Share this article on:

00005768-199712000-0001000005768_1997_29_1609_williams_menstrual_12miscellaneous-article< 141_0_7_6 >Medicine & Science in Sports & Exercise©1997The American College of Sports MedicineVolume 29(12)December 1997pp 1609-1618Menstrual cycle phase and running economy[Applied Sciences: Biodynamics]WILLIAMS, TRACY J.; KRAHENBUHL, GARY S.Exercise and Sport Research Institute, Arizona State University, Tempe, AZ 85287Submitted for publication March 1996Accepted for publication May 1997.ABSTRACTMenstrual cycle phase and running economy. Med. Sci. Sports Exerc., Vol. 29, No. 12, pp. 1609-1618, 1997. To further elucidate the relationship between RE and menstrual cycle phase, eight eumenorrheic moderately-trained female runners were studied throughout their menstrual cycles, which were divided into five phases: early follicular (EF), late follicular (LF), early luteal (EL), mid-luteal (ML), and late luteal (LL). Subjects were studied at rest and while running at speeds initially corresponding to 55% and 80% maximal oxygen consumption (˙VO2max). Ventilation (L·min-1) was significantly (P < 0.05) higher in ML compared with EF during all three conditions (mean ± SE)(rest: 12.4 ± 0.7 vs 10.3 ± 0.8; 55% ˙VO2max: 46.2± 0.9 vs 42.2 ± 1.4; and 80% ˙VO2max: 68.8 ± 3.0 vs 63.3 ± 2.0 L·min-1, respectively). Resting˙VO2 (mL·kg-1·min-1) was significantly(P < 0.05) higher in ML (4.8 ± 0.1) compared with EF (3.9± 0.2). Profile of Mood States (POMS) total mood disturbance (TMD) and three subscale (depression, fatigue, and confusion) scores were also significantly higher during ML compared with EF; TMD: 127 ± 6.0 vs 104± 6.0; depression: 6 ± 1.4 vs 3 ± 1.4; fatigue: 9± 1.0 vs 4 ± 0.9; and confusion: 7 ± 0.9 vs 5 ± 1.2, respectively. The POMS vigor subscale score was significantly lower during ML (11 ± 1.5) when compared with EF (19 ± 0.7). RE at speeds corresponding to 55% ˙VO2max was not significantly different between phases of the menstrual cycle. RE at speeds corresponding to 80%˙VO2max was, however, significantly less (higher ˙VO2) during ML (41.4 ± 0.8 mL·kg-1min-1) than EF (40.2± 0.5 mL·kg-1·min-1). It was concluded that RE at speeds corresponding to 80% ˙VO2max in moderately-trained female runners covaries independently with ventilatory drive changes and with fluctuations in mood state which occur throughout the menstrual cycle.Maximal oxygen consumption (˙VO2max) has been shown to be important in determining one's ability to succeed competitively in endurance running events (13). One factor that differentiates the performance of runners homogeneous with respect to ˙VO2max is running economy (RE), which is defined as the rate of oxygen consumption(˙VO2) during a given submaximal steady-state running velocity(12). A person with a higher submaximal ˙VO2 at a given speed is considered less economical than one with a lower˙VO2.Previous research has reported a number of variables that appear to be associated with differences in RE, such as: intraindividual variation(16), age (30), temperature(34), fatigue (7), training(40), psychological mood states(49), numerous biomechanical parameters(48), and gender (4,15). However, the results of some of these studies are equivocal. In the area of gender differences, some researchers have reported that males are more economical (lower ˙VO2) than females (4) while others have reported no difference (15). Unfortunately, neither of these researchers studying female subjects have taken into account the possible influence of the menstrual cycle on RE.It is well established that progesterone levels can increase 10-fold from the follicular to the luteal phase of a eumenorrheic (ovulatory cycle with normal hormonal levels) menstrual cycle. Progesterone has been shown to be associated with hyperventilation and a decrease in alveolar carbon dioxide tension during pregnancy (32) and during the luteal phase of the menstrual cycle (21,41,43). Research has also indicated that during the luteal phase of the menstrual cycle, eumenorrheic women have an increase in hypercapnic ventilatory response(HCVR) as compared with that in the follicular phase(43). This is believed to be because of progesterone's ability to lower the threshold of the medullary respiratory center and increase its excitability (32). The effect of progesterone on ventilation was verified by Tok and Loeschcke's(47) animal research which showed a marked increase in ventilation when progesterone was administered directly on the medulla of a cat.There is a certain oxygen demand associated with ventilation, such that an increase in ventilation is associated with an increase in oxygen consumption(1). Progesterone itself, as well as its thermogenic effect, is associated with an increase in ventilation; therefore, its elevation should be accompanied by an increase in oxygen demand. This would imply that during the luteal phase of the menstrual cycle, when progesterone levels are high and there is an increase in ventilatory drive, there should be an additional oxygen requirement to support heightened ventilation. If the increase in ventilation (compared with that in the follicular phase) persists during exercise, an increased oxygen demand should also be exhibited, thereby elevating the oxygen consumption during a given steady-state submaximal running velocity (RE).Endurance-trained athletes differ from nonathletes in that they exhibit decreased hypoxic and hypercapnic ventilatory drives at rest(6) and during exercise (35). In 1981, Schoene et al. (43) hypothesized that if there were a functional advantage for athletes to possess diminished ventilatory drives, then any factor resulting in an increase in ventilatory drive may impair exercise performance. They studied three groups of females: eumenorrheic athletes and nonathletes, and amenorrheic athletes to determine the influence of menstrual cycle phase (follicular vs luteal) on maximal cycle ergometer performance and ventilatory drive. Amenorrheic athletes studied 2 wk apart did not exhibit any significant differences between the two testing periods, while the eumenorrheic subjects demonstrated higher resting VE and HCVR slopes during the luteal phase. The nonathletes, however, were the only group to exhibit a significant decrease in maximal exercise performance. A recent study by Lebrun et al. (31) reported slightly lower treadmill ˙VO2max (mL·kg-1·min-1) values in the mid-luteal phase as compared to the mid-follicular phase in 16 trained eumenorrheic women, although the values were not statistically significant. While subsequent studies using eumenorrheic nonathletes(19) and nonathletic males administered medroxyprogesterone acetate (MPA) (2) did not show significant differences in maximal cycle ergometer performance between phases or differing levels of MPA, a study on eumenorrheic athletes by DeSouza et al.(18) supports the finding that athletes do not appear to experience significant differences in ˙VO2max between follicular and luteal phases on the menstrual cycle.While the majority of data supports the view that maximal exercise does not appear to be affected by menstrual cycle phase(18,19,43), the literature concerning submaximal exercise responses contains conflicting results. It has been reported that cycle ergometer performance at intensities of 60%˙VO2max or less (25) and 70%˙VO2max (24) are associated with no significant difference in ˙VO2 during the luteal phase; that performance of high-intensity (90% ˙VO2max) exercise is enhanced in the luteal phase (29); and that there is no difference in high intensity treadmill endurance capacity (90% ˙VO2max) between phases (31).Much of the research comparing various parameters and menstrual cycle phases involves data collected on only one day during each of the phases(19,25,31,43). The time course of ventilatory changes throughout the menstrual cycle has yet to be determined; therefore, important patterns of change significant to endurance exercise may have been missed.The purpose of the present study was to provide more detailed information of the patterns of change in ventilatory drive and the association between menstrual cycle phase and running economy in moderately-trained eumenorrheic female runners. The following factors contributed to the research design: 1) research has shown that progesterone increases ventilatory drive and ventilation; 2) ventilation requires oxygen; 3) though results of submaximal exercise responses during the menstrual cycle are equivocal regarding differences between cycle phase, some research has indicated that there are response differences at intensities less than 60% ˙VO2max (2). Such differences have not generally been observed at intensities greater than 60%˙VO2max. It was therefore hypothesized that the mid-luteal increases in ventilatory drive, ventilation, ˙VO2, and a more negative mood state would be strong enough to influence (increase) aerobic demands during a treadmill run of modest intensity at steady-state velocity(55% ˙VO2max). However, the greater demand for oxygen when exercising at a higher intensity (80% ˙VO2max) was expected to obliterate the changes observed at the lower intensity.METHODSSubjects. Ten moderately trained female runners volunteered to serve as subjects and completed all test sessions. A moderately trained runner was defined as a runner who had been engaged in a systematic endurance running program (4-8 miles·d-1, at least 3 d·wk-1) for at least 1 yr before the start of this study and had a current (within the last 6 months) 10 km race time between 40 and 48 min. Each subject was asked to maintain a consistent training program throughout the study to avoid training-induced alterations in RE. Each subject completed daily training logs for verification of adherence to this requirement. Weekly mileage was held constant for each subject and ranged from 12 to 26 miles·wk-1 across subjects, with a group mean of 15.75 miles·wk-1.The subjects were initially considered to be eumenorrheic (11-14 cycles per year), based upon the previous two year's menstrual cycle history. Confirmation of menstrual status during the study was obtained through analysis of daily basal body temperature charts, serum estradiol and progesterone levels, and mid-cycle luteinizing hormone (LH) surge. Females who had been pregnant or had taken oral contraceptives (or any other type of hormonal drug) within the 6 months preceding data collection were excluded from participating. In addition, only nonsmokers and females with no known cardiopulmonary disease were eligible for participation.Written informed consent was obtained from each subject before testing. The protocol was explained thoroughly to each subject in a manner that would not bias the results.Preparatory data collection (Month 1). A sample timeline outlining the entire three months of data collection (Preparatory, Primary, and Post) is located in Table 1. During the follicular phase of the month before the primary data collection, each subject performed the following preparatory tests: ˙VO2max treadmill test (which was also repeated as a post test during the follicular phase following the period of primary data collection), two level submaximal treadmill accommodation runs(before the collection of the submaximal data), four level submaximal runs for the construction of a personalized economy curve, and a 3-site skinfold test. All preparatory test sessions (Month 1) were conducted during the follicular phase of the subject's menstrual cycle. In addition, during both the preparatory testing (Month 1) and the primary test sessions (Month 2), each subject recorded her daily basal body temperature.TABLE 1. Sample data collection timeline.Maximal oxygen consumption. The ˙VO2max protocol used in this study was similar to that employed by Bransford and Howley(4). Following treadmill accommodation, each subject warmed up on the treadmill for 3 min at 2.68 m·s-1 while at a 0% grade. At minute three, the treadmill speed was increased to 3.13 m·s-1 while the grade remained level. Every 2 min thereafter, the grade was increased by 2.5% until the subject reached volitional exhaustion.Expired air was sampled using an on-line SensorMedics 2900 Metabolic Measurement System (SensorMedics, Yorba Linda, CA). Expired flow, O2 concentration, and CO2 concentration were continuously sampled during the test. The analyzers were calibrated before and after each test using standard gases of known concentration. ˙VO2max was defined as the value that was achieved concurrently with the occurrence of a plateau in˙VO2 (a change of ˙VO2 of 2.1 mL·kg-1·min-1 or less) with an increase in workload (46).Basal body temperature. At the time the subject signed the informed consent, she was supplied with a basal thermometer, instruction sheet, a basal body temperature chart, and a pencil (Serophene Patient Starter Kit, Serono Laboratories, Inc., Norwell, MA). Subjects were told that they should start charting their basal body temperatures beginning with the first day of their menstrual cycle. They were asked to use an “X” for the days of menses for which there was menstrual flow. Subjects were told that their basal temperature was to be recorded every morning immediately upon awakening. They were advised that activities such as going to the bathroom, eating, or drinking could increase body temperature and thus should be avoided before taking their temperature.Subjects were instructed to place the basal thermometer under the tongue for 5 min. The temperature was then to be recorded by placing a dot (or an“X”, if during menses) at the corresponding temperature on the graph. It was also explained to the subjects that it was important to note on the graph anything that may have increased the temperature, such as being ill, having a restless night, or forgetting to take the temperature before getting up. Subjects also had an orientation session in which they were taught how to shake down and read the thermometer.At the completion of the study a visual inspection of the charts was made to determine whether the charts were “biphasic,” in that there was an approximate mid-cycle sustained (7-14 d) rise in temperature of at least 0.2 °C (17).Skinfolds. At least 24 h after the ˙VO2max test, each subject returned to the laboratory to have her percent body fat estimated by a three-site skinfold test. The three sites were the triceps, the abdomen, and the suprailiac. The measurements were performed in triplicate, with the average values then entered into the Jackson-Pollock 3-site skinfold equation for females (27). Even though 6-site skinfolds are generally considered to be slightly more accurate than the 3-site, the 3-site skinfolds were considered adequate for the descriptive purposes of the present study.The decision to use skinfolds to estimate percent body fat, as opposed to hydrostatic weighing, was based upon recent evidence that hydrostatic weighing may not be appropriate to use with female athletic populations because of differences in body density found in this population(5,26). A recent study by Lebrun et al.(31) reported no differences between follicular and luteal phases in either skinfold sums or percent body fat as determined by hydrostatic weighing.Treadmill accommodation. Following the skinfold test each subject participated in the first of 2 d of level treadmill accommodation runs. The objective of these two sessions was to ensure the establishment of stable gait mechanics and reduce novelty stress before the RE primary data collection. Both sessions consisted of 15 min of level running at each of two speeds which were estimated to approximate 55% and 80% ˙VO2max (30 min each session). Each subject was therefore exposed to a total of 60 min (two 30-min sessions) of treadmill running before the primary data collection sessions. This amount of time has been shown to be adequate for novice treadmill runners to acquire consistent gait mechanics on the treadmill(8,42) and was therefore assumed to be adequate for the subjects in the present study.Economy curve. The final preliminary test session occurred on the day following the last treadmill accommodation run. Each subject ran for 6 min at each of four speeds (2.23, 2.68, 3.13, and 3.58 m·s-1) to construct an economy curve. The metabolic and ventilatory data were obtained using the same equipment as described previously (under the ˙VO2max section).By knowing the ˙VO2max and the corresponding submaximal˙VO2 values from four velocities, it was possible to construct an economy curve for each subject. Economy curves are based on the knowledge that within a range of running speeds a linear regression equation provides an adequate description of the relationship between velocity and ˙VO2(14). The SEE for the present study ranged from 0.1-1.0 mL·kg-1·min-1. With this linear equation, it was possible to estimate the speeds at which each subject would run at relative workloads of 55% and 80% ˙VO2max. These absolute speeds were then used throughout the RE tests run during the primary data collection period. These two relative workloads allowed the subjects to be tested: (a) at an intensity at which previous research had demonstrated differences in various exercise parameters: ˙VO2, ventilation, ratings of perceived exertion, and time to exhaustion (<60% ˙VO2max); and (b) at an intensity that would feature steady-state conditions and more closely approximated an actual race pace (80% ˙VO2max).Primary data collection (Month 2). Beginning on Day 7 of the follicular phase of the second month each subject came to the laboratory for data collection every third day for the duration of her menstrual cycle. The temperature, relative humidity, and barometric pressure of the lab were held constant at 21-23 °C, 55-60%, and 730-738 mm Hg, respectively. All of the running tests for a given subject were conducted at the same time of day to eliminate potential confounding of primary data as a result of circadian fluctuations. Each subject was asked to refrain from exercise for at least 8 h before coming to the lab. In addition, each subject came to the laboratory in a fasted state (12 h), since consumed food can influence oxygen consumption and testing results (3). Each subject also performed all running sessions (personal training and laboratory testing) in the same pair of running shoes since footwear design and mass loading of the foot have been shown to influence the aerobic demands of running(23).Hormonal assays. Each subject had basal serum estradiol and progesterone samples taken between 0600-0800 h on the morning of her exercise tests. Following an overnight fast, each subject had 5 mL of blood drawn from an antecubital vein. The vacutainer was immediately put on ice. Following centrifugation, the serum was stored at -20 °C. All samples were analyzed at the end of the study in one assay to minimize interassay variability. Estradiol and progesterone concentrations were analyzed in duplicate using radioimmunoassay (double antibody) estradiol and progesterone kits from ICN Biomedicals, Inc. (Costa Mesa, CA). The intra-assay coefficient of variation was 10% and 13% for serum estradiol and progesterone, respectively. Hormone values are reported as the average of duplicate analyses.For the purposes of this study, for the subjects to be considered eumenorrheic they must have met the following hormonal requirements: 1) follicular estradiol levels of 350-2000 pmol·L-1 and luteal levels of 70-1100 pmol·L-1, and 2) follicular progesterone values of <1-3 nmol·L-1 and luteal values of 9-95 nmol·L-1 (26,44).Psychological test. Before each exercise session, each subject completed a Profile of Mood States Inventory (POMS) (33) to assess her mood state on that day. The subjects were given the verbal instructions to complete the questionnaire according to how they felt“Right Now.” The POMS is a 65-adjective questionnaire that yields scores for the following six mood states: tension, depression, anger, vigor, fatigue, and confusion. With the exception of vigor, low values are indicative of a more positive mood state. In addition, a Total Mood Disturbance (TMD) score was obtained in the standard way (33) (by subtracting the vigor subscore from the total of the other five constructs and then adding the constant of 100 to the difference to avoid negative scores). A TMD score gives an indication of overall mood state. The higher the TMD score, the worse the overall mood state.Resting ˙VO2. Following the completion of the POMS, breath-by-breath values of ˙VO2 and ventilation (VE) were measured by a SensorMedics 2900 Measurement System. Subjects were seated quietly for 30 min and data was collected once steady-state VE and˙VO2 were attained (resting VE and ˙VO2).Hypercapnic ventilatory response test. Once resting˙VO2 values had been attained, each subject performed a hyperoxic, progressive hypercapnic ventilatory response (HCVR) test in a manner similar to that described by Clark and Read (9). Breath-by-breath values of end-tidal CO2 (PETCO2) and ventilation (VE) were measured by a SensorMedics 2900 Measurement System. Subjects were seated and initially breathed room air until steady-state VE and ˙VO2 values were attained. Breathing was then shunted to a bag initially containing 94% O2 and 6% CO2. Subjects rebreathed the air from this bag until PETCO2 values reached 55-65 mm Hg (usually 4 min), or sooner if the subject requested. Music from a radio was played during the HCVR test to distract the subjects from the test. From the data obtained, a linear regression of VE = S(PETCO2 - B) was performed to determine the slope (S) and the PETCO2 intercept (B) values(9). The values for S were taken as indicators of ventilatory drive sensitivity (9,43).Running economy test. The RE sessions consisted of having each subject run for 6 min at each of two absolute speeds representing the appropriate relative intensities (55% and 80% ˙VO2max), which had been determined using the running velocity/oxygen consumption regression equation developed during the preliminary test sessions. After the first 6-min bout, the treadmill belt speed was increased to the next pace. Treadmill belt speed was calibrated for each subject during the initial minute of each speed by recording the time that elapsed during 10 revolutions of the treadmill belt; if necessary, adjustments were made during the second minute to achieve the intended pace.Running economy and other metabolic data were determined using a SensorMedics 2900 Measurement System, which collects information continuously and provides output every 20 s. Values obtained over the last 2 min of each 6-min run were averaged to represent the value for that speed on that day(average of six 20-s values per speed).Luteinizing hormone. Each subject was given two FIRST RESPONSE Ovulation Predictor Test Kits (5-d) (Hygeia Sciences, Inc, a subsidary of Carter-Wallace, Inc., New York). This kit measures the subject's urinary luteinizing hormone (LH) concentration. Each kit contained instructions that were both written and pictorial.The subject was asked to read the instructions that were included with the kits. An investigator then verbally explained the instructions to the subjects and then observed while the subject demonstrated the procedure with a practice kit. Once the subject demonstrated proficiency, she was allowed to take the kits, containing the written instructions, home.Based on previous menstrual cycle length history and kit instructions, each subject was verbally instructed and then given a sheet of paper indicating on which day she was to begin daily testing of her morning urine using the LH kits. The subject was to continue daily testing until the test indicated that she had a high concentration of LH, reflecting an LH surge, which presumably predicted the subsequent occurrence of ovulation.Statistics. Pre and postexperiment ˙VO2max values were analyzed with a paired t-test to determine whether a training or detraining effect had occurred. Consistent with the experimental strategy of isolating and comparing points in the cycle associated with extremes in progesterone concentrations, the resting and exercise variables were analyzed using planned contrasts ANOVA (SPSS-X, Chicago, IL). The planned comparisons contrasted the means of the EF and ML data points, since these were the two that were expected to exhibit the largest differences.Because each subject had a different cycle length (27.9 ± 4.8 d), and therefore, a different number of tests (range: 7-10 tests), the data were aggregated such that each subject had a value for the following phases: early follicular (EF), late follicular (LF), early luteal (EL), mid-luteal (ML), and late luteal (LL). With the aid of the basal body temperature (BBT) charts, LH kits and serum hormone levels, each subject's menstrual cycle was divided into a follicular and luteal phase. The first day of menses and those days up to and including the day of the LH surge were considered the follicular phase. The day after the LH surge up to the first day of the next menses were considered the luteal phase. Each subject's follicular phase was then divided in half, excluding the first 6 d. The results of tests that occurred in the first half were averaged to give a value for the early follicular (EF) phase. Those results of tests that occurred during the second half of the follicular phase were also averaged to give a value for the late follicular phase. The luteal phase was divided into thirds with tests in the first third being averaged to give a value for the early luteal phase, the second third being the mid luteal phase, and the last third being the late luteal phase. Statistical significance was accepted at P < 0.05. Testing 10 subjects would have been sufficient for detecting a significant (P = 0.05, power = 0.80) difference for an effect size (SD units) of 1.25 mL·kg-1-1min-1 (50).RESULTSMenstrual cycle phase. Serum estradiol and progesterone levels were used in conjunction with basal body temperature charts and urinary LH surge predictor kits to confirm menstrual cycle phase and status. Hormonal analysis indicated that two of the 10 subjects did not have normal cycles during data collection. One subject had extremely low levels of estradiol and progesterone (112-287 pmol·L-1 and < 1-2 nmol·L-1, respectively). The other subject did not exhibit a follicular increase in estradiol and her progesterone levels started to increase but then decreased before increasing again, indicating a failure to ovulate. The eight subjects with eumenorrheic hormone profiles exhibited typical biphasic basal body temperature charts and documented the proper color change, white to medium pink, necessary to indicate an LH surge according to the ovulatory predictor kits, while the two women with atypical hormone profiles demonstrated no distinct patterns to their basal body temperature charts and though their urinary kits changed color, the color specified by the kit as an indication of an LH surge was not attained. Both subjects were therefore excluded from the data analysis, which resulted in a final subject population of eight. The subsequent loss of two subjects slightly decreased the probability of detecting significant differences (Power = 0.7 with eight subjects). Planned contrasts revealed that serum progesterone values from both EF and LF phases were significantly (P < 0.05) lower than the values from the EL, ML, and LL phases (Table 2). In addition, ML values were significantly higher than both EL and LL values.TABLE 2. Resting values (mean (SE)) of the eight moderately trained female distance runners.The results revealed that EF serum estradiol values were significantly lower than LF values, and LF values were significantly higher than those from EL, ML, and LL phases (Table 2). In addition, ML serum estradiol values were significantly higher than EF, EL, and LL values.Subject characteristics. Physical characteristics of the subjects are summarized in Table 3. The results of a pairedt- test indicated that the mean difference (± SE) of -1.1(± 0.6) mL·kg-1·min-1 between the pre- and postexperiment ˙VO2max values was not statistically significant.TABLE 3. Descriptive characteristics of the eight moderately trained female distance runners.The estimations of relative body fat are also similar to those reported in the literature, in that sedentary women have higher and elite runners have lower percent body fat values than those exhibited by the moderately trained runners in the present study.Resting responses. Resting ˙VO2 during the EF phase was significantly lower than in the ML phase (Table 2). In the current investigation VE was significantly higher during the ML phase when compared with that during the EF phase (Table 2). Planned contrasts indicated that the HCVR slopes during the EL, ML, and LL phases were significantly higher than during the EF phase. In addition, the HCVR slopes during the EL and ML phases were significantly higher than the values from the LF phase (Table 2).Psychological responses. Examination of the psychological parameters revealed the following results: TMD values during EF phase were significantly lower than in the ML and LL phases, and the LF TMD values were also significantly lower than the ML values (Table 4). Scores for depression, fatigue, and confusion during the EF phase were significantly lower than in the ML phase; and vigor scores during the EF phase were significantly higher than scores from EL, ML, and LL phases(Table 4).TABLE 4. Profile of Mood States scores (mean (SE)) during five phases of the menstrual cycle for eight eumenorrheic moderately trained female distance runners.Exercise responses. The results of the planned contrasts for values obtained during treadmill running at the absolute speeds corresponding to 55% and 80% ˙VO2max intensities indicated that EF VE was significantly lower than ML VE (Table 5).TABLE 5. Exercise (55% and 80% V2max) values (mean (SE)) of the eight moderately trained female distance runners.The results from the present study revealed that VE during the ML phase continued to be significantly higher than EF at both submaximal running intensities. Even though VE was significantly elevated during the ML phase, there was no difference in RE at speeds corresponding to 55%˙VO2max. At the speed corresponding to 80% ˙VO2max a significant increase (greater oxygen demand) in RE occurred between the EF and ML phase (Table 5).DISCUSSIONIn this study 10 moderately trained female distance runners were tested during their menstrual cycle to determine whether there was a relationship between menstrual cycle phase and running economy. A critical aspect of this study, therefore, was the confirmation of a normal menstrual cycle. While two of the subjects were subsequently excluded from the study, the eight remaining subjects displayed hormone profiles and basal body temperatures that fit the requirements of this experiment and were consistent with those reported in the literature for eumenorrheic adult women (26,44). The results of all three methods of detecting an ovulatory menstrual cycle(basal body temperatures, LH surge, and concentrations of estradiol and progesterone) were therefore in close agreement, which increased confidence that both eumenorrheic and atypical menstrual cycles were detected and properly identified in the present study.The assumption that the subjects in this study were moderately trained was supported by the ˙VO2max values attained. The ˙VO2max values reported in the present study are typical of those reported in the literature; values were higher than those reported for sedentary women(4,20), but lower than those obtained for elite runners (15,22,51). Another important component of the study was that the pre- and post-˙VO2max values were not significantly different (Table 3). This was important because of the possibility for running training to result in an improvement in RE (4), which would have confounded the results. In addition, as most studies have shown that ˙VO2max does not change throughout the menstrual cycle(18,19,43), it was assumed that˙VO2max remained stable throughout the study and, therefore, did not affect RE.The association between resting ˙VO2(mL·kg-1·min-1) and menstrual cycle phase remains ill defined (25,45). While numerous factors have the potential to influence resting ˙VO2 measures, it is quite possible that those who have reported no difference in resting ˙VO2 between phases may have confounded their results by including anovulatory subjects. For example, Stephenson et al. (45) reported that oxygen uptake was not statistically different at rest between phases in six adult female subjects who were considered to be eumenorrheic based solely upon cycle length. In addition, it appears that values presented were the averages of three different menstrual cycles. Not only is it possible to have a cycle which is normal in length but is anovulatory, it is also possible to confound the results by averaging across menstrual cycles since it is known that hormone concentrations vary from cycle to cycle(28). When studying the relationship between variables and the menstrual cycle, it is critical to verify menstrual status.The results of the present study do, however, agree with those of Hessemer and Bruck (25), in which 10 eumenorrheic women (athletes and nonathletes) were studied at rest during the follicular and luteal phases of menstrual cycle. The authors reported that ˙VO2 was increased at rest in the luteal phase compared with that in the follicular phase. These subjects also had ovulation determined via basal body temperature and phase later confirmed by serum progesterone levels. Data was, however, collected on only one follicular day (4-7 d after menses) and one luteal day (4-8 d after rise in BBT). The combination of a slightly larger subject population(N = 10) in the Hessemer and Bruck (25) study versus the Stephenson et al. (N = 6) study(45) and the range of subject cycle lengths and testing days may have allowed the authors (25) to detect the differences in ˙VO2 at rest between the phases.Alhough a few studies have reported no difference in resting VE between phases (18,45), the majority of research supports the present finding of an increase in VE during the ML phase or when progesterone levels are high(2,19,21,43). Given the stimulatory effect of progesterone on ventilation, one would expect to observe an increase in VE during the luteal phase, in eumenorrheic women, when progesterone levels are high.Those studies showing no difference in resting VE between phases may have inadvertently included anovulatory subjects (45), or may have missed important observations by testing only once during the luteal phase (18). In 1990, DeSouza et al.(18) examined resting, submaximal (40 min run at 80%˙VO2max), and maximal treadmill runs in eumenorrheic and amenorrheic well-trained collegiate runners in one early follicular and one mid-luteal day of the menstrual cycle. Subjects may have been at different points in the luteal phase, with some subjects being tested when progesterone was starting to rise and VE was not yet being stimulated maximally, and others during progesterone's peak or decline. When such values are averaged, the increase in VE could easily be obscured such that there appeared to be no differences. The frequent testing schedule in the present study allowed data to be collected in the same relative phase of the cycle. The subjects' values for VE were therefore being averaged during periods of similar hormonal profiles. This increased the homogeneity of the phases and the likelihood that similar differences would be detected.The increase in resting VE during the luteal phase is typically explained as being the result of the stimulatory effect of progesterone on chemoreceptors. This effect has traditionally been determined by measuring the slope (S) of the ventilatory response to hypercapnia (HCVR) or hypoxia (HVR). The luteal phase results in both the Schoene et al. (43) study and the present study indicate that there was a significant increase in ventilatory drive in the ML phase (high progesterone) compared with that in the EF phase. This indicates that the subjects in the present study exhibited a typical response to the stimulatory effect of progesterone.It has been reported that male elite athletes exhibit an“iceberg” profile when the POMS subscales are plotted, with vigor being the tip of the iceberg (38). Similar findings have been reported for female elite distance runners (37); however, menstrual cycle phase did not appear to be taken into consideration. The results of the present study indicate that in the follicular phase, when the average of the subject's individual subscales of the POMS was plotted, it resembled the “iceberg” profile (Fig. 1) similar to those reported for athletes (37); however, during the ML phase the subjects experienced an altered mood state featuring significantly higher mean values for TMD, depression, fatigue, and confusion scales (Table 4). In addition, the vigor subscale score was significantly lower during the luteal phase than the follicular phase(Table 4). These changes portray a less positive mental health profile during the ML phase. These findings appear to be in agreement with those previously reported (10,11), in that there is an increase in negative moods during the luteal phase, although a direct correlation between hormonal fluxes and mood alterations has not previously been demonstrated.Figure 1-Profile sheet of the six subscales from the Profile of Mood States (POMS) for the Early Follicular and Mid Luteal dates for the eight eumenorrheic moderately trained female distance runners.In summary, the results of the variables analyzed at rest are in agreement with the menstrual cycle literature, indicating that the subjects were eumenorrheic and demonstrated typical menstrual cycle related changes.A number of studies have been conducted since it was first suggested that the diminished ventilatory drive exhibited by athletes may be beneficial for performance and thus any factor that increased ventilatory drive would be detrimental for performance. The results have been conflicting, and as mentioned previously, much of the variation could be a result of differences in the point in the phase at which data collection occurred. It is generally accepted that exercise responses to submaximal cycle ergometry are fairly consistent among and within subjects, and although running economy is fairly stable in terms of day-to-day variability (within subjects)(50), a number of factors are known to influence it(36). In addition to the ventilatory changes that occur with increased concentrations of progesterone, which have the potential to affect submaximal exercise responses, alterations in basal body temperature and changes in mood states are two factors postulated to influence running economy. Both factors have been reported to change during different phases of the menstrual cycle; however, the association between menstrual cycle phase and running economy has yet to be determined.The present finding of no significant difference in ˙VO2 between phases at speeds corresponding to 55% ˙VO2max supports the results of previous studies using cycle ergometry at intensities ≤ 70%˙VO2max (29,45). The present findings do not, however, support the findings of Hessemer and Bruck(25) that there is an increase in ˙VO2 during the luteal phase compared with that during the follicular phase during submaximal exercise. One possible explanation for not finding a difference in˙VO2 at the speed corresponding to 55% ˙VO2max may have been because the intensity was very close to the subjects' transitional speed between walking and running. There appears to be some evidence that requiring a person to jog at a speed at which they prefer to walk increases their perceived exertion (39) and may increase their˙VO2 at that speed. If the subjects were less economical during the study because of the speed, it may have masked differences and made it difficult to detect a change in economy because of the menstrual cycle. RE is a variable that reflects the total demand for oxygen during submaximal running, regardless of source; therefore, any increase in oxygen demand will result in a decreased RE.An important consideration for the athlete is the practical importance of the differences observed in RE at a speed corresponding to 80%˙VO2max. Since most distance events are run at or above paces that correspond to 85% ˙VO2max, by plotting the economy curves for the EF and ML phases, the difference in 10 km times can be established. Using this approach and assuming that the association persists at higher intensities, it can be shown that the subjects would have predicted 10 km times of 47.02 min during the ML phase and 45.51 min during the EF phase. In most 10 km races, a difference of 90 s on a runner's time can mean the difference between many places in order of finish.The finding of an elevated ˙VO2 (less economical) and VE during running at the 80% ˙VO2max intensity in the ML phase (compared with that in the EF phase) is in contrast to results reported by DeSouza et al.(18) in which no difference in VE or ˙VO2 was found between follicular and luteal phases during submaximal (80%˙VO2max) and maximal running. It is possible that by testing only once in each phase, a difference in VE and ˙VO2 may have been undetected. The testing procedure in the present study allowed VE and˙VO2 to be tested during phases of the cycle when progesterone levels were more likely to be at their lowest (EF) and highest (ML) levels, which increased the probability of detecting differences if they existed.Correlation coefficients were calculated to determine the relationship between all variables (physiological and psychological) and hormone(progesterone and estradiol) levels. Serum estradiol concentrations were not significantly (P > 0.05) correlated to any of the variables. Serum progesterone values, however, were significantly (P < 0.01) correlated with all variables that were significantly different between the EF(low progesterone) and ML (high progesterone) phases of the menstrual cycle. The correlations were positive for all variables except vigor, which was negative. These results indicate that the differences observed between phases in the variables examined are significantly correlated with changes in the progesterone levels rather than the estradiol levels.CONCLUSIONSThe results of the present study demonstrate that while resting˙VO2, VE, and ventilatory drive are increased during the luteal phase when compared with the follicular phase, and RE at speeds corresponding to 80% ˙VO2max is decreased (less economical) during the luteal phase, RE at speeds corresponding to 55% ˙VO2max appears to be unaltered by menstrual cycle phase in moderately trained female runners. This would indicate that menstrual cycle phase should be considered when testing resting and submaximal (80% ˙VO2max) ventilatory parameters in females. The results of the POMS data would also suggest that menstrual cycle phase be included when psychological data are being obtained on females. Although the generalizability of these findings are limited because of the sample (a nonrandom group of moderately trained eumenorrheic runners), it is concluded that RE at speeds corresponding to 80% ˙VO2max in moderately trained female distance runners is associated with changes in ventilatory drive that occur in synchronization with basal serum progesterone fluctuations during the menstrual cycle.The present study was not designed to assess whether menstrual cycle changes influence actual endurance performance, which is the variable of interest for most competitive athletes. It should be noted that amenorrheic female endurance runners will not undergo the changes in progesterone and, therefore, will most likely not experience the performance decrement associated with the luteal phase of the menstrual cycle. In a society in which the emphasis is placed on winning, this raises the concern that a female competitive runner might intentionally attempt to become amenorrheic and thus risk her health. It should be emphasized, however, that the subjects in the present study were moderately trained, and one should use caution when attempting to relate the results of this study to a different population, such as elite female distance runners.Additional studies with larger sample sizes are needed to assess whether the changes that occurred in the present study also occur during actual competitive situations. In addition, future studies using female subjects need to address the possibility that menstrual cycle phase may be a confounding variable that should be controlled.The authors wish to thank the four phlebotomists for their assistance in the collection of the blood, Dr. Kathleen Matt and Rhoda Nelson for their expertise in the hormonal analysis of the blood samples, the subjects for their willingness to participate in this study, Dr. Bruce Staats and Mike Witzke for their technical assistance, and Hygeia Sciences, Inc. and Serono Laboratories, Inc. for their donations of the luteinizing hormone kits and basal body thermometers, respectively, used in this study.This study was supported in part by the Douglas Conley Running Research Fund.Address for correspondence: Dr. Tracy J. Williams, Training Division, Indiana State Police, 100 North Senate Avenue, IGC-N, Indianapolis, IN 46204.REFERENCES1. Aaron, E. A., C. K. Seow, B. D. Johnson, and J. A. Dempsey. Oxygen cost of exercise hyperpnea: Implications for performance.J. Appl. Physiol. 72:1818-1825, 1992. [Context Link]2. Bonekat, H. W., M. L. Dombovy, and B. A. Staats. Progesterone-induced changes in exercise performance and ventilatory response.Med. Sci. Sports Exerc. 19:118-123, 1987. [CrossRef] [Full Text] [Medline Link] [Context Link]3. Bonen, A., F. J. Haynes, W. Watson-Wright, et al. Effects of menstrual cycle on metabolic responses to exercise. J. Appl. Physiol.: Respir. Environ. Exerc. Physiol. 55:1506-1513, 1983. [Context Link]4. Bransford, D. R. and E. T. Howley. Oxygen cost of running in trained and untrained men and women. Med. Sci. Sprots 9:41-44, 1977. [Context Link]5. Bunt, J. C., S. B. Going, C. H. Westfall, C. Perry, R. Pamenter, and T. G. Lohman. Impact of bone mineral content on body fat estimations in females. Med. Sci. Sports Exerc. 20:S60, 1988. [Medline Link] [Context Link]6. Byrne-Quinn, E., J. V. Weil, I. E. Sodal, G. F. Filley, and R. F. Grover. Ventilatory control in the athlete. J. Appl. Physiol. 30:91-98, 1971. [Context Link]7. Cananagh, P. R., G. C. Andrew, R. Kram, M. M. Rodgers, D. J. Sanderson, and E. M. Henning. An approach to biomechanical profiling of elite distance runners. Int. J. Sports Biomech. 1:36-62, 1985. [Context Link]8. Cavanagh, P. R. and K. R. Williams. The effect of stride length variation on oxygen uptake during distance running. Med. Sci. Sports Exerc. 14:30-35, 1982. [CrossRef] [Full Text] [Medline Link] [Context Link]9. Clark, T. J. H. and D. J. C. Read. A rebreathing method for studying the ventilatory response to CO2. J. Physiol. 184:41-42, 1966. [CrossRef] [Context Link]10. Cockerill, I. M., A. M. Nevill, and N. C. Byrne. Mood, mileage and the menstrual cycle. Br. J. Sports Med. 26:145-150, 1992. [CrossRef] [Medline Link] [Context Link]11. Collins, A., P. Eneroth, and B-M. Landgren. Psychoneuroendocrine stress responses and mood as related to the menstrual cycle. Psychosom. Med. 47:512-527, 1985. [Medline Link] [Context Link]12. Conley, D. L. and G. S. Krahenbuhl. Running economy and distance running performance of highly trained athletes. Med. Sci. Sports Exerc. 12:357-360, 1980. [CrossRef] [Full Text] [Medline Link] [Context Link]13. Costill, D. L. The relationship between selected physiological variables and distance running performance. J. Sports Med. 7:61-66, 1967. [Context Link]14. Daniels, J. and J. Gilbert. Oxygen Power:Performance Tables for Distance Runners. Tempe, AZ: Oxygen Power, 1979, pp. 95-107. [Context Link]15. Daniels, J., G. Krahenbuhl, C. Foster, J. Gilbert, and S. Daniels. Aerobic responses of female distance runners to submaximal and maximal exercise. Ann. N. Y. Acad. Sci. 301:726-733, 1977. [CrossRef] [Medline Link] [Context Link]16. Daniels, J., N. Scardina, J. Hayes, and P. Foley. Variations in ˙VO2 submax during treadmill running. Med. Sci. Sports 16:108, 1984. [Context Link]17. DeMouzon, J., J. Testart, B. Lefevre, J-L. Pouly, and R. Frydmay. Time relationships between basal body temperature and ovulation or plasma progestins. Fertil. Steril. 41:254-259, 1984. [Context Link]18. Desouza, M. J., M. S. Maguire, K. Rubin, and C. M. Maresh. Effects of menstrual phase and amenorrhea on exercise responses in runners. Med. Sci. Sports Exerc. 22:575-580, 1990. [Context Link]19. Dombovy, M. L., H. W. Bonekat, T. J. Williams, and B. A. Staats. Exercise performance and ventilatory response in the menstrual cycle. Med. Sci. Sports Exerc. 19:111-117, 1987. [CrossRef] [Full Text] [Medline Link] [Context Link]20. Drinkwater, B. L. Physiological responses of women to exercise. In: Exercise and Sport Science Reviews, J. H. Wilmore(Ed.). New York: Academic Press, 1973, pp. 125-153. [Context Link]21. England, S. J. and L. E. Farhi. Fluctuations in alveolar CO2 and in base excess during the menstrual cycle.Respir. Physiol. 26:157-161, 1976. [CrossRef] [Medline Link] [Context Link]22. Fay, L., B. R. Londeree, T. P. Lafontaine, and M. R. Volek. Physiological parameters related to distance running performance in female athletes. Med. Sci. Sports Exerc. 21:319-324, 1989. [CrossRef] [Full Text] [Medline Link] [Context Link]23. Frederick, E. C., E. T. Howley, and S. K. Powers. Lower oxygen demands of running in soft-soled shoes. Res. Q. Exerc. Sport 57:174-177, 1986. [Context Link]24. Hackney, A. C., C. S. Curley, and B. J. Nicklas. Physiological responses to submaximal exercise at the mid-follicular, ovulatory and mid-luteal phases of the menstrual cycle. Scand. J. Med. Sci. Sports 1:94-98, 1991. [CrossRef] [Medline Link] [Context Link]25. Hessemer, V. and K. Bruck. Influence of menstrual cycle on thermoregulatory, metabolic, and heart rate responses to exercise at night.J. Appl. Physiol. 59:1911-1917, 1985. [Medline Link] [Context Link]26. Highet, R. Athletic amenorrhea: An update on aetiology, complications, and management. Sports Med. 7:82-108, 1989. [CrossRef] [Full Text] [Medline Link] [Context Link]27. Jackson, A. S. and H. L. Pollock. Practical assessment of body composition. Phys. Sportsmed. 13:76-90, 1985. [Medline Link] [Context Link]28. Johansson, E. D. B., L. Wide, and C. Gemzell. Luteinizing hormone (LH) and progesterone in plasma and LH and oestrogens in urine during 42 normal menstrual cycles. Acta Endocrin. 68:502-512, 1971. [Context Link]29. Jurkowski, J. E. H., N. L. Jones, C. J. Toews, and J. R. Sutton. Effects of menstrual cycle on blood lactate, O2 delivery, and performance during exercise. J. Appl. Physiol:Respir. Environ. Exerc. Physiol. 51:1493-1499, 1981. [Context Link]30. Krahenbuhl, G. S. and T. J. Williams. Running economy: changes with age during childhood and adolescence. Med. Sci. Sports Exerc. 24:462-466, 1992. [CrossRef] [Full Text] [Medline Link] [Context Link]31. Lebrun, C. M., D. C. McKenzie, J. C. Prior, and J. E. Taunton. Effects of menstrual cycle phase on athletic performance. Med. Sci. Sports Exerc. 27:437-444, 1995. [CrossRef] [Full Text] [Medline Link] [Context Link]32. Lyons, H. A. Centrally acting hormones and respiration.Pharmacol. Ther. B. 2:743-751, 1976. [CrossRef] [Medline Link] [Context Link]33. McNair, D. M., M. Lorr, and L. F. Droppleman.EITS Manual for the Profile of Mood States. San Diego, CA: Educational and Industrial Testing Service, 1971. [Context Link]34. MacDougall, J., W. Reddan, C. Layton, and J. Dempsey. Effects of metabolic hyperthermia on performance during heavy prolonged exercise. J. Appl. Physiol. 36:538-544, 1974. [Context Link]35. Martin, B. J., K. E. Sparks, C. W. Zwillich, and J. V. Weil. Low exercise ventilation in endurance athletes. Med. Sci. Sports 11:181-185, 1979. [Full Text] [Medline Link] [Context Link]36. Morgan, D. W., P. E. Martin, and G. S. Krahenbuhl. Factors affecting running economy. Sports Med. 7:310-330, 1989. [CrossRef] [Full Text] [Medline Link] [Context Link]37. Morgan, W. P., P. J. O'Connor, P. B. Sparling, and R. R. Pate. Psychological characterization of the elite female distance runner.Int. J. Sports Med. 8:124-131, 1987. [Medline Link] [Context Link]38. Morgan, W. P. and M. L. Pollock. Psychologic characterization of the elite distance runner. Ann. N. Y. Acad. Sci. 301:382-403, 1977. [CrossRef] [Medline Link] [Context Link]39. Noble, B. J., K. F. Metz, K. B. Pandolf, and E. Cafarelli. Perceived exertion during walking and running. II. Med. Sci. Sports 5:116-120, 1973. [Context Link]40. Patton, J. and J. Vogel. Cross-sectional and longitudinal evaluations of an endurance training program. Med. Sci. Sports 9:100-103, 1977. [Full Text] [Medline Link] [Context Link]41. Plass, E. D. and F. W. Oberst. Respiration and pulmonary ventilation in normal nonpregnant, pregnant, and puerperal women.Am. J. Obstet. Gynecol. 35:441-452, 1938. [Context Link]42. Schieb, D. A. Kinematic accommodation of novice treadmill runners. Res. Q. Exerc. Sport 57:1-7, 1986. [CrossRef] [Context Link]43. Schoene, R. B., H. T. Robertson, D. J. Pierson, and A. P. Peterson. Respiratory drives and exercise in menstrual cycles of athletic and nonathletic women. J. Appl. Physiol. 50:1300-1305, 1981. [Context Link]44. Shangold, M. M. Menstruation: Women and Exercise Physiology and Sports Medicine. M. Shangold and G. Mikin (Eds.). Philadelphia: F.A. Davis, 1988, pp. 129-144. [Context Link]45. Stephenson, L. A., M. A. Kolka, and J. E. Wilkerson. Metabolic and thermoregulatory responses to exercise during the human menstrual cycle. Med. Sci. Sports Exerc. 14:270-275, 1982. [CrossRef] [Full Text] [Medline Link] [Context Link]46. Taylor, H. L., E. Buskirk, and A. Henschel. Maximal oxygen intake as an objective measure of cardio-respiratory performance.J. Appl. Physiol. 8:73-80, 1955. [Medline Link] [Context Link]47. Tok, G. and H. H. Loeschcke. The effect of progesterone on the medullary respiratory chemosensitivity in the cat. Pflugers Arch. Suppl. 382:R18, 1979. [Context Link]48. Williams, K. R. and P. R. Cavanagh. Relationship between distance running mechanics, running economy, and performance.J. Appl. Physiol. 63:1-9, 1987. [Context Link]49. Williams, T. J., G. S. Krahenbuhl, and D. W. Morgan. Mood states and running economy in moderately-trained male runners.Med. Sci. Sports Exerc. 23:727-731, 1991. [CrossRef] [Full Text] [Medline Link] [Context Link]50. Williams, T. J., G. S. Krahenbuhl, and D. W. Morgan. Daily variation in running economy of moderately-trained male runners.Med. Sci. Sports Exerc. 23:944-948, 1991. [CrossRef] [Full Text] [Medline Link] [Context Link]51. Wilmore, J. H. and C. H. Brown. Physiological profiles of women distance runners. Med. Sci. Sports 6:178-181, 1974. [Full Text] [Medline Link] [Context Link]ESTRADIOL; EUMENORRHEA; OXYGEN CONSUMPTION; PROFILE OF MOOD STATES INVENTORY;|00005768-199712000-00010#xpointer(id(R2-10))|11065213||ovftdb|00005768-198704000-00009SL0000576819871911811065213P83[CrossRef]|00005768-199712000-00010#xpointer(id(R2-10))|11065404||ovftdb|00005768-198704000-00009SL0000576819871911811065404P83[Full Text]|00005768-199712000-00010#xpointer(id(R2-10))|11065405||ovftdb|00005768-198704000-00009SL0000576819871911811065405P83[Medline Link]|00005768-199712000-00010#xpointer(id(R5-10))|11065405||ovftdb|SL00005768198820s6011065405P86[Medline Link]|00005768-199712000-00010#xpointer(id(R8-10))|11065213||ovftdb|00005768-198201000-00006SL000057681982143011065213P89[CrossRef]|00005768-199712000-00010#xpointer(id(R8-10))|11065404||ovftdb|00005768-198201000-00006SL000057681982143011065404P89[Full Text]|00005768-199712000-00010#xpointer(id(R8-10))|11065405||ovftdb|00005768-198201000-00006SL000057681982143011065405P89[Medline Link]|00005768-199712000-00010#xpointer(id(R9-10))|11065213||ovftdb|SL0000524519661844111065213P90[CrossRef]|00005768-199712000-00010#xpointer(id(R10-10))|11065213||ovftdb|SL0000241219922614511065213P91[CrossRef]|00005768-199712000-00010#xpointer(id(R10-10))|11065405||ovftdb|SL0000241219922614511065405P91[Medline Link]|00005768-199712000-00010#xpointer(id(R11-10))|11065405||ovftdb|SL0000684219854751211065405P92[Medline Link]|00005768-199712000-00010#xpointer(id(R12-10))|11065213||ovftdb|00005768-198012050-00010SL0000576819801235711065213P93[CrossRef]|00005768-199712000-00010#xpointer(id(R12-10))|11065404||ovftdb|00005768-198012050-00010SL0000576819801235711065404P93[Full Text]|00005768-199712000-00010#xpointer(id(R12-10))|11065405||ovftdb|00005768-198012050-00010SL0000576819801235711065405P93[Medline Link]|00005768-199712000-00010#xpointer(id(R15-10))|11065213||ovftdb|SL00000620197730172611065213P96[CrossRef]|00005768-199712000-00010#xpointer(id(R15-10))|11065405||ovftdb|SL00000620197730172611065405P96[Medline Link]|00005768-199712000-00010#xpointer(id(R19-10))|11065213||ovftdb|00005768-198704000-00008SL0000576819871911111065213P100[CrossRef]|00005768-199712000-00010#xpointer(id(R19-10))|11065404||ovftdb|00005768-198704000-00008SL0000576819871911111065404P100[Full Text]|00005768-199712000-00010#xpointer(id(R19-10))|11065405||ovftdb|00005768-198704000-00008SL0000576819871911111065405P100[Medline Link]|00005768-199712000-00010#xpointer(id(R21-10))|11065213||ovftdb|SL0000697519762615711065213P102[CrossRef]|00005768-199712000-00010#xpointer(id(R21-10))|11065405||ovftdb|SL0000697519762615711065405P102[Medline Link]|00005768-199712000-00010#xpointer(id(R22-10))|11065213||ovftdb|00005768-198906000-00016SL0000576819892131911065213P103[CrossRef]|00005768-199712000-00010#xpointer(id(R22-10))|11065404||ovftdb|00005768-198906000-00016SL0000576819892131911065404P103[Full Text]|00005768-199712000-00010#xpointer(id(R22-10))|11065405||ovftdb|00005768-198906000-00016SL0000576819892131911065405P103[Medline Link]|00005768-199712000-00010#xpointer(id(R24-10))|11065213||ovftdb|SL00013584199119411065213P105[CrossRef]|00005768-199712000-00010#xpointer(id(R24-10))|11065405||ovftdb|SL00013584199119411065405P105[Medline Link]|00005768-199712000-00010#xpointer(id(R25-10))|11065405||ovftdb|SL00004560198559191111065405P106[Medline Link]|00005768-199712000-00010#xpointer(id(R26-10))|11065213||ovftdb|00007256-198907020-00002SL00007256198978211065213P107[CrossRef]|00005768-199712000-00010#xpointer(id(R26-10))|11065404||ovftdb|00007256-198907020-00002SL00007256198978211065404P107[Full Text]|00005768-199712000-00010#xpointer(id(R26-10))|11065405||ovftdb|00007256-198907020-00002SL00007256198978211065405P107[Medline Link]|00005768-199712000-00010#xpointer(id(R27-10))|11065405||ovftdb|SL000013171985137611065405P108[Medline Link]|00005768-199712000-00010#xpointer(id(R30-10))|11065213||ovftdb|00005768-199204000-00012SL0000576819922446211065213P111[CrossRef]|00005768-199712000-00010#xpointer(id(R30-10))|11065404||ovftdb|00005768-199204000-00012SL0000576819922446211065404P111[Full Text]|00005768-199712000-00010#xpointer(id(R30-10))|11065405||ovftdb|00005768-199204000-00012SL0000576819922446211065405P111[Medline Link]|00005768-199712000-00010#xpointer(id(R31-10))|11065213||ovftdb|00005768-199503000-00022SL0000576819952743711065213P112[CrossRef]|00005768-199712000-00010#xpointer(id(R31-10))|11065404||ovftdb|00005768-199503000-00022SL0000576819952743711065404P112[Full Text]|00005768-199712000-00010#xpointer(id(R31-10))|11065405||ovftdb|00005768-199503000-00022SL0000576819952743711065405P112[Medline Link]|00005768-199712000-00010#xpointer(id(R32-10))|11065213||ovftdb|SL000064931976274311065213P113[CrossRef]|00005768-199712000-00010#xpointer(id(R32-10))|11065405||ovftdb|SL000064931976274311065405P113[Medline Link]|00005768-199712000-00010#xpointer(id(R35-10))|11065404||ovftdb|00005756-197911020-00012SL0000575619791118111065404P116[Full Text]|00005768-199712000-00010#xpointer(id(R35-10))|11065405||ovftdb|00005756-197911020-00012SL0000575619791118111065405P116[Medline Link]|00005768-199712000-00010#xpointer(id(R36-10))|11065213||ovftdb|00007256-198907050-00003SL000072561989731011065213P117[CrossRef]|00005768-199712000-00010#xpointer(id(R36-10))|11065404||ovftdb|00007256-198907050-00003SL000072561989731011065404P117[Full Text]|00005768-199712000-00010#xpointer(id(R36-10))|11065405||ovftdb|00007256-198907050-00003SL000072561989731011065405P117[Medline Link]|00005768-199712000-00010#xpointer(id(R37-10))|11065405||ovftdb|SL000043551987812411065405P118[Medline Link]|00005768-199712000-00010#xpointer(id(R38-10))|11065213||ovftdb|SL00000620197730138211065213P119[CrossRef]|00005768-199712000-00010#xpointer(id(R38-10))|11065405||ovftdb|SL00000620197730138211065405P119[Medline Link]|00005768-199712000-00010#xpointer(id(R40-10))|11065404||ovftdb|00005756-197709020-00005SL000057561977910011065404P121[Full Text]|00005768-199712000-00010#xpointer(id(R40-10))|11065405||ovftdb|00005756-197709020-00005SL000057561977910011065405P121[Medline Link]|00005768-199712000-00010#xpointer(id(R42-10))|11065213||ovftdb|SL00006966198657111065213P123[CrossRef]|00005768-199712000-00010#xpointer(id(R45-10))|11065213||ovftdb|00005768-198204000-00003SL0000576819821427011065213P126[CrossRef]|00005768-199712000-00010#xpointer(id(R45-10))|11065404||ovftdb|00005768-198204000-00003SL0000576819821427011065404P126[Full Text]|00005768-199712000-00010#xpointer(id(R45-10))|11065405||ovftdb|00005768-198204000-00003SL0000576819821427011065405P126[Medline Link]|00005768-199712000-00010#xpointer(id(R46-10))|11065405||ovftdb|SL00004560195587311065405P127[Medline Link]|00005768-199712000-00010#xpointer(id(R49-10))|11065213||ovftdb|00005768-199106000-00012SL0000576819912372711065213P130[CrossRef]|00005768-199712000-00010#xpointer(id(R49-10))|11065404||ovftdb|00005768-199106000-00012SL0000576819912372711065404P130[Full Text]|00005768-199712000-00010#xpointer(id(R49-10))|11065405||ovftdb|00005768-199106000-00012SL0000576819912372711065405P130[Medline Link]|00005768-199712000-00010#xpointer(id(R50-10))|11065213||ovftdb|00005768-199108000-00010SL0000576819912394411065213P131[CrossRef]|00005768-199712000-00010#xpointer(id(R50-10))|11065404||ovftdb|00005768-199108000-00010SL0000576819912394411065404P131[Full Text]|00005768-199712000-00010#xpointer(id(R50-10))|11065405||ovftdb|00005768-199108000-00010SL0000576819912394411065405P131[Medline Link]|00005768-199712000-00010#xpointer(id(R51-10))|11065404||ovftdb|00005756-197423000-00010SL000057561974617811065404P132[Full Text]|00005768-199712000-00010#xpointer(id(R51-10))|11065405||ovftdb|00005756-197423000-00010SL000057561974617811065405P132[Medline Link]4437345Menstrual cycle phase and running economyWILLIAMS, TRACY J.; KRAHENBUHL, GARY S.Applied Sciences: Biodynamics1229