In women taking the oral contraceptive pill (OC users), plasma ethinyl estradiol suppresses the production of 17β-estradiol (5) and appears in concentrations 6 times that of naturally occurring estrogen present in ovulating women (non-OC users). Although relatively smaller fluctuations in endogenous female sex hormones across the menstrual cycle can influence the cardiovascular system (1,8), ventilation (2,18), thermoregulation (14), and substrate metabolism (4) during exercise, less is known about the effect of exogenous ethinyl estradiol on the exercise response in OC users. Indeed, early research demonstrates a significant reduction in peak O2 uptake (V[Combining Dot Above]O2) in women following 4–6 months of OC use (7,15,21).
In contrast to earlier studies (7,15,21), recently published work (16,23) reported no differences in V[Combining Dot Above]O2peak between OC and non-OC users. It is difficult to interpret these opposing findings because of the differences in OC preparations and duration of administration between studies, as well as variations in experimental design. For example, one recent study used a cross-sectional design comparing women who had been using OC for at least 18 months with non-OC users and reported no differences in peak V[Combining Dot Above]O2 (16). In contrast, earlier studies reporting a lowered peak V[Combining Dot Above]O2 delivered short-term (4–6 months) administration of ethinyl estradiol in previously ovulating women (7,15,21). Therefore, it is possible that the deleterious effect of OC on peak V[Combining Dot Above]O2 is short term (i.e., <6 months) in nature.
Although the possible reduction in peak V[Combining Dot Above]O2 with OC use is intriguing, the effect of OC on endurance performance has not been thoroughly examined. Previous studies that have demonstrated a reduction in peak V[Combining Dot Above]O2 with the OC use (7,15,21) have not investigated other determinants of endurance, such as the anaerobic threshold (AT), exercise economy, and time-trial performance. Therefore, the purpose of this investigation was to comprehensively examine the effects of long-term OC use on endurance performance, as measured by peak V[Combining Dot Above]O2, AT, exercise economy, and time to exhaustion during heavy-intensity cycling. It was hypothesized that long-term OC use would be associated with (a) a decrement in markers of endurance performance (i.e., peak V[Combining Dot Above]O2, AT, and exercise economy) and (b) a decrement in endurance performance (i.e., time to exhaustion during heavy-intensity cycling). Based on the results of this study, female endurance athletes and their coaches will be able to make informed decisions about OC use with the view of optimizing performance. It is currently estimated that the prevalence of OC use in athletic females matches that within the general community and could be as high as 80% (3). Therefore, the findings of this study have the potential to provide valuable information to a large proportion of female athletes and impact their current OC practices.
Experimental Approach to the Problem
To investigate the effect of long-term OC use on endurance performance, one group of women using OC (OC group) and another group of normally menstruating women, who were not using OC (CON group), completed the study protocols in a paired approach. Baseline blood samples were collected from participants for the subsequent determination of serum 17β-estradiol and progesterone concentrations. Later, participants performed an incremental exercise test to exhaustion to determine peak V[Combining Dot Above]O2 for cycling and to estimate the AT. The subjects then completed 2 continuous submaximal cycling tests each comprising 3 distinct power outputs (two 6-minute work stages below AT, and 1 above AT performed to exhaustion). Pulmonary gas exchange, heart rate (HR), blood lactate concentration ([La−1]), ratings of perceived exertion (RPE), and time to exhaustion were measured, and cycling economy was calculated and compared between the groups.
A total of 16 women (OC group, n = 8; CON group, n = 8) participated in the present study. The subjects were apparently healthy and did not have any documented history of pulmonary, cardiovascular, or metabolic disorders. All women were recreationally active individuals (exercised >3 days per week for 30 minutes per session) who currently participated in formally scheduled team sports (e.g., netball, touch football or soccer) on a weekly basis, but they were not participating in regular cycling exercise. Each subject in the OC group was matched with a subject in the CON group for weekly duration of physical activity, age of menarche, and body mass index. Volunteers who could not be matched with an appropriate partner to form a pair were excluded from the study. Subjects were informed of the experimental procedures and provided written informed consent before any data collection, and the Griffith University Human Research Ethics Committee approved the experimental procedures.
The OC group were using combined monophasic OC for at least 12 months (mean use, 32.3 ± 24.2 months; range, 12–72 months) before starting the study and continued OC use throughout the experimental period. The CON group had regular menstrual cycles (i.e., 28-day to 30-day cycle) for at least 12 months before the study and throughout the experimental period. All subjects had never knowingly been pregnant.
Incremental Exercise Test to Exhaustion. To account for cyclic fluctuations in blood hormone concentrations, the incremental exercise test to exhaustion took place on the second day of menstruation for women in the CON group and on the second day of withdrawal (2 days after active pill cessation) for women in the OC group.
The incremental exercise comprised 3 minutes of cycling at 30 W, followed by power output increments of 10 W every 30 seconds until volitional exhaustion. Subjects were required to maintain a pedal cadence of 70 revolutions per minute throughout the test. The test was terminated when the subjects could not maintain a cadence greater than 65 revolutions per minute, despite strong verbal encouragement. The incremental exercise test was performed on an electronically braked cycle ergometer (Excalibur Sport 925900; Lode BV, Groningen, the Netherlands). Gas exchange parameters were measured breath-by-breath using a calibrated metabolic measurement system (MedGraphics Ultima CardiO2; Medical Graphics Corporation, St Paul, MN, USA). Cardiac rhythm and HR were measured continuously using a 5-lead electrode configuration (X12+; Mortara Instrument, Milwaukee, WI, USA).
Metabolic parameters were averaged over 30-second intervals to determine peak V[Combining Dot Above]O2 and to estimate the AT. Peak exercise values were determined as the highest completed 30-second stage before termination of the exercise. The AT was estimated noninvasively using the modified V-slope and ventilatory equivalents methods (20,24).
Submaximal Exercise Tests. The subjects completed a total of 4 submaximal exercise tests on separate days. The first 2 submaximal exercise tests were used as practice trials (Practice T1 and Practice T2) and performed 2 and 4 days after the incremental exercise test. One month after the incremental exercise test, subjects completed the first experimental submaximal exercise test (experimental T1), which took place on the second day of menstruation for the CON group and on the second day of withdrawal for the OC group. Three days after experimental T1, subjects completed the second submaximal exercise test (experimental T2).
The submaximal exercise tests comprised continuous cycling at a pedal cadence of 70 revolutions per minute, across 3 constant-load work stages. The first and second work stages were each 6 minutes in duration and performed at power outputs equal to 40% (40%AT) and 80% (80%AT) of the power output achieved at the AT. The third work stage was performed to exhaustion at a power output equal to 50% of the difference between the power output achieved at the AT and peak power output ([INCREMENT]50%) obtained during the incremental exercise test. The workloads were chosen to ensure that power outputs were distinctly below (moderate intensity) and above (heavy/severe intensity) the AT (6).
Subjects provided an RPE at the end of each work stage using the Borg scale (6–20). Blood was sampled from a hyperemic earlobe 5 minutes after the beginning of each work stage and 1 minute after exhaustion. Blood [La−1] was determined using an automated lactate analyzer (Lactate Pro; ARKRAY, Inc., Kyoto, Japan). The experimental equipment (i.e., electronically braked cycle ergometer, metabolic measurement system, and electrocardiograph system) used in the submaximal exercise tests were the same as that used in the incremental exercise test.
The breath-by-breath V[Combining Dot Above]O2 data from 5 to 6 minutes of each work stage were averaged and used to determine the O2 demand of cycling for each power output. Cycling economy was calculated as the change in V[Combining Dot Above]O2 for a given change in power output between the consecutive stages ([INCREMENT]V[Combining Dot Above]O2/[INCREMENT]W; ml·min−1·W−1) (13). Time to exhaustion for the final work stage was recorded to the nearest second. Timing started when the power output for the final work stage was applied and the test terminated when the subject could not maintain a cadence greater than 65 revolutions per minute, despite strong verbal encouragement. Always the same investigator performed timing of the final work stage.
Experimental Controls. Throughout the 2-month testing period, subjects were instructed to continue their normal diet and physical activity regimen. However, they were asked to avoid intense physical activity and refrain from the consumption of caffeine and alcohol for a 24-hour period preceding each exercise test. Subjects were asked to record and duplicate their food intake for 24 hours before the V[Combining Dot Above]O2peak test and before experimental T1 and T2. The participants were in a postabsorptive state having eaten a meal approximately 2 hours before each exercise test. All testing took place at the same time of that day (±2 hours) for each subject.
Determination of Hormone Concentrations
Resting venous blood samples were collected before the incremental cycling test (month 1, day 2) for the determination of serum 17β-estradiol and progesterone concentrations. Serum samples were frozen at −80°C until analysis. Serum samples were analyzed in duplicate by a commercial pathology laboratory (Sullivan Nicolaides Pathology, Gold Coast, Australia) using a radioimmunoassay technique that required no sample extraction. The intraassay and interassay coefficients of variation for 17β-estradiol ranged from 2.1 to 7.2% and 2.9 to 8.7%, respectively, and progesterone ranged from 6.5 to 8.2% and 4.5 to 8.1%, respectively (Sullivan Nicolaides Pathology).
Data and Statistical Analyses
Descriptive statistics (mean ± SD) were used to characterize group data. A priori power analysis was performed to compute the required sample size given an alpha level of 0.05 and a power of 0.80. Based on the results of Casazza et al (7), an effect size of 1.6 was used, and a total sample size of 16 was determined. The repeatability of time to exhaustion and physiological responses during experimental T1 and T2 were determined using a paired t-tests and interclass correlation coefficient (ICC). All dependent variables were analyzed using a 2 (group) × 3 (work stage) analysis of variance with repeated measures. Where statistically significant F values were detected, pairwise comparisons using Fisher’s least significant difference were performed to determine differences among the work stages and between the groups. Where statistical significance was observed between primary group variables, Cohen’s effect size (d) was calculated. Statistical Package for the Social Sciences (release 19.0; SPSS, Inc., Chicago, IL, USA) was used for all statistical analyses, and significance was accepted at p ≤ 0.05.
No significant differences (p > 0.05) in age (22 ± 3 vs. 20 ± 2 years), height (1.68 ± 0.05 vs. 1.65 ± 0.05 m), body mass (63.0 ± 7.9 vs. 60.1 ± 5.7 kg), body mass index (22.3 ± 2.2 vs. 22.1 ± 1.5 kg·m−1), age of menarche (12.9 ± 1.7 vs. 13.5 ± 1.2 years), and weekly duration of physical activity (2.9 ± 1.0 vs. 2.8 ± 1.0 h·wk−1) were observed between the CON group and the OC group, respectively. Six women in the OC group were using an OC formula containing 35 µg of ethinyl estradiol and 200 µg of cyproterone acetate and the remaining two 30 µg of ethinyl estradiol and 150 µg of levonorgestrel. Serum 17β-estradiol concentration was higher (F = 16.27; p < 0.001) in the CON group (123.5 ± 19.6 p·mol−1) compared with the OC group (47.8 ± 26.5 p·mol−1). There was no difference (F = 2.25; p = 0.13) in progesterone concentration between the CON (1.0 ± 0.3 n·mol−1) and the OC (0.6 ± 0.2 n·mol−1) groups.
Peak Exercise and Anaerobic Threshold Values
Table 1 summarizes the cardiorespiratory parameters measured during the incremental exercise test. Peak V[Combining Dot Above]O2 (F = 5.89; p = 0.03; d = −1.22) and V[Combining Dot Above]O2 at the AT (F = 6.41; p = 0.02; d = −1.24) were higher in the CON group compared with the OC group. However, when AT was expressed as a percentage of V[Combining Dot Above]O2peak, there were no differences between the 2 groups (F = 0.27; p = 0.61). Peak respiratory exchange ratio (RER) (F = 0.102; p = 0.10), peak minute expired ventilation (V[Combining Dot Above]E) (F = 0.40; p = 0.85), peak HR (F = 2.24; p = 0.16), and peak power output (F = 3.54; p = 0.08) were not different between the 2 groups.
Physiological Responses to Submaximal Exercise
No significant difference (n = 16; t = 0.159; p = 0.88) was observed in the mean cycling time to exhaustion between experimental T1 (8.23 ± 3.7 minutes) and T2 (8.27 ± 3.4 minutes). Furthermore, the ICC value of 0.95 (p < 0.01) determined in the present study indicates a high level of repeatability for time to exhaustion recorded during the submaximal cycling tests. Furthermore, no statistical differences between experimental T1 and T2 were detected for any other dependent variables (p > 0.05) and all ICCs were >0.8. Therefore, values for each variable were averaged and reported as one value to represent the 2 trials.
The HR, V[Combining Dot Above]E, [La−1], and RPE responses measured during the submaximal exercise test are listed in Table 2. Heart rate, V[Combining Dot Above]E, and RPE increased significantly with each progressive stage (p < 0.01). Blood [La−1] measured at 80%AT, [INCREMENT]50%, and at exhaustion was significantly higher (p < 0.01) than that measured at 40%AT. There were no differences in HR, V[Combining Dot Above]E, [La−1], and RPE measured at any stage between the CON group and OC group (p > 0.05). There were no differences in cycling economy (p > 0.05; see Table 3) or time to exhaustion (CON: 7.54 ± 3.01 minutes; OC: 9.48 ± 3.98 minutes; F = 1.48; p = 0.25) between the 2 groups.
The present study demonstrates that long-term OC use (>12 months) is associated with reduced peak V[Combining Dot Above]O2 and V[Combining Dot Above]O2 at the AT. However, we did not find differences in cycling economy or time to exhaustion for a submaximal cycling test between women using OC and normally menstruating women. Our findings provide evidence that despite a reduction in peak V[Combining Dot Above]O2, long-term OC use does not affect all measures of endurance.
Previous studies have reported a significant reductions in peak V[Combining Dot Above]O2 ranging from 5% to 17% after 4–6 months of OC administration (7,15,21). In agreement, we found peak V[Combining Dot Above]O2 to be 22% lower in the OC group compared with the CON group. It is important to note that in the current study, women in the OC group used contraceptives for an average of 32 months. Although the impact of OC use on O2 delivery during exercise has generally been discounted (7,10,12), one previous study (7) suggests that reduced activation of the sympathetic nervous system may be responsible for the observed reduction in peak V[Combining Dot Above]O2 with OC use. Lower plasma epinephrine, norepinephrine, and lactate concentrations at peak exercise in amenorrheic women compared with normally menstruating women (19) provides experimental evidence that reduced plasma ovarian hormone concentrations are associated with reduced sympathetic nervous system activity during maximal exercise. It was also observed that V[Combining Dot Above]O2 at the AT was 23.5% lower in the OC group compared with the CON group. The AT is thought to be influenced by muscle blood flow, lactate removal, fiber-type recruitment, and mitochondrial respiration (22). Therefore, the decrement in peak V[Combining Dot Above]O2 and V[Combining Dot Above]O2 at the AT after OC use might be because of mechanisms at the level of the muscle. Despite the differences in V[Combining Dot Above]O2 at the AT, the relative intensity at which AT occurred (i.e., percent of V[Combining Dot Above]O2) was similar between the OC and CON groups.
Given that we used a cross-sectional experimental design similar to Rebelo et al (16), who found no differences in V[Combining Dot Above]O2peak between non-OC and OC users, it is worth considering why our findings are different. Although women in both studies were of similar age, height, and body mass, peak V[Combining Dot Above]O2 was approximately 36% lower in the physically active women described by Rebelo et al (16), raising the question of an interactive effect between OC use and training status. Alternatively, given that peak exercise HR in the Rebelo et al (16) study was 15–23 b·min−1 below the age-predicted maximum, and the AT (65% V[Combining Dot Above]O2peak) was particularly high for recreationally active females (17), it is possible that these subjects did not reach a genuine peak, and this could have potentially masked the effect of OC use on peak V[Combining Dot Above]O2.
We observed no differences in cycling economy between the OC group and CON group. Additionally, there were no differences in the physiological responses during submaximal exercise (i.e., HR, V[Combining Dot Above]E, RER and blood [La−1]) between the OC group and the CON group. In contrast to our findings, an earlier study (9) reported improved running economy after 21 days of OC use without changes in the physiological responses to submaximal exercise. Therefore, the effect of OC use on exercise economy may be dependent on the mode of exercise and the duration of OC use.
Performance may be the most important factor when considering the effect of OC on endurance. We demonstrated no difference in time to exhaustion during a submaximal cycling test between the OC group and the CON group. Similarly, an earlier study (11) demonstrated that 2 months of OC use was associated with a 4.7% decrease in peak V[Combining Dot Above]O2 with no change in time to exhaustion during severe-intensity (approximately 90% V[Combining Dot Above]O2peak) running. Such observations suggest that if OC use does have a deleterious effect on peak V[Combining Dot Above]O2 and V[Combining Dot Above]O2 at the AT, the mechanisms responsible for these differences are unrelated to performance determinants.
In conclusion, the present study demonstrated that compared with normally menstruating women, long-term OC users have a lower peak V[Combining Dot Above]O2 and V[Combining Dot Above]O2 at the AT, although relative intensity at which AT occurred did not differ between non-OC and OC users. We also demonstrated that exercise economy and performance during an endurance challenge (i.e., time to exhaustion) was not different between normally menstruating women and long-term OC users. Therefore, despite a reduction in peak V[Combining Dot Above]O2 and V[Combining Dot Above]O2 at the AT, long-term OC use did not alter endurance exercise performance (when exercising at relative intensity, i.e., percent of AT and peak V[Combining Dot Above]O2), as indicated by exercise economy and time to exhaustion. We recognize that comprehensive longitudinal investigations examining several determinants of endurance will be required to confirm our findings. In addition, further work is required to elucidate the mechanisms underlying the observed decrement in peak V[Combining Dot Above]O2 and V[Combining Dot Above]O2 at the AT in OC users in comparison to normally menstruating women, and in OC users over time.
The findings of this study add to the increasing interest in the effect of OC use on performance in female athletes. Our data suggest that despite a reduction in peak V[Combining Dot Above]O2 and V[Combining Dot Above]O2 at the AT, long-term monophasic OC use does not affect cycling economy or time to exhaustion during heavy-intensity cycling. Based on the findings of this study, we propose that long-term OC use in female endurance athletes will not negatively affect endurance performance. Therefore, female athletes do not need to be concerned about the effect of long-term OC use on endurance performance.
The results do not constitute endorsement by the authors or the National Strength and Conditioning Association. No external funding was used for the completion of this study.
1. Altemus M, Roca C, Galliven E, Romanos C, Deuster P. Increased vasopressin and adrenocorticotropin responses to stress in the midluteal phase of the menstrual cycle. J Clin Endocrinol Metab 86: 2525–2530, 2001.
2. Behan M, Zabka AG, Thomas CF, Mitchell GS. Sex steroid hormones and the neural control of breathing. Respir Physiol Neurobiol 136: 249–263, 2003.
3. Bennell K, White S, Crossley K. The oral contraceptive pill: A revolution for sportswomen. Br J Sports Med 33: 231–238, 1999.
4. Bunt JC. Metabolic actions of estradiol: Significance for acute and chronic exercise responses. Med Sci Sports Exerc 22: 286–290, 1990.
5. Burrows M, Peters CE. The influence of oral contraceptives on athletic performance in female athletes. Sports Med 37: 557–574, 2007.
6. Carter H, Pringle J, Jones A, Doust J. Oxygen uptake kinetics during treadmill running across exercise intensity domains. Eur J Appl Physiol 86: 347–354, 2002.
7. Casazza GA, Suh SH, Miller BF, Navazio FM, Brooks GA. Effects of oral contraceptives on peak exercise capacity. J Appl Physiol 93: 1698–1702, 2002.
8. Chan NN, MacAllister RJ, Colhoun HM, Vallance P, Hingorani AD. Changes in endothelium-dependent vasodilatation and alpha-adrenergic responses in resistance vessels during the menstrual cycle in healthy women. J Clin Endocrinol Metab 86: 2499–2504, 2001.
9. Giacomoni M, Falgairette G. Decreased submaximal oxygen uptake during short duration oral contraceptive use: A randomized cross-over trial in premenopausal women. Ergonomics 43: 1559–1570, 2000.
10. Larsson G, Milsom I, Lindstedt G, Rybo G. The influence of a low-dose combined oral contraceptive on menstrual blood loss and iron status. Contraception 46: 327–334, 1992.
11. Lebrun CM, Petit MA, McKenzie DC, Taunton JE, Prior JC. Decreased maximal aerobic capacity with use of a triphasic oral contraceptive in highly active women: A randomised controlled trial. Br J Sports Med 37: 315–320, 2003.
12. Lehtovirta P, Kuikka J, Pyorala T. Hemodynamic effects of oral contraceptives during exercise. Int J Gynaecol Obstet 15: 35–37, 1977.
13. Loveless DJ, Weber CL, Haseler LJ, Schneider DA. Maximal leg-strength training improves cycling economy in previously untrained men. Med Sci Sports Exerc 37: 1231–1236, 2005.
14. Marsh SA, Jenkins DG. Physiological responses to the menstrual cycle: Implications for the development of heat illness in female athletes. Sports Med 32: 601–614, 2002.
15. Notelovitz M, Zauner C, McKenzie L, Suggs Y, Fields C, Kitchens C. The effect of low-dose oral contraceptives on cardiorespiratory function, coagulation, and lipids in exercising young women: A preliminary report. Am J Obstet Gynecol 156: 591–598, 1987.
16. Rebelo AC, Zuttin RS, Verlengia R, Cesar MC, de Sa MF, da Silva E. Effect of low-dose combined ora contraception on aerobic capacity and anaerobic threshold level in active and sedentary young women. Contraception 81: 309–315, 2010.
17. Robbins JL, Duscha BD, Bensimhon DR, Wasserman K, Hansen JE, Houmard JA, Annex BH, Kraus WE. A sex-specific relationship between capillary density and anaerobic threshold. J Appl Physiol 106: 1181–1186, 2009.
18. Saaresranta T, Polo O. Hormones and breathing. Chest 122: 2165–2182, 2002.
19. Schaal K, Van Loan MD, Casazza GA. Reduced catecholamine response to exercise in amenorrheic athletes. Med Sci Sports Exerc 43: 34–43, 2011.
20. Sue DY, Wasserman K, Moricca RB, Casaburi R. Metabolic acidosis during exercise in patients with chronic obstructive pulmonary disease. Use of the V-slope method for anaerobic threshold determination. Chest 94: 931–938, 1988.
21. Suh SH, Casazza GA, Horning MA, Miller BF, Brooks GA. Effects of oral contraceptives on glucose flux and substrate oxidation rates during rest and exercise. J Appl Physiol 94: 285–294, 2003.
22. Svedahl K, MacIntosh BR. Anaerobic threshold: The concept and methods of measurement. Can J Appl Physiol 28: 299–323, 2003.
23. Vaiksaar S, Jurimae J, Maestu J, Purge P, Kalytka S, Shakhlina L, Jurimae T. No effect of menstrual cycle phase and oral contraceptive use on endurance performance in rowers. J Strength Cond Res 25: 1571–1578, 2011.
24. Wasserman K, Hansen J, Sue D, Stringer W, Whipp B. Principles of Exercise Testing and Interpretation: Including Pathophysiology and Clinical Applications. Philadelphia, PA: Lippincott, Williams, and Wilkins, 2005.