Menstrual cycle phase affects temperature during endurance exercise JAMES M. PIVARNIK, CARLOS AND JAMES R. MORROW, JR.
J. MARICHAL,
THOMAS
regulation
SPILLMAN,
Department of Health and Human Performance, University of Houston, Houston 77204; and Department of Obstetrics and Gynecology, Baylor College of Medicine, Houston, Texas 77030 PIVARNIK,JAMESM., CARLOSJ. MARICHAL,THOMASSPILLMAN, ANDJAMESR. MORROW,JR. Menstrual cycle phase affects temperature regulation during endurance exercise. J. Appl. Physiol. 72(2): 543~548,1992.-We investigated whether menstrual cycle phasewould affect temperature regulation during an endurance exercisebout performed at room temperature (T,) of 22°C and 60% relative humidity. Nine eumenorrheic women [age 27.2 * 3.7 yr, peak 0, uptake (TO,) 2.52 t 0.35 l/min] performed 60 min of cycle exerciseat 65% of peak vo2. Subjects were tested in both midfollicular (F) and midluteal (L) phases, although one woman did not show a rise in serumprogesterone (P,) that is typically evident 1 wk after ovulation. vo2, rectal (T,) and skin (Tsk)temperatures, heart rates (HR), and ratings of perceived exertion (RPE) were measuredthroughout exercise.Sweat loss(SL) was estimated from pre- and postexercise body weight differences. vo2, SL, and Tskwere not affected by menstrual cycle phase.Preexercise T, was 0.3”C higher during L than during F conditions, and this difference increased to 0.6”C by the end of exercise (P < 0.01). Comparedwith F, HRs during L were -10 beats/min greater (P < 0.001) at all times, whereasRPE responseswere significantly greater (P < 0.01) by 50 min of cycling. No differences in any measuredvalues were found in the subject whoseP, was low in both test conditions. Resultsindicate that thermoregulation (specifically, regulation of T,), aswell as cardiovascular strain and perception of exercise,was adversely affected during the L phase. follicular; luteal; progesterone;women
BASAL BODY TEMPERATURE is typically
elevated during the luteal phase of the menstrual cycle in eumenorrheic women (19). This increase in thermoregulatory set point has been related to elevated progesterone (P,J levels or changes in estrogen- (E,) to-P, ratios coincident with the luteal phase (11,21). Recent findings suggest that differences in resting core temperature are typically maintained during exercise (27). However, most studies of menstrual cycle phase effects on temperature regulation during physical activity have involved high ambient temperatures that necessitated low-intensity (8, 9, 15) or short-duration (2) exercise sessions. Also, in some instances, clear documentation of ovulation and menstrual cycle phase has not been demonstrated (2,8,28,29,31). In contrast, it is not well known whether a woman’s thermoregulatory response to exercise may be altered by menstrual cycle phase during an extended session of aerobic exercise performed in moderate temperatures. Increases in core temperature during endurance exercise 0161-7567/92
$2.00
Copyright
are driven mainly by relative exercise intensity if ambient temperatures fall within a “prescriptive zone” of 5 and 35OC (23,24). Therefore it is possible that a woman’s ability to maintain thermal balance may be negatively affected during the luteal phase of the menstrual cycle if she begins exercise at an elevated core temperature and her activity is of sufficient duration and intensity. Our purpose was to determine whether menstrual cycle phase affects temperature regulation during a lengthy bout (60 min) of steady-state endurance exercise performed at room temperature. We designed our study so that physiological and perceptual responses would be mainly a result of exercise intensity and duration rather than ambient heat load. METHODS
All experiments were conducted in the summer months in Houston, TX (average daily ambient temperature 31”C, relative humidity 75%), where subjects (n = 9) had consistently performed outdoor exercise (running, cycling) for 21 -2 mo before and during the study period. Therefo re the women were considered to be aerobically trained and heat acclimatized. All experimental procedures were reviewed and approved by the University Committee for the Protection of Human Subjects. The experimental protocol was explained, and written consent was obtained from each subject before she began the study. All subjects were eumenorrheic, with regular menstrual cycles that ranged from 27 to 32 days. None had taken any oral contraceptives for at least the previous 12 mo. Each subject performed a peak 0, consumption (VO,) test on a cycle ergometer (Schwinn EX-II) so that exercise intensity of subsequent submaximal tests could be accurately determined. Beginning with 25 W, power output was increased by 25 W at 2-min intervals. Subjects continued to cycle until volitional exhaustion. Peak 'Ijo was taken as the highest value recorded during the test. Ina all cases, respiratory exchange ratio exceeded 1.10. . VO, and CO, production (VCO,) were measured continuously by the open-circuit method. Volumes of inspired gases were measured with a Parkinson-Cowan flowmeter, and expired fractions of 0, and CO, were measured by Applied Electrochemistry S-3A and Cavitron Anarad analyzers, respectively. Before each test, the an-
0 1992 the American
Physiological
Society
543
Downloaded from www.physiology.org/journal/jappl by ${individualUser.givenNames} ${individualUser.surname} (192.236.036.029) on August 14, 2018. Copyright © 1992 American Physiological Society. All rights reserved.
544
MENSTRUAL
CYCLE AND TEMPERATURE
alyzers were calibrated with certified standard gases of known concentrations. Calipers (Lafayette Instruments) were used to measure skinfold thicknesses at each woman’s triceps, suprailium, and anterior thigh. Percent body fat was estimated by use of the sum-o.f-skinfolds equation developed by Jackson et al. (16). The endurance exercise protocol required each subject to perform on two occasions continuous 60.min rides on the cycle ergometer. Power outputs were set so that each woman would exercise at ~65-70% of her peak TO,. The study was designed so that a subject performed her tests ~7 days before (midfollicular phase, F) and 7 days after (midluteal phase, L) ovulation. The order of submaximal tests was randomized among the subjects. All experiments were conducted in Isimilar ambient conditions (22.3 t 0.9OC T,, 60 t 6% relative humidity, wind speed ~0.2 m/s). Each subject wore the same clothes (shortsleeved shirt, shorts) for each test. Seven subjects performed their tests between 0700 and 0800. Because of scheduling conflicts, the remaining two subjects were tested between 1200 and 1300 on both study days. To ensure that subjects were adequately and uniformly hydrated, each was instructed to drink an amount of water equal to 1% of her body weight the night before the test. An additional 300 ml of water were ingested on the morning of the test, but subjects were not allowed to drink during the exercise sessions. On arrival at the laboratory, each participant undressed, and a nude body weight was taken. The subject then inserted a rectal thermistor -10-15 cm beyond the anal sphincter to allow continuous monitoring of core temperature (T,). Surface thermistors were attached to four sites (right arm, right calf, right thigh, and chest) so that mean skin temperature (Tek) could be calculated according to Ramanathan’s method (20). A pulse monitor (CIC) was attached to the subject’s chest and was used to monitor heart rate (HR) via telemetry during each exercise test. An incremental 150point Borg scale (4) was used to determine subjects’ ratings of perceived exertion (RPE). Possible numerical ratings ranged from 6 to 20, with verbal anchors varying from “very, very light” to “very, very hard” for scores of 7 and 19, respectively. Subjects were instructed to give an overall body rating of their perception of effort required during the exercise test. All the above measures were obtained at lo-min intervals. Beginning at 6 min into the test, indirect calorimetry was used to obtain subjects’ VO, values. Fiveminute samples were taken on four occasions (minutes 6-10,21-25,36-40, 51-55) during the test. Values were averaged to give an overall vo2 estimate for the 60 min of cycling. After exercise, the subjects disrobed and dried themselves as quickly as possible before a nude postexercise body weight was obtained. Sweat loss (SL) during the test was calculated as the difference between pre- and postexercise body weights. Day of ovulation was predicted from evidence of luteinizing hormone (LH) surge in the urine (OvuSTICK, Monoclonal Antibodies). This test has been shown to be a valid ovulation indicator when compared with serum hormone (LH and P4) levels (7). To further document whether ovulation had occurred in response to the LH
REGULATION
TABLE 1. Selected subject characteristics Age,
yr
27.2t3.7
Ht, cm
166t5
W
kg
59.3t5.8
% Fat
Peak iTop, Umin
15.8t5.5
2.52t0.35
Values are means -t SD for 9 subjs.
surge, P, assays were performed on venous blood samples drawn immediately before each submaximal exercise test. P, levels were determined in duplicate by radioimmunoassay with polypropylene tubes coated with antibody (Diagnostic Products). The P, assay is sensitive to 0.05 rig/ml and shows ~0.2% cross-reactivity with estradiol, cortisol, or testosterone. Intra-assay variation between duplicate samples did not exceed 7.5%. Dependent t tests were used to determine whether average VO, (averaged during exercise), SL, and preexercise T,, values differed because of menstrual cycle phase. A multiple treatments-by-subjects analysis of variance (ANOVA) design was used to determine whether changes occurred in T,, Tsk, HR, and RPE between experimental conditions (2 menstrual phases) and/or during the exercise bouts (6 time periods). Because of the number of comparisons, the
J. MARICHAL,
THOMAS
regulation
SPILLMAN,
Department of Health and Human Performance, University of Houston, Houston 77204; and Department of Obstetrics and Gynecology, Baylor College of Medicine, Houston, Texas 77030 PIVARNIK,JAMESM., CARLOSJ. MARICHAL,THOMASSPILLMAN, ANDJAMESR. MORROW,JR. Menstrual cycle phase affects temperature regulation during endurance exercise. J. Appl. Physiol. 72(2): 543~548,1992.-We investigated whether menstrual cycle phasewould affect temperature regulation during an endurance exercisebout performed at room temperature (T,) of 22°C and 60% relative humidity. Nine eumenorrheic women [age 27.2 * 3.7 yr, peak 0, uptake (TO,) 2.52 t 0.35 l/min] performed 60 min of cycle exerciseat 65% of peak vo2. Subjects were tested in both midfollicular (F) and midluteal (L) phases, although one woman did not show a rise in serumprogesterone (P,) that is typically evident 1 wk after ovulation. vo2, rectal (T,) and skin (Tsk)temperatures, heart rates (HR), and ratings of perceived exertion (RPE) were measuredthroughout exercise.Sweat loss(SL) was estimated from pre- and postexercise body weight differences. vo2, SL, and Tskwere not affected by menstrual cycle phase.Preexercise T, was 0.3”C higher during L than during F conditions, and this difference increased to 0.6”C by the end of exercise (P < 0.01). Comparedwith F, HRs during L were -10 beats/min greater (P < 0.001) at all times, whereasRPE responseswere significantly greater (P < 0.01) by 50 min of cycling. No differences in any measuredvalues were found in the subject whoseP, was low in both test conditions. Resultsindicate that thermoregulation (specifically, regulation of T,), aswell as cardiovascular strain and perception of exercise,was adversely affected during the L phase. follicular; luteal; progesterone;women
BASAL BODY TEMPERATURE is typically
elevated during the luteal phase of the menstrual cycle in eumenorrheic women (19). This increase in thermoregulatory set point has been related to elevated progesterone (P,J levels or changes in estrogen- (E,) to-P, ratios coincident with the luteal phase (11,21). Recent findings suggest that differences in resting core temperature are typically maintained during exercise (27). However, most studies of menstrual cycle phase effects on temperature regulation during physical activity have involved high ambient temperatures that necessitated low-intensity (8, 9, 15) or short-duration (2) exercise sessions. Also, in some instances, clear documentation of ovulation and menstrual cycle phase has not been demonstrated (2,8,28,29,31). In contrast, it is not well known whether a woman’s thermoregulatory response to exercise may be altered by menstrual cycle phase during an extended session of aerobic exercise performed in moderate temperatures. Increases in core temperature during endurance exercise 0161-7567/92
$2.00
Copyright
are driven mainly by relative exercise intensity if ambient temperatures fall within a “prescriptive zone” of 5 and 35OC (23,24). Therefore it is possible that a woman’s ability to maintain thermal balance may be negatively affected during the luteal phase of the menstrual cycle if she begins exercise at an elevated core temperature and her activity is of sufficient duration and intensity. Our purpose was to determine whether menstrual cycle phase affects temperature regulation during a lengthy bout (60 min) of steady-state endurance exercise performed at room temperature. We designed our study so that physiological and perceptual responses would be mainly a result of exercise intensity and duration rather than ambient heat load. METHODS
All experiments were conducted in the summer months in Houston, TX (average daily ambient temperature 31”C, relative humidity 75%), where subjects (n = 9) had consistently performed outdoor exercise (running, cycling) for 21 -2 mo before and during the study period. Therefo re the women were considered to be aerobically trained and heat acclimatized. All experimental procedures were reviewed and approved by the University Committee for the Protection of Human Subjects. The experimental protocol was explained, and written consent was obtained from each subject before she began the study. All subjects were eumenorrheic, with regular menstrual cycles that ranged from 27 to 32 days. None had taken any oral contraceptives for at least the previous 12 mo. Each subject performed a peak 0, consumption (VO,) test on a cycle ergometer (Schwinn EX-II) so that exercise intensity of subsequent submaximal tests could be accurately determined. Beginning with 25 W, power output was increased by 25 W at 2-min intervals. Subjects continued to cycle until volitional exhaustion. Peak 'Ijo was taken as the highest value recorded during the test. Ina all cases, respiratory exchange ratio exceeded 1.10. . VO, and CO, production (VCO,) were measured continuously by the open-circuit method. Volumes of inspired gases were measured with a Parkinson-Cowan flowmeter, and expired fractions of 0, and CO, were measured by Applied Electrochemistry S-3A and Cavitron Anarad analyzers, respectively. Before each test, the an-
0 1992 the American
Physiological
Society
543
Downloaded from www.physiology.org/journal/jappl by ${individualUser.givenNames} ${individualUser.surname} (192.236.036.029) on August 14, 2018. Copyright © 1992 American Physiological Society. All rights reserved.
544
MENSTRUAL
CYCLE AND TEMPERATURE
alyzers were calibrated with certified standard gases of known concentrations. Calipers (Lafayette Instruments) were used to measure skinfold thicknesses at each woman’s triceps, suprailium, and anterior thigh. Percent body fat was estimated by use of the sum-o.f-skinfolds equation developed by Jackson et al. (16). The endurance exercise protocol required each subject to perform on two occasions continuous 60.min rides on the cycle ergometer. Power outputs were set so that each woman would exercise at ~65-70% of her peak TO,. The study was designed so that a subject performed her tests ~7 days before (midfollicular phase, F) and 7 days after (midluteal phase, L) ovulation. The order of submaximal tests was randomized among the subjects. All experiments were conducted in Isimilar ambient conditions (22.3 t 0.9OC T,, 60 t 6% relative humidity, wind speed ~0.2 m/s). Each subject wore the same clothes (shortsleeved shirt, shorts) for each test. Seven subjects performed their tests between 0700 and 0800. Because of scheduling conflicts, the remaining two subjects were tested between 1200 and 1300 on both study days. To ensure that subjects were adequately and uniformly hydrated, each was instructed to drink an amount of water equal to 1% of her body weight the night before the test. An additional 300 ml of water were ingested on the morning of the test, but subjects were not allowed to drink during the exercise sessions. On arrival at the laboratory, each participant undressed, and a nude body weight was taken. The subject then inserted a rectal thermistor -10-15 cm beyond the anal sphincter to allow continuous monitoring of core temperature (T,). Surface thermistors were attached to four sites (right arm, right calf, right thigh, and chest) so that mean skin temperature (Tek) could be calculated according to Ramanathan’s method (20). A pulse monitor (CIC) was attached to the subject’s chest and was used to monitor heart rate (HR) via telemetry during each exercise test. An incremental 150point Borg scale (4) was used to determine subjects’ ratings of perceived exertion (RPE). Possible numerical ratings ranged from 6 to 20, with verbal anchors varying from “very, very light” to “very, very hard” for scores of 7 and 19, respectively. Subjects were instructed to give an overall body rating of their perception of effort required during the exercise test. All the above measures were obtained at lo-min intervals. Beginning at 6 min into the test, indirect calorimetry was used to obtain subjects’ VO, values. Fiveminute samples were taken on four occasions (minutes 6-10,21-25,36-40, 51-55) during the test. Values were averaged to give an overall vo2 estimate for the 60 min of cycling. After exercise, the subjects disrobed and dried themselves as quickly as possible before a nude postexercise body weight was obtained. Sweat loss (SL) during the test was calculated as the difference between pre- and postexercise body weights. Day of ovulation was predicted from evidence of luteinizing hormone (LH) surge in the urine (OvuSTICK, Monoclonal Antibodies). This test has been shown to be a valid ovulation indicator when compared with serum hormone (LH and P4) levels (7). To further document whether ovulation had occurred in response to the LH
REGULATION
TABLE 1. Selected subject characteristics Age,
yr
27.2t3.7
Ht, cm
166t5
W
kg
59.3t5.8
% Fat
Peak iTop, Umin
15.8t5.5
2.52t0.35
Values are means -t SD for 9 subjs.
surge, P, assays were performed on venous blood samples drawn immediately before each submaximal exercise test. P, levels were determined in duplicate by radioimmunoassay with polypropylene tubes coated with antibody (Diagnostic Products). The P, assay is sensitive to 0.05 rig/ml and shows ~0.2% cross-reactivity with estradiol, cortisol, or testosterone. Intra-assay variation between duplicate samples did not exceed 7.5%. Dependent t tests were used to determine whether average VO, (averaged during exercise), SL, and preexercise T,, values differed because of menstrual cycle phase. A multiple treatments-by-subjects analysis of variance (ANOVA) design was used to determine whether changes occurred in T,, Tsk, HR, and RPE between experimental conditions (2 menstrual phases) and/or during the exercise bouts (6 time periods). Because of the number of comparisons, the
