Journal Logo

APPLIED SCIENCES: Physical Fitness and Performance

Dose Response of Inhaled Salbutamol on Exercise Performance and Urine Concentrations


Author Information
Medicine & Science in Sports & Exercise: January 2008 - Volume 40 - Issue 1 - p 149-157
doi: 10.1249/mss.0b013e3181591df7
  • Free


Asthma and exercise-induced bronchospasm are pulmonary disorders that result in a narrowing of the airways and some measurable reduction in lung function. Prevalence rates are estimated to be between 5 and 10% in the general population; however, they have been reported to be much higher in athletic populations, with 50% of some subgroups estimated to experience compromised lung function (36). This limitation may have a negative effect on exercise performance for these athletes in comparison with their unaffected competitors (4).

Currently, four β2-agonists, salbutamol (SAL), formoterol, salmeterol, and turbutaline, have been approved by the World Anti-Doping Agency (WADA) for use via inhalation by asthmatics, as long as the athlete obtains a therapeutic use exemption (TUE). Applications for use of β2-agonists have been increasing during the past 20 yr in both seasonal games, with 6.6% and 4.6% of all participants at the 2002 (Salt Lake City) (2) and 2004 (Athens) (3) Olympic games, respectively. Of the four β2-agonists allowed, SAL is most commonly used, and there is growing concern that nonasthmatic athletes are using inhaled SAL in an attempt to gain a competitive advantage (2). Anecdotal evidence suggests that asthmatic and nonasthmatic athletes believe in its ability to enhance performance and are using doses that substantially exceed therapeutic recommendations (200 μg). This is despite the fact that the current research overwhelmingly suggests that acute inhaled SAL does not enhance endurance performance in nonasthmatics (5,6,8,10,21,22,25,29).

One major criticism of the majority of studies examining the ergogenic effects of inhaled SAL is that they did not evaluate changes in performance using a sport-specific test with competitive athletes. The validity of a test to be representative of performance is an important factor when evaluating the ergogenic effects of a treatment (13). Only two studies have investigated the effects of inhaled SAL using a simulated sport-specific performance test (22,32). Norris and colleagues (22) have shown no difference in 20-km cycling time-trial performance time after a dose of 400 μg. In contrast, van Baak et al. (32), using four times the therapeutic dose (800 μg), have shown inhaled SAL to be effective in decreasing time to complete a set amount of work relative to individual maximal workload (~60 min). If inhaled SAL has an ergogenic effect on performance, it may be related to dose, which is the case with respect to enhanced ventilation in both asthmatics and nonasthmatics (14,17). Decreased work of breathing may augment exercise performance because of increases in cardiac output available to working muscles (11). However, the only examination of the dose-related ergogenic potential of inhaled SAL would suggest otherwise. Goubault and colleagues (10) have shown no effect of dose (placebo, 200 μg, and 800 μg) on cycle time to exhaustion, even though FEV1 was enhanced after SAL inhalation. As with most of the studies, the use of a nonspecific test of performance limits the applicability of Goubault et al.'s findings to sport. In cycling, races are not won by the athlete holding a constant power until exhaustion; rather, they require the athlete to complete a set distance first. The literature is currently lacking a controlled examination of the dose-response effects of inhaled salbutamol using a sport-specific performance test in competitive athletes.

Because β2-agonists, particularly with oral doses, have the potential to have anabolic-like effects, unauthorized use of SAL is closely monitored through doping control. Even for athletes possessing a TUE, a urine concentration of nonsulphated SAL (cSAL) greater than 1000 ng·mL−1 is considered an adverse analytical finding resulting from oral administration, and this can result in suspension. This cutoff point has been questioned of late, with recent reports of positive test results using inhaled therapeutic doses, all with urine concentrations well over 1000 ng·mL−1 after exercise (19,27). Although the majority of urine samples reported in the literature rarely exceed 500 ng·mL−1 (24,34), it has been suggested that with individual variations in dose, changes in hydration status after competition, and the ability to absorb, metabolize, and excrete salbutamol, the possibility for elevated concentrations exists (19). Previous findings from our laboratory (unpublished) have shown that at rest, cSAL as reported by WADA is related to and increases in variability with dose (up to 800 μg), although none of the subjects exceeded the doping control limit of 1000 ng·mL−1. Because of the effects of exercise on renal function (9), responses after intense efforts may differ. An examination of the dose-response effect of inhaled salbutamol on urine concentrations after exercise as used in doping control is lacking.

Although research to date has shown no significant improvement in performance with the use of inhaled SAL, the dose-response effect on performance has not been evaluated in a homogenous group of highly trained athletes with a sport-specific performance test. Therefore, the purpose of this study was to examine the effects of increasing doses of SAL on 20-km time-trial performance and urine concentrations after exercise in competitive athletes. We hypothesized that there would be no dose-response effect of inhaled SAL on time-trial performance, but that urine concentrations would be related to dose and highly variable.



Healthy, competitive male cyclists and triathletes (N = 37) were recruited for this study. An a priori power calculation was performed, using 1.5 times the coefficient of variance (CV) for mean power through 20 km as the minimum improvement that would make a competitive difference (13). We have previously demonstrated a CV of approximately 2% in mean power, using the Velotron cycle ergometer for this performance test in this population (28). It was calculated that approximately 30 subjects were required, with an estimated standard deviation of 20 W, to identify significance at 0.05 with a power of 0.80. All athletes were competing at a provincial (state) level or higher in the elite categories for their respective sports and disciplines. Exclusion criteria included a maximal aerobic power (V˙O2max) of less than 60 mL·kg−1·min−1 and 5 L·min−1, previous history or diagnosis of asthma, abnormal resting spirometry, or a positive eucapnic voluntary hyperpnea (EVH) test indicative of exercise-induced bronchospasm (EIB). Written informed consent was obtained from all subjects and the methods and protocol were approved by the University of British Columbia clinical research ethics board.

Experimental design.

A randomized, double-blind, repeated-measures design was used with four different treatment protocols (placebo (DP), 200 μg (D2), 400 μg (D4), and 800 μg (D8) of inhaled SAL). Each subject came to the lab on five different occasions, with a minimum of 72 h between visits. The first visit included medical screening, measurement of height and weight, pulmonary function, and an EVH test. Only subjects with a negative EVH test result were included in the dose-response portion of this study. Qualifying subjects then performed a ramped exercise test to determine maximal oxygen consumption on the same day. The remaining four sessions involved a simulated 20-km cycling time trial after one of the four treatments. The maximum time between two time trials was 14 d, because two subjects had scheduling conflicts. All other subjects completed the four trials within a period of 21 d. At the end of each time trial, athletes were required to provide a urine sample that was analyzed for concentration of nonsulfated salbutamol. See Figure 1 for a timeline of the performance testing sessions.

Timeline for treatment and time trials.

Lung function and airway hyperresponsiveness.

Before completing the EVH test, subjects performed baseline pulmonary function measures. This was achieved via a flow-volume loop, using a Medical Graphics CPX-D Metabolic cart (St. Paul, MN) with 1070 Pulmonary Function Software. Calibration was performed before each testing session, and subjects were familiarized with the procedure before actual testing. Each subject performed three trials, with the highest valid FEV1 recorded. A trial was considered valid if it was greater than 80% of the predicted value and was reproducible using ATS criteria (1). Subjects were then screened for susceptibility to bronchospasm, using the EVH challenge test. This test has previously been described in detail (2) and is one of the allowable tests by the International Olympic Committee to provide evidence for use of asthma medication during competition. Briefly, each subject was required to breathe a hypercapnic gas mixture (5% CO2, 21% O2, balance nitrogen) for a period of 6 min at a target ventilation that was calculated as 30 times the individual's pretest FEV1 (~85% maximal voluntary ventilation (MVV)). Spirometry was measured immediately afterward and at 5, 10, 15, and 20 min afterward. A decrease in FEV1 of greater than 10% from baseline measure was considered a positive test for bronchospasm, usually observed in the first 10 min. For the purposes of this study, the maximum decrease in FEV1 at any time point was identified and recorded as a percentage drop from pretest FEV1.

Maximal exercise test.

A maximal exercise test was performed on the Velotron Pro cycle ergometer (Racermate Inc., Seattle, WA). Before each test, factory calibration was verified, using the Accuwatt "run down" verification program (Racermate Inc., Seattle, WA) accompanying the ergometer software. Subjects were fitted to the ergometer according to the setup of their own bicycle. All settings were recorded and used in subsequent time trials. Bike settings included both seat and handle bar height and horizontal position, as well as crank length. Subjects were instructed to remain seated throughout the test. A 30-W·min−1 ramp protocol was used and controlled via the Velotron Coaching Software (version 1.5.186, RacerMate Inc., Seattle, WA), with expired gases collected and analyzed every 15 s (TrueOne 2400, Parvo Medics, Sandy, UT). Oxygen consumption (V˙O2), minute ventilation (E), production of carbon dioxide (V˙CO2), and respiratory exchange ratio (RER) were recorded. Flow and gas calibrations were performed before each test, using a 3-L calibration syringe and gases of known concentrations, respectively. Standard indicators for achieving V˙O2max were used, including volitional fatigue, a plateau in V˙O2 with increasing work rate, HR ≥ 90% of age-predicted maximum, and RER ≥ 1.15. V˙O2max was recorded as the mean of the two highest consecutive 15-s samples. Heart rate (HR) was measured by telemetry (Polar Vantage XL, Kempele, Finland) and recorded. Peak power was recorded as the highest completed 15-s interval, with power recorded in 7.5-W intervals.

Dose-response evaluation: exercise protocol.

A timeline of events for days 2-5 is depicted in Figure 1. Subjects were encouraged to prepare for each time trial as they would for a competitive event, with no strenuous exercise in the previous 24 h. Warm-up was self-selected, and although this varied between individuals, it was the same for each subject for all trials. Immediately after the warm-up, subjects were weighed and began receiving a treatment. A total of eight inhalations were administered each day from three different coded MDI, for a dose equal to one dose of DP, D2, D4, or D8. Spacers were used to optimize delivery of the medication, and subjects were trained in its proper use before participation. At 10 min after inhalation, a mask (Hans Rudolph 8930 Series, Kansas City, MO) and two-way breathing valve (Hans Rudolph 2700 Series, Kansas City, MO) were fitted to the subject and connected to a metabolic cart (TrueOne 2400, Parvo Medics, Sandy, UT). A complete seal of the mask was confirmed before testing. At 15 min after inhalation, subjects began the simulated 20-km time trial and were instructed to complete the distance as quickly as possible. All time trials were performed on the Velotron Pro cycle ergometer, which was calibrated before each test. This performance test has been described previously and is highly reproducible in trained cyclists, with a coefficient of variance (CV) of < 1% for time and < 2% for mean power (28). Approximately half of the subjects were familiar with this protocol in our laboratory; those who were not performed a familiarization trial after a rest period at the end of day 1. Subjects were required to perform two laps of a 10-km course, which was designed using the Velotron 3D software accompanying the ergometer (version NB04.1.0.2101, RacerMate Inc., Seattle, WA). The course was flat with no active wind effect. Resistance was adjustable by the ergometer's electronic gearing system. A gearing system simulating a 53-39 front-chain-ring setup, and a 23-21-19-17-16-15-14-13-12-11 rear-cog set was used. Throughout the time trial, subjects were able to watch themselves racing the course on the computer monitor. Distance traveled and gears selected were displayed, but all other feedback was blinded to the subject. Power, speed, and time were recorded by the ergometer software and downloaded afterward for analysis. Because the Velotron is an ergometer that calculates speed from power, mean power was the primary variable of analysis. The sampling rate for all ergometer variables was 1 Hz. Heart rate was also recorded by the ergometer and confirmed by telemetry (Polar Vantage XL, Kempele, Finland) throughout the time trial. Subjects did not receive any information as to how well they performed until all trials were completed. Throughout the time trial, expired gases were collected, with metabolic parameters averaged every 20 s. Every 2 km, subjects were asked to rate the perceived exertion (RPE) for leg (RPEL) and breathing (RPED) effort, using a 10-point Borg RPE scale. On completion of the time trial, subjects were requested to cool down until the 55-min mark after inhalation.

Urine collection and analysis.

At the 1-h mark after inhalation (T60), subjects were requested to provide a urine sample of about 15 mL. It was requested that the sample be obtained midstream and that the bladder be voided of urine afterward. Once the urine sample was obtained, specific gravity (SG) was measured, using a refractometer (Pocket PAL-10S, Atago, Bellevue, WA). All samples were then frozen to −20° C until laboratory analysis. Samples were analyzed by a third-party laboratory for total cSAL, accounting for free and glucuronized forms only. This is the value that is reported by WADA at the time of the study. Concentrations were determined by liquid chromatography-mass spectrometry. Urine was incubated with glucuronidase (from Helix pomatia, Sigma-Aldrich Co., St. Louis, MO) at 37°C for 2 h before addition of the internal standard. The internal standard (propionylprocanamide) was added to 1 mL of the urine specimen. The mixture was acidified with 0.5 mL of 10% trichloroacetic acid and 8 mL of chloroform added. The mixture was vortexed and centrifuged, and the aqueous phase was recovered for SAL assay. The instrument used was an Agilent model 1100 MSD coupled to an Agilent model 1090 liquid chromatograph; both instruments are controlled by Agilent ChemStation software. The mobile phase used for the chromatography was 10 mM aqueous ammonium acetate adjusted to pH 3.2 and acetonitrile (95:5 ramping to 75:25), and the column employed was an Eclipse XDB-C8 (4.6 mm × 30 mm × 3.5 μm) (Agilent Technologies, Wilmington, DE). Primary ions used for the quantitation were 240 m/z (SAL) and 292 m/z (propionylprocanamide). Flow rate for LC MS was 0.3 mL·min−1, with a retention time for SAL of 2.30 min. Concentrations were determined by comparison with a standard curve of the relative intensities of the SAL ion against that of the internal standard ion for standard solutions of the drugs prepared in drug-free urine.

Data analysis.

Means and standard deviations (SD) were calculated for descriptive variables. A Shapiro-Wilks W test for normality and Mauchley's test for equal variances were used to confirm that the assumptions for analysis of variance were met. A repeated-measures analysis of variance was used to determine statistical significance across treatments for all performance variables and urine concentrations measured. Post hoc analyses were performed, using Tukey's test for significance when a main effect was found. Pearson product-moment correlations were used to examine relationships between urine concentrations and specific gravity. Statistical procedures were completed using Statistica Software (version 5.0, Statsoft Inc., Tulsa, OK). For all tests, α was set at 0.05. Values reported are means ± SD unless otherwise noted.


Subject characteristics and airway hyperresponsiveness.

Subjects with positive (N = 30) and negative (N = 7) responses to the EVH test are shown in Table 1. A total of seven subjects produced a positive EVH test, resulting in an incidence rate of approximately 19% for airway hyperresponsiveness. Maximum drop in FEV1 after the EVH test was 27.7%. All positive responders were excluded from the remainder of the study. Baseline performance characteristics of the remaining subjects (N = 30) are shown in Table 2.

Subject characteristics and lung function.
Baseline performance characteristics of negative EVH subjects (N = 30).

Time-trial performance.

Three subjects were unable to complete all conditions because of illness and/or personal reasons; therefore, results of only 27 subjects are presented here. Time to completion did not differ between trials (DP = 30.72 ± 1.06 min; D2 = 30.56 ± 1.03 min; D4 = 30.67 ± 1.06 min; D8 = 30.70 ± 1.04 min; P = 0.16). Mean power (Pmean) through the 20 km for each of the conditions ranged between 306 and 310 W, with no effect (P = 0.12) of SAL observed between conditions (Fig. 2). This was approximately 67% of max power (Pmax) and 4.05 W·kg−1. Similarly, there was no effect of SAL on mean V˙O2 (~55 mL·kg−1·min−1; P = 0.32) and heart rate (~172 bpm; P = 0.52) throughout the time trials (Fig. 2). This equated to approximately 82% and 92% of the respective peak values achieved on day 1. Inhaled SAL did not affect exercise ventilation or patterns of breathing, because breathing frequency (~45 bpm) and tidal volume (~2.9 L) were both similar across conditions (Fig. 3). During each time trial, subjects were asked to assess their rates of perceived exertion for both leg and breathing efforts. Mean values through 20 km were unaffected by SAL (Table 3), and there was no difference at any distance between conditions for RPEL and RPED (Fig. 4).

The effects of salbutamol dose on mean power (A), oxygen consumption (B), and heart rate (C) during a 20-km time trial. DP, placebo; D2, 200 μg; D4, 400 μg; D8, 800 μg.
The effects of salbutamol dose on mean ventilation (A), respiratory rate (B), and tidal volume (C) during a 20-km time trial. DP, placebo; D2, 200 μg; D4, 400 μg; D8, 800 μg.
Mean ratings of perceived exertion for legs (RPEL) (A) and breathing (RPED) (B) at 2-km intervals. Rating of difficulty ranged from 1 (nothing at all) to 10 (maximal).
The effects of salbutamol on rate of perceived exertion for legs (RPEL) and breathing (RPED); values are means(SD).

Urine concentrations of salbutamol.

There was no difference in SG across conditions (DP = 1.012 ± 0.008, D2 = 1.013 ± 0.008, D4 = 1.013 ± 0.008, D8 = 1.012 ± 0.007; P = 0.90), with minimum and maximum values obtained across all trials of 1.002 and 1.032, respectively. As shown in Table 4, cSAL increased as dose increased, with D4 being greater than DP (P < 0.01), and D8 being significantly greater than all other conditions (P < 0.01). Large variability existed in cSAL across all doses, with a minimum of 0 ng·mL−1 and a maximum of 831 ng·mL−1 (Table 4). Figure 5 shows the variability in individual samples; of note is that no samples exceeded 1000 ng·mL−1 when uncorrected for specific gravity. A significant relationship between SG and cSAL was observed in conditions D4 (N = 28) and D8 (N = 30) (r = 0.42 and 0.37, respectively, P < 0.05) (Fig. 6). Alternatively, SG was not related to cSAL in the DP (N = 30) or D2 (N = 29) conditions (r = 0.18 and 0.11, respectively).

Urine concentrations of nonsulphated salbutamol (cSAL). The dashed line represents the doping control limit of 1000 ng·mL−1. DP, placebo; D2, 200 μg; D4, 400 μg; D8, 800 μg.
Relationships between specific gravity and urine concentrations of salbutamol (cSAL) after inhalation of 400-μg (A) and 800-μg (B) doses.
Urine concentrations of salbutamol (nonsulfated) at 60 min (T60) after inhalation values are reported in nanograms per milliliter.


The main purposes of this study were to examine the dose-response effect of inhaled SAL on exercise performance and urine concentration in competitive, nonasthmatic athletes. The primary findings were that SAL had no effect on 20-km time-trial performance, as measured by time and mean sustainable power, nor did it have effects on metabolic and ventilatory parameters during exercise. Urine concentrations of SAL after exercise at 1 h after inhalation increased with dose and were highly variable, although no subjects exceed the WADA cutoff of 1000 ng·mL−1. Additionally, it was found that approximately 19% of cyclists and triathletes tested in this study were susceptible to bronchospasm.

Ergogenic properties of inhaled salbutamol.

This is the first study to use a sport-specific evaluation method while examining the dose response of inhaled SAL on performance in nonasthmatic athletes. Previous research has used standard laboratory evaluations such as maximal aerobic power, anaerobic threshold, or time to exhaustion (8,10,20-22,25,30). Overwhelmingly, these studies have found inhaled SAL to have no performance-enhancing effects in athletes from a variety of different sporting backgrounds (8,10,20-22,25,30). Furthermore, Goubalt et al. (10) have shown a lack of a dose response with doses up to 800 μg in a time-to-exhaustion test at 85% of maximal oxygen consumption. Although in agreement with our current findings, the applicability of nonspecific test results to sport performance enhancement is questionable. The validity of a test to be representative of performance is an important factor when evaluating the ergogenic effects of a treatment (13). Only two studies have used sport-specific protocols, and they have provided conflicting results (22,32). After a dose of 400 μg, Norris and colleagues (22) have shown no effect on 20-km time-trial performance. At higher doses (800 μg), however, van Baak et al. (32) have demonstrated an improvement in time to complete a set amount of cycling work, suggesting that the ergogenic effects of SAL may be related to dose. Our findings do not support this concept and are in agreement with the majority of other investigations that have failed to show an ergogenic effect. The difference in findings could be attributable to the length of the protocol used (> 1 h vs approximately 30 min) and the effects of SAL on substrate use; longer protocols would require more reliance on fatty acid metabolism. However, this seems unlikely, because van Baak and colleagues (32) have shown no differences between conditions in lactate measures or substrate availability during exercise. Furthermore, we observed no differences in oxygen consumption or carbon dioxide production across doses, which is in agreement with previous findings (6,8,10,12,25). More likely is the explanation provided by Kindermann (15), who suggests that the significant difference found by van Baak and colleagues (32) is likely attributable to the influence of the worst-performing subjects, who seem to have experienced an 8-10% improvement after SAL inhalation.

One potential mechanism for SAL to have ergogenic properties may be related to its ability to act as a potent bronchodilator. Even in nonasthmatic individuals, SAL has the ability to increase airway caliber, resulting in a measurable increase in airway function (6,10,12,21,32). Theoretically, this may lead to enhanced alveolar ventilation and/or a reduced work of breathing, thereby increasing available oxygen for working muscles. However, previous reports have shown that during physical activity, SAL does not have an accumulative effect on the normal bronchodilatory response to exercise (6,10,12,21,32), nor does it reduce respiratory resistance (23). Hence, our finding that exercise ventilation was unchanged with SAL and unaffected by dose was not surprising, and it is similar to previous findings at both maximal and submaximal intensities (6,10,22,25,30). The finding that RPED was similar and that the pattern of ventilation (tidal volume and breathing frequency) did not change between conditions further supports the notion that SAL inhalation in nonasthmatics has minimal impact on ventilation during exercise. Two other studies in which subjects subjectively rated dyspnea during exercise found similar results (8,10). It has also been postulated that SAL may alter substrate use by mobilizing fatty acids and sparing glucose. Indeed, SAL has a stimulatory effect on lipolysis at rest and leads to increased fatty acid mobilization (26). However, evidence to support that acute SAL treatment augments any normal response to exercise is lacking (10,32,33).

Lastly, the present study only examined the effect of acute administrations of inhaled SAL on exercise performance. Our findings cannot preclude the possibility that short-term (~3 wk) use by this means will not have an ergogenic effect. Continued oral administration of SAL for 3 wk has resulted in enhanced endurance performance (7) and increases in peak and mean power during high-intensity cycling (16). Unlike acute administrations of SAL, short-term oral use has been shown to alter substrate availability and use during exercise (7), along with increasing strength capabilities (18). Although oral administration is currently banned by WADA, it should not be assumed that continued inhaled administration is nonergogenic. The likelihood that athletes would use the drug regularly in training as part of an overall management program is high. A constant presence of SAL in the plasma after inhalation may lead to some of the adaptations that have been associated with oral administration. Further studies of the short-term use of inhaled SAL on performance are necessary to examine this possibility.

Urine concentrations.

To our knowledge, this is the first study reporting the dose-response effect of inhaled SAL on urine concentrations after exercise. Previously reported postexercise values after low (200 μg) and high (1600 μg) inhaled doses show large variability between subjects, with the majority of samples being less than 500ng·mL−1 (34). Our findings are similar in both regards, and as dose increased, so did the variability between subjects, particularly after inhalation of 800 μg (Fig. 4). This was not surprising, because intrasubject variability of urine recovery of SAL is high (~38%) (31). It has been suggested that this variability may be attributable to individual differences in lung absorption, metabolism, renal clearance, and hydration (19). Our finding of the low but significant correlations between SG and cSAL at the two highest doses (Fig. 6) suggests that hydration may indeed play a minor role in the observed variability. This role may be more pronounced after longer events in warm environments that result in significant fluid shifts and dehydration. Samples in this study were taken at 60 min after inhalation, after only 30 min of exercise. Furthermore, exercise can also adversely effect renal function as glomerular filtration rate, osmotic clearance, and urine flow are compromised after 30 min of exercise at 85% V˙O2max (9). Considering the multiple organs and processes that are involved before the excretion of SAL, it is difficult to isolate a single reason to explain the variability in urine concentrations observed between subjects. More work specifically addressing this issue is needed.

Currently, in international sport, any urine sample containing more than 1000 ng·mL−1 of SAL is considered an adverse analytical finding by WADA. Even after four times the recommended therapeutic dose, none of the subjects in this study exceeded the limit. That said, several subjects did exceed values previously reported in the literature, both at rest and after exercise (24,34). Additionally, Schweizer and colleagues (27) have reported an in-competition measurement of 8000 ng·mL−1 in a male athlete with a TUE, and they were able to reproduce this positive test in a nonexercising trial after a dose of 900 μg administered during 5 h. Although values exceeding the WADA limit are plausible, our data would suggest that they are not the norm, even after multiple doses. Recent changes by WADA, providing the opportunity for an athlete with a TUE that has exceeded this limit to prove that values were the result of therapeutic use of inhaled salbutamol, seem appropriate.

Susceptibility to bronchospasm.

Although applications for a TUE to use SAL during the Olympic Games have been increasing during the past 20 yr (2), the percentage of athletes with asthma competing at the games is within the range of the prevalence of asthma in the general population (~4-7%). Nonetheless, there are certain sports where this rate is much higher-specifically, cycling and triathlon, where the percentages of athletes requesting TUE for SAL at the Sydney Olympic Games (2000) were approximately 17 and 20%, respectively (2). These values are similar to the incidence of bronchospasm susceptibility in our study. Of the 37 cyclists and triathletes who participated in this study, 7 had reductions > 10% in FEV1 after an EVH test, equating to approximately 19% testing positive for airway hyperresponsiveness. Furthermore, subjects were selected from a group of athletes who had not previously been diagnosed with asthma, highlighting that several athletes were competing with compromised airway function unbeknownst to them. Previous data from the 1996 U.S. Olympic Cycling team show that up to 50% of the athletes experienced asthma or airway hyperresponsiveness (35). Our findings suggest that there may be need for further education of competitive cyclists and triathletes regarding the symptoms and complications of asthma and exercise-induced bronchospasm in sport. Although several factors are known to contribute to airway hyperresponsiveness, the reasons why some groups of athletes have an increased rate of incidence are not clear and require further investigation.


In conclusion, this study has failed to demonstrate any effects of SAL on time-trial performance and ventilatory/metabolic parameters. Furthermore, the use of multiple doses up to 800 μg did not reveal trends related to dose, strengthening the consensus that acute administration of inhaled SAL does not enhance performance in endurance sports. From a doping control standpoint, although urine cSAL will generally fall under 500 ng·mL−1, with no individuals exceeding the WADA limit after doses up to 800 μg, individual responses are highly variable. This is possibly related to hydration status, but it likely depends more so on individual differences in absorption, metabolism, and renal function. Lastly, the incidence of airway hyperresponsiveness in the cyclists and triathletes evaluated in this study is significantly higher than that normally reported for the general population. Because all athletes were previously undiagnosed with asthma, further education is suggested for athletes, coaches, and medical professionals to increase the awareness and/or education with respect to the symptoms, proper diagnosis, and consequences of airway sensitivity with respect to sport.

This project was funded by the World Anti-Doping Agency. A.W. Sheel was supported by a Scholar Award from the Michael Smith Foundation for Health Research and a New Investigator award from the Canadian Institutes of Health Research.

The authors would also like to acknowledge Diana Jespersen and Wendy Pethick for their invaluable technical and lab assistance.


1. American Thoracic Society. Standardization of spirometry - 1994 update. Am J Respir Crit Care Med. 1995;152:1107-36.
2. Anderson SA, Fitch KD, Perry CP, et al. Responses to bronchial challenge submitted for approval to use inhaled B2-agonists before an event at the 2002 Winter Olympics. J Allergy Clin Immunol. 2003;111:45-50.
3. Anderson SD, Sue-Chu M, Perry CP, et al. Bronchial challenges in athletes applying to inhale a beta2-agonist at the 2004 Summer Olympics. J Allergy Clin Immunol. 2006;117:767-73.
4. Beck KC, Suman OE, Scanlon PD. Asthma: before, during, and after exercise. In: Rundell KW, Wilber RL, Lemanske RF Jr, editors. Exercise-Induced Asthma. Windsor (Canada): Human Kinetics; 2002. p. 163-80.
5. Carlsen KH, Hem E, Stensrud T, Held T, Herland K, Mowinckel P. Can asthma treatment in sports be doping? The effect of the rapid onset, long-acting inhaled beta2-agonist formoterol upon endurance performance in healthy well-trained athletes. Respir Med. 2001;95:571-6.
6. Carlsen KH, Ingjer F, Kirkegaard H, Thyness B. The effect of inhaled salbutamol and salmeterol on lung function and endurance performance in healthy well-trained athletes. Scand J Med Sci Sports 1997;7:160-5.
7. Collomp K, Candau R, Lasne F, Labsy Z, Prefaut C, De Ceaurriz J. Effects of short-term oral salbutamol administration on exercise endurance and metabolism. J Appl Physiol. 2000;89:430-6.
8. Fleck SJ, Lucia A, Storms WW, Wallach JM, Vint PF, Zimmerman SD. Effects of acute inhalation of albuterol on submaximal and maximal V˙O2 and blood lactate. Int J Sports Med. 1993;14:239-43.
9. Freund BJ, Shizuru EM, Hashiro GM, Claybaugh JR. Hormonal, electrolyte, and renal responses to exercise are intensity dependant. J Appl Physiol. 1991;70:900-6.
10. Goubault C, Perault MC, Leleu E, et al. Effects of inhaled salbutamol in exercising non-asthmatic athletes. Thorax. 2001;56:675-9.
11. Harms CA, Wetter TJ, St. Croix CM, Pegelow DF, Dempsey JA. Effects of respiratory muscle work on exercise performance. J Appl Physiol. 2000;89:131-8.
12. Heir T, Stemshaug H. Salbutamol and high-intensity treadmill running in nonasthmatic highly conditioned athletes. Scand J Med Sci Sports. 1995;5:231-6.
13. Hopkins WR, Hawley JA, Burke LM. Design and analysis of research on sport performance enhancement. Med Sci Sports Exerc. 1999;31:472-85.
14. Houghton CM, Woodcock AA, Singh D. A comparison of lung function methods for assessing dose-response effects of salbutamol. Br J Clin Pharmacol. 2004;58:134-41.
15. Kindermann W. Do inhaled beta(2)-agonists have an ergogenic potential in non-asthmatic competitive athletes? Sports Med. 2007;37:95-102.
16. Le Panse B, Collomp K, Portier H, et al. Effects of short-term salbutamol ingestion during a Wingate test. Int J Sports Med. 2005;26:518-23.
17. Lipworth BJ, Clark RA, Dhillon DP, Brown RA, McDevitt DG. B-adrenoreceptor responses to high doses of inhaled salbutamol in patients with bronchial asthma. Br J Clin Pharmacol. 1988;26:527-33.
18. Martineau L, Horan MA, Rothwell NJ, Little RA. Salbutamol, a beta 2-adrenoceptor agonist, increases skeletal muscle strength in young men. Clin Sci. 1992;83:615-21.
19. McKenzie DC. Salbutamol and the competitive athlete. Clin J Sport Med. 2004;14:316.
20. McKenzie DC, Rhodes EC, Stirling DR, et al. Salbutamol and treadmill performance in non-atopic athletes. Med Sci Sports Exerc. 1983;15:520-2.
21. Meeuwisse WH, McKenzie DC, Hopkins SR, Road JD. The effect of salbutamol on performance in elite nonasthmatic athletes. Med Sci Sports Exerc. 1992;24:1161-6.
22. Norris SR, Petersen SR, Jones RL. The effect of salbutamol on performance in endurance cyclists. Eur J Appl Physiol. 1996;73:364-8.
23. Pichon A, Roulaud M, Denjean A, de Bisschop C. Airway tone during exercise in healthy subjects: effects of salbutamol and ipratropium bromide. Int J Sports Med. 2005;26:321-6.
24. Pichon A, Venisse N, Krupka E, Perault MC, Denjean A. Urinary and blood concentrations of β2-agonists in trained subjects: comparison between routes of use. Int J Sports Med. 2006;27:187-92.
25. Sandsund M, Sue-Chu M, Helgerud J, Reinertsen RE, Bjermer L. Effect of cold exposure (−15 degrees C) and salbutamol treatment on physical performance in elite nonasthmatic cross-country skiers. Eur J Appl Physiol. 1998;77:297-304.
26. Schiffelers SLH, van Harmelen VJA, de Grauw HAJ, Saris WHM, van Baak MA. Dobutamin as selective B1-adrenoceptor agonist in in vivo studies on human thermogenesis and lipid utilization. J Appl Physiol. 1999;87:977-81.
27. Schweizer C, Saugy M, Kamber M. Doping test reveals high concentrations of salbutamol in a swiss track and field athlete. Clin J Sport Med. 2004;14:312-5.
28. Sporer BC, McKenzie DC. Reproducibility of a laboratory based 20-km time trial evaluation in competitive cyclists using the Velotron Pro ergometer. Int J Sports Med. 2007;28:940-4.
29. Stewart IB, Labreche JM, McKenzie DC. Acute formoterol administration has no ergogenic effect in nonasthmatic athletes. Med Sci Sports Exerc. 2002;34:213-7.
30. Stewart IB, Labreche JM, McKenzie DC. Effect of a long-and short-acting B2-agonist on exercise-induced arterial hypoxemia. Med Sci Sports Exerc. 2003;35:603-7.
31. Tomlinson HS, Corlett SA, Chrystyn H. Dose-response relationship and reproducibility of urinary salbutamol excretion during the first 30 min after an inhalation. Br J Clin Pharmacol. 2003;56:225-7.
32. van Baak MA, de Hon OM, Hartgens F, Kuipers H. Inhaled salbutamol and endurance cycling performance in non-asthmatic athletes. Int J Sports Med. 2004;25:533-8.
33. van Baak MA, Mayer LH, Kempinski RE, Hartgens F. Effect of salbutamol on muscle strength and endurance performance in nonasthmatic men. Med Sci Sports Exerc. 2000;32:1300-6.
34. Ventura R, Segura J, Berges R, et al. Distinction of inhaled and oral salbutamol by urine analysis using conventional screening procedures for doping control. Ther Drug Monit. 2000;22:277-82.
35. Weiler JM, Layton T, Hunt M. Asthma in United States Olympic athletes who participated in the 1996 Summer Games. J Allergy Clin Immunol. 1998;102:722-6.
36. Wilber RL. Incidence of asthma and exercise-induced asthma. In: Rundell KW, Wilber RL, Lemanske RF Jr. editors. Exercise-Induced Asthma. Windsor (Canada): Human Kinetics; 2002. p. 39-68.


© 2008 American College of Sports Medicine