JOURNAL OF APPLIED PHYSIOLOGY Vol. 40, No. 2, February 1976. Printed
in U.S.A.
Effects of exercise on excretion of urinary free cortisol
rates
A. BONEN Physical Fitness Research Laboratory, Department University of Illinois, Champaign, Illinois 61820
of Physical
Education,
BONEN, A, Effects of exercise on excretion rates of urinary free cortisol. J. Appl. Physiol. 40(2): 1X-158. 1976. -Excretion rates of urinary free cortisol were studied in 20 men assigned to four treadmill exercise groups: walking at 3 mph for 10 min or 30 min, or running at 7.5 mph for 10 min or 30 min. Free cortisol in urine was measured before and 30, 60, and 90 min after exercise, and again on a control day. Patterns of free-cortisol excretions after exercise at 7.5 mph for 10 and 30 min were significantly different from the control day (P < O.OS>, with the largest changes occurring in the 30-min group. Exercise and control patterns were not different for the other two conditions (P > 0.05). Within the 7.5 mph-30 min group the postexercise cortisol excretion rates were directly related to the relative intensity of exercise (%Vop max) and the respiratory exchange ratio. It is concluded that changes in free cortisol excretion rates depend on the duration as well as the intensity of exercise.
The exercise experiments were repeated on two separate occasions (4-35 days apart). A control day was interposed between the exercise days. All conditions were scheduled between midday and 10 p.m., when plasma and urinary cortisol levels are relatively stable (113,14). Each subject acted as his own control, and his 3 test sessionswere at the same time of day @lo-min difference). Subjects had not eaten for at least 3 h prior to the experiments. Water intake during the experimental periods was not controlled. After the subject had voided he sat quietly for 1 h to standardize the preexercise activity. During that time urine was collected at 30 and 60 min; the first sample was discarded, and the second was designated the preexercise sample. Urine samples were also collected 30, 60, and 90 min after exercise was terminated. Identical sampling periods were maintained during the control sessions when the men rested oxygen consumption; treadmill; walking; running for the entire experimental period. Urine volumes were measured, and aliquots were stored at -20°C. Oxygen consumption (Vo,> was measured during PLASMA CONCENTRATIONS of total cortisol are increased minutes ,9 and 10 of the submaximal exercise bouts. when exercise is extremely intense (lo), or when it Expired air (approximately 1,000 ml/min) was withexceeds 60% of the maximal oxygen intake (VO, max) drawn by vacuum pump from a calibrated dry gas meter (8). Yet, measurements of total plasma cortisol provide into evacuated polyethylene bags, and the gas samples only an approximate index of the physiologically active were analyzed for 0, (paramagnetic model E-2, Beckcortisol, since about 90% of the circulating cortisol is man) and CO, (Lira infrared analyzer, Mine Safety bound. In man, the excreted free cortisol provides a Appliances) (4). During the maximal oxygen intake test sensitive index of the circulating, physiologically active (7) the vo, was monitored each minute. cortisol (2, 3), since there is a rather constant fraction of Cortisol assay. All urine samples were assayed in free cortisol that is filtered by the kidney (2). The pur- triplicate by a modification of the competitive proteinpose of this study, therefore, was to investigate the binding method for urinary free cortisol (1, 15, 16). relative effects of the duration and the intensity of Unbound urinary cortisol was extracted in 10 ml methylexercise on the excretion rates of urinary free cortisol. ene chloride, and 2.0 ml of the organic phase were evaporated to dryness at 45°C; 1.0 ml of a 0.8% plasmaMETHOD CBG solution with tritium labeled cortisol was added to Twenty well-conditioned males (ages 25.1 k 3.3 yr the dried residue and the samples were left overnight at and vo 2 rnax55.6 k 4.2 ml/kg per ml) participated in this 4°C. Protein-bound and unbound cortisol fractions were study. To minimize the psychological effects of the exper- separated with 31 mg of Fuller’s earth. After centrifugaiment on adrenocortical secretory activity a 20-min fa- tion 250 ~1 of the supernatant were added to 5 ml of miliarization period on the treadmill, and two maximal scintillation cocktail (Aquasol; New England Nuclear). oxygen uptake tests (7) were completed several weeks All samples were counted for IO min in a liquid-scintillation counter (Isocap model 300; Nuclear Chicago). Assay before the study. Subjects were assigned at random to four categories of recoveries were 91.3 k 5.9%. The cortisol excretion rates treadmill exercise: walking for 10 or 30 min at 3.0 mph, Statistical analyses or running for 10 or 30 min at 7.5 mph. Excretion rates (ng/min) were transformed to logarithms, and the trends of the excretions over the four collection periods of urinary free cortisol were measured before and after exercise and were compared with excretion rates on a were analyzed by the least squares, multiple regression nonexercise control day. procedures described by Kerlinger and Pedhazur (12). l
Downloaded from www.physiology.org/journal/jappl by ${individualUser.givenNames} ${individualUser.surname} (130.237.165.040) on December 4, 2018. Copyright © 1976 the American Physiological Society. All rights reserved.
156
A. BONEN
For analyses within each group, the regression coefficients, or trends, of the cortisol excretion rates (log) vs. time on the exercise days were compared against the control day trends. For comparisons among groups, the first and second day exercise data were averaged and intergroup differences in the preexercise cortisol excretion rates were adjusted with covariance analysis. RESULTS
Intersubject cortisol excretion rates were quite variable (Table 1). Therefore, the cortisol excretion rates were transformed to logarithms and the regression analyses for trends were based on these transformed data (Fig. 1). Exercise at 3 mph for 10 min and at 3 mph for 30 min did not alter the cortisol excretion rate. In both groups, the trends of excretion over the four sampling periods during the exercise days were not significantly different (P > 0.05) from the control day (Fig. 1). In group c, exercising for 10 min at 7.5 mph, the control day cortisol excretion rate decreased linearly during the period of observation (P < O-05), whereas it remained virtually constant during the exercise days (Fig. 1). This difference in trends between control and exercise patterns was significant (P < 0.05). Excretion patterns on the two exercise days were not significantly different (P >
0.05). In group d, the control day cortisol excretion rates were constant during the experimental period (Fig. l), since the slope of the regression line, calculated for the GROUP
A
TABLE 1. Excretion rates of urinary free cortisol (nglmin) at rest and after exercise 1
Exercise
Day 1
1
Exercise
Day 2
Control
Day
I
-1 x8 ‘ds
Postexerci se time, min
& 5 r 30
Group a 3.0mph for
10
(18.6% N=5 Group for
mini
b 3.0 mph 30 min
2
E 30 t
60
90
k7.21 k4.1
k8.9
20.7 14.9 24.8 23.2
25.2 16.0 k3.4 23.2
17.6 ~2.4
20.1 29.9 ~4.0 k6.2
21.8 15.1
32.1 24.9 e5.9 54.1
33.3 56.8
26.7 30.1 ~5.2 46.0
24.0 27.0
d,‘:
Postrest time, min
60
90
17.0 24.7
17.4 52.5
13.7 k2.5
19.5 11.9 14.2 15.4 57.2 e2.1 k2.1 k2.7
23.2 kO.8
16.5 k2.1
32.2 22.9 25.2 17.6 k9.7 k6.6 k9.2 k5.7
vo, max) c 7.5 mph 10 min
(75.7% N=5 Group for
27.51 22.6 28.7
1 90
Postexercise time, min
i'o, max)
t17.59 N==5 Group for
60
T
, % z .C1
vo, ma&
d 7.5 mph 30 min
(80.4%
Vo, &
N=5
I
Values
123.4 89.2 58.8 143.3 k33.4C12.9
51.6 36.4 *15.7 10.7
are means
I * SE.
GROUP B 30 min at 3 mph
10 min at 3 mph
40 30
four sampling periods, was not significantly different from zero (P > 0.05). On the days of exercise, 30 min at 7.5 mph, the cortisol excretion rates during the 30-60min and the 60-90-min periods after exercise were greatly increased when compared to the control day (Fig. 1). From the preexercise sample to the 30-min postexercise sample there was a decrease in the excretion rate, followed by a large increase during the next 30 min and a slight decrease during the last 30 min. These exercise day trends (Fig. 1) conformed to cubic patterns
20
-_
GROUP
C 10 min at 7.5 mph
GROUP D 30 min at 7.5 mph /PI
Prr c + m
loL
/’
/ 1’ / J‘-, 1’ / / ,J/ 1’
I,
I
1
1
PRE -
30
60
90
PRE -
1
EXERCISE
EXERCISE
POST EXERCISE (MIN
1
II
1
30
-\ .. .. ‘* -. -. --. 1.
1
60
FIG. 1. Urinary free cortisol excretion rates before and after exercise and during similar periods on a control day (@ = exercise day 1; n = exercise day 2 ; 0 = controz day).
I
90
POST EXERCISE (MIN)
Downloaded from www.physiology.org/journal/jappl by ${individualUser.givenNames} ${individualUser.surname} (130.237.165.040) on December 4, 2018. Copyright © 1976 the American Physiological Society. All rights reserved.
CORTISOL
EXCRETION
RATES
AFTER
EXERCISE
Q’ < 0.05) and were significantly different from the control day trend (P < 0.05). Excretion rate patterns on the two exercise days were not significantly different (P > 0.05). Comparison among the four groups of the mean postexercise cortisol excretion rates, adjusted for preexercise excretion rates, showed significant interaction of intensity vs. duration of exercise (P < 0.05). Mean increase in the postexercise excretion rates in group d was significantly greater than in the other three groups (P < 0.05); other intergroup differences were not significantly different (P > 0.05). The mean rate of postexercise excretion of all groups, covaried for preexercise excretion rates, was lower during exercise day 2 than day 1 (P < 0.05). Cortisol excretion rates were not correlated with age, height, weight or vu, MBXon the control and exercise days, nor were the excretion rates correlated with the urine volume (r = 0.04) when the exercise and control data were combined (n = 240). In group d, the postexercise cortisol excretion rates correlated highly with the exercise intensity and the respiratory exchange ratio (Table 2). Urine excretion rates on the exercise and control days were 1.81 t 2.45 mllmin (n = 240). DISCUSSION
Significant differences from control day cortisol excretions were found in the two 7.5-mph groups (75-80% . VO 2 max) but not in the 3-mph groups (l&19% ‘iio, max). Results of this study agree with those studies in which total plasma cortisol concentrations have been measured. Changes in plasma cortisol concentrations have not been observed after moderate exercise of 60-min duration at 50% vo, max (8) or 8 min at 75% vo2 max (IO), but significant increments have been reported for intense exercise of 5-min duration at 98% v0, max (10) or for 1 hour of exercise at 70-80% vo, rnax(8, 11). Considering the delay between changes in plasma concentrations and urinary excretions, the observed postexercise excretion patterns in group d are quite similar to those reported for postexercise plasma cortisol concentrations after prolonged, intensive exercise. Peak plasma cortisol concentrations have been reported to occur 10-30 min (8, 9, 10) and 30-60 min after exercise (IS). Ingroup d the urinary excretion of free cortisol was greatest during the 30-60-min period after exercise. The large preexercise cortisol excretion rate in group d was the result of an extremely high preexercise excretion rate in one of the subjects, 101.8 and 98.9 ng/min on exercise days 1 and 2, respectively. Although preexercise anxiety might explain these results, this seems unlikely, since this individual was more familiar with treadmill running than most other subjects. Excreted free cortisol is presumed to be an index of the circulating, physiologically active cortisol (12). Results of cortisol turnover studies in man (9) suggest that the increased postexercise cortisol excretion rate may be attributed to an increased cortisol production during exercise, a decreased rate of cortisol utilization after exercise, a cortisol efflux from the tissues after exercise, or a combination of these factors. Exercise-induced incre-
157 2. Correlations bet ween corfisol rates (log) in 7.5 mph30 min group (n = 5), relative intensity of exercise, and respiratory exchange ratio
TABLE
Postexercise
% vO2 max
R
Cortisol
excretion
Excretions
30 min
60 min
90 min
0.94*
0.98* 0.97*
0.94* 0.99*
037*
Preexercise cortisol excretion rates are statistically of the postexercise rates. *P < 0.01.
partialed
out
ments in the cortisol excretion rate may also be derived from the CBG-bound cortisol fraction, since the CBGcortisol dissociation constant is quite temperature sensitive. At temperatives exceeding 37”C, such as during intense exercise (9), the CBG binding affinity is sharply reduced and the quantity of free cortisol is rapidly increased (21). In exercised guinea pigs, the bound cortisol fraction remained unaltered while the unbound plasma cortisol concentrations were more than doubled (19). In man, large increments in total plasma cortisol concentrations occurred when the tympanic temperatures exceeded 37.2”C (8). It is not known whether this change represents mainly an increase in the free cortisol fraction. Conflicting reports of adrenocortical responses to exercise (6, 17, 20) probably resulted from a failure to relate alterations in glucocorticoid concentrations to relative workloads. Davies and Few (8) found that the change in total plasma cortisol content was positively correlated with %iro, MaX, providing exercise intensity exceeded 60% ire, max. The present results indicate that urinary free cortisol excretion rates are also positively correlated with %iro, max when exercise is sufficiently intense. However, the duration of exercise must also be considered since cortisol excretion rates and exercise intensity were only correlated in group d, 80% vo2 max for 30 min, but not in group c, 75% vo, max for 10 min. Therefore, it is difficult to define a lower limit of exercise intensity that alone increases cortisol concentrations, since cortisol excretion rates depend on the interactive effects of exercise intensity and duration. Measurement of postexercise urinary cortisol excretions may not provide an accurate index of plasma cortisol responses during exercise, since cortisol is more rapidly utilized than at rest (9), and blood flow to the kidney is also reduced. Calculations of cortisol excretions are particularly dependent on the urine volume. Therefore, incomplete emptying of the bladder may cause significant errors, especially when the excreted urine volumes are small, such as after severe exercise (5). Urine c o 11ec t’ions in the present study were not a problem; excreted volumes ranged from 9,5 to 530 ml. Despite the limitations associated with assessing cortisol excretion rates, the results of the present study agree well with the results of studies in which plasma cortisol concentrations were measured after exercise (9, 12). In summary, after 10 or 30 min of mild exercise (approximately 20% i70, ,,,) urinary free cortisol excretion rates were not different from control, but after exercise
Downloaded from www.physiology.org/journal/jappl by ${individualUser.givenNames} ${individualUser.surname} (130.237.165.040) on December 4, 2018. Copyright © 1976 the American Physiological Society. All rights reserved.
158
A. BBNEN
at 7.5 mph (approximately 80% vo, max) for 10 or 30 min they were significantly greater than on the control day. Only in group d (30 min, 7.5 mph), however, were the postexercise excretion rates significantly different from those of the other three groups. Thus there was a strong interaction of exercise intensity and duration on the excretion rates of urinary free cortisol.
This study was supported in part by a grant from the University of Illinois Research Board. Portions of this paper were presented at the American College of Sports Medicine meeting, Knoxville, Tennessee, 9-11 May, 1974. Present address of A. Bonen: Human Performance Research Laboratory, School of Physical Education, Dalhousie University, Halifax, Nova Scotia, Canada B3H 355. Received
for publication
14 July
1975.
REFERENCES C. G., C. W. BURKE, AND C. L. COPE. Urinary free cortisol measured by competitive protein binding. J. EndocrinoZ. 42: 79-89, 1968. BEISEL, W. R., J. J. Cos, R. HORTON, P. Y. CHAO, AND P. H. FORSHAM. Physiology of urinary cortisol excretion. J. Clin. Endocrinol. Metab. 24: 887-893, 1964. BJEISEL, W. R., V. C. DIRAIMONDO, AND P, H. FORSHAM, Cortisol transport and disappearance. Arm. InternaL Med. 60: 641-652, 1964. BONEN, A., AND A. N. BELCASTRO. The accuracy of a dry gas meter to monitor ventilation during exercise. Brit. J. Sports Med. 8: 181-182, 1974. CASTENFORS, J. Renal function during exercise. With special reference to exercise proteinuria and the release of renin. Acta Physiol. Stand. Suppl. 293: l-44, 1967. CORNIL, A., A. DE COSTER, G. COPINSCHI, AND J. R. M. FRANCKSON. Effect of muscular exercise on the plasma level of cortisol in man. Acta EndocrinoZ. 48: 163-168, 1965. COSTILL, D. L., AND E. L. Fox. Energetics of marathon running. Med. Sci. Sports 1: 81-86, 1969. DAVIES, C. T. M., AND J. D. FEW. Effects of exercise on adrenocortical function. J. Appl. PhysioZ. 35: 887-891, 1973. FEW, J. D. The effect, of exercise on the secretion and metabolism of cortisol in man. J. EndocrinoZ. 62: 341-353, 1974. HARTLEY, L. H., J. W. MASON, R. P. HOGAN, L. G. JONES, T. A. KOTCHEN, E. H. MOUGEY, F. E. WHERRY, L. L. PENNINGTON, AND P. T. RICKETTS. Multiple hormonal responses to grad.ed exercise in relation to physical training. J. AppZ. Physiol. 33: 602-606, 1972. HARTLEY, L. H., J. W. MASON, R. P. HOGAN, L. G. JONES, T. A. KOTCHEN, E. G. MOUGEY, F. E. WHERRY, L. L. PENNINGTON, AND P. T. RICKETTS. Multiple hormonal responses to prolonged exercise in relation to physical training. J. Appl. PhysioZ. 33: 607610, 1972. KERLINGER, F. N., AND E. J. PEDHAZUR. MuZtiDZe Repression in
1. BEARDWELL,
2.
3.
4.
5.
6.
7. 8. 9. 10.
11.
12.
Behavioral 1973.
Research.
Toronto:
Holt,
Rinehart
and
Winston,
13. K~BBERLING, J., AND A. VON ZUR M~HLEN. The influence of diphenylhydantoin and carbamazepine on the circadian rhythm of free urinary corticoids and on the suppressibility of the basal and the “impulsive” activity by dexamethasone. Acta Endocrinol. 72: 308-318, 1973. 14. MILLS, J. N. Human circadian rhythms. Physiol. Rev. 46: 128171, 1966. 15. MURPHY. B. E. P. Some studies on the protein-binding of steroids and their application to the routine micro and ultramicro measurement of various steroids in body fluids by competitive protein-binding radioassay. J. CZin. Endocrinol. Metab. 27: 973990, 1967. 16. MURPHY, B. E. P. Clinical evaluation of urinary cortisol determinations by competitive protein-binding radioassay. J. CZin. EndocrinoZ. Metab. 28: 343-348, 1968. 17. ROSE, L. I., H. S. FRIEDMAN, S. C, BEERING, AND K. H. COOPER. Plasma cortisol changes following a mile run in conditioned subjects. J. CZin, EndocrinoZ. Metab. 33: 339-341, 1970, 18. SUTTON, J. R., J. D. YOUNG, L. LAZARUS, 5. B. HICKIE, AND J. MAKSVYTIS. The hormonal response to physical exercise. AustraZian Ann. Med. 18: 84-90, 1969. 19. VIRU, A., AND M. 0~s. Effects of physical exertion on the blood level of bound and free corticosteroids. EndocrinoZ. Exp., BratisLava 6: 227-230, 1972. 20. WEGMAN, H. M., K. E. KLEIN, AND H. BRUNER. Submaximale Belastung und maximale Belastarbeit, I, Biochemische Untersuchungen an Untrainierten unter korperlicher Arbeit. Intern. 2. Angew. PhysioZ. 26: 4-12, 1968. 21. WESTPHAL, U. Steroid-Protein Interactions. Monographs on EndocrinoZogy, edited by F. Gross, A. Labhart, T. Mann, L. T. Samuels, and J. Zander. New York: Springer-Verlag, Vol. IV, 1971.
Downloaded from www.physiology.org/journal/jappl by ${individualUser.givenNames} ${individualUser.surname} (130.237.165.040) on December 4, 2018. Copyright © 1976 the American Physiological Society. All rights reserved.
in U.S.A.
Effects of exercise on excretion of urinary free cortisol
rates
A. BONEN Physical Fitness Research Laboratory, Department University of Illinois, Champaign, Illinois 61820
of Physical
Education,
BONEN, A, Effects of exercise on excretion rates of urinary free cortisol. J. Appl. Physiol. 40(2): 1X-158. 1976. -Excretion rates of urinary free cortisol were studied in 20 men assigned to four treadmill exercise groups: walking at 3 mph for 10 min or 30 min, or running at 7.5 mph for 10 min or 30 min. Free cortisol in urine was measured before and 30, 60, and 90 min after exercise, and again on a control day. Patterns of free-cortisol excretions after exercise at 7.5 mph for 10 and 30 min were significantly different from the control day (P < O.OS>, with the largest changes occurring in the 30-min group. Exercise and control patterns were not different for the other two conditions (P > 0.05). Within the 7.5 mph-30 min group the postexercise cortisol excretion rates were directly related to the relative intensity of exercise (%Vop max) and the respiratory exchange ratio. It is concluded that changes in free cortisol excretion rates depend on the duration as well as the intensity of exercise.
The exercise experiments were repeated on two separate occasions (4-35 days apart). A control day was interposed between the exercise days. All conditions were scheduled between midday and 10 p.m., when plasma and urinary cortisol levels are relatively stable (113,14). Each subject acted as his own control, and his 3 test sessionswere at the same time of day @lo-min difference). Subjects had not eaten for at least 3 h prior to the experiments. Water intake during the experimental periods was not controlled. After the subject had voided he sat quietly for 1 h to standardize the preexercise activity. During that time urine was collected at 30 and 60 min; the first sample was discarded, and the second was designated the preexercise sample. Urine samples were also collected 30, 60, and 90 min after exercise was terminated. Identical sampling periods were maintained during the control sessions when the men rested oxygen consumption; treadmill; walking; running for the entire experimental period. Urine volumes were measured, and aliquots were stored at -20°C. Oxygen consumption (Vo,> was measured during PLASMA CONCENTRATIONS of total cortisol are increased minutes ,9 and 10 of the submaximal exercise bouts. when exercise is extremely intense (lo), or when it Expired air (approximately 1,000 ml/min) was withexceeds 60% of the maximal oxygen intake (VO, max) drawn by vacuum pump from a calibrated dry gas meter (8). Yet, measurements of total plasma cortisol provide into evacuated polyethylene bags, and the gas samples only an approximate index of the physiologically active were analyzed for 0, (paramagnetic model E-2, Beckcortisol, since about 90% of the circulating cortisol is man) and CO, (Lira infrared analyzer, Mine Safety bound. In man, the excreted free cortisol provides a Appliances) (4). During the maximal oxygen intake test sensitive index of the circulating, physiologically active (7) the vo, was monitored each minute. cortisol (2, 3), since there is a rather constant fraction of Cortisol assay. All urine samples were assayed in free cortisol that is filtered by the kidney (2). The pur- triplicate by a modification of the competitive proteinpose of this study, therefore, was to investigate the binding method for urinary free cortisol (1, 15, 16). relative effects of the duration and the intensity of Unbound urinary cortisol was extracted in 10 ml methylexercise on the excretion rates of urinary free cortisol. ene chloride, and 2.0 ml of the organic phase were evaporated to dryness at 45°C; 1.0 ml of a 0.8% plasmaMETHOD CBG solution with tritium labeled cortisol was added to Twenty well-conditioned males (ages 25.1 k 3.3 yr the dried residue and the samples were left overnight at and vo 2 rnax55.6 k 4.2 ml/kg per ml) participated in this 4°C. Protein-bound and unbound cortisol fractions were study. To minimize the psychological effects of the exper- separated with 31 mg of Fuller’s earth. After centrifugaiment on adrenocortical secretory activity a 20-min fa- tion 250 ~1 of the supernatant were added to 5 ml of miliarization period on the treadmill, and two maximal scintillation cocktail (Aquasol; New England Nuclear). oxygen uptake tests (7) were completed several weeks All samples were counted for IO min in a liquid-scintillation counter (Isocap model 300; Nuclear Chicago). Assay before the study. Subjects were assigned at random to four categories of recoveries were 91.3 k 5.9%. The cortisol excretion rates treadmill exercise: walking for 10 or 30 min at 3.0 mph, Statistical analyses or running for 10 or 30 min at 7.5 mph. Excretion rates (ng/min) were transformed to logarithms, and the trends of the excretions over the four collection periods of urinary free cortisol were measured before and after exercise and were compared with excretion rates on a were analyzed by the least squares, multiple regression nonexercise control day. procedures described by Kerlinger and Pedhazur (12). l
Downloaded from www.physiology.org/journal/jappl by ${individualUser.givenNames} ${individualUser.surname} (130.237.165.040) on December 4, 2018. Copyright © 1976 the American Physiological Society. All rights reserved.
156
A. BONEN
For analyses within each group, the regression coefficients, or trends, of the cortisol excretion rates (log) vs. time on the exercise days were compared against the control day trends. For comparisons among groups, the first and second day exercise data were averaged and intergroup differences in the preexercise cortisol excretion rates were adjusted with covariance analysis. RESULTS
Intersubject cortisol excretion rates were quite variable (Table 1). Therefore, the cortisol excretion rates were transformed to logarithms and the regression analyses for trends were based on these transformed data (Fig. 1). Exercise at 3 mph for 10 min and at 3 mph for 30 min did not alter the cortisol excretion rate. In both groups, the trends of excretion over the four sampling periods during the exercise days were not significantly different (P > 0.05) from the control day (Fig. 1). In group c, exercising for 10 min at 7.5 mph, the control day cortisol excretion rate decreased linearly during the period of observation (P < O-05), whereas it remained virtually constant during the exercise days (Fig. 1). This difference in trends between control and exercise patterns was significant (P < 0.05). Excretion patterns on the two exercise days were not significantly different (P >
0.05). In group d, the control day cortisol excretion rates were constant during the experimental period (Fig. l), since the slope of the regression line, calculated for the GROUP
A
TABLE 1. Excretion rates of urinary free cortisol (nglmin) at rest and after exercise 1
Exercise
Day 1
1
Exercise
Day 2
Control
Day
I
-1 x8 ‘ds
Postexerci se time, min
& 5 r 30
Group a 3.0mph for
10
(18.6% N=5 Group for
mini
b 3.0 mph 30 min
2
E 30 t
60
90
k7.21 k4.1
k8.9
20.7 14.9 24.8 23.2
25.2 16.0 k3.4 23.2
17.6 ~2.4
20.1 29.9 ~4.0 k6.2
21.8 15.1
32.1 24.9 e5.9 54.1
33.3 56.8
26.7 30.1 ~5.2 46.0
24.0 27.0
d,‘:
Postrest time, min
60
90
17.0 24.7
17.4 52.5
13.7 k2.5
19.5 11.9 14.2 15.4 57.2 e2.1 k2.1 k2.7
23.2 kO.8
16.5 k2.1
32.2 22.9 25.2 17.6 k9.7 k6.6 k9.2 k5.7
vo, max) c 7.5 mph 10 min
(75.7% N=5 Group for
27.51 22.6 28.7
1 90
Postexercise time, min
i'o, max)
t17.59 N==5 Group for
60
T
, % z .C1
vo, ma&
d 7.5 mph 30 min
(80.4%
Vo, &
N=5
I
Values
123.4 89.2 58.8 143.3 k33.4C12.9
51.6 36.4 *15.7 10.7
are means
I * SE.
GROUP B 30 min at 3 mph
10 min at 3 mph
40 30
four sampling periods, was not significantly different from zero (P > 0.05). On the days of exercise, 30 min at 7.5 mph, the cortisol excretion rates during the 30-60min and the 60-90-min periods after exercise were greatly increased when compared to the control day (Fig. 1). From the preexercise sample to the 30-min postexercise sample there was a decrease in the excretion rate, followed by a large increase during the next 30 min and a slight decrease during the last 30 min. These exercise day trends (Fig. 1) conformed to cubic patterns
20
-_
GROUP
C 10 min at 7.5 mph
GROUP D 30 min at 7.5 mph /PI
Prr c + m
loL
/’
/ 1’ / J‘-, 1’ / / ,J/ 1’
I,
I
1
1
PRE -
30
60
90
PRE -
1
EXERCISE
EXERCISE
POST EXERCISE (MIN
1
II
1
30
-\ .. .. ‘* -. -. --. 1.
1
60
FIG. 1. Urinary free cortisol excretion rates before and after exercise and during similar periods on a control day (@ = exercise day 1; n = exercise day 2 ; 0 = controz day).
I
90
POST EXERCISE (MIN)
Downloaded from www.physiology.org/journal/jappl by ${individualUser.givenNames} ${individualUser.surname} (130.237.165.040) on December 4, 2018. Copyright © 1976 the American Physiological Society. All rights reserved.
CORTISOL
EXCRETION
RATES
AFTER
EXERCISE
Q’ < 0.05) and were significantly different from the control day trend (P < 0.05). Excretion rate patterns on the two exercise days were not significantly different (P > 0.05). Comparison among the four groups of the mean postexercise cortisol excretion rates, adjusted for preexercise excretion rates, showed significant interaction of intensity vs. duration of exercise (P < 0.05). Mean increase in the postexercise excretion rates in group d was significantly greater than in the other three groups (P < 0.05); other intergroup differences were not significantly different (P > 0.05). The mean rate of postexercise excretion of all groups, covaried for preexercise excretion rates, was lower during exercise day 2 than day 1 (P < 0.05). Cortisol excretion rates were not correlated with age, height, weight or vu, MBXon the control and exercise days, nor were the excretion rates correlated with the urine volume (r = 0.04) when the exercise and control data were combined (n = 240). In group d, the postexercise cortisol excretion rates correlated highly with the exercise intensity and the respiratory exchange ratio (Table 2). Urine excretion rates on the exercise and control days were 1.81 t 2.45 mllmin (n = 240). DISCUSSION
Significant differences from control day cortisol excretions were found in the two 7.5-mph groups (75-80% . VO 2 max) but not in the 3-mph groups (l&19% ‘iio, max). Results of this study agree with those studies in which total plasma cortisol concentrations have been measured. Changes in plasma cortisol concentrations have not been observed after moderate exercise of 60-min duration at 50% vo, max (8) or 8 min at 75% vo2 max (IO), but significant increments have been reported for intense exercise of 5-min duration at 98% v0, max (10) or for 1 hour of exercise at 70-80% vo, rnax(8, 11). Considering the delay between changes in plasma concentrations and urinary excretions, the observed postexercise excretion patterns in group d are quite similar to those reported for postexercise plasma cortisol concentrations after prolonged, intensive exercise. Peak plasma cortisol concentrations have been reported to occur 10-30 min (8, 9, 10) and 30-60 min after exercise (IS). Ingroup d the urinary excretion of free cortisol was greatest during the 30-60-min period after exercise. The large preexercise cortisol excretion rate in group d was the result of an extremely high preexercise excretion rate in one of the subjects, 101.8 and 98.9 ng/min on exercise days 1 and 2, respectively. Although preexercise anxiety might explain these results, this seems unlikely, since this individual was more familiar with treadmill running than most other subjects. Excreted free cortisol is presumed to be an index of the circulating, physiologically active cortisol (12). Results of cortisol turnover studies in man (9) suggest that the increased postexercise cortisol excretion rate may be attributed to an increased cortisol production during exercise, a decreased rate of cortisol utilization after exercise, a cortisol efflux from the tissues after exercise, or a combination of these factors. Exercise-induced incre-
157 2. Correlations bet ween corfisol rates (log) in 7.5 mph30 min group (n = 5), relative intensity of exercise, and respiratory exchange ratio
TABLE
Postexercise
% vO2 max
R
Cortisol
excretion
Excretions
30 min
60 min
90 min
0.94*
0.98* 0.97*
0.94* 0.99*
037*
Preexercise cortisol excretion rates are statistically of the postexercise rates. *P < 0.01.
partialed
out
ments in the cortisol excretion rate may also be derived from the CBG-bound cortisol fraction, since the CBGcortisol dissociation constant is quite temperature sensitive. At temperatives exceeding 37”C, such as during intense exercise (9), the CBG binding affinity is sharply reduced and the quantity of free cortisol is rapidly increased (21). In exercised guinea pigs, the bound cortisol fraction remained unaltered while the unbound plasma cortisol concentrations were more than doubled (19). In man, large increments in total plasma cortisol concentrations occurred when the tympanic temperatures exceeded 37.2”C (8). It is not known whether this change represents mainly an increase in the free cortisol fraction. Conflicting reports of adrenocortical responses to exercise (6, 17, 20) probably resulted from a failure to relate alterations in glucocorticoid concentrations to relative workloads. Davies and Few (8) found that the change in total plasma cortisol content was positively correlated with %iro, MaX, providing exercise intensity exceeded 60% ire, max. The present results indicate that urinary free cortisol excretion rates are also positively correlated with %iro, max when exercise is sufficiently intense. However, the duration of exercise must also be considered since cortisol excretion rates and exercise intensity were only correlated in group d, 80% vo2 max for 30 min, but not in group c, 75% vo, max for 10 min. Therefore, it is difficult to define a lower limit of exercise intensity that alone increases cortisol concentrations, since cortisol excretion rates depend on the interactive effects of exercise intensity and duration. Measurement of postexercise urinary cortisol excretions may not provide an accurate index of plasma cortisol responses during exercise, since cortisol is more rapidly utilized than at rest (9), and blood flow to the kidney is also reduced. Calculations of cortisol excretions are particularly dependent on the urine volume. Therefore, incomplete emptying of the bladder may cause significant errors, especially when the excreted urine volumes are small, such as after severe exercise (5). Urine c o 11ec t’ions in the present study were not a problem; excreted volumes ranged from 9,5 to 530 ml. Despite the limitations associated with assessing cortisol excretion rates, the results of the present study agree well with the results of studies in which plasma cortisol concentrations were measured after exercise (9, 12). In summary, after 10 or 30 min of mild exercise (approximately 20% i70, ,,,) urinary free cortisol excretion rates were not different from control, but after exercise
Downloaded from www.physiology.org/journal/jappl by ${individualUser.givenNames} ${individualUser.surname} (130.237.165.040) on December 4, 2018. Copyright © 1976 the American Physiological Society. All rights reserved.
158
A. BBNEN
at 7.5 mph (approximately 80% vo, max) for 10 or 30 min they were significantly greater than on the control day. Only in group d (30 min, 7.5 mph), however, were the postexercise excretion rates significantly different from those of the other three groups. Thus there was a strong interaction of exercise intensity and duration on the excretion rates of urinary free cortisol.
This study was supported in part by a grant from the University of Illinois Research Board. Portions of this paper were presented at the American College of Sports Medicine meeting, Knoxville, Tennessee, 9-11 May, 1974. Present address of A. Bonen: Human Performance Research Laboratory, School of Physical Education, Dalhousie University, Halifax, Nova Scotia, Canada B3H 355. Received
for publication
14 July
1975.
REFERENCES C. G., C. W. BURKE, AND C. L. COPE. Urinary free cortisol measured by competitive protein binding. J. EndocrinoZ. 42: 79-89, 1968. BEISEL, W. R., J. J. Cos, R. HORTON, P. Y. CHAO, AND P. H. FORSHAM. Physiology of urinary cortisol excretion. J. Clin. Endocrinol. Metab. 24: 887-893, 1964. BJEISEL, W. R., V. C. DIRAIMONDO, AND P, H. FORSHAM, Cortisol transport and disappearance. Arm. InternaL Med. 60: 641-652, 1964. BONEN, A., AND A. N. BELCASTRO. The accuracy of a dry gas meter to monitor ventilation during exercise. Brit. J. Sports Med. 8: 181-182, 1974. CASTENFORS, J. Renal function during exercise. With special reference to exercise proteinuria and the release of renin. Acta Physiol. Stand. Suppl. 293: l-44, 1967. CORNIL, A., A. DE COSTER, G. COPINSCHI, AND J. R. M. FRANCKSON. Effect of muscular exercise on the plasma level of cortisol in man. Acta EndocrinoZ. 48: 163-168, 1965. COSTILL, D. L., AND E. L. Fox. Energetics of marathon running. Med. Sci. Sports 1: 81-86, 1969. DAVIES, C. T. M., AND J. D. FEW. Effects of exercise on adrenocortical function. J. Appl. PhysioZ. 35: 887-891, 1973. FEW, J. D. The effect, of exercise on the secretion and metabolism of cortisol in man. J. EndocrinoZ. 62: 341-353, 1974. HARTLEY, L. H., J. W. MASON, R. P. HOGAN, L. G. JONES, T. A. KOTCHEN, E. H. MOUGEY, F. E. WHERRY, L. L. PENNINGTON, AND P. T. RICKETTS. Multiple hormonal responses to grad.ed exercise in relation to physical training. J. AppZ. Physiol. 33: 602-606, 1972. HARTLEY, L. H., J. W. MASON, R. P. HOGAN, L. G. JONES, T. A. KOTCHEN, E. G. MOUGEY, F. E. WHERRY, L. L. PENNINGTON, AND P. T. RICKETTS. Multiple hormonal responses to prolonged exercise in relation to physical training. J. Appl. PhysioZ. 33: 607610, 1972. KERLINGER, F. N., AND E. J. PEDHAZUR. MuZtiDZe Repression in
1. BEARDWELL,
2.
3.
4.
5.
6.
7. 8. 9. 10.
11.
12.
Behavioral 1973.
Research.
Toronto:
Holt,
Rinehart
and
Winston,
13. K~BBERLING, J., AND A. VON ZUR M~HLEN. The influence of diphenylhydantoin and carbamazepine on the circadian rhythm of free urinary corticoids and on the suppressibility of the basal and the “impulsive” activity by dexamethasone. Acta Endocrinol. 72: 308-318, 1973. 14. MILLS, J. N. Human circadian rhythms. Physiol. Rev. 46: 128171, 1966. 15. MURPHY. B. E. P. Some studies on the protein-binding of steroids and their application to the routine micro and ultramicro measurement of various steroids in body fluids by competitive protein-binding radioassay. J. CZin. Endocrinol. Metab. 27: 973990, 1967. 16. MURPHY, B. E. P. Clinical evaluation of urinary cortisol determinations by competitive protein-binding radioassay. J. CZin. EndocrinoZ. Metab. 28: 343-348, 1968. 17. ROSE, L. I., H. S. FRIEDMAN, S. C, BEERING, AND K. H. COOPER. Plasma cortisol changes following a mile run in conditioned subjects. J. CZin, EndocrinoZ. Metab. 33: 339-341, 1970, 18. SUTTON, J. R., J. D. YOUNG, L. LAZARUS, 5. B. HICKIE, AND J. MAKSVYTIS. The hormonal response to physical exercise. AustraZian Ann. Med. 18: 84-90, 1969. 19. VIRU, A., AND M. 0~s. Effects of physical exertion on the blood level of bound and free corticosteroids. EndocrinoZ. Exp., BratisLava 6: 227-230, 1972. 20. WEGMAN, H. M., K. E. KLEIN, AND H. BRUNER. Submaximale Belastung und maximale Belastarbeit, I, Biochemische Untersuchungen an Untrainierten unter korperlicher Arbeit. Intern. 2. Angew. PhysioZ. 26: 4-12, 1968. 21. WESTPHAL, U. Steroid-Protein Interactions. Monographs on EndocrinoZogy, edited by F. Gross, A. Labhart, T. Mann, L. T. Samuels, and J. Zander. New York: Springer-Verlag, Vol. IV, 1971.
Downloaded from www.physiology.org/journal/jappl by ${individualUser.givenNames} ${individualUser.surname} (130.237.165.040) on December 4, 2018. Copyright © 1976 the American Physiological Society. All rights reserved.
