remove organic matter and does not seem to ensure the total elimination of viral pathogens from water, even if its use is combined with low levels of chlorine.[111] Little is known on the possible effects on the respiratory health of copper/silver ionization. Because oxidative stress has been shown to be strongly involved in airway damage caused by chlorine,[58,84] an antioxidant supplementation has been suggested to have a protective role in swimmers,[58,112] and especially on AHR.[113,114] However, further studies are needed to support this recommendation. 6. Areas for Future Research Further research is necessary to better understand the mechanisms involved in the development of rhinitis, asthma and AHR in swimmers, particularly on how inflammation and remodelling, especially those induced by environmental exposures, can lead to these changes. Sports Med 2009; 39 (4)
The Respiratory Health of Swimmers
Asthma and AHR may develop through the swimming career, but their prevalence seems to decrease in swimmers who stop high-level training, suggesting a reversibility of alterations in airways function.[12] Indeed, in a study of Helenius et al.[12] the prevalence of AHR went from 31% during swimming activity to 12% in former swimmers. This is in agreement with Potts,[9] who found a reduction in respiratory symptoms in swimmers who did not exercise in a swimming pool for several days, although the mechanism of improvement may be different. Similarly, SimonRigaud et al.[67] reported a significant reduction of AHR in eight competitive swimmers after the annual swimming-pool cleaning compared with before. This observation should be confirmed, but could help with understanding to what extent AHR is reversible and what the mechanisms involved in swimmers are. The reversibility of epithelial damage in the case of long-term chlorine exposure also remains to be studied. Eventually, airway protective measures will be developed for competitive swimmers either in the form of medications or modifications of the pool environment. The efficacy and safety of different disinfection processes should be tested as well as the effects of better ventilation that could remove the contaminants present in the air above the water. It should be determined if some medications can prevent the effects of chlorine sub-products on airway function, and possibly inflammation or remodelling. 7. Conclusions Swimming is an enjoyable sport that could help maintain physical fitness and therefore should be promoted. However, evidence has shown that chlorine derivatives may create irritant effects to the upper and lower airways. Although low level or infrequent exposure to these agents may not be detrimental, alterations in airway function are common in swimmers and various factors may promote the development of airway dysfunction in susceptible populations such as competitive athletes. Chlorine sub-products seem to be the main agents responsible for swimming pool-induced rhinitis, asthma and AHR. The clinical signiª 2009 Adis Data Information BV. All rights reserved.
309
ficance, time-course and reversibility of the observed airways changes are not well documented. However, the potential risks from exposure to chlorination by-products in well managed swimming pools should be set against the benefits of physical activity and the risks of microbial disease in the absence of disinfection. More information is needed to better understand the optimal management of respiratory problems in swimmers, particularly high-level athletes, and how to prevent changes in airway function and asthma. Acknowledgements Vale´rie Bougault was supported by Universite´ Laval (GESER), Quebec, Canada. The authors have no conflicts of interest that are directly relevant to the content of this review.
References 1. Stephens T, Craig CL. Le mieux-eˆtre des Canadiens et des Canadiennes: faits saillants de l’enqueˆte Campbell de 1988. Ottawa (ON): Institut Canadien de la Recherche sur la Condition Physique et le Mode de Vie, 1990 2. Pieron M, De Knop P. Socie´te´ et Sport: Gestion et organisation du sport en Belgique. Rapport a` la Fondation du roi Baudouin, Bruxelles, 2000 [online]. Available from URL: http://www.kbs-frb.be/files/db/fr/PUB_1101_Gestion_ et_organisation_du_sport_en_Belgique%20.pdf [Accessed 2007 Aug 21] 3. Bar-Yishay E, Gur I, Inbar O, et al. Difference between swimming and running as stimuli for exercise-induced asthma. Eur J Appl Physiol 1982; 48 (3): 387-97 4. Bundgaard A, Schmidt A, Ingemann Hansen T, et al. Exercise-induced asthma after swimming and bicycle exercise. Eur J Respir Dis 1982 May; 63 (3): 245-8 5. Fitch KD, Morton AR, Blansky BA. Effects of swimming training on children with asthma. Arch Dis Child 1976 Mar; 51 (3): 190-4 6. World Health Organization. International programme on chemical safety, environmental health criteria 216: disinfectants and disinfectant by-products. Geneva: WHO, 2000 7. Deitmer T, Scheffler R. Nasal physiology in swimmers and swimmers’ sinusitis. Acta Otolaryngol 1990 Sep-Oct; 110 (3-4): 286-91 8. Zwick H, Popp W, Budik G, et al. Increased sensitization to aeroallergens in competitive swimmers. Lung 1990; 168 (2): 111-5 9. Potts JE. Adverse respiratory health effects of competitive swimming: the prevalence of symptoms, illness, and bronchial responsiveness to methacholine and exercise [dissertation]. Vancouver (BC): University of British Columbia, 1994 10. Langdeau JB, Turcotte H, Bowie DM, et al. Airway hyperresponsiveness in elite athletes. Am J Respir Crit Care Med 2000 May; 161 (5): 1479-84
Sports Med 2009; 39 (4)
310
11. Momas I, Brette F, Spinasse A, et al. Health effects of attending a public swimming-pool: follow-up of a cohort of pupils in Paris. J Epidemiol Community Health 1993 Dec; 47 (6): 464-8 12. Helenius IJ, Rytila P, Sarna S, et al. Effect of continuing or finishing high-level sports on airway inflammation, bronchial hyperresponsiveness, and asthma: a 5-year prospective follow-up study of 42 highly trained swimmers. J Allergy Clin Immunol 2002 Jun; 109 (6): 962-8 13. Turcotte H, Langdeau JB, Thibault G, et al. Prevalence of respiratory symptoms in an athlete population. Respir Med 2003 Aug; 97 (8): 955-63 14. Levesque B, Duchesne JF, Gingras S, et al. The determinants of prevalence of health complaints among young competitive swimmers. Int Arch Occup Environ Health 2006 Oct; 80 (1): 32-9 15. Gelardi M, Bonini M, Bonini S, et al. Non allergic rhinitis in competitive swimmers [abstract]. J Allergy Clin Immunol 2007 Jan; 119 Suppl. 1; S163 16. Pyne DB, McDonald WA, Gleeson M, et al. Mucosal immunity, respiratory illness, and competitive performance in elite swimmers. Med Sci Sports Exerc 2000; 33 (3): 348-3 17. Passali D, Damiani V, Passali GC, et al. Alterations in rhinosinusal homeostasis in a sportive population: our experience with 106 athletes. Eur Arch Otorhinolaryngol 2004 Oct; 261 (9): 502-6 18. Katelaris CH, Carrozzi FM, Burke TV, et al. Patterns of allergic reactivity and disease in Olympic athletes. Clin J Sports Med 2006 Sep; 16 (5): 401-5 19. Helenius IJ, Rytila P, Metso T, et al. Respiratory symptoms, bronchial responsiveness, and cellular characteristics of induced sputum in elite swimmers. Allergy 1998 Apr; 53 (4): 346-52 20. Helenius I, Tikkanen HO, Sarna S, et al. Asthma and increased bronchial responsiveness in elite athletes: atopy and sport event as risk factors. J Allergy Clin Immunol 1998 May; 101 (5): 646-52 21. Alaranta A, Alaranta H, Helio¨vaara M, et al. Allergic rhinitis and pharmacological management in elite athletes. Med Sci Sports Exerc 2005 May; 37 (5): 707-11 22. Bousquet J, Jeffery P, Busse W, et al. Asthma: from bronchoconstriction to airways inflammation and remodelling. Am J Respir Crit Care Med 2000 May; 161 (5): 1720-45 23. Helenius I, Haahtela T. Allergy and asthma in elite summer sport athletes. J Allergy Clin Immunol 2000 Sep; 106 (3): 444-52 24. Tattersfield AE, Knox AJ, Britton JR, et al. Asthma. Lancet 2002 Oct; 360 (9342): 1313-22 25. Bonsignore MR, Morici G, Riccobono L, et al. Airway cells after swimming outdoors or in the sea in nonasthmatic athletes. Med Sci Sports Exerc 2003 Jul; 35 (7): 1146-52 26. Rundell KW, Joohee IM, Mayers L, et al. Self-reported symptoms and exercise-induced asthma in the elite athletes. Med Sci Sports Exerc 2001 Feb; 33 (2): 208-13 27. Dickinson JW, Whyte GP, McConnell AK, et al. Impact of changes in the IOC-MC asthma criteria: a British perspective. Thorax 2005 Aug; 60 (8): 629-32
ª 2009 Adis Data Information BV. All rights reserved.
Bougault et al.
28. Jansen DF, Timens W, Kraan J, et al. (A) Symptomatic bronchial hyper-responsiveness and asthma. Respir Med 1997 Mar; 91 (3): 121-34 29. Potts J. Factors associated with respiratory problems in swimmers. Sports Med 1996 Apr; 21 (4): 256-61 30. Boulet LP, Turcotte H, Langdeau JB, et al. Lower airway inflammatory responses to high-intensity training in athletes. Clin Invest Med 2005 Feb; 28 (1): 15-22 31. Carlsen KH, Oseid S, Odden H, et al. The response of children with and without bronchial asthma to heavy swimming exercise. In: Oseid S, Carlsen KH, editors. Children and exercise XIII. Champaign (IL): Human Kinetics Publishers Inc., 1989: 351-60 32. Holzer K, Anderson SD, Douglass J. Exercise in elite summer athletes: challenges for diagnosis. J Allergy Clin Immunol 2002 Feb; 110 (4): 374-80 33. Turcotte H, Boulet LP. Perception of breathlessness during early and late asthmatic response. Am Rev Respir Dis 1993 Aug; 148 (2): 514-18 34. Anderson SD, Holzer K. Exercise-induced asthma: Is it the right diagnosis in elite athletes? J Allergy Clin Immunol 2000 Sep; 106 (3): 419-28 35. Spence L, Brown WJ, Pyne DB, et al. Incidence, etiology, and symptomatology of upper respiratory illness in elite athletes. Med Sci Sports Exerc 2007 Apr; 39 (4): 577-86 36. Nieman DC. Exercise infection, and immunity. Int J Sports Med 1994 Oct; 15 Suppl. 3: S131-41 37. Maiolo C, Fuso L, Todar A, et al. Prevalence of asthma and atopy in Italian Olympic athletes. Int J Sports Med 2003; 24: 139-44 38. Bonini S, Bonini M, Bousquet J, et al. Rhinitis and asthma in athletes: an ARIA document in collaboration with GA2LEN. Allergy 2006 Jun; 61 (6): 681-92 39. Weiler JM, Layton T, Hunt M. Asthma in United States Olympic athletes who participated in the 1996 summer games. J Allergy Clin Immunol 1998 Nov; 102 (5): 722-6 40. Karjalainen EM, Laitinen A, Sue-Chu M, et al. Evidence of airway inflammation and remodeling in ski athletes with and without bronchial hyperresponsiveness to methacholine. Am J Respir Crit Care Med 2000 Jun; 161 (6): 2086-91 41. Bernard A, Carbonnelle S, Michel O, et al. Lung hyperpermeability and asthma prevalence in schoolchildren: unexpected associations with the attendance at indoor chlorinated swimming pools. Occup Environ Med 2003 Jun; 60 (6): 385-94 42. Bernard A, Nickmilder M. Respiratory health and baby swimming. Arch Dis Child 2006 Jul; 91 (7): 620-1 43. Bernard A. Chlorination products: emerging links with allergic diseases. BMC Emerg Med 2007; 14 (16): 1689-99 44. Massin N, Bohadana AB, Wild P, et al. Respiratory symptoms and bronchial responsiveness in lifeguards exposed to nitrogen trichloride in indoor swimming pools. Occup Environ Med 1998 Apr; 55 (4): 258-63 45. Jacobs JH, Spaan S, van Rooy GBGJ, et al. Exposure to trichloramine and respiratory symptoms in indoor swimming pool workers. Eur Respir J 2007 Apr; 29 (4): 690-8 46. Nemery B, Hoet PHM, Nowak D. Indoor swimming pools, water chlorination and respiratory health. Eur Respir J 2002 May; 19 (5): 790-3
Sports Med 2009; 39 (4)
The Respiratory Health of Swimmers
47. Thickett KM, McCoach JS, Gerber JM, et al. Occupational asthma caused by chloramines in indoor swimmingpool air. Eur Respir J 2002 May; 19 (5): 827-32 48. Drobnic F, Freixa A, Casan P, et al. Assessment of chlorine exposure in swimmers during training. Med Sci Sports Exerc 1996 Feb; 28 (2): 271-4 49. Helenius IJ, Lumme A, Haahtela T. Asthma, inflammation and treatment in elite athletes. Sports Med 2005; 35 (7): 565-74 50. Schelegle ES, Adams WC. Reduced exercise time in competitive simulations consequent to low level ozone exposure. Med Sci Sports Exerc 1986 Aug; 18 (4): 408-14 51. Gong H. Effects of ozone on exercise performance. J Sports Med Phys Fitness 1987 Mar; 27 (1): 21-9 52. Carlisle AJ, Sharp NC. Exercise and outdoor ambient air pollution. Br J Sports Med 2001 Aug; 35 (4): 214-22 53. Flouris AD. Modelling atmospheric pollution during the games of the XXVIII Olympiad: effects on elite competitors. Int J Sports Med 2006; 27: 137-42 54. Das R, Blanc PD. Chlorine gas exposure and the lung: a review. Toxicol Ind Health 1993 May-Jun; 9 (3): 439-55 55. World Health Organization. Guidelines for safe recreational water environments. Vol. 2. Swimming pools and similar environments. Geneva: WHO, 2006 56. Barrow CS, Alarie Y, Warrick JC, et al. Comparison of the sensory irritation in mice to chlorine and hydrogen chloride. Arch Environ Health 1977 Mar-Apr; 32 (2): 68-76 57. Jones RN, Hughes JM, Glindmeyer H, et al. Lung function after acute chlorine exposure. Am Rev Respir Dis 1986 Dec; 134 (6): 1190-95 58. Varraso R, Massin N, Hery M, et al. Not only training but also exposure to chlorinated compounds generates response to oxidative stimuli in swimmers. Toxicol Ind Health 2002 Jul; 18 (6): 269-78 59. Agabiti N, Ancona C, Forastiere F, et al. Short term respiratory effects of acute exposure to chlorine due to a swimming pool accident. Occup Environ Med 2001 Jun; 58 (6): 399-404 60. Holzwarth G, Balmer RG, Sony L. The fate of chlorine and chloramines in cooling towers: Henry’s law constants for flashoff. Water Res 1984; 18 (11): 1421-7 61. Hery M, Hecht G, Gerber JM, et al. Exposure to chloramines in the atmosphere of indoor swimming pools. Ann Occup Hyg 1995; 39: 427-39 62. Shusterman D, Murphy MA, Balmes J. Influence of age, gender, and allergy status on nasal reactivity to inhaled chlorine. Inhal Toxicol 2003 Oct; 15 (12): 1179-89 63. Shusterman D, Balmes J, Murphy MA, et al. Chlorine inhalation produces nasal airflow limitation in allergic rhinitic subjects without evidence of neuropeptide release. Neuropeptides 2004 Dec; 38 (6): 351-58 64. Schins RPF, Emmen H, Hoogendijk L, et al. Nasal inflammatory and respiratory parameters in human volunteers during and after repeated exposure to chlorine. Eur Respir J 2000 Oct; 16 (4): 626-32 65. Silvers WS, Poole JA. Exercise-induced rhinitis: a common disorder that adversely affects allergic and nonallergic athletes. Ann Allergy Asthma Immunol 2006 Feb; 96 (2): 334-40
ª 2009 Adis Data Information BV. All rights reserved.
311
66. Kohlhammer Y, Do¨ring A, Scha¨fer T, et al. Swimming pool attendance and hay fever rates later in life. Allergy 2006 Nov; 61 (11): 1305-9 67. Simon-Rigaud ML, Eechout C, Bourdin H, et al. Bronchial hyperreactivity and swimming: influence of training in a chlorinated swimming-pool [in French]. Sci Sports 1997; 12 (2): 142-7 68. Nickmilder M, Bernard A. Ecological association between childhood asthma and availability of indoor chlorinated swimming pools in Europe. Occup Environ Med 2007 Jan; 64 (1): 37-46 69. Rimmer J, Ruhno JW. Rhinitis and asthma united airway disease. Med J Aust 2006 Nov; 185 (10): 565-71 70. Polosa R, Ciamarra I, Mangano G, et al. Bronchial hyperresponsiveness and airway inflammation markers in nonasthmatics with allergic rhinitis. Eur Respir J 2000 Jan; 15 (1): 30-5 71. Carbonnelle S, Francaux M, Doyle I, et al. Changes in serum pneumoproteins caused by short-term exposures to nitrogen trichloride in indoor chlorinated swimming pools. Biomarkers 2002 Nov-Dec; 7 (6): 464-78 72. Hermans C, Bernard A. Lung epithelium-specific proteins: characteristics and potential applications as markers. Am J Respir Crit Care Med 1999 Feb; 159 (2): 646-78 73. Lagerkvist BJ, Bernard A, Blomberg A, et al. Pulmonary epithelial integrity in children: relationship to ambient ozone exposure and swimming pool attendance. Environ Health Perspect 2004 Dec; 112 (17): 1768-71 74. Bernard AM, Roels HA, Buchet JP, et al. Serum clara cell protein: an indicator of bronchial cell dysfunction caused by tobacco smoking. Environ Res 1994 Jul; 66 (1): 96-104 75. Bernard AM, Gonzalez-Lorenzo JM, Siles E, et al. Early decrease of serum clara cell protein in silica-exposed workers. Eur Respir J 1994 Nov; 7 (11): 1932-7 76. Bonsignore MR, Morici G, Riccobono L, et al. Airway inflammation in nonasthmatic amateur runners. Am J Physiol Cell Mol Physiol 2001 Sep; 281 (3): L668-76 77. Lumme A, Haahtela T, O¨unap J, et al. Airway inflammation, bronchial hyperresponsiveness and asthma in elite ice hockey players. Eur Respir J 2003 Jul; 22 (1): 113-7 78. Morici G, Bonsignore MR, Zangla D, et al. Airway cell composition at rest and after an all-out test in competitive rowers. Med Sci Sports Exerc 2004 Oct; 36 (10): 1723-29 79. Sue-Chu M, Henriksen AH, Bjermer L. Non-invasive evaluation of lower airway inflammation in hyper-responsive elite cross-country skiers and asthmatics. Respir Med 1999 Oct; 93 (10): 719-25 80. Verges S, Devouassoux G, Flore P. Bronchial hyperresponsiveness, airway inflammation, and airflow limitation in endurance athletes. Chest 2005 Jun; 127 (6): 1935-41 81. Bougault V, Turmel J, St-Laurent J, et al. Asthma, airway inflammation and epithelial damage in swimmers and cold-air athletes. Eur Respir J 2009; 33 (4): 1-7 82. Chimenti L, Morici G, Paterno A, et al. Endurance training damages small airway epithelium in mice. Am J Respir Crit Care Med 2007 Mar; 175 (5): 442-9 83. Bonsignore MR, Chimenti L, Paterno A, et al. Longitudinal changes and apoptosis in induced cells in longdistance runners. Proceedings of ATS Meeting; 2006 May 19-24; San Diego (CA), A485
Sports Med 2009; 39 (4)
Bougault et al.
312
84. Martin JG, Campbell HR, Lijima H, et al. Chlorine-induced injury to the airways in mice. Am J Respir Crit Care Med 2003 Sep; 168 (5): 568-74 85. Menaouar A, Anglade D, Baussand P, et al. Chlorine gas induced acute lung injury in isolated rabbit lung. Eur Respir J 1997 May; 10 (5): 1100-7 86. Yildirim C, Kocoglu H, Goksu S, et al. Long-term pulmonary histopathologic changes in rats following acute experimental exposure to chlorine gas. Inhal Toxicol 2004 Dec; 16 (14): 911-15 87. Lemiere C, Malo JL, Boutet M. Reactive airways dysfunction syndrome due to chlorine: sequential bronchial biopsies and functional assessment. Eur Respir J 1997 Jan; 10 (1): 241-4 88. Parimon T, Kanne JP, Pierson DJ. Acute inhalation injury with evidence of diffuse bronchiolitis following chlorine gas exposure at a swimming-pool. Respir Care 2004 Mar; 49 (3): 291-4 89. Deschamps D, Soler P, Rosenberg N, et al. Persistent asthma after inhalation of a mixture of sodium hypochlorite and hypochloric acid. Chest 1994 Jun; 105 (6): 1895-6 90. Martinez TT, Long C. Explosion risk from swimming pool chlorinators and review of chlorine toxicity. J Toxicol Clin Toxicol 1995; 33 (4): 349-54 91. Bonetto G, Corradi M, Carraro S, et al. Longitudinal monitoring of lung injury in children after acute chlorine exposure in a swimming-pool. Am J Respir Crit Care Med 2006 Sept; 174 (5): 545-9 92. Mustchin CP, Pickering CAC. ‘Coughing water’: bronchial hyperreactivity induced by swimming in a chlorinated pool. Thorax 1979 Oct; 34 (5): 682-3 93. Ploysongsang Y, Beach BC, DiLisio RE. Pulmonary function changes after acute inhalation of chlorine gas. South Med J 1982 Jan; 75 (1): 23-6 94. D’Alessandro A, Kuschner W, Wong, et al. Exaggerated responses to chlorine inhalation among persons with nonspecific airway hyperreactivity. Chest 1996 Feb; 109 (2): 331-7 95. Sue-Chu M, Karjalainen EM, Altraja A, et al. Lymphoid aggregates in endobronchial biopsies from young elite cross-country skiers. Am J Respir Crit Care Med 1998 Aug; 158 (2): 597-601 96. Francis HC, Prys-Picard CO, Fishwick D, et al. Defining and investigating occupational asthma: a consensus approach. Occup Environ Med 2007 June; 64 (6): 361-5 97. Rotman HH, Fliegelman MJ, Moore T, et al. Effects of low concentrations of chlorine on pulmonary function in humans. J Appl Physiol 1983 Apr; 54 (4): 1120-4 98. Brooks SM, Weiss MA, Bernstein IL. Reactive airways dysfunction syndrome (RADS): persistent asthma syndrome after high level irritant exposures. Chest 1985 Sep; 88 (3): 376-84 99. Global Strategy for Asthma Management and Prevention, Global Initiative for Asthma (GINA) 2006 [online]. Available from URL: http://www.ginasthma.org/ [Accessed 2007 Aug 21] 100. Sue-Chu M, Karjalainen EM, Laitinen A, et al. Placebocontrolled study of inhaled budesonide on indices of airway inflammation in broncholaveolar lavage fluid and
ª 2009 Adis Data Information BV. All rights reserved.
101.
102. 103.
104.
105.
106.
107.
108.
109.
110.
111.
112.
113.
114.
bronchial biopsies in cross-country skiers. Respiration 2000; 67 (4): 417-25 Helenius I, Lumme A, Obase Y, et al. No effect of montelukast on asthma-like symptoms in elite ice hockey players. Allergy 2004 Jan; 59 (1): 39-44 World Anti-doping Agency [online]. Available from URL: http://www.wada-ama.org/en/ [Accessed 2007 Aug 21] Rundell KW. High levels of airborne ultrafine and fine particulate matter in indoor ice arenas. Inhal Toxicol 2003 Mar; 15 (3): 237-50 Katelaris CH, Carrozzi FM, Burke T, et al. Effects of intranasal budesonide on symptoms, quality of life, and performance in elite athletes with allergic rhinoconjunctivitis. Clin J Sport Med 2002 Sep; 12 (5): 296-300 Bousquet J, Khaltaev N, Cruz AA, et al. Allergic rhinitis and its impact on asthma (ARIA) 2008 update (in collaboration with the World Health Organization, GA(2)LEN and AllerGen). Allergy 2008 Apr; 563 Suppl. 86: 8-160 Anderson SD, Sue-Chu M, Perry CP, et al. Bronchial challenges in athletes applying to inhale a b2-agonist at the 2004 summer Olympics. J Allergy Clin Immunol 2006 Apr; 117 (4): 767-73 Crapo RO, Casaburi R, Coates AL, et al. Guidelines for methacholine and exercise challenge testing-1999. Am J Respir Crit Care Med 2000 Jan; 161 (1): 309-29 International Olympic Committee. Beta2 adrenoceptor agonists and the Olympic games in Turin [online]. Available from URL: http://multimedia.olympic.org/pdf/en_ report_981.pdf/ [Accessed 2007 Aug 21] Malo JL, Pineau L, Cartier A, et al. Reference values of the provocative concentrations of methacholine that cause 6% and 20% changes in forced expiratory volume in one second in a normal population. Am Rev Respir Dis 1983 Jul; 128 (1): 8-11 Thole RT, Sallis RE, Rubin AL, et al. Exercise-induced bronchospasm prevalence in collegiate cross-country skiers. Med Sci Sports Exerc 2001 Oct; 33 (10): 1641-6 Bosch A, Diez JM, Abad FX. Disinfection of human enteric viruses in water by copper: silver and reduced levels of chlorine. Water Sci Technol 1993; 27: 351-6 Cavas L, Tarhan L. Effects of vitamin-mineral supplementation on cardiac marker and radical scavenging enzymes, and MDA levels in young swimmers. Int J Sport Nutr Exerc Metab 2004 Apr; 14 (2): 133-46 Mickleborough TD, Murray RL, Ionescu AA, et al. Fish oil supplementation reduces severity of exercise-induced bronchoconstriction in elite athletes. Am J Respir Crit Care Med 2003 Nov; 168 (10): 1181-9 Baumann JM, Rundell KW, Evans TM, et al. Effects of cysteine donor supplementation on exercise-induced bronchoconstriction. Med Sci Sports Exerc 2005 Sep; 37 (9): 1468-73
Correspondence: Dr Louis-Philippe Boulet, Centre de Recherche de l’Institut Universitaire de Cardiologie et de Pneumologie de Que´bec, 2725 chemin Sainte-Foy, Que´bec, QC, Canada G1V 4G5. E-mail:
[email protected]
Sports Med 2009; 39 (4)
Sports Med 2009; 39 (4): 313-329 0112-1642/09/0004-0313/$49.95/0
REVIEW ARTICLE
ª 2009 Adis Data Information BV. All rights reserved.
The Placebo Effect in Sports Performance A Brief Review Christopher J. Beedie and Abigail J. Foad Canterbury Christ Church University, Canterbury, UK
Contents Abstract. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1. The Placebo Effect in Sport . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2. Findings of Intervention Studies . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.1 Ariel and Saville (1972) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.2 Maganaris et al. (2000) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.3 Clark et al. (2000) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.4 Foster et al. (2004). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.5 Porcari et al. (2006). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.6 Beedie et al. (2006) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.7 Beedie et al. (2007) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.8 McClung and Collins (2007) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.9 Kalasountas et al. (2007) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.10 Benedetti et al. (2007) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.11 Pollo et al. (2008) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.12 Foad et al. (2008) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3. Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Abstract
313 314 314 314 318 319 320 320 321 321 322 323 324 325 325 327
The placebo effect, with its central role in clinical trials, is acknowledged as a factor in sports medicine, although until recently little has been known about the likely magnitude and extent of the effect in any specific research setting. Even less is known about the prevalence of the effect in competitive sport. The present paper reviews 12 intervention studies in sports performance. All examine placebo effects associated with the administration of an inert substance believed by subjects to be an ergogenic aid. Placebo effects of varying magnitudes are reported in studies addressing sports from weightlifting to endurance cycling. Findings suggest that psychological variables such as motivation, expectancy and conditioning, and the interaction of these variables with physiological variables, might be significant factors in driving both positive and negative outcomes. Programmatic research involving the triangulation of data, and investigation of contextual and personality factors in the mediation of placebo responses may help to advance knowledge in this area.
The interaction of the mind and body has intrigued philosophers and scientists for centuries. Alternative models have prevailed at different times,
but current emphasis is on the unity of the two.[1-3] With the development in medicine of interdisciplinary fields such as psychoneuroimmunology,[4]
Beedie & Foad
314
research is demonstrating that the effects of an individual’s beliefs may in fact have some scientific basis.[5,6] One such belief is the placebo effect, a positive outcome resulting from the belief that a beneficial treatment has been received.[7] The placebo effect has an interesting history, one that exemplifies McGuire’s[8] model of the life of an artefact; first it is ignored, then its presumed contaminating effects are controlled for, finally it is studied in its own right. In medicine, the placebo effect has long been acknowledged, and has been controlled for in clinical trials for over 50 years. More recently, a substantial body of research has also studied the effect directly. For both a comprehensive review of this research and an interesting theory of the effect see, respectively, Price and co-workers[9] and Evans.[10] 1. The Placebo Effect in Sport Arguably mirroring the situation several decades ago in medical research, there is more speculation than hard evidence relating to the placebo effect in sport. The placebo effect has been implicated in the use of nutritional ergogenic aids,[11] anabolic steroids,[12] creatine-monohydrate,[13] vitamin E,[14] mandibular orthopaedic devices,[15] pre-competition anatomical manipulation,[16] sports hypnotism[17] and pre-competition fasting,[18] as well as in phenomena such as the runner’s high.[19] Accounts of deliberate or inadvertent use of the placebo effect by coaches or athletes have been published in autobiographical texts,[20] training manuals[21] and newspaper articles.[22] Anecdotal accounts of many examples of what might legitimately be described as placebo effects in sport are presented in a comprehensive overview of extraordinary human performance,[23] and a survey of athletes’ experiences.[24] The authors suggest that placebo effects accounted for observed effects in studies of, for example, carbohydrate feeding,[25] respiratory training devices,[26] cooling protocols,[27] fructose and glucose supplementation,[28] ice water immersion,[29] magnetic therapy,[30] knee surgery[31] and super-oxygenated water.[32] Such data support the idea that the placebo effect impacts on sports performance, although the ª 2009 Adis Data Information BV. All rights reserved.
empirical evidence required to move beyond speculation was in fact lacking until recently, with only one published study prior to 2000.[33] Since then, a further 11 experimental studies have been published.[7,32,34-42] This review focuses on the methods and findings of these studies. Data are reported as originally presented, whether in terms of statistical significance or magnitude-based inferences.[43] An indication of the relative magnitude of effects in terms of percentage change relative to baseline or control conditions is presented in table I. 2. Findings of Intervention Studies 2.1 Ariel and Saville (1972)
In 1972, and preceding further research by almost 30 years, Ariel and Saville[33] investigated the placebo effect of anabolic steroids. Fifteen experienced weightlifters (»5 sessions/week for »2 years) were recruited to a study of the effects of the oral anabolic steroid methandrostenolone. Baseline maximal strength data were collected for four tasks: bench press, military press, seated press and squat. Subjects were informed that they would receive methandrostenolone 10 mg/day for the duration of the study, and the likely positive effects of the drug on performance were described (a feature of placebo effect studies in sport and elsewhere is the catalysing or reinforcement of an expectation of the intervention via such a ‘belief intervention’). Six subjects received an inert placebo throughout the study. Strength data for these six subjects were collected for two 4-week periods, the ‘pre-placebo period’ in which no intervention was administered, and the ‘placebo period’ in which a placebo capsule was administered. Subjects exhibited strength gains over baseline in the pre-placebo period (3.4%, 0.8%, 2.7% and 2.0% for bench press, military press, seated press and squat, respectively), and again in the placebo period (9.6%, 8.5%, 6.2% and 13.8%, respectively). Change scores between pre-placebo and placebo periods reached statistical significance at p < 0.05 in all but the seated press. The authors summarized by stating that significantly greater strength gains were exhibited when subjects believed that they were ingesting Sports Med 2009; 39 (4)
Authors (y)
Sample size
Sample characteristicsa
Design
% change
Performance measure
Intervention informed
received
Ariel and Saville[33] (1972)
6
Sub-elite weightlifters
Within-subjects design
Strength (bench press, military press, seated press, squat)
Anabolic steroid
Placebo
9.5
Maganaris et al.[39] (2000)
11
Sub-elite weightlifters
Betweensubjects design
Strength (bench press, dead lift, squat)
Anabolic steroid
Placebo
3.8
Anabolic steroid then placebob
Placebo
1.7
Carbohydrate
Placebo (50% of subjects), carbohydrate (50% of subjects)
4.3
Placebo
Placebo (50% of subjects) carbohydrate (50% of subjects)
0.5
50/50 chance of receiving carbohydrate or placebo
Placebo (50% of subjects), carbohydrate (50% of subjects)
-1.1
Clark et al.[7] (2000)
43
Sub-elite endurance cyclists
Betweensubjects Latinsquare design (6-cell)
Endurance (40 km cycling power)
Overall placebo effect
3.8
16
Sub-elite runners
Within-subjects design
Endurance (5 km running time)
New ergogenic aid
Placebo
1.1
Porcari et al.[32] (2006)
32
Sub-elite runners
Betweensubjects design
Endurance (5 km running time)
Superoxygenated water
Placebo
8.0
Continued next page
315
Sports Med 2009; 39 (4)
Foster et al.[37] (2004)
The Placebo Effect in Sport
ª 2009 Adis Data Information BV. All rights reserved.
Table I. Characteristics and findings of placebo effect research in sports performance
Authors (y)
Beedie et al.[34] (2006)
McClung and Collins[40] (2007)
Beedie et al.[35] (2007)
Kalasountas et al.[38] (2007)
316
ª 2009 Adis Data Information BV. All rights reserved.
Table I. Contd Sample size 6
16
43
42
Sample characteristicsa
Design
Sub-elite cyclists
Within-subjects design
Sub-elite endurance athletes
Sub-elite athletes
Betweensubjects design
Betweensubjects design
Intervention informed
received
Endurance (10 km cycling power)
0 mg/kg caffeine
Placebo
-1.4
4.5 mg/kg caffeine
Placebo
1.3
9.0 mg/kg caffeine
Placebo
3.1
Overall placebo effect
2.2
Sodium bicarbonate
Sodium bicarbonate
1.7
Sodium bicarbonate
Placebo
1.5
No treatment
Sodium bicarbonate
No treatment
No treatment
0.0
Overall placebo effect
1.8
Positive ergogenic aid
Placebo
0.0
Negative ergogenic aid
Placebo
-1.6
Strength (bench press, seated leg press)
Amino acids
Placebo
19.6
Strength (bench press, seated leg press)
Amino acids then placebob
Placebo
6.3
Endurance (1000 m running time)
Anaerobic (30 m running speed)
-0.3
Continued next page
Beedie & Foad
Sports Med 2009; 39 (4)
Untrained students
Within-subjects Latin square/ balanced placebo design (4-cell)
% change
Performance measure
Authors (y)
Benedetti et al.[41] (2007)
Pollo et al.[42] (2008)
Sample size
Sample characteristicsa
Design
40
Sub-elite athletes
Mixed design
44
Sub-elite athletes
Mixed design
Intervention informed
Pain tolerance
No treatment Morphine Morphine (after conditioning procedure) Morphine (after conditioning procedure)
No treatment Placebo Placebo
Caffeine
Placebo
11.8
Caffeine (after conditioning procedure)
Placebo
22.1
Strength (leg extension)
Perceived fatigue
Foad et al.[36] (2008)
14
Sub-elite cyclists
Within-subjects Latinsquare/balanced placebo design (4 cell)
% change
Performance measure
Endurance (40 km cycling power)
received
Naloxone
7.5 17.6 50.7
6.2
Placebo
-0.3
Placebo
-7.8
Caffeine
Caffeine
2.3
Caffeine
Placebo
0.1
No treatment
Caffeine
No treatment
No treatment Overall placebo effect
2.9 -1.9 0.7
a
Subject classifications are derived from their descriptions in the original papers: untrained = no regular training; sub-elite = regularly training but not above national status; elite = international status.
b
Subjects were initially informed that they were receiving the drug. Halfway through the trials, subjects were correctly informed that they were receiving a placebo.
317
Sports Med 2009; 39 (4)
Caffeine Caffeine (after conditioning procedure)
The Placebo Effect in Sport
ª 2009 Adis Data Information BV. All rights reserved.
Table I. Contd
Beedie & Foad
318
methandrostenolone than when they believed they were not. No inferences regarding the mechanisms underlying the increases in performance during the placebo period – e.g. increased effort or motivation – were offered. They concluded that future investigators must be cautious when assessing the effects of ergogenic aids on performance because the assumption that the dependent measure has been isolated may well be erroneous. This may be the case, although Ariel and Saville[33] did not test this assumption on subjects blind to treatment – the design most commonly employed in ergogenics research. The effects reported by Ariel and Saville[33] were substantial and, in all but one comparison, statistically significant. These data are surprising given the relatively small improvement that would be expected from subjects reported to be highly trained and experienced weightlifters. Expectation of the likely effects of methandrostenolone might have been high, and these might have resulted in increased motivation and perhaps a greater than usual vigilance to training and recovery. Such responses, although behavioural, satisfy the definition of a placebo effect above. They might also have augmented more apparently physiological placebo responses of the type reported elsewhere,[9,10,34] such as enhanced pain tolerance and fatigue resistance. It is possible that subjects did not perform to volitional maximum at baseline (an issue discussed below), and on this basis the degree to which the observed effects were driven by the experimental manipulation could be questioned. It is unfortunate that data for experimental subjects who did receive methandrostenolone were not reported, as this would have facilitated direct comparison of the magnitude of placebo and drug effects. Furthermore, given the improvements observed in the pre-placebo period, a control group for the placebo period was warranted. 2.2 Maganaris et al. (2000)
The work of Ariel and Saville[33] is frequently cited in the strength and conditioning, weightlifting and body-building media. While studies were subsequently published in exercise[44] and ª 2009 Adis Data Information BV. All rights reserved.
sports-related surgery,[31] it was over 25 years before another empirical study of the placebo effect on sports performance was published. Maganaris et al.[39] investigated the deceptive administration of a placebo anabolic steroid among 11 national-level power lifters. The authors used a more complex design than had Ariel and Saville[33] and proposed two hypotheses: firstly, that subjects would show substantial increases in performance, and secondly, that when the deception was revealed, performance would return to baseline. Baseline data were collected in competitive conditions for the bench press, dead lift and squat. One week later, and prior to the first experimental trial, subjects were administered the placebo but informed that they were receiving a fast-acting anabolic steroid. Mean percentage improvements in maximal weight lifted over baseline were 3.5%, 4.2% and 5.2% for the bench press, dead lift and squat, respectively (p < 0.01). Subjects were given another dose to take during the following week. One week later, and prior to a second experimental trial, all subjects reported improved training performance since taking the tablets. At this stage, however, six subjects were correctly informed of the experimental deception. In the second experimental trial, while improvements over baseline were maintained in the group who believed they had ingested steroids (3.2%, 4.0% and 4.4%, respectively; p < 0.01), performances of the six subjects correctly informed of the deception were reduced significantly (1.7%, -0.4% and 0.4%, respectively relative to baseline). In fact the authors described the performance of the second group as having returned to baseline. Maganaris et al.[39] discussed how subjects had ‘‘picked up on the street reputation of anabolic steroids’’ (p. 277) and that the expectancy driven by this reputation generated substantial improvements in performance that were reversed once this expectancy was removed. The authors suggested that evidence indicative of a substantial placebo effect associated with a treatment widely considered to be ergogenic could be used in antidoping initiatives. They added that their investigation might have been more convincing had they used a Latin-square design in which steroids Sports Med 2009; 39 (4)
The Placebo Effect in Sport
had been administered alongside the placebo. The Latin square design, also referred to as the balanced placebo design,[45] commonly comprises four conditions: (i) inform drug/receive drug; (ii) inform drug/receive placebo; (iii) inform no-treatment/receive drug; and (iv) inform placebo/receive placebo. This design facilitates assessment of the independent effects of placebo and pharmacology, and their interactions, and would have enabled Maganaris et al.[39] to better assess both the placebo and pharmacological effects of steroid supplementation. While the investigations of Ariel and Saville[33] and Maganaris et al.[39] were arguably limited in terms of sample size – and in the former, the lack of a suitable control group with which to differentiate ‘pure’ placebo effects from other factors – the effects were substantial for weightlifters. In fact Maganaris et al. stated that all but one subject would have gained international status as the result of the intervention. The authors of both studies suggested that subjects in strength-based sports expect certain drugs to enhance performance, and thus a component of such enhancements may be placebo driven. 2.3 Clark et al. (2000)
In the same year as Maganaris et al.[39] published their findings, Clark et al.[7] published an investigation of the placebo and real effects of carbohydrate supplementation on 40 km cycling performance among 41 male and two female competitive cyclists. The authors used a somewhat more complex balanced repeated-measures design than had previous studies. Also unlike previous studies, the authors administered an active substance alongside the placebo. To encourage positive beliefs about the intervention, subjects were advised that those who received carbohydrate would probably show an improvement in performance compared with those who received the placebo. Baseline data were collected, and then 1 week later subjects performed an experimental time trial during which they consumed a drink containing non-caloric sweetener either with or without carbohydrate. Subjects were randomized to one of three treatment ª 2009 Adis Data Information BV. All rights reserved.
319
groups: (i) informed carbohydrate; (ii) informed non-caloric sweetener; and (iii) informed 50/50 chance of receiving carbohydrate. Without their knowledge, half of the subjects in each group were further randomized to receive carbohydrate, while the others received placebo. The experimenters were thus able to analyse the effects of six different combinations of placebo and nutritional supplementation. Informed carbohydrate subjects showed an improvement in power over baseline (4.3 – 4.8%); this improvement was, surprisingly, greater for those who received the placebo than for those who received the carbohydrate. Informed 50/50 subjects showed little change in performance (-1.1 – 8.5%) compared with informed placebo subjects (0.5 – 5.8%), irrespective of the actual substance administered. The authors estimated that the placebo effect, calculated as the change for the informed carbohydrate group minus the change for the informed placebo group, was 3.8% (7.9 to -0.2%; p = 0.06), and that the real effect of carbohydrate, calculated as change in power for carbohydrate minus the change in power for placebo, was a slight reduction in power of 0.3% (4.4 to -3.8%; p = 0.87). The coefficient of variation for the informed 50/50 group was 1.6 times larger than the combined coefficients of variation of the other two groups. The authors speculated that uncertainty about the treatment caused some subjects to make a greater effort than at baseline, whereas others resigned themselves to poorer performance. Clark et al.[7] argued that the existence of a placebo effect with a sham treatment, and the existence of individual differences with a blind treatment, both imply that at least some subjects do not perform at volitional maximum in performance research. They speculated further that, if in competition these athletes perform at a higher percentage of volitional maximum, the effect of a treatment in a laboratory test might be substantially different from the effect of the same treatment in competition. They suggested that a treatment might operate in the zone between submaximal and maximal effort in a performance test, a margin that may be substantially reduced or even absent in competition. The authors, as Sports Med 2009; 39 (4)
Beedie & Foad
320
had Ariel and Saville[33] before them, advised caution in extrapolating enhancements observed in the laboratory to the real world. Clark et al.[7] made several methodological recommendations in relation to placebo effect research in sport, for example the use of Latinsquare designs and assessment of personality towards better understanding placebo mechanisms. While the first of these recommendations has been heeded by several investigators,[36,40] the role of personality in placebo responses has only just begun to be explored,[46] and remains a promising avenue for future research.
2.4 Foster et al. (2004)
In 2004, Foster et al.[37] investigated the effects of a placebo treatment on 5 km running performance. The study was presented at a conference and it has not been published in full text in a peer-reviewed journal. Sixteen well trained and task-habituated . recreational runners (VO2peak = 58 – 8 mL/kg/min) were recruited to a study of a ‘new’ ergogenic aid. The authors showed subjects a video designed to promote the value of the substance in endurance performance. Subjects performed random-ordered 5 km time trials after consuming either normal water or water falsely purporting to contain the ergogenic aid. Measures included total time, lap times, ratings of perceived exertion (RPE), heart rate and blood lactate. Competitively meaningful differences were observed in 5-km time trial performance between the control and placebo water conditions (21 : 54 vs 21 : 40; p = 0.11), with 12 of the 16 subjects running faster when they believed they had ingested the ergogenic aid. The authors also noted a competitively meaningful mean 2.5-second improvement in performance over the final 400 m when subjects believed they had ingested the ergogenic aid. No significant differences were observed between conditions in RPE (8.2 – 1.0 vs 8.4 – 1.2), peak heart rate (177 – 5 vs 177 – 6 beats/min) or blood lactate (12.2 – 3.2 vs 11.4 – 2.2 mmol/L). The authors concluded that while they were not statistically significant, the pattern of observed effects was clear enough to warrant further research. ª 2009 Adis Data Information BV. All rights reserved.
2.5 Porcari et al. (2006)
In a follow-up conference abstract, Porcari and co-workers[32] reported the findings of an investigation of the placebo effect in 5 km running performance. This study expanded on a previous paper[47] that reported no differences between ‘super-oxygenated’ water and placebo in several variables (e.g. heart rate, blood lactate, ratings of perceived exertion) associated with performance. Thirty-two experienced runners ranging from recreational to competitive completed an exercise . test to measure fitness level (VO2max = 60.8 – 8.2 mL/kg). Subjects then watched a video suggesting the ergogenic qualities of super-oxygenated water. Each subject subsequently ran three 5 km time trials on an indoor 200 m track (1 · habituation and 2 · counterbalanced experimental). Runs were completed at least 3 days apart. Prior to experimental trials, subjects drank either 475 mL of bottled water that was correctly identified as such, or water that they were told was super-oxygenated. Measures were total time, heart rate, RPE and blood lactate. Significant differences between control (water) and experimental (placebo) trials for total time (21:04 – 3:34 vs 19:41 – 2:32) were reported. No significant differences in heart rate (data not presented), RPE (7.7 – 1.4 vs 7.7 – 1.2) or lactate (9.8 – 3.9 vs 10.2 – 3.7 mmol/L) were reported. Twenty-seven of 32 subjects (84%) ran faster when they believed they had received the super-oxygenated water. The authors reported that the observed improvement over baseline in the experimental conditions was largely attributable to the performances of less accomplished runners (2:22 minutes as opposed to 0:28 minutes for the more accomplished runners). In describing the study on the American Council for Exercise website,[48] one of the authors, Otto, reported that several of the less accomplished subjects claimed that they ‘felt lighter on their feet’ and wanted to know where they could buy the product, while more experienced runners asserted that they didn’t feel any different after the run and ‘didn’t think that stuff works’. This hint of a relationship between status and placebo responsiveness has been alluded to elsewhere.[7,34] Sports Med 2009; 39 (4)
The Placebo Effect in Sport
2.6 Beedie et al. (2006)
In 2006, Beedie et al.[34] examined the possibility of a dose-response relationship to placebos presented as ‘zero-’, ‘low-’ and ‘high’-dose caffeine among seven well-trained competitive cyclists. Measures in the first phase of the study were power, heart rate, oxygen uptake and blood lactate. Subsequently, qualitative data were derived through interview. Subjects were provided with literature reviewing research findings in caffeine and cycling performance and detailing anecdotal evidence of the use of caffeine amongst elite cyclists. Following habituation and baseline trials, subjects were informed that, over three experimental trials, they would receive a placebo, caffeine 4.5 and 9.0 mg/kg double-blind and randomly assigned. However, a placebo was administered in all experimental conditions. Post-experimental baseline trials were also conducted. One subject failed to complete one trial and was excluded from further statistical analysis. The authors reported their findings in terms of magnitude-based inference.[43] A likely trivial increase in mean power of 1.0% (-1.4% to 3.6%) over baseline was associated with experimental trials, rising to a likely beneficial 2.2% (-0.8% to 5.4%) increase in power associated with experimental trials in which subjects believed they had ingested caffeine. A dose-response relationship was evident in experimental trials, with subjects producing 1.4% (-4.6% to 1.9%) less power than at baseline when they believed they had ingested a placebo, 1.3% (-1.4% to 4.1%) more power than at baseline when they believed they had ingested caffeine 4.5 mg/kg, and 3.1% (0.4–6.7%) more power than at baseline when they believed they had ingested caffeine 9.0 mg/kg. The authors concluded that when subjects were administered a placebo capsule believing it to be caffeine, their performance was substantially enhanced. They noted that the effects were similar in magnitude to those associated with the administration of caffeine reported elsewhere. Of further interest was the fact that no substantial differences in any measured physiological variables between baseline and experimental conditions were observed, ª 2009 Adis Data Information BV. All rights reserved.
321
suggesting that the mechanism underlying the observed effects was not a substantial change in effort. Wishing to investigate the potential mechanisms of the observed effects, Beedie et al.[34] conducted follow-up interviews with each subject. Interviews were conducted in two parts, the first before revealing the experimental deception to the subject, the second afterwards. Interview data were reported for all seven original subjects. Data were consistent with the performance of some subjects but less so with others: five subjects believed that they had experienced a placebo effect in one or more of the three experimental trials, and proposed mechanisms such as pain reduction, fatigue resistance, changes in strategy and reduced arousal. One subject reported experiencing a substantial negative effect on performance that he attributed to the high dose of caffeine. The two subjects who reported the least confidence in having experienced a placebo effect produced the highest mean power overall, and the subject who produced the lowest mean power overall reported arguably the largest and least ambiguous placebo effect. The authors suggested that the findings supported the previously proposed relationship between training status and placebo responsiveness.[7,32] 2.7 Beedie et al. (2007)
Having observed potentially negative placebo (nocebo) effects associated with the administration of caffeine,[34] Beedie et al.[35] designed a study to investigate whether, following ingestion of a placebo ergogenic aid, subjects who possessed positive beliefs about the substance would perform to a higher level than subjects who had negative beliefs about the substance. Forty-two team sports athletes were randomly allocated to one of two groups, positive belief and negative belief. Both groups were informed that they would be completing a 30 m repeat-sprint protocol in two conditions, baseline and experimental. They were further informed that prior to the experimental condition they would be administered a new ergogenic aid. To minimize the potential Sports Med 2009; 39 (4)
Beedie & Foad
322
for experimenter effects, a highly experienced researcher not associated with the original research project was responsible for delivering the intervention and managing the data collection process. The two groups were also isolated from one another to eliminate the possibility of any spillover. Subjects performed 3 · 30 m sprints with 2 minutes recovery between each. They were then administered a placebo capsule. Each group was provided with a different description of the placebo: the positive belief group that the substance had been found to enhance both repeat sprint and endurance performance in team sport players, the negative belief group that the substance had been found to enhance endurance performance while having a negative impact on repeat sprint performance. Twenty minutes subsequent to the intervention, subjects undertook experimental trials in an identical manner to the baseline trials. The speed of both groups diminished progressively in successive baseline trials. In experimental trials, however, while the trend towards reduced speed in consecutive trials continued for the negative belief group, mean speed per trial for the positive belief group increased. Although no change in mean speed from baseline to experimental trials was evident for the positive belief group (p = 0.96), a significant linear trend of greater speed with each successive experimental trial suggested that positive belief exerted a positive impact on performance (p < 0.01). Data for the negative belief group indicated that they ran on average 0.08 seconds (1.7%) slower than baseline in experimental trials (p = 0.01). Furthermore, mean delta score (baseline to experimental) for the positive belief group was 0.00 (standard deviation [SD] = 0.09), while that for the negative belief group it was 0.09 (SD = 0.10) and differed significantly (p = 0.01). The authors concluded that both positive and negative beliefs were associated with placebo effects of opposite polarity that significantly affected performance. Beedie et al.[35] also investigated interindividual variability in the placebo response. In the positive belief group, 26% of individual performances fell outside individual 95% limits of agreement, and 50% of these were faster than the ª 2009 Adis Data Information BV. All rights reserved.
upper limit. In the negative belief group, 33% of performances fell outside the limits of agreement, and all of these were slower than the lower limit. The authors concluded that both subjects’ beliefs about whether or not they have ingested a substance and subjects’ beliefs about the potential efficacy of that substance influence placebo responding. They also speculated that, if a negative belief about a placebo treatment exerts a negative impact on performance, negative beliefs about a legitimate treatment could offset some percentage of the beneficial pharmacological or physiological effects of that treatment. 2.8 McClung and Collins (2007)
In a further study of running performance, McClung and Collins[40] used a Latin-square design to evaluate the physiological and psychological effects of sodium bicarbonate among 16 track athletes (12 men and four women). The authors reported extensive piloting prior to the study. Subjects ran 5 · 1000 m time trials, one habituation and one trial per counterbalanced condition of (i) informed drug/received drug, (ii) informed drug/received placebo, (iii) informed no-treatment/received drug, and (iv) informed no-treatment/received no-treatment. Measures were time, RPE, blood lactate and heart rate (no heart rate data or analyses were reported). The authors hypothesized that not only would subjects who received sodium bicarbonate run faster and report a lower RPE than in conditions in which they did not receive the drug, but that the expectation of receipt of the drug in the informed drug/received placebo condition would result in improved performance and lower RPE than in the informed no-treatment/received no-treatment condition. McClung and Collins[40] informed subjects that the study was to examine the effects of sodium bicarbonate and a new additive that would reduce gastric discomfort associated with sodium bicarbonate. The authors used this information to explain why, before the informed notreatment/received sodium bicarbonate condition, subjects had to ingest a lemon-flavoured drink. That is, subjects were informed that the strong, Sports Med 2009; 39 (4)
The Placebo Effect in Sport
lemon-tasting drink was the additive, and that the trial in question was a test of the additive alone. Sodium bicarbonate was in fact deceptively administered in this solution. Results of a 2 · 2 (drug · belief) ANOVA with final time as the dependent variable indicated a statistically significant main effect of belief (p < 0.001). No statistically significant main effect for drug, or interaction between drug and belief, was observed. Similar effects were observed with RPE as the dependent variable. In relation to lactate, although the authors reported that pre-trial lactate was significantly lower in the two experimental conditions in which sodium bicarbonate was administered than in the two in which it was not – a factor that they interpret as suggestive of the efficacy of sodium bicarbonate as a lactate buffer – they did not discuss the posttrial lactate data presented in their table, which are somewhat less easy to interpret. McClung and Collins[40] summarized by stating that not only does the overt administration of sodium bicarbonate improve performance by a competitively meaningful degree over notreatment (1.7%), but that the expectation of receiving sodium bicarbonate improves performance in the absence of that substance by a not dissimilar amount (1.5%). These findings, they suggested, hint at the possibility that some of the well documented benefits of sodium bicarbonate may be gained through expectancy effects alone. The authors noted the lack of a performance effect when subjects had ingested sodium bicarbonate but believed that they had not, suggesting what they termed a biochemical ‘failure’. However, in this experimental condition, subjects believed that they had ingested the ‘new additive’ and might have suspected that this could have an effect on performance. A valid test of the biological effects of sodium bicarbonate uncontaminated by psychological factors would require that subjects believe they have received no treatment at all. Post-intervention manipulation checks suggested that four subjects were suspicious that the study was not what it appeared to be, although these subjects could not explain which aspect they were suspicious of. The authors concluded by reiterating the suggestion of Maganaris ª 2009 Adis Data Information BV. All rights reserved.
323
and colleagues[39] that evidence of a placebo effect of an ergogenic treatment provides a strong argument against the use of performance-enhancing drugs and might therefore contribute to educational anti-doping strategies. 2.9 Kalasountas et al. (2007)
Again in 2007, and citing the work of Maganaris et al.[39] as a catalyst, Kalasountas et al.[38] examined placebo effects on the weightlifting performance of college students. The authors tested the same hypotheses as had Maganaris et al.: (i) that subjects who received a placebo ergogenic aid would show greater increases in performance than controls; and (ii) that the performance of subjects informed that the substance is no longer effective will return to control levels. Forty-two subjects were randomly allocated to one of three groups of 14: two experimental (placebo/placebo and placebo/no-placebo) and control. Subjects in the placebo/placebo group received a placebo ergogenic aid during both experimental trials, while subjects in the placebo/no-placebo group received a placebo in the first experimental trial only. Controls did not receive the placebo in either trial. Subjects were requested not to engage in weight-training activity other than that required for the experiment and to cease use of dietary supplements for 10 days prior to and during the study. Five trials were conducted. The first three were baseline, each trial separated by 48 hours. Subjects performed single lifts on the bench press and a seated leg press, performing one lift per minute and increasing the resistance until a repetition could not be completed with correct form. Resistance on the final completed attempt was recorded as the maximum. Experimental trials were carried out the following week and were also separated by 48 hours. Prior to the first trial, subjects in both experimental groups were given placebo tablets and informed that the substance was a combination of amino acids likely to produce immediate strength effects. Two more tablets were given 8–10 minutes after the trial. In the second experimental trial, the same process was followed for placebo/placebo subjects while Sports Med 2009; 39 (4)
Beedie & Foad
324
placebo/no-placebo subjects were provided with negative information about the substance and informed that no tablets would be administered. Subsequently, subjects in both experimental groups were interviewed about their beliefs regarding the effectiveness of the pill. All were then informed of the true nature of the study. In the first experimental trial, both experimental groups improved significantly over controls on both measures (p < 0.01). In the second experimental trial, revealing the deception to the placebo/no-placebo group resulted in performance on both measures dropping to a level not significantly different from controls (p > 0.05). Performance of controls did not improve from baseline during the experimental period. The authors suggested on this basis that placeboassociated expectancy played a significant part in the observed performance. They speculated that, because participants were largely untrained, alterations in neurobiological factors, set in motion by expectancies, may offer some explanation of the underlying mechanisms. Follow-up interviews revealed that 67% and 56% of subjects in the placebo/placebo and placebo/no-placebo groups, respectively, reported positive expectations resulting from consuming the pill, 75% and 56% reported increased vigour and energy levels after taking the pill, and 58% and 56% reported feeling better the day between experimental trials. No subjects reported feeling worse after taking the pill, although 67% of the placebo/no-placebo group reported feeling disappointed, less enthusiastic or that their performance suffered as a result of the negative news about the supplement. The authors suggested that their results generally supported those of Maganaris et al.[39] In fact, they indicated that their use of a control group was an advance on the latter’s method, although in both studies the subjects arguably acted as their own controls. Kalasountas et al.[38] noted that their second hypothesis was only partially supported. That is, although force outputs declined on average in the placebo/no-placebo group in the second experimental trial, they did not reach baseline levels, suggesting that a placebo intervention subsequently revealed might still exert a positive ª 2009 Adis Data Information BV. All rights reserved.
effect, or alternatively that the initial placebo intervention resulted in increased motivation in the first trial and a subsequent greater training effect carried through to the second. Expressing a similar sentiment to Maganaris et al.[39] and McClung and Collins,[40] Kalasountas et al.[38] concluded that, as corticosteroid use is on the rise among adolescents, their study could serve as a starting point for coaches and teachers in educating young persons about the risks of doping and using ineffective nutritional supplements, while encouraging them to concentrate on the psychological aspects of enhancing performance. In relation to performance interventions, the authors speculated that manipulations resulting in the positive performance outcomes in the placebo/placebo group could possibly serve as an appropriate intervention to assist amateur lifters and fitness enthusiasts though performance slumps, or as a method to demonstrate the importance of psychological factors in successful performance. 2.10 Benedetti et al. (2007)
The question of whether placebo responses could or should be used to enhance performance in training and competition was raised in two studies by Benedetti and colleagues.[41,42] In the first of these studies, Benedetti et al.[41] investigated the placebo analgesic effects of morphine on a pain endurance test designed to simulate sport competition. Morphine or placebo (saline or naloxone) was administered to 40 recreationally active males in a randomized, double-blind design. During pre-competition training, teams A and B received no pharmacological substance; teams C and D were trained with morphine. During competition, team A received no treatment while teams B and C were given placebo morphine 1 hour before competition. Team D also received a placebo and was told that it was morphine; however, they actually received naloxone, an opioid antagonist. Subjects had a tourniquet wrapped around their forearm and were required to repeatedly squeeze a hand spring exerciser until they could no longer continue. The time before stopping was recorded and team Sports Med 2009; 39 (4)
The Placebo Effect in Sport
averages calculated. The largest placebo effect was seen in team C who received the morphine preconditioning (p < .001). Naloxone negated the morphine preconditioning effects in team D indicating the activation of endogenous opioids after placebo administration. A correlation between morphine and placebo was, however, still present after naloxone treatment, suggesting the possible contribution of non-opioid mechanisms. The placebo analgesic responses were obtained after two morphine administrations that were separated as long as 1 week from each other. These long time intervals indicate that pharmacological conditioning procedures have longlasting effects, with potentially interesting implications for the use of drugs in training and competition. That is, could the placebo response be used to enhance performance in competition, and if so, would it be ethically acceptable to do so?
325
The two studies by Benedetti and colleagues[41,42] indicate that either pharmacological or non-pharmacological conditioning procedures can be effective in eliciting a placebo response. The authors suggested that these procedures could be employed in the field; for example, athletes could be pre-conditioned with a performance-boosting drug and then given a placebo prior to competition to avoid illegal drug administration on competition day. These studies therefore raise important and timely questions: are such procedures legally and ethically acceptable ways to enhance performance, or should they be considered as doping? As the authors noted, if such procedures were performed, many illegal drugs in sport would be neither discoverable nor would they violate the anti-doping rules. 2.12 Foad et al. (2008)
2.11 Pollo et al. (2008)
In a subsequent study involving two of the previous three authors, Pollo et al.[42] investigated the effects of an ergogenic placebo on quadriceps muscle performance and perceived fatigue. Fortyfour recreationally active males were divided into four groups, two control and two placebo (n = 11). In the first experiment, a placebo was deceptively administered with the suggestion that it was a high dose of caffeine. This resulted in a significant increase in mean muscle work (11.8 – 16.1%, p < .01) but no perceived decrease in muscle fatigue (p > .05). In the second experiment, placebo caffeine administration was accompanied by a conditioning procedure whereby the weight to be lifted was surreptitiously reduced. The load was then restored to the original weight and placebo caffeine administered again. Compared with the first experiment, the placebo effect was larger, with a significant increase in muscle work (22.1 – 23.5%, p < .01) and a decrease in perceived muscle fatigue (-7.8 – 10.1, p < .01). These findings, the authors suggested, indicate a central mechanism of top-down modulation of the global performance of muscles by placebos, and underscore the role of learning in the placebo response. ª 2009 Adis Data Information BV. All rights reserved.
Using a design similar to the Latin squares design employed by McClung and Collins,[40] Foad et al.[36] used the balanced placebo design[45] to examine the placebo and pharmacological effects of caffeine in cycling performance. Fourteen well trained competitive cyclists were informed that they were participating in a study examining the effects of caffeine on 40 km laboratory cycling performance. The authors reported piloting several aspects of the design before the experimental phase. Subjects performed two 40 km time trials in each of four experimental conditions: (i) informed caffeine/received caffeine; (ii) informed no-treatment/received caffeine; (iii) informed caffeine/received placebo; and (iv) informed no-treatment/received notreatment. Trials were conducted once per week per subject. No feedback other than distance covered was provided to subjects during trials. Measures were power, oxygen uptake, blood lactate and heart rate. To avoid alerting subjects to potential changes in subjective symptoms during trials, the authors chose not to measure RPE, although they acknowledged this as a potential limitation. Caffeine was administered in a chilled saline solution that had been shown in a pilot study to mask the taste of caffeine. Subjects Sports Med 2009; 39 (4)
326
were informed that the saline solution was administered to maintain hydration. In the two conditions in which caffeine was administered, it was administered in this solution. In the two conditions in which subjects were informed they were receiving caffeine, a placebo capsule was administered with the saline solution to maintain this belief. The authors reported their findings in terms of magnitude-based inference.[43] A very likely beneficial main effect on mean power of receiving caffeine (3.5 – 2.0%), and a possibly beneficial main effect of being informed of caffeine (0.7 – 1.4%), was observed. A substantial interaction between belief and pharmacology (2.6 – 3.3%) indicated that caffeine exerted a greater effect on performance when subjects were informed that they had not ingested it (a similar finding to Clark et al.[7] in relation to carbohydrate), while belief exerted a greater influence on performance in the absence of caffeine, a finding counter to the greater effect of belief in the presence of the active substance reported by McClung and Collins.[40] A possibly harmful nocebo effect relative to baseline was present when subjects were correctly informed that they had ingested no caffeine (-1.9 – 2.2%). No substantial changes relative to baseline were observed in mean heart rate, although clear and substantial increases in blood lactate were evident following the receipt of caffeine. Data for mean oxygen uptake were unclear. The authors reported that the within-subject coefficient of variation (CV) for power in deceptive conditions at 2.8% was 1.7 times larger than the CV when subjects were truthfully informed that they were receiving caffeine, indicating the possibility of some disparity between internal sensations and instructions amongst some subjects. This finding adds to that of Clark et al.,[7] who reported that the CV for their not-informed group was 1.6 times larger than for informed subjects. Both ratios suggest that either a lack of information or a disparity between information and experience might reduce the reliability of experimental trials. In summarizing, Foad et al.[36] suggested that their data supported the ergogenic efficacy of ª 2009 Adis Data Information BV. All rights reserved.
Beedie & Foad
caffeine and argued that such an improvement is highly likely to be worthwhile to a competitive cyclist. They suggested that, consistent with the findings of Beedie et al.,[35] both positive and negative expectations likely impact on performance and that this effect might vary between individuals. In considering their findings in the context of previous research, the authors noted the failure to observe a clear placebo effect in the informed caffeine/received placebo condition. Certainly, as the authors argued, given that subjects produced greater power in that condition than in the informed no-treatment/received notreatment condition, a substantial placebo effect could be inferred. Nonetheless, performance in the former condition was only marginally better than at baseline, suggesting that, in the absence of caffeine, the negative effect of negative belief on performance was somewhat more substantial than the positive effect of positive belief. In light of their data and of previous findings, Foad et al.[36] speculated that placebo/nocebo effects might operate somewhat differently in the presence of an active substance than in its absence. They added that, in some cases, the placebo and biological effects of a substance might share the same space, i.e. the potential to improve performance from sub-maximal to maximal, an argument made by Clark et al.[7] and arguably supported by the effects reported by McClung and Collins.[40] They concluded that, all other things being equal, the placebo effect observed in a study in which an active substance is administered and in which beliefs are also manipulated, might be somewhat different in magnitude to the placebo effect observed in a study in which only the beliefs are manipulated and no active substance is administered. In discussing their findings, Foad et al.[36] also addressed the issue of expectancy. They contrasted their design with previous research in which subjects have been unsure as to whether they would receive caffeine or a placebo. Although it has often been suggested in the literature that uncertainty about treatment allocation likely reduces the magnitude of observed effects,[49] the authors cited recent research that suggests that a degree of uncertainty might in fact Sports Med 2009; 39 (4)
The Placebo Effect in Sport
be required to elicit a placebo effect. For example, Fiorillo and co-workers[50] demonstrated that placebo-induced dopamine activation is maximal when the probability of experiencing a beneficial outcome is 0.5. The authors suggested that this somewhat counter-intuitive idea may lend support, in sports performance at least, to the idea suggested by Clark et al.[7] that the placebo effect might be more of a factor in laboratory research than in the real world. Arguably the main finding of Foad et al.[36] was that caffeine exerted an ergogenic effect whether subjects believed it had been ingested or not. Still, the observed interactions between pharmacology and psychology are of interest. Furthermore, the finding that caffeine exerted a greater effect in trials in which subjects were falsely informed they had not ingested it than in trials in which they were correctly informed that they had is as counter-intuitive as that of Clark et al.,[7] who reported that subjects produced more power in the informed carbohydrate/received placebo condition than in the informed carbohydrate/received carbohydrate condition. Although possibly anomalous, these findings warrant further investigation. 3. Conclusion This review addresses 12 intervention studies from the sports literature. Of these, one was published in 1972, and the remaining 11 were published after 2000. It is evident that systematic research has been a feature of only the last few years. In six studies the dependent variable was endurance performance, in four strength performance, in one, anaerobic performance and in one, pain tolerance. Over and above performance data, several studies report physiological or psychological data. Both positive and negative placebo effects on performance were reported, with magnitudes varying from -1.9% to 50.7% of baseline/control performance, the majority falling between 1% and 5%. All but one study reported either a statistically or clinically significant effect. Several authors have suggested potential applications of their findings. As is the case in ª 2009 Adis Data Information BV. All rights reserved.
327
medicine, use of the placebo effect by practitioners might prove ethically problematical; issues of trust between practitioner and client, or between scientist and subject, should be paramount. Beyond these issues, given the evidence for nocebo responses above, the assumption that such practical application would always elicit positive results is questionable. Thus, the question as to whether placebo responsiveness, if indeed it is a generalized trait, represents a desirable or undesirable characteristic in terms of athletic personality is perhaps one of the key questions to be addressed by future research. The placebo effect is still a little understood phenomenon. This statement is true of many sports psychological phenomena, for example flow states and emotion, although unlike such phenomena the placebo effect, given its central role in the estimation of effects in placebocontrolled studies, is fundamental to sports science research and evidence-based practice. The placebo effect also warrants further investigation as a mind-body phenomenon of interest in its own right. As is suggested both explicitly and implicitly by several authors above, the logical conclusion from any study in which an athlete performs to a higher level as the result of receiving a sham treatment is that there is untapped psychological potential in that athlete. Whatever the mechanisms underlying placebo effects in sport, it certainly seems incumbent on sports scientists to further investigate the potential for placebo effects to enhance performance. Over and above this, that some authors have reported placebo effects similar in magnitude to those reported for the drug the placebo purported to represent, suggests that there is potential for future placebo effect research to be targeted at informing anti-doping initiatives. An opportunity to examine the placebo effect passes unused in many research environments. Incorporating a fully balanced placebo design may not always be feasible, or indeed appropriate, but by incorporating a baseline measure or non-placebo control group into a study, researchers might better elucidate both the biological and psychological effects of the intervention under examination. While this approach Sports Med 2009; 39 (4)
Beedie & Foad
328
is still more costly than the standard two-condition design, there is also an economy in the approach, i.e. findings might inform two domains. Given that the placebo effect has arguably transitioned from the role of artefact to that of legitimate area of study, it is possible to envisage a point in time at which stronger justification might be required for not incorporating a no-placebo condition than for incorporating one. By comparison with placebo effect research in medicine, placebo effect research in sport is in its infancy. Potential mechanisms have only recently been addressed. Physiological data have provided few clues, and although qualitative data suggest that effects might be related to expectancy-driven changes in pain sensation, fatigue resistance and anxiety, such data are retrospective, and, even if they were collected in real time, might reflect faulty perceptual processing. There is, however, sufficient empirical evidence from sport to warrant more concerted and consistent research into the placebo effect from within the discipline. Programmatic research involving the triangulation of data to establish, clarify and elucidate the nature of placebo effects in sport could be instructive. Investigation of contextual and personality factors in the mediation of placebo responses, and a thorough exploration of designs and methodologies, may also help to advance knowledge in this area. By elucidating the nature of this effect in sports performance, research may contribute not only to an elaboration of placebo phenomena per se, but to a greater understanding of the mechanisms and methods for best unravelling experimental effects in sports research and understanding real world performance. 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. Di Blasi Z. The placebo effect: the crack in the biomedical box. Psychologist 2003; 16 (Pt 2): 72-5 2. Ogden J. Health psychology. Berkshire (UK): Open University Press, 2004
ª 2009 Adis Data Information BV. All rights reserved.
3. Taylor SE. Health psychology. New York: McGraw-Hill, 2003 4. Ader R, Cohen N. Psychoneuroimmunology: conditioning and stress. Annu Rev Psychol 1993; 44: 53-5 5. de la Fuente-Ferna´ndez R, Phillips AG, Zamburlini M, et al. Dopamine release in human ventral striatum and expectation of reward. Behav Brain Res 2002; 136 (Pt 2): 359-63 6. Yang EV, Bane CM, MacCallum RC, et al. Stress-related modulation of matrix metalloproteinase expression. J Neuroimmunol 2002; 133 (Pt 1-2): 144-50 7. Clark VR, Hopkins WG, Hawley JA, et al. Placebo effect of carbohydrate feeding during a 40-km cycling time trial. Med Sci Sports Exerc 2000; 32: 1642-7 8. McGuire WJ. The nature of attitudes and attitudes change. In: Lindzey G, Aronson E, editors. The handbook of social psychology, vol. III. Reading (MA): Addison-Wesley, 1969: 136-314 9. Price DD, Finniss DG, Benedetti F. A comprehensive review of the placebo effect: recent advances and current thought. Annu Rev Psychol 2008 Jan 59 [online]. Available from URL: http://arjournals.annualreviews.org/action/ showJournals [Accessed 2008 Feb 10] 10. Evans D. Placebo: the belief effect. London: HarperCollins, 2003 11. Bonci L. Nutritional ergogenics: performance enhancers vs. the placebo effect indications and contraindications. Proceedings of the National Athletic Trainers’ Association. 49th Annual Meeting and Clinical Symposia; 1998 Jun 17–20, Baltimore (MA). Champaign (IL): Human Kinetics, 1998: 270-2 12. Yesalis CE, Bahrke MS. Anabolic-androgenic steroids: current issues. Sports Med 1995; 19: 326-40 13. Gutirrez-Sancho O, Moncada-Jimenez J, Robinson E, et al. The effects of creatine supplementation on biochemical, body composition, and performance outcomes in humans: a meta-analysis. Int J Appl Sports Sci 2006; 18 (Pt 2): 12-38 14. Shephard RJ. Vitamin E and athletic performance. J Sports Med 1983; 23: 461-70 15. Kerr IL. Mouth guards for the prevention of injuries in contact sports. Sports Med 1986 Nov-Dec; 3 (Pt 6): 415-27 16. Brolinson PG. Precompetition manipulation: placebo or performance enhancer? Clin J Sport Med 2003 Mar; 13 (Pt 2): 69-70 17. Liggett DR. Sports hypnosis. Champaign (IL): Human Kinetics, 2000 18. Aragon-Vargas LF. Effects of fasting on endurance exercise. Sports Med 1993; 16: 255-65 19. Hinton ER, Taylor S. Does placebo response mediate runner’s high? Percept Mot Skills 1986 June; 62 (Pt 3): 789-90 20. Vogt W. Breaking the chain: drugs and cycling, the true story (trans. William Fotherington). London: Random House, 1999 21. Gallagher H. On swimming. London: Pelham, 1970: 33-8 22. World champions or soccer cheats? The Daily Telegraph, United Kingdom [online]. Available from URL: http:// www.dailytelegraph.com/world champions or soccer cheats.htm [Accessed 2004 Apr 1] 23. Murphy MM, White RA. In the zone: transcendent experience in sport. New York: Penguin, 1995: 34-102
Sports Med 2009; 39 (4)
The Placebo Effect in Sport
24. Beedie CJ. The placebo effect in competitive sport: qualitative data. J Sport Sci Med 2007; 6: 21-8 25. Burke LM, Hawley JA, Schabort EJ, et al. Carbohydrate loading failed to improve 100-km cycling performance in a placebo-controlled trial. J Appl Physiol 2000; 88: 1284-90 26. Sonetti DA, Wetter TJ, Pegelow DF, et al. Effects of respiratory muscle training versus placebo on endurance exercise performance. Respir Physiol 2001; 127 (Pt 2-3): 185-99 27. Hornery DJ, Papalia S, Mujika I, et al. Physiological and performance benefits of halftime cooling. J Sci Med Sport 2005 Mar; 8 (Pt 1): 15-25 28. McMurray RG, Wilson JR, Kitchell BS. The effects of fructose and glucose on high intensity endurance performance. Res Q Exerc Sport 1983; 54: 156-62 29. Sellwood KL, Brukner P, Williams D, et al. Ice-water immersion and delayed-onset muscle soreness: a randomised controlled trial. Br J Sports Med 2007; 41: 392-7 30. Reeser JC, Smith DT, Fischer V, et al. Static magnetic fields neither prevent nor diminish symptoms and signs of delayed onset muscle soreness. Arch Phys Med Rehabil 2005; 86 (Pt 3): 565-70 31. Moseley Jr JB, Wray NP, Kuykendall D, et al. Arthroscopic treatment of osteoarthritis of the knee: a prospective, randomized, placebo-controlled trial: results of a pilot study. Am J Sports Med 1996 Jan-Feb; 24 (Pt 1): 28-34 32. Porcari JP, Otto J, Felker H, et al. The placebo effect on exercise performance [abstract]. J Cardiopulmon Rehabil Prev 2006 Jul/Aug; 26 (Pt 4): 269 33. Ariel G, Saville W. Anabolic steroids: the physiological effects of placebos. Med Sci Sports Exerc 1972; 4: 124-6 34. Beedie CJ, Stuart EM, Coleman DA, et al. Placebo effect of caffeine in cycling performance. Med Sci Sports Exerc 2006; 38: 2159-64 35. Beedie CJ, Coleman DA, Foad AJ. Positive and negative placebo effects resulting from the deceptive administration of an ergogenic aid. Int J Sport Nutr Exerc Metab 2007; 17: 259-69 36. Foad AJ, Beedie CJ, Coleman DA. Pharmacological and psychological effects of caffeine ingestion in 40 km cycling performance. Med Sci Sports Exerc 2008; 40 (Pt 1): 158-65 37. Foster C, Felker H, Porcari JP, et al. The placebo effect on exercise performance [abstract]. Med Sci Sports Exerc 2004 May; 36 Suppl. 5: S171 38. Kalasountas V, Reed J, Fitzpatrick J. The effect of placeboinduced changes in expectancies on maximal force production in college students. J Appl Sport Psychol 2007; 19 (Pt 1): 116-24
ª 2009 Adis Data Information BV. All rights reserved.
329
39. Maganaris CN, Collins D, Sharp M. Expectancy effects and strength training: do steroids make a difference? Sport Psychologist 2000; 14 (Pt 3): 272-8 40. McClung M, Collins D. ‘‘Because I know it will!’’: placebo effects of an ergogenic aid on athletic performance. J Sport Exerc Psychol 2007 Jun; 29 (Pt 3): 382-94 41. Benedetti F, Pollo A, Colloca L. Opioid-mediated placebo responses boost pain endurance and physical performance: is it doping in sport competitions? J Neurosci 2007 Oct; 27 (Pt 44): 11934-9 42. Pollo A, Carlino E, Benedetti F. The top-down influence of ergogenic placebos on muscle work and fatigue. Eur J Neurosci 2008; 28: 379-88 43. Batterham AM, Hopkins WG. Making meaningful inferences about magnitudes. Int J Sport Physiol Perf 2006; 1: 50-7 44. Desharnais R, Jobin J, Cote C, et al. Aerobic exercise and the placebo effect: a controlled study. Psychosom Med 1993; 55 (Pt 2): 149-54 45. Marlatt GA, Rohsenow DJ. Cognitive processes in alcohol use: expectancy and the balanced placebo design. In: Mello N, editor. Advances in substance abuse (1). Greenwich (CT): JAI Press, 1980: 159-99 46. Beedie CJ, Foad AJ, Coleman DA. Identification of placebo responsive participants in 40 km cycling performance. J Sport Sci Med 2008; 7 (Pt 1): 166-75 [online]. Available from URL: http://www.jssm.org/vol7/n1/24/v7n1-24text. php [Accessed 2008 Apr 2] 47. Wilmert N, Porcari JP, Foster C, et al. The effects of oxygenated water on exercise physiology during incremental exercise and recovery. J Exerc Physiol 2002 Nov; 5 (Pt 4) [online]. Available from URL: http://faculty.css.edu/ tboone2/asep/Porcari.pdf [Accessed 2007 Nov 8] 48. Porcari J, Foster C. Mind over body: ACE fitness matters 2006 May/June [online]. Available from URL: http://www.acefitness.org/getfit/PlaceboStudy2006.pdf [Accessed 2006 Dec 5] 49. Kirsch I, Weixel LJ. Double-blind versus deceptive administration of a placebo. Behav Neurosci 1988; 102 (Pt 22): 319-23 50. Fiorillo CD, Tobler PN, Schultz W. Discrete coding of reward probability and uncertainty by dopamine neurons. Science 2003; 299: 1898-902
Correspondence: Dr Christopher Beedie, Department of Sport Science, Tourism and Leisure, Canterbury Christ Church University, North Holmes Road, Canterbury, Kent, CT1 1QU, UK. E-mail:
[email protected]
Sports Med 2009; 39 (4)
Sports Med 2009; 39 (4): 331-337 0112-1642/09/0004-0331/$49.95/0
REVIEW ARTICLE
ª 2009 Adis Data Information BV. All rights reserved.
Serum Creatinine Concentration and Creatinine-Based Estimation of Glomerular Filtration Rate in Athletes Giuseppe Banfi,1,2 Massimo Del Fabbro1,2 and Giuseppe Lippi3 1 IRCCS Galeazzi, Milan, Italy 2 Dipartimento di Tecnologie per la Salute, School of Medicine, University of Milan, Milan, Italy 3 University of Verona, Dipartimento di Scienze Morfologico-Biomediche, Sezione di Chimica Clinica, Verona, Italy
Contents Abstract. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1. Serum Creatinine Metabolism and Measurement. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2. Serum Creatinine Concentration in Athletes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3. Serum Creatinine Values in Relation to Training and Competitions. . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4. Laboratory Methods for Measuring Creatinine in Athletes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5. Equations for Estimating Glomerular Filtration Rate (GFR) in Athletes . . . . . . . . . . . . . . . . . . . . . . . . . . . 5.1 Equations for Assessing GFR Changes in Athletes during Exercise . . . . . . . . . . . . . . . . . . . . . . . . . . 5.2 Equations for Assessing GFR in Athletes at Rest. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6. Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Abstract
331 332 332 334 334 335 335 335 336
Since serum creatinine concentration test results in athletes can sometimes be abnormal, correct test interpretation needs to take account of the sportsman’s status, as well as the transient exercise-induced changes in creatinine and the marked differences in creatinine levels among athletes competing in different sport disciplines. Serum creatinine concentration in athletes is related to body mass index, so the use of general population reference ranges should not be recommended in sports medicine. This does not necessarily imply, however, that specific creatinine reference ranges for athletes need to be defined. The individuality index (the ratio between intra- and interindividual variability) for creatinine is about half (0.33 vs 0.60) the generally considered lower limit for a population-based reference interval to be classified as useful. Prediction of glomerular filtration rate in athletes by means of creatininebased equations is also questionable because of discrepancies among formulae, owing to the particular anthropometric characteristics of athletes (high body mass index). Furthermore, differences in muscle mass need to be entered into the recommended equations for calculating the estimated glomerular filtration rate (eGFR) in athletes. Based on current knowledge about creatinine measurement and eGFR calculation, we suggest that athletes be monitored periodically by consecutive
Banfi et al.
332
creatinine level assessments, comparing the values reported during the training and competition season with the baseline levels recorded during the recovery period.
Serum creatinine concentration values in athletes can sometimes be abnormal on testing; therefore, correct interpretation will need to take the sportsman’s status into account. Test results are usually interpreted by applying the reference range established for the general population, but this approach is not completely correct owing to exercise-induced changes in creatinine levels, which are transient and promptly normalize during recovery. Moreover, the interindividual variability of the parameter is wide, producing an individuality index (the ratio between intra- and interindividual variability) of 0.33, which is much lower than the value of 0.60 generally considered the lower limit for a population-based reference interval to be classified as useful. We collected creatinine data from studies on athletes in order to determine whether correlations existed between creatinine levels, body mass index (weight [kg] divided by height [m] squared [BMI]), training and competitions, and analytical methods. We also report the existing data about the use of estimated glomerular filtration rate (eGFR) equations in athletes, because these formulae are recommended instead of creatinine values for evaluating kidney disease. Since these equations have not been validated in healthy subjects, we thought it would be of interest to assess their application to athletes. 1. Serum Creatinine Metabolism and Measurement Creatinine is non-enzymatically derived from creatine. The creatine turnover rate in healthy men is constant, accounting for 1.6% of the total creatine pool per day.[1] The creatinine quantity produced and released daily could be calculated from the total muscle mass, which is the most important determinant of the creatine pool size because it contains about 98% of the body’s total stores of creatine. Serum creatinine concentration is the most widely used and commonly accepted measure of ª 2009 Adis Data Information BV. All rights reserved.
renal function in clinical medicine.[1] It can be influenced by body mass, diet (meat content in diet) and analytical methods. The common reference range for creatinine in the general adult male population, as obtained by the Jaffe reaction in automated systems, is 0.9–1.3 mg/dL (76–115 mmol/L).[2] This range has been recently narrowed to 0.68–1.13 mg/dL by using various analytical methods and based on a wide population of adult males.[3] The Jaffe method, a colorimetric reaction based on picrate, is commonly used for measuring creatinine; it is simple, inexpensive and easily adapted to automated systems. However, molecules other than creatinine can interfere with sample processing, altering the results by up to 20%.[1] For this reason, enzymatic methods and the calibration of all methods against gas chromatography-isotope dilution mass spectrometry have recently been recommended.[4] The accuracy of creatinine measurement will eventually be improved by standardizing all assays to the values obtained by using reference methods. Mass spectrometry employing a stable isotope-labelled internal standard has the highest accuracy, but it is labour intensive and expensive; it could be used for obtaining reference materials. A network of laboratories coordinated by the US National Kidney Disease Education Program is working to reduce interassay variability due to different calibrations. A traceable isotope dilution mass spectrometry reference method releasing a reference material with traceable creatinine values (NIST SRM 967) will soon be available. The recalibration of current methods for determining serum creatinine levels will likely lead to a revision of reference values.[4] 2. Serum Creatinine Concentration in Athletes In sports medicine, creatinine is widely used to evaluate the general health status of athletes, Sports Med 2009; 39 (4)
Estimating Glomerular Filtration Rate in Athletes
333
Table I. Serum creatinine values (mean – SD) of male athletes compared with the concentrations in age-matched male sedentary subjects Sport
Sport level
No. of athletes
Serum creatinine (mg/dL)
No. of sedentary subjects
Serum creatinine (mg/dL)
p-Value
100
1.00 – 0.10
<0.01
Reference
220
1.10 – 0.20
Triathlon
Italian national team
15
0.99 – 0.07
6
Basketball
Italian first division team
29
1.15 – 0.07
6
Cycling
Two professional teams
35
0.93 – 0.07
6
Total group
6
Motorcycling
Professional team
13
0.92 – 0.09
6
Soccer
Italian first division team
27
1.27 – 0.09
6
Sailing
America’s Cup yacht crew
23
1.08 – 0.11
6
Alpine skiing
Italian national team
34
1.15 – 0.10
6
Rugby
Italian national team
44
1.30 – 0.11
6
Nordic skiing
Italian national team
37
0.88 – 0.10
60
0.94 – 0.12
<0.05
7
Cycling
Professionals
80
0.83 – 0.12
60
0.94 – 0.12
<0.01
7
Cycling
Professionals
50
0.93 – 0.14
35
0.98 – 0.10
<0.05
8
60
0.95 – 0.20
Cycling
Non-professionals
71
0.86 – 0.20
Cycling
Professionals
76
0.80 – 0.20
Thai boxing
Professionals
20
1.02 – 0.04
10
0.99 – 0.02
<0.01
9
<0.01
9
NS
10
NS = not significant.
particularly when water-electrolytic balance is crucial. Reference values for biochemical parameters of specific importance to sportsmen have never been established, and those used for the general population, including serum creatinine, are routinely applied to athletes. However, because this procedure can lead to misinterpretation of these values in athletes, which sometimes fall outside the limits established for the general population, further study is needed on the behaviour of serum creatinine in order to determine more appropriate reference intervals. The origin of creatinine from creatine, which is stored mainly in muscle tissue, is usually claimed as justifying the differences in creatinine levels between men and women.[1] This concept is also applied to athletes when compared with nonathletes. Surprisingly, the relationship between muscle mass and creatinine has not been extensively studied so far. Studies based on the classic Jaffe method have reported a correlation ª 2009 Adis Data Information BV. All rights reserved.
between the two in the general population, and in the elderly in particular.[5] Serum creatinine concentration is generally higher in athletes than in sedentary people, although some differences in creatinine levels among athletes from different sports disciplines have been reported (table I). These differences were demonstrated for professional athletes in eight different sports disciplines, characterized by different aerobic/ anaerobic metabolism, different training loads and frequency of competitions, different length of competitions, and different annual periods of training and competitions. The same study also showed significant differences in serum creatinine levels between physically active and sedentary subjects.[6] It should be underlined that the distribution of serum creatinine concentrations in the athlete population is unusual, showing either values lower than those observed in sedentary people (<1 mg/dL) or much higher values (well above 1 mg/dL). In brief, the distribution is unequal.[6] Sports Med 2009; 39 (4)
Banfi et al.
334
In a recent study investigating the relationship between muscle mass, serum creatinine and BMI in athletes,[11] serum creatinine was measured in 151 professional athletes from the Italian national rugby team (n = 44), the Italian national triathlon team (n = 9), the Italian first division soccer team (n = 27), the America’s Cup yacht crew (n = 22), the Italian alpine ski national team (n = 34), and a ProTour cycling team (n = 24). The method was the Jaffe reaction on an Abbott Aeroset c8000 (Abbott Diagnostics, Abbott Park, IL, USA). The age range was 17–35 years. Blood samples were collected before the start of training and competitions season, closely following preanalytical precautions. A positive correlation was found between BMI and serum creatinine (r = 0.48; p < 0.001). The rugby players had the highest BMI values (28.83 – 2.41 kg/m2) and the highest values of serum creatinine (1.31 – 0.12 mg/dL), whereas the cyclists had a low BMI (21.33 – 1.21 kg/m2) and the lowest concentration of serum creatinine (0.91 – 0.07 mg/dL). Notably, the correlation was found in sports characterized by different kinds of training, competitive season and involvement of aerobic and anaerobic metabolic pathways, and it was related to the specific status of professional sport activities. For instance, cyclists and triathletes typically have a low percentage of fat tissue and the lowest creatinine values, whereas rugby players have relatively high amounts of fat tissues[12] and have the highest values. Serum creatinine concentrations lower than those observed in controls have also been found in different groups of professional cyclists compared with sedentary people (table I) by using an enzymatic method on a Roche Modular system (Roche, Basel, Switzerland).[8,9] These findings argue for the need to interpret creatinine values in sportsmen according to the specific type of sport an athlete performs. 3. Serum Creatinine Values in Relation to Training and Competitions Serum creatinine concentration is minimally affected by training and competition,[10,13] even in sports involving extreme physical effort.[14] ª 2009 Adis Data Information BV. All rights reserved.
In a study on Thai boxers (n = 20; age range 14–17 years), the creatinine values during normal training, intensive training and after a match were not statistically different from those of the control group. Creatinine clearance did not differ, except for the post-match values, which were significantly lower than those in the control group and compared with those measured during the training period of the athletes.[10] In 16 volunteers participating in the First Race Across the Alps – an ultraendurance cycling race of 509 km at an altitude of 300–2750 m, including 11 mountain passes – there was a statistically significant increase in serum creatinine values as measured by the Jaffe method immediately after the end of the race versus those observed before the start of the competition. However, the mean values consistently fell within the reference range. The authors stressed the importance of regular training and adequate fluid replacement for professional ultramarathon cyclists, who required, on average, a total fluid intake of 17 L during the race.[15] Physical exercise will acutely change serum creatinine level depending on the severity and duration of exercise, as well as on the age of athletes. It was observed that the estimated creatinine clearance in ultraendurance mountain runners increased with increasing age.[16] A marked rise in serum creatinine level is transient, even in athletes undergoing strenuous effort.[14,17] In line with these observations, it can be assumed that creatinine level changes in athletes are minimal and transient during and after training and competitions. 4. Laboratory Methods for Measuring Creatinine in Athletes Laboratory methods may be crucial for correctly interpreting creatinine concentrations. Most of the published data on athletes were determined using the traditional Jaffe reaction. Yet even the introduction of enzymatic methods has not substantially improved test accuracy. Serum creatinine concentrations in 127 sera samples from 57 toplevel rugby players during the 2005–6 competitive season were measured using both the Jaffe method Sports Med 2009; 39 (4)
Estimating Glomerular Filtration Rate in Athletes
and an enzymatic method. Paradoxically, the enzymatic method produced higher values than the traditional method. This was probably the result of recalibration procedures manufacturers have developed to minimize inter-method differences.[17] The potential of detecting intra- and intergroup differences, if any exist, appears similar for any of the methods used to evaluate serum creatinine levels. 5. Equations for Estimating Glomerular Filtration Rate (GFR) in Athletes Recently, the use of equations for estimating glomerular filtration rate (GFR) was recommended.[4] The equations include creatinine concentration besides additional variables known to influence creatinine measurement and interpretation. The Cockcroft and Gault (CG) formula proposed some years ago and widely used[18] has been replaced by the modification of diet in renal disease (MDRD) formula.[4,19] The MDRD equation could be of particular value in sports medicine because it is not influenced by body mass.[19] A general consensus on the use of such equations in the healthy population has yet to be established: the National Kidney Disease Education Program currently recommends that eGFR >60 mL/min/1.73 m2 be reported simply as ‘>60 mL/min/1.73 m2’ rather than a discrete numeric value.[4] There are few reports of GFR estimated by an equation in athletes. 5.1 Equations for Assessing GFR Changes in Athletes during Exercise
In a study on ultramarathon cyclists, creatinine clearance was calculated by the CG formula: the study highlighted a significant decrease in eGFR immediately after the race (85 – 19 mL/min) compared with the baseline value (114 – 27 mL/min), but it returned to baseline values within 24 hours after the end of the race (113 – 28 mL/min). A similar decline in creatinine clearance (18%) in amateur cyclists after a competition was also described.[20] In 27 marathon runners, GFR, as estimated by the MDRD formula, significantly increased after the race, but returned to baseline (pre-race) values ª 2009 Adis Data Information BV. All rights reserved.
335
within 24 hours. The behaviour of eGFR and creatinine was identical, whereas urea output further increased 1 day after the competition.[21] In ultraendurance mountain runners, the estimated creatinine clearance (CG formula) did not vary with the percentage in weight loss.[16] The average intensity of daily physical exercise is positively associated with eGFR, as calculated by the MDRD formula, whilst it is inversely associated with serum creatinine levels.[9] The effect of the body mass factor on the equations is crucial, since eGFR values change markedly depending on whether the CG or the MDRD formula is used. For example, in 19 toplevel rugby players with high BMI values (28.9 – 2.5 kg/m2), measured before the start of training and competitions, the CG equation gave a mean eGFR value of 123 – 17 mL/min, whilst the MDRD yielded a mean of 72 – 8 mL/min.[17] 5.2 Equations for Assessing GFR in Athletes at Rest
The effect of training on homeostatic renal function in elite cyclists was evaluated by estimating GFR by means of an equation based on age, sex, creatinine, urea nitrogen and albumin.[22] The authors underlined that the duration, intensity and volume of cycling may affect renal function, particularly when the training workload is very high.[22] Professional cyclists have significantly lower bodyweight, BMI and serum creatinine concentration than healthy sedentary individuals. In a specific study comparing creatinine-based estimations of GFR in cyclists at rest versus sedentary controls, a significantly higher MDRD-estimated GFR was reported in the athletes compared with the controls (119 vs 104 mL/min/1.73 m2; p < 0.001).[23] Conversely, the GFR values estimated by both the Mayo Clinic quadratic equation (MCQE)[24] and the CG formulae did not differ significantly (137 vs 135 mL/min/1.73 m2 [p = 0.128] and 127 vs 127 mL/min m2 [p = 0.490], respectively). Compared with the MDRD values, the mean GFR calculated by the MCQE and CG formulae was overestimated by 29% and 23% in the sedentary population and by 17% and 7% in Sports Med 2009; 39 (4)
Banfi et al.
336
the athletes, respectively. A lower bias was observed when comparing CG versus MCQEestimated values in both the sedentary subjects (mean -5%; 95% CI -32, 22) and the athletes (mean -6%; 95% CI -41, 29). The correlation between the different equations in the healthy sedentary individuals was consistently significant, whereas the only significant association in the athletes was that between the MCQE and the CG values. The results of this study indicate that the three most widely used creatinine-based formulae yield significant variations in the estimated GFR in a population of endurance athletes at rest. Probably, the CG or the MCQE equation might have been more suitable than the MDRD in this context, since either globally appears more robust against variations in training regimen.[23]
6. Conclusions The usual reference intervals could be misleading for correct interpretation of serum creatinine concentrations in athletes, especially when the serum creatinine value is higher than the upper limit of the reference range. However, we believe that specific creatinine reference ranges should not be determined for athletes. The individuality index (the ratio between intra- and interindividual variability) of creatinine is 0.33,[25] nearly half the index value of 0.60 generally considered the threshold for classifying a reference interval useful in a population. Instead, athletes should be individually monitored by consecutive creatinine assessments, using as the baseline value the one calculated before the start of training and competitions, while taking into due account the type of sport and the athlete’s BMI. Moreover, a physiological model for studying and validating eGFR equations should be specifically developed for athletes. The significantly higher GFR values estimated by the MDRD equation in athletes at rest does not permit the use of the current reference range of >60 mL/min/1.73 m2, since the results of this equation could be unreliable when the athletes are unaccustomed to the training load. ª 2009 Adis Data Information BV. All rights reserved.
Acknowledgements We are indebted to Mr Kenneth Britsch, who reviewed the language. No funding was used in the preparation of this review. The authors have no conflicts of interest directly relevant to the content of the review.
References 1. Perrone RD, Madias NE, Levey AS. Serum creatinine as an index of renal function: new insights into old concepts. Clin Chem 1992; 38: 1933-53 2. Burtis CA, Ashwood ER, editors. Tietz textbook of clinical chemistry. 2nd ed. Philadelphia (PA): Saunders, 1994: 2184 3. Rustad P, Felding P, Franzson L, et al. The Nordic Reference Interval Project 2000: recommended reference intervals for 25 common biochemical properties. Scand J Clin Lab Invest 2004; 64: 271-84 4. Myers GL, Miller WG, Coresh J, et al. Recommendations for improving serum creatinine measurement: a report from the Laboratory Working Group of the National Kidney Disease Education Program. Clin Chem 2006; 52: 5-18 5. Vikse BE, Vollset SE, Tell GS, et al. Distribution and determinants of serum creatinine in the general population: the Hordaland Health Study. Scand J Clin Lab Invest 2004; 64: 709-22 6. Banfi G, Del Fabbro M. Serum creatinine values in elite athletes competing in 8 different sports: comparison with sedentary people. Clin Chem 2006; 52: 330-1 7. Lippi G, Brocco G, Franchini M, et al. Comparison of serum creatinine, uric acid, albumin and glucose in male professional endurance athletes compared with healthy controls. Clin Chem Lab Med 2004; 42: 644-7 8. Lippi G, Salvagno GL, Montagnana M, et al. Influence of physical exercise and relationship with biochemical variables of NT-pro-brain natriuretic peptide and ischemia modified albumin. Clin Chim Acta 2006; 367: 175-80 9. Lippi G, Banfi G, Salvagno GL, et al. Glomerular filtration rate in endurance athletes. Clin J Sport Med 2008; 18: 286-8 10. Saengsirisuwan V, Phadungkij S, Pholpramool C. Renal and liver functions and muscle injuries during training and after competition in Thai boxers. Br J Sports Med 1998; 32: 304-8 11. Banfi G, Del Fabbro M, Lippi G. Relation between serum creatinine and body mass index in elite athletes of different sport disciplines. Br J Sports Med 2006; 40: 675-8 12. Nicholas CW. Anthropometric and physiological characteristics of rugby union football players. Sports Med 1997; 6: 375-96 13. Kratz A, Lewandrowski KB, Siegel AJ, et al. Effect of marathon running on hematologic and biochemical laboratory parameters, including cardiac markers. Am J Clin Pathol 2002; 118: 856-63 14. Fallon KE, Sivyer G, Sivyer K, et al. The biochemistry of runners in a 1600 km ultramarathon. Br J Sports Med 1999; 33: 264-9 15. Neumayr G, Pfister R, Hoertnagl H, et al. Renal function and plasma volume following ultramarathon cycling. Int J Sports Med 2005; 26: 2-8
Sports Med 2009; 39 (4)
Estimating Glomerular Filtration Rate in Athletes
16. Page AJ, Reid SA, Speedy DB, et al. Exercise-associated hyponatremia, renal function, and nonsteroidal antiinflammatory drug use in an ultra-endurance mountain run. Clin J Sport Med 2007; 17: 43-8 17. Del Fabbro M, Banfi G, Monopoli L, et al. Creatinine values in elite rugby players measured by means of cynetic Jaffe and enzymatic methods [abstract]. Bioch Clin 2006; 30: 367 18. Cockcroft DW, Gault MH. Prediction of creatinine clearance from serum creatinine. Nephron 1976; 16: 31-41 19. Levey AS, Greene T, Kusek JW, et al. A simplified equation to predict glomerular filtration rate from serum creatinine [abstract]. J Am Soc Nephrol 2000; 11 Suppl.: A0828 20. Neumayr G, Pfister R, Hoertnagl H, et al. The effect of marathon cycling on renal function. Int J Sports Med 2003; 24: 131-7 21. Leers MPG, Schepers R, Baumgarten R. Effects of a longdistance run on cardiac markers in healthy athletes. Clin Chem Lab Med 2006; 44: 999-1003
ª 2009 Adis Data Information BV. All rights reserved.
337
22. Touchberry CD, Ernsting M, Haff G, et al. Training alterations in elite cyclists may cause transient changes in glomerular filtration rate. J Sports Sci Med 2004; 3: 28-36 23. Lippi G, Banfi G, Salvagno GL, et al. Comparison of creatinine-based estimations of glomerular filtration rate in endurance athletes at rest. Clin Chem Lab Med 2008; 46: 235-9 24. Rule AD, Larson TS, Bergstrahl EJ, et al. Using serum creatinine to estimate glomerular filtration rate: accuracy in good health and in chronic kidney disease. Ann Intern Med 2004; 141: 929-34 25. Fraser CG. Biological variation: from principles to practice. Washington, DC: AACC Press, 2001
Correspondence: Prof. Giuseppe Banfi, IRCCS Galeazzi, via Galeazzi 4, 20161 Milano, Italy. E-mail:
[email protected] or giuseppe.banfi1@ unimi.it
Sports Med 2009; 39 (4)
