Sex Differences in Response to Maximal Eccentric Exercise : Medicine & Science in Sports & Exercise (original) (raw)
Unaccustomed eccentric exercise, which involves exerting force as the muscle lengthens, typically results in muscle damage. Although muscle tissue analysis provides direct evidence of morphological changes resulting from the exercise stimulus, indirect measurements associated with muscle damage are often used, such as losses in muscle force, muscle soreness, and increased levels of muscle proteins such as creatine kinase (CK) and myoglobin (Mb) in the blood (6). The time course of changes for these four markers after high-intensity eccentric exercise are well documented, and although the values of each marker generally peak at differing time points within 5 d after the exercise stimulus, they all usually return to baseline or near baseline within 7-10 d (6).
Animal research has demonstrated sex differences in markers of muscle damage after eccentric exercise. Analyses of inflammatory markers, CK response, and muscle tissue morphology have shown that female animals experience an attenuated muscle damage response to eccentric exercise (1). This has led to speculation that women may experience less muscle damage after eccentric exercise than men (27). However, research in humans has been equivocal regarding sex differences in measures of muscle damage (6).
Studies have shown no significant sex differences in force loss and recovery between men and women (25). Sayers and Clarkson (25) report no significant differences between the sexes in mean force loss; however, they found that a larger number of women experienced profound decreases immediately after exercise in MVC (> 70%) than men. Women register higher cutaneous pain responses than men (21), but cutaneous pain does not significantly correlate with muscle pain in humans (14). Although one study (10) found that women reported lower pain intensities than men via a visual analog scale (VAS) 2 d after eccentric exercise, the literature largely shows no significant difference between men and women in the degree of delayed-onset muscle soreness (9). Studies involving moderate eccentric exercise, such as downhill running, have found no sex differences in the CK or Mb response (11), whereas findings from studies employing high-intensity eccentric exercise have not been consistent (6,15,27). The lack of consistency in findings could be attributable to differences in the exercise used, or it may be attributable to the variability in the measurements. As Sayers and Clarkson (25) point out, the mean values may not accurately reflect a possible sex difference in response. We believe, on the basis of our previous work (25), that there may be sex-specific differences in the distribution of response to eccentric exercise with regard to high responders, but that these differences may be masked by only examining the mean values.
The purpose of this report is to examine strength loss, muscle soreness, and serum CK and Mb after high-intensity eccentric exercise in a large group of men and women. To our knowledge, no study has investigated sex differences in all four of these standard indirect markers of muscle damage in a large population of subjects. On the basis of our previous work in which mean values may have masked sex differences (25), we hypothesized that, on average, there would be no significant differences between men and women for the means of these markers after exercise; instead, there would be differences between men and women in terms of the variability and distribution of the responses.
METHODS
The data used in this study were derived from a larger clinical trial from our laboratory that used a standard exercise protocol to evaluate the efficacy of an analgesic on muscle pain. Written informed consent, as approved by the human subjects review committee at the University of Massachusetts-Amherst, was obtained from all subjects. Of the 208 subjects who participated in the clinical trial, 100 participants (42 men, 58 women) received only the placebo treatment. The data collected from the subjects in the placebo group were analyzed for the current study. Renal function data and genetic data from this study are published elsewhere (5,7).
Subjects agreed to refrain from analgesic use, muscle treatments, strenuous or new physical activity, and alcohol during the study. Exclusion criteria included an occupation that required heavy weight lifting (strenuous lifting of packages or equipment), participation in a resistance training program in the previous 6 months, use of dietary supplements associated with muscle gain, baseline blood values outside of normal range, known muscle disorders, existing myopathy, diabetes mellitus, or hypothyroidism. Data regarding previous physical activity level (independent of resistance training status) and menstrual cycle were not collected. Use of hormonal contraceptives was reported. Of the 58 women in the study, 28 were hormonal contraceptive users, 23 were nonusers, and 7 did not report. All subjects completed a physical exam by a physician to determine that they were healthy and eligible for the study. Age, height, weight, body mass index, and baseline blood values are reported in Table 1.
Subject characteristics: mean ± SD.
Subjects performed 50 maximal eccentric (muscle lengthening) actions of the elbow flexor muscles of their nondominant arm. Two sets of 25 contractions were separated by a 5-min rest period. Each contraction was 3 s long, followed by 12 s of rest. Each subject was seated at a modified preacher curl bench with two cushions holding the arm in place. Starting from an elbow joint angle of approximately 45°, each subject was instructed (and verbally encouraged throughout the entire exercise) to attempt to maximally contract the elbow flexor muscles, resisting the forward motion of the lever as an experienced investigator moved the lever until the arm was extended to approximately 180°. Although it was not possible to guarantee maximal contraction in the absence of a superimposed electrical stimulus during the contraction (13), the mechanical advantage of the lever system allowed the investigator to easily overcome the force exerted by each subject, strongly encouraging maximal effort through the entire range of motion. Subjects were instructed to drink water before and during the exercise visit. They were encouraged to maintain hydration throughout the study and to monitor their urine color (no subject experienced darkened urine).
Before and immediately after the exercise, subjects performed three maximal voluntary isometric contractions (MVC) (3 s each, with 1 min between trials). For this test, subjects were positioned on a preacher bench with their elbow at 90°, and they were instructed to maximally use their elbow flexors alone to pull against a force transducer (model 3268CTL, Lafayette Instrument Company), as described by Thompson et al. (28). An experienced investigator provided verbal encouragement throughout the entire trial. Although 90° may not have been the optimal angle for all subjects performing MVC (19), and optimal angle may have changed after eccentric exercise (20), this was an intrasubject design, and the measurement was collected to investigate relative strength loss, which was achieved through the consistent use of the same angle. The average of the peak forces from each of three trials was used as a maximal isometric strength value. Up to five trials were performed to achieve three trials with force measurements within 5% of each other, to increase the likelihood that maximal contraction was reached. Subjects returned to the laboratory at 0.5 (12-14 h), 3, 4, 7, and 10 d after exercise when strength testing was performed.
Blood was collected before exercise and 4, 7, and 10 d after exercise to be analyzed for serum CK activity and Mb concentration. Samples were taken via venipuncture at an antecubital vein and collected into a vacutainer tube containing no anticoagulant. The samples were allowed to clot at room temperature and then centrifuged. The serum was immediately transferred to a plastic vial and stored at −20°C until transport for analysis. The time points were chosen because 1) CK peaks at 4 d after elbow flexion exercise (6) and is commonly used clinically as an indicator of the risk for renal failure (7), and 2) to ensure that CK had returned to baseline before subjects could be released from the study. These samples were taken as a precautionary measure to ensure that subjects in the clinical trial were not in danger of renal failure, and, therefore, each day, serum samples were transported to Holyoke Medical Center Laboratory (Holyoke, MA) for CK analysis, and to Quest Diagnostics, Inc. (Cambridge, MA) for Mb analysis. CK was measured by spectrophotometry on a Hitachi 917 (Boehringer Mannheim, Indianapolis, IN). Mb was measured by nephelometry on a BN2 Analyzer (Dade Behring, Deerfield, IL). Normal ranges for baseline CK and Mb values for men and women are listed in Table 1.
Muscle soreness/pain was evaluated before exercise and 0.5, 3, 4, 7, and 10 d after exercise, using a VAS after conducting two full-range-of-motion biceps curls, holding a 0.45-kg (for a subject whose weight was < 58.97 kg) or 0.90-kg (for subjects whose weight was ≥ 58.97 kg) dumbbell with the nondominant arm. An experienced investigator demonstrated the motion in a slow and controlled manner (approximately 4 s in duration), beginning with the elbow fully flexed and the dumbbell even with the shoulder, then extending the elbow until it was fully extended by the side of the body. The investigator then supervised the subject as he or she performed the test. The VAS was a 100-mm horizontal line on which the subject placed a vertical mark corresponding to his or her degree of peak soreness during the aforementioned movement. A mark measured as 0 mm would correspond to "no soreness," whereas a mark measured as 100 mm would correspond to "unbearable pain."
Statistical analysis.
Differences between men and women were tested at baseline by independent _t_-tests for each variable in question. The effect of exercise on muscle strength, soreness, CK, and Mb were assessed using repeated-measures ANOVA. In the case of CK, where baseline values were significantly different between men and women, ANCOVA was used. When a significant interaction was detected, a post hoc analysis was performed. Because different mechanisms may underlie immediate strength loss (e.g., fatigue) versus strength loss in the days after the exercise (0.5-10 d after exercise), independent _t_-tests were performed between men and women for strength loss immediately after exercise. Furthermore, _t_-tests between groups were also performed for time points during recovery (0.5-10 d after exercise), with Bonferroni adjustment to protect for multiple tests. Because there was a significant difference in baseline strength (force) between men and women, a Pearson product-moment correlation was performed to ascertain whether the differences in relative strength loss between men and women were associated with baseline strength rather than sex. Pearson product-moment correlations were also calculated to examine the relationship between strength loss at each time point (immediately after, and 0.5 d, 4 d, 7 d, and 10 d after) and to examine the relationship between dependent variables.
The time points for each dependent variable at which the mean response was the greatest were determined. These time points were used to examine variability within each sex, and Levene's tests were used to test for equality of variance between sexes. The distributions for these variables were described using skewness and kurtosis, and they are graphed as histograms when relevant. In a normal distribution, both skewness and kurtosis values are expected to equal 0. A distribution is positively skewed if there is a tendency towards the higher end of values, and it is negatively skewed if there is a tendency towards the lower end of values. A distribution displays positive kurtosis if there is a tendency towards the mean, showing a more peaked distribution (referred to as leptokurtic); a distribution displays negative kurtosis if there is a lesser tendency towards the mean, showing a flatter distribution (referred to as platykurtic). These time points also were used to perform Pearson product-moment correlations to determine the relationship among all four dependent variables in terms of magnitude of response. Significance was determined as P ≤ 0.05.
RESULTS
Subject characteristics and baseline measures.
Subject characteristics are provided in the Table 1. The men in this study were significantly taller (P ≤ 0.05) and heavier (P ≤ 0.05) than the women. They also had significantly higher baseline CK values (P ≤ 0.05) and greater baseline strength (P ≤ 0.05). There were no significant differences between the sexes for age or baseline Mb. Baseline values for strength, CK, and Mb were consistent with previous work in our laboratory (5).
Muscle strength.
Relative strength loss and recovery during the 10 d after exercise within each sex are depicted in Figure 1. A two-way ANOVA of relative strength loss and recovery at time points immediately after exercise and at 0.5, 3, 4, 7, and 10 d after exercise detected significant main effects of sex (P ≤ 0.05) and time (P ≤ 0.05). However, the interaction term (sex × time) did not determine a significant difference in the pattern of relative strength-loss values between men and women, showing that there were no sex-dependent differences in strength recovery during the 10 d after exercise.
Relative strength loss and recovery over time (at baseline, immediately after exercise, and 0.5, 3, 4, 7, and 10 d after exercise) expressed as mean ± SEM. Women exhibit significantly greater strength loss immediately after exercise, but there is no significant difference between the sexes at any other time point. * P ≤ 0.05.
The time point at which the subjects in this study exhibited the greatest degree of strength loss was immediately after exercise. An independent _t_-test for the effect of the exercise on relative immediate strength loss revealed that women experienced significantly greater mean ± SD strength loss (−57.8 ± 19.1%) than men (−50.4 ± 16.9%) (P ≤ 0.05). A greater percentage of women experienced more than 70% strength loss at this time point than men; 34.4% of women experienced greater than 70% strength loss immediately after exercise (N = 20), whereas 7.1% of men experienced greater than 70% strength loss immediately after exercise (N = 3). The Pearson product-moment correlation for baseline strength and immediate postexercise strength loss was not significant, suggesting that the strength-loss differences between men and women were likely not related to differences in baseline strength. Furthermore, there was no difference in immediate strength loss (mean ± SD) between women who used hormonal contraceptives (59.2 ± 19.7%) and those who did not (59.8 ± 18.7%). Independent _t_-tests (with Bonferroni adjustments for multiple tests) between men and women during recovery time points (0.5-10 d after exercise) showed no significant differences (P ≤ 0.05). Pearson product-moment correlations for strength loss among time points for both sexes are shown in Table 2. In men, immediate strength loss was highly significantly correlated (P < 0.01) with all recovery time points, in agreement with Nosaka et al. (17). All correlations were also highly significant in women (_P_ < 0.01). When outliers (> 2 SD away from the mean) for both sexes were excluded from the analysis, significance was not affected. When female high responders for immediate strength loss (> 70% strength loss) were excluded from the analysis, the correlation between all time points remained highly significant, with the exception of the correlation between immediate strength loss and 10 d after exercise, which, although not highly significant, was still significant (P ≤ 0.05).
Correlations among strength loss at all time points.
To confirm our findings of a significant sex difference for immediate postexercise strength loss, we decided to expand the analysis to include all subjects in the study (101 men, 107 women). Because strength loss immediately after exercise was taken before treatment, there was no concern for a treatment effect when considering the entire sample of 208 subjects. The mean ± SD strength loss for men was −48.6 ± 18.7%, and for women it was −56.2 ± 19.6% (P ≤ 0.05); 27% (N = 29) of the women had greater than 70% strength loss, whereas only 7.9% (N = 8) of men did. The postexercise strength-loss results for the entire cohort confirm our findings for the placebo group. A histogram of relative strength loss immediately after exercise in the placebo group for each sex is depicted in Figure 2. Levene's test found no differences between men and women for relative strength-loss variance at this time point. However, women exhibited positive skewness greater than twice the standard error of skewness (SES), whereas men exhibited skewness within twice the SES. These data confirm our findings and suggest that the distribution of immediate strength-loss values for women skews to the higher end of relative strength-loss values to a greater degree than that of men (Table 3). When a histogram was examined for hormonal contraceptive users and nonusers, there was no difference in the distribution of responses, and Levene's test found no differences in variance between hormonal contraceptive users and nonusers.
Distribution of peak (immediate postexercise) relative strength loss. The distribution of strength-loss values for women are skewed towards the higher end of values compared with men.
Indirect markers of muscle damage: time points listed for each marker are those for which the highest mean value was found.
Serum CK and Mb.
CK measurements at baseline and at 4, 7, and 10 d after exercise are depicted for each sex in Figure 3. An ANCOVA for CK detected significant effects for time (P ≤ 0.05), but not for sex. The interaction term (sex × time) detected a significant difference in the pattern of response between the sexes (P ≤ 0.05), and a post hoc analysis determined that CK response for men was significantly greater than for women at 4 d after exercise (men = 10276.3 ± 11740.3; women = 6594.7 ± 7216.4). This is also the time point at which the greatest mean CK response for both sexes was exhibited. A histogram of CK responses at this time point within each sex is depicted in Figure 4. Furthermore, there was no difference between women who used hormonal contraceptives (6818.5 ± 7235.0) and those who did not (7094.0 ± 6889.2). When the histogram was examined for hormonal contraceptive users and nonusers, there was no difference in the distribution of responses. Although men in this study were significantly heavier and taller than women (Table 1), no significant correlations were found between CK response and height, weight, or BMI, suggesting that this difference in body size between the sexes did not contribute to the significant differences in CK response in this study.
Creatine kinase (CK) response over time (at baseline and at 4, 7, and 10 d after exercise) expressed as mean ± SEM. When covaried for baseline sex differences in CK, the interaction (sex X time) is still significant. Post hoc analysis showed that men exhibit significantly greater CK response at 4 d after exercise. * P ≤ 0.05.
Distribution of peak (4 d after exercise) creatine kinase (CK). The distribution of CK values for men are skewed towards the higher end of values compared with women.
Levene's test found a trend towards a difference between men and women for variance in CK values at 4 d after exercise (P = 0.06), with CK values for men displaying a trend towards greater variance then women. Levene's test found no differences in variance between hormonal contraceptive user and nonusers. Skewness and kurtosis of 4-d postexercise CK values are shown in Table 3. Men and women showed positive kurtosis (tendency towards the mean), with the kurtosis values higher for men (and greater than twice the standard error of kurtosis (SEK)) than women (whose kurtosis values were well within twice the SEK) (Table 3).
A two-way ANOVA for Mb at baseline and at 4, 7, and 10 d after exercise detected significant effects of time (P ≤ 0.05) but not of sex. Furthermore, the interaction term (sex × time) did not determine a significant difference in the pattern for Mb measurements between men and women (Fig. 5). The time point for which data were collected in this study that showed the greatest mean response was 4 d after exercise. In terms of distribution of values for Mb at this time point, both sexes displayed positive skewness values greater than twice the SES and positive kurtosis values greater than twice the SEK (Table 3). Thus, the distribution of responses was similar for men and women. A histogram of Mb values is depicted in Figure 6.
Myoglobin (Mb) response over time (at baseline and at 4, 7, and 10 d after exercise) expressed as mean ± SEM. There were no significant differences between men and women in their Mb response.
Distribution of peak (4-d postexercise) myoglobin (Mb).
Muscle soreness.
A two-way ANOVA for soreness at 0.5, 3, 4, 7, and 10 d after exercise detected significant effects for time (P ≤ 0.05) but not for sex. Furthermore, the interaction term (sex × time) did not determine a significant difference in the pattern of soreness measurements between men and women. The time point for which data were collected in this study that showed the greatest degree of mean soreness was 3 d after exercise. There was no significant difference in means between men and women at this time point in their pain responses. Furthermore, Levene's test found no difference between men and women for variance, and although men exhibit a flatter distribution (negative kurtosis) compared with women, neither sex shows kurtosis values greater than twice the SEK. The mean ± SD, as well as the skewness and kurtosis values for men and women at 3 d after exercise, are listed in Table 3.
Relationships among dependent variables.
Correlation matrices for men and women are shown in Table 4. For men, significant correlations were found between all dependent variables (strength loss, CK, Mb, and soreness), suggesting that changes in these measures after eccentric exercise are related. For women, significant, but lower, correlations were found among strength loss and the biochemical measures (CK, Mb). However, for women, no significant correlations were found between soreness and any of the other measures (strength loss, CK, Mb).
Correlations among indirect markers of muscle damage.
DISCUSSION
In addition to differences in mean response over time for the four standard indirect markers of muscle damage (strength loss, soreness, CK, and Mb), this study also examined potential differences in variance and distribution between the sexes for these markers. The most interesting findings from this study were that 1) women demonstrated greater strength loss immediately after eccentric exercise than men (showing a distribution of values that was skewed towards the higher end of strength loss), whereas men demonstrated a normal distribution at this time point; 2) men showed a larger CK response 4 d after exercise than women and exhibited a trend towards greater variability at this time point than women, with a markedly more leptokurtic distribution; 3) there were no differences found between men and women for soreness at 3 d after exercise or for serum Mb response at 4 d after exercise, nor were there significant interaction terms detected for either soreness or serum Mb over time; and 4) in men, all four indirect markers of muscle damage were significantly correlated, whereas in women the correlations were lower compared with those for men, with soreness showing no correlation with the other measures. A significant limitation of this study was that the timing of the measurements may have missed the peak responses. Even so, the data collected further elucidate potential sex differences in these four widely used markers of muscle damage.
Women in this study experienced significantly greater relative strength loss immediately after exercise than did men. Furthermore, in agreement with previous literature (13,25), a larger percentage of women than men experienced profound strength loss immediately after exercise. Sayers and Clarkson (25) report that in a sample of 192 subjects, 24 of the 32 subjects experiencing more than 70% strength loss immediately after exercise were women (17% of all women vs 8% of all men); and in a recent study (13), 14 out of 21 high responders for strength loss were women. In a sample of 100 subjects in the present study, 20 of the 23 subjects who experienced more than 70% strength loss immediately after exercise were women (34% of all women vs 7% of all men). As a result, the distribution of relative strength-loss values immediately after exercise for women in the present study was skewed (> 2 SES) towards the higher end of immediate strength-loss values, whereas men exhibited skewness values within 2 X SES distribution.
Interestingly, two women showed strength losses of less than 10%. One explanation for this could be that they were unable to contract maximally, even though they tried, and therefore, they did not exhibit the immediate strength losses (> 50%) typically seen with maximal eccentric exercise (6). In the absence of a superimposed electrical stimulus, it cannot be known whether these subjects failed to achieve maximal contraction during the eccentric exercise (13). However, one of these subjects had the lowest baseline strength (6.7 N·m) of all subjects, which may be an indication that this was the case. Her relative strength loss immediately after exercise was 2.3%. The second subject also exhibited minor strength loss immediately after exercise (9.4%) and a greater degree of strength loss at 0.5 d after exercise (27.4%), which was still less than that typically seen with strenuous eccentric exercise. This suggests that she, too, may have been unable to exert a maximal effort.
Various reasons have been put forth to explain the potential relative strength-loss differences between the sexes, including differences in historical physical activity patterns and protocol compliance (25). However, it is doubtful that relative strength-loss differences between the sexes in this study can be explained by differences in fitness levels or degrees of effort. No subject in this study had participated in resistance training in the 6 months leading up to their participation, and the maximal effort of the subjects was encouraged by an experienced investigator during both the exercise and testing protocols. It is possible that men perform more submaximal eccentric contractions as part of daily life compared with women, and this might explain the lower strength loss; it is known that the "repeated-bout effect" can be incurred such that submaximal eccentric exercise attenuates muscle damage during subsequent eccentric exercise (3). However, the most dramatic evidence of the repeated-bout effect is typically an attenuated CK response (8), and we found that men had higher CK activity than women after the eccentric exercise. Although women experienced greater immediate strength loss than men, they were not significantly different from men in terms of strength loss at other time points. In fact, by 0.5 d after exercise, men and women no longer differed significantly for this measure. Therefore, because differences between the sexes were only evident immediately after exercise, but not during recovery (0.5-10 d after exercise), it is likely that the differences in strength loss immediately after exercise were attributable, at least in part, to differences in fatigue rather than only muscle damage. It is interesting to note that the correlation of immediate strength loss with strength loss at 0.5 d after exercise was lower for women (r = 0.75) compared with men (r = 0.87). Also, correlations among relative strength loss at all time points for both men and women were highly significant (P < 0.01). When female high responders for immediate strength loss (> 70% strength loss) were excluded from the analysis, the correlation between all time points remained highly significant, with the exception of the correlation between immediate strength loss and strength loss at 10 d after exercise, which, although not highly significant, was still significant (P ≤ 0.05). Therefore, the correlations are not substantially influenced by high responders.
Muscle fatigue is the decline in a muscle's ability to exert force. Sex differences in fatigue have been previously demonstrated for time to task failure during both maximal and submaximal isometric contractions such that women were more fatigue resistant than men (6). However, the changes in muscle tissue that contribute to fatigue, such as those associated with metabolism, blood flow, or excitation-contracting coupling (e.g., intracellular calcium), may differ with contraction type (4), and Clark et al. (4) found that whereas women were more fatigue resistant during a submaximal isometric time-to-task-failure test than men, there was no difference between the sexes during an isotonic time-to-task-failure test. Therefore, it is possible that sex-dependent differences in metabolism, blood flow, or intracellular calcium specific to eccentric exercise could explain the fact that women in this study exhibited greater degrees of immediate relative strength loss than men. Evidence has shown that calcium release and uptake are both depressed after intense dynamic exercise (12). Another potential mechanism for this difference in fatigue between men and women could be an immediate superimposition of high-frequency fatigue attributable to conduction block, or the differences may be attributable to a decreased ability to recruit motor units on the part of women compared with men. Yet, another possible reason for this difference could be a sex-dependent, fatigue-related alteration in central nervous system activity; it has been proposed that fatigue is regulated, in part, by sensory-mediated changes in central activation (13), though, to our knowledge, no sex differences in this process have been discovered. Although these explanations are speculative, findings from the present study would suggest that, for some women, maximal eccentric exercise leads to profound strength loss immediately after exercise, resulting from potential factors such as a greater degree of excitation-contraction uncoupling or greater decreases in central activation.
The expected strength loss after strenuous eccentric exercise is between 50 and 60% (6). Why some women in this study showed > 70% strength loss immediately after eccentric exercise is not clear; they did not differ from the other women in terms of physical characteristics or initial strength. However, these results suggest that in studies using both men and women, it is important to recognize that some women may experience greater degrees of fatigue after exercise. Therefore, strength loss should always be assessed throughout the recovery period if it is going to be used as an indirect marker for muscle damage; immediate strength-loss data could be misleading in this regard.
The ANCOVA for CK response showed a significant interaction (sex × time). Post hoc analysis revealed that men showed significantly higher mean CK values than women at 4 d after exercise, the time at which CK is expected to peak after eccentric exercise (6). This finding is consistent with the intriguing work of Stupka et al. (26), who have reported a more attenuated rise in plasma CK activity in women after an acute bout of submaximal eccentric exercise of the knee extensors compared with men. Sex differences in CK response to exercise have also been previously demonstrated in animals (30).
Interestingly, more men than women were CK high responders at this time point (Fig. 4). This is also somewhat consistent with the findings of Stupka et al. (26), who show a greater range of CK response in men than in women, although the distribution of values is not reported. Although it could be argued that these high responders experienced more muscle damage or had greater membrane permeability, resulting in greater CK response, a direct relationship between CK response and degree of muscle damage has not been demonstrated in the literature (26,27,30).
Although men in this study were significantly heavier than women, suggesting that they also had increased muscle mass, it is doubtful that this increase in muscle mass was responsible for the differences in the magnitude of CK response after eccentric exercise of the elbow flexors. With the maximal exercise model used in this study, it has been shown that increased muscle mass exposed to the eccentric exercise of the elbow flexors does not result in increased CK response (18). It may be speculated that inherently higher CK enzyme activity in skeletal muscle in men compared with women could contribute to the differences in baseline serum CK response as well as those after exercise; however, this has never been documented. Further research is needed to discover the mechanism underlying the difference in CK response between the sexes and the existence of more male high responders compared with women.
It has been argued in previous literature (2,23,27,29) that differences in postexercise CK response in women may be related to differences in circulating estradiol resulting from differing menstrual cycle phase or from the use of hormonal contraceptives at the time of the exercise. By extension, this argument would possibly support the idea that differences in CK response between men and women are also hormone dependent, because premenopausal women have higher levels of circulating estradiol than men. Studies have shown that circulating estradiol and/or hormonal contraceptive use may affect CK response in women after maximal eccentric exercise of the knee extensors (23), downhill running (2), and endurance exercise (29). However, in the current study, there were no significant differences found in either the mean CK response at 4 d after exercise or in the distribution of these values. This is in agreement with the work of Miles and Schneider (16), who have shown that circulating estradiol does not affect CK response in women after eccentric exercise of the leg muscles, and with previous work in our laboratory that has shown that hormonal contraceptive use did not affect CK response after 50 maximal eccentric contractions of the elbow flexors (the protocol used in the current study) (24). Although women in this study exhibited an attenuated CK response on average, there are still female high responders whose CK values at 4 d after exercise were above 20,000 U·L−1 (Fig. 4). CK is an extremely variable measurement (6), and, therefore, although men have a higher CK response than women on average, there will still be women who are CK high responders. Furthermore, there was higher variability in response for men (Fig. 4), which cannot be explained by levels of estrogen. Thus, it seems unlikely that differences in circulating estradiol contributed to the differences in CK response between the sexes or in CK variability for women after maximal eccentric exercise of the elbow flexors. The variability in CK response is still poorly understood.
For Mb, both sexes exhibited similar skewness and kurtosis patterns (> 2 SEK and SES) for distribution at 4 d after exercise. These data suggest that there may be no difference between the sexes in terms of Mb response after eccentric exercise. CK and Mb responses generally correlate after acute exercise (22); however, differences between the sexes were found in this study for CK, but not for Mb. Correlations between 4-d Mb (the peak time point for which data were collected in this study) and earlier time points (2 and 3 d after exercise) have been shown previously to be very strong (r ~ 0.90) (22).Therefore, although the peak value may have been missed, it is unlikely that it resulted in the failure to detect a difference between the sexes. Still, this limitation restricts conclusions we can reach; further investigation is warranted to confirm a lack of significant difference between men and women in Mb response to eccentric exercise and to determine why this may be so.
There were no differences found between the sexes in terms of muscle soreness response after the exercise. Although soreness generally peaks at 2 d after exercise, the current study did not include a measurement at this time point, because of the demands of the clinical trial from which the data were derived. Although this is a limitation in the current study, unpublished data in our laboratory have shown very strong correlations between 2-d and 3-d soreness for men (r = 0.94) and women (r = 0.96). Therefore, because sex differences were not detected at 3 d after exercise, it is unlikely that they existed at 2 d after exercise, though further examination of additional time points is warranted. Furthermore, the findings in this study are consistent with previous literature regarding soreness after eccentric exercise. Although sex differences in cutaneous pain have been reported, with women being more responsive to painful stimuli than men (14), these difference are apparently not found with exercise-induced muscle soreness (10). The evidence from the current study further strengthens this assertion.
Significant correlations among the four indirect markers of muscle damage were found for men but were weaker in women. This may be attributable, in part, to a heightened contribution of fatigue to immediate strength loss in some women, a factor unrelated to CK, Mb, or soreness response. In the case of soreness measures, no significant correlations with the other three markers were found in women. The mechanism for this sex-specific response is not known, and further investigation is warranted.
CONCLUSION
This is the first study to systematically investigate sex differences in four of the standard indirect markers of muscle damage (strength loss, CK response, Mb response, and soreness) in a large population of subjects. The large sample size allowed for variability analysis within sex, which would not be possible in smaller studies; and the relationships between all four markers could be determined for both men and women, allowing for detection of differences between the sexes not previously described. First, women demonstrated greater degrees of relative strength loss immediately after exercise compared with men, with a distribution skewed towards the higher end of strength-loss values. This greater skewness in values for women was attributable to a higher percentage of women experiencing profound (> 70%) immediate strength loss. These results show that differences between the sexes for immediate strength loss were driven, in part, by female high responders. Furthermore, because these differences in relative strength loss were found immediately after exercise, but not at later time points, they may be related to differences in fatigue rather than muscle damage. Second, 4 d after exercise, men exhibited a larger CK response compared with women, a trend towards greater variance, and a more peaked distribution of values. These results show that differences between the sexes for CK response were driven, in part, by male high responders. Thus, women were more likely to be high responders for immediate strength loss, and men were more likely to be high responders for increased serum CK in response to eccentric exercise. Furthermore, these findings indicate that investigators should take sex into account when analyzing data in future studies involving maximal exercise and indirect markers of muscle damage such as CK and strength loss. When comparing means for groups that include both sexes, care should be taken during the statistical analysis to ensure that differences between groups are not driven by female high responders (in the case of strength loss) or male high responders (in the case of CK values).
This study was funded by Medinova, Inc.
REFERENCES
1. Amelink GJ, Bar PR. Exercise-induced muscle protein leakage in the rat. Effects of hormonal manipulation. J Neurol Sci. 1986;76:61-8.
2. Carter A, Dobridge J, Hackney AC. Influence of estrogen on markers of muscle tissue damage following eccentric exercise. Fiziol Cheloveka. 2001;27:133-7.
3. Chen TC, Nosaka K, Sacco P. Intensity of eccentric exercise, shift of optimum angle, and the magnitude of repeated-bout effect. J Appl Physiol. 2007;102:992-9.
4. Clark BC, Manini TM, The DJ, Doldo NA, Ploutz-Snyder LL. Gender differences in skeletal muscle fatigability are related to contraction type and EMG spectral compression. J Appl Physiol. 2003;94:2263-72.
5. Clarkson PM, Hoffman EP, Zambraski E, et al. ACTN3 and MLCK genotype associations with exertional muscle damage. J Appl Physiol. 2005;99:564-9.
6. Clarkson PM, Hubal MJ. Exercise-induced muscle damage in humans. Am J Phys Med Rehabil. 2002;81:S52-69.
7. Clarkson PM, Kearns AK, Rouzier P, Rubin R, Thompson PD. Serum creatine kinase levels and renal function measures in exertional muscle damage. Med Sci Sports Exerc. 2006;38(4):623-7.
8. Clarkson PM, Litchfield P, Graves J, Kirwan J, Byrnes WC. Serum creatine kinase activity following forearm flexion isometric exercise. Eur J Appl Physiol Occup Physiol. 1985;53:368-71.
9. Dannecker EA, Hausenblas HA, Kaminski TW, Robinson ME. Sex differences in delayed onset muscle pain. Clin J Pain. 2005;21:120-6.
10. Dannecker EA, Koltyn KF, Riley JL 3rd, Robinson ME. Sex differences in delayed onset muscle soreness. J Sports Med Phys Fitness. 2003;43:78-84.
11. Eston RG, Lemmey AB, McHugh P, Byrne C, Walsh SE. Effect of stride length on symptoms of exercise-induced muscle damage during a repeated bout of downhill running. Scand J Med Sci Sports. 2000;10:199-204.
12. Hill CA, Thompson MW, Ruell PA, Thom JM, White MJ. Sarcoplasmic reticulum function and muscle contractile character following fatiguing exercise in humans. J Physiol. 2001;531:871-8.
13. Hubal M, Rubenstein SR, Clarskon PM. Mechanisms of variability in strength loss after muscle-lengthening actions. Med Sci Sports Exerc. 2007;39(3):461-8.
14. Janal MN, Glusman M, Kuhl JP, Clark WC. On the absence of correlation between responses to noxious heat, cold, electrical and ischemic stimulation. Pain. 1994;58:403-11.
15. Miles M, Clarkson PM, Smith LL, Howell JN, McCammon MR. Serum creatine kinase activity in males and females following two bouts of eccentric exercise. Med Sci Sports Exerc. 1994;26(5 Suppl):S68.
16. Miles MP, Schneider CM. Creatine kinase isoenzyme MB may be elevated in healthy young women after submaximal eccentric exercise. J Lab Clin Med. 1993;122:197-201.
17. Nosaka K, Chapman D, Newton M, Sacco P. Is isometric strength loss immediately after eccentric exercise related to changes in indirect markers of muscle damage? Appl Physiol Nutr Metab. 2006;31:313-9.
18. Nosaka K, Clarkson PM. Relationship between post-exercise plasma CK elevation and muscle mass involved in the exercise. Int J Sports Med. 1992;13:471-5.
19. Prasartwuth O, Allen TJ, Butler JE, Gandevia SC, Taylor JL. Length-dependent changes in voluntary activation, maximum voluntary torque and twitch responses after eccentric damage in humans. J Physiol. 2006;571:243-52.
20. Prasartwuth O, Taylor JL, Gandevia SC. Maximal force, voluntary activation and muscle soreness after eccentric damage to human elbow flexor muscles. J Physiol. 2005;567:337-48.
21. Riley JL 3rd, Robinson ME, Wise EA, Myers CD, Fillingim RB. Sex differences in the perception of noxious experimental stimuli: a meta-analysis. Pain. 1998;74:181-7.
22. Rodenburg JB, Bar PR, De Boer RW. Relations between muscle soreness and biochemical and functional outcomes of eccentric exercise. J Appl Physiol. 1993;74:2976-83.
23. Roth SM, Gajdosik R, Ruby B. Effects of circulating estradiol on exercise-induced creatine kinase activity. J Exerc Physiol. 2001;4:10-7.
24. Savage KJ, Clarkson PM. Oral contraceptive use and exercise-induced muscle damage and recovery. Contraception 2002;66:67-71.
25. Sayers SP, Clarkson PM. Force recovery after eccentric exercise in males and females. Eur J Appl Physiol. 2001;84:122-6.
26. Stupka N, Lowther S, Chorneyko K, Bourgeois JM, Hogben C, Tarnopolsky MA. Gender differences in muscle inflammation after eccentric exercise. J Appl Physiol. 2000;89:2325-32.
27. Stupka N, Tarnopolsky MA, Yardley NJ, Phillips SM. Cellular adaptation to repeated eccentric exercise-induced muscle damage. J Appl Physiol. 2001;91:1669-78.
28. Thompson PD, Moyna N, Seip R, et al. Functional polymorphisms associated with human muscle size and strength. Med Sci Sports Exerc. 2004;36(7):1132-9.
29. Timmons BW, Hamadeh MJ, Devries MC, Tarnopolsky MA. Influence of gender, menstrual phase, and oral contraceptive use on immunological changes in response to prolonged cycling. J Appl Physiol. 2005;99:979-85.
30. Van der Meulen JH, Kuipers H, Drukker J. Relationship between exercise-induced muscle damage and enzyme release in rats. J Appl Physiol. 1991;71:999-1004.
Keywords:
MUSCLE FATIGUE; MUSCLE DAMAGE; RESISTANCE EXERCISE; STRENGTH LOSS
©2008The American College of Sports Medicine