The Effect of Strength Training at the Same Time of the Day on the Diurnal Fluctuations of Muscular Anaerobic Performances

Abstract

Chtourou, H, Driss, T, Souissi, S, Gam, A, Chaouachi, A, Souissi, N. The effect of strength training at the same time of the day on the diurnal fluctuations of muscular anaerobic performances. J Strength Cond Res 26(1): 217–225, 2012—The aim of this study was to examine the effects of training at the same time of the day on the diurnal variations of anaerobic performances to provide some recommendations to adjust training hours with the time of the day of competitive events. Thirty participants underwent 8 weeks of lower-extremity progressive resistance training performed 3 times per week designed to promote muscular strength and power. These subjects were randomly assigned to a morning training group (MTG, 07:00–08:00 hours, n = 10), an evening training group (ETG, 17:00–18:00 hours, n = 10), and a control group (CG, completed all tests but did not train, n = 10). Performance in the squat jump, the countermovement jump, the Wingate and 1 repetition maximum (1RM) during leg extension, leg curl, and squat tests was recorded just before and 2 weeks after an 8-week course of regular training. For all the subjects, the morning and evening tests were scheduled at the same time of the day as for the morning and evening training sessions. Before training, the results indicated a significant increase in performance from morning to evening tests (ca. 2.84–17.55% for all tests) for all groups. After training, the diurnal variations in anaerobic performances were blunted in the MTG. In fact, there was no significant difference in muscular power or strength between morning and evening tests. However, these intradaily variations in anaerobic performances persisted in the ETG and CG. From a practical point of view, adaptation to strength training is greater at the time of the day at which training was scheduled than at other times.

Introduction

Sports performances are dependent on human biological rhythms that influence both mental and physical activities (¹). To date, the effect of time of the day on short-term anaerobic performances has been well established, with early morning nadirs and peak performances in the late afternoon (^14,16,21-23). This intradaily variation has been consistently reported either when tests were conducted over the entire 24 hours to study circadian rhythm effects or when they were conducted only during the daytime typical for sports event or exercise training to study the diurnal effects (¹⁵). However, the intradaily variations of performances can be influenced by several factors such as sleep deprivation, age, morningness-eveningness chronotype, jet lag, and the time of the day of training (²³).

Enhanced morning anaerobic performance may occur as a result of either (a) a potential shift of physiological rhythms induced by routinely waking up and being physically active early in the day (⁸) or (b) a time-of-day effect on the responses to strength training which can be observed via enhanced performance at the specific time of the day when training is undertaken (^6,17-19,22). Up to now, few studies have been reported on the influence of time of the day on exercise training. Souissi et al. (²²) showed that 6 weeks of strength training in the morning hours can typically improve poor morning performance to the same or an even higher level as their normal daily peak typically observed in the late afternoon. Similarly, Sedliak et al. (^17-19) observed, after 10 weeks of strength training, an improvement in power during loaded squat jumps (SJs), isometric unilateral knee extension peak torque, and half-squat 1RM in both morning and evening training groups (ETGs). After training, the diurnal variations were reduced in the morning training group (MTG). However, they persisted in the ETG and the control group (CG). Thus, both authors concluded that strength training repeatedly at a particular time of the day may be beneficial when daily peak of maximum strength needs to be achieved at that time of the day, especially in the morning hours. However, Souissi et al. found stronger temporal specificity of the morning group than did Sedliak et al. Furthermore, Blonc et al. (⁶) showed that 5 weeks of training (sprints, jumps, and others exercises) can improveperformance for both the MTG and the ETG but they failed to show any time-of-day effect on either performance or training benefit. They suggested that the passive warm-up effect of the environment might be the cause of there being no effect of the training at a specific hour. Thus, data about the effects of strength training at the same time of the day on the diurnal fluctuations of anaerobic performances remain inconsistent. Furthermore, information about this issue may potentially provide redundant information and may help sport science and strength and conditioning professionals in establishing the best time of the day of training to enhance muscular strength and power.

In view of the above considerations, the aim of this study was to analyze if strength training at the same time of the day modifies the diurnal fluctuations of performances during the Wingate, the vertical jump, and 1RM tests.

Methods

Experimental Approach to the Problem

For the purpose of this report, the subjects performed maximal brief SJ, countermovement jump (CMJ), the Wingate test, and the 1RM test during leg extension, leg curl, and squat before and 2 weeks after 8 weeks of strength training. They were randomly assigned to either an MTG (that trained only between 07:00 and 08:00 hours, n = 10), an ETG (that trained only between 17:00 and 18:00 hours, n = 10), or a CG (that did not train but participated in all tests, n = 10). They performed the evaluation tests in both the morning (between 07:00 and 08:00 hours) and the evening (between 17:00 and 18:00 hours). Morning and evening tests and training hours were chosen according to the peak and the minimum of circadian rhythm of anaerobic performances and with regard to the findings in the literature (^6,13,17-19). The test sessions were conducted from January to March, so the pretraining tests were carried out in January and the posttraining tests in March 2009.

Subjects

Thirty male physical education students took part in this study. Their mean (±SD) age, height, and weight were 22.9 ± 1.3 years, 1.8 ± 0.05 m, and 72.0 ± 8.8 kg, respectively. After receiving a thorough explanation of the protocol, they gave written consent to participate in this study. Ethical approval for the investigation was secured from the University Ethics Committee. The subjects were considered healthy and had no medical contraindications that could affect the results of this study. They were categorized as either “moderately morning type” (n = 8) or “neither type” (n = 22), on the basis of their answers to Horne and Ôstberg's (¹⁰) self-assessment questionnaire, which assesses morningness eveningness. They had also regular sleeping schedules. Mean (±SD) estimated sleep duration was 7.2 ± 0.4 hours (hours ± fraction). Initially, all the participants were physically active but had no history of lower-extremity strength training program during the 6 months before the study. They had exactly the same time schedule at the university from sunrise to sunset under the control of the experimental team. Thereafter, these participants were randomly assigned to an MTG, an ETG, or a CG.

Procedures

During the week before the experiment, the participants were familiarized with the equipment and the experimental procedures to minimize the learning effect during the course of the study. Then, they were tested at 2 moments, a week before and 2 weeks after the 8-week training period. The second test session was performed 2 weeks after the end of the training to take account of the “overcompensation” phenomenon (²²). Each participant completed strictly identical test sessions both in the morning (07:00–08:00 hours) and in the evening (17:00–18:00 hours). They performed only 1 test session per day. The time interval between 2 consecutive tests was at least 24 hours. The morning and evening test sessions were completed in a counterbalanced design. However, power tests in both sessions were performed in the same order: SJ, CMJ, and Wingate test, with at least 15 minutes of recovery between 2 successive tests. The subjects of the MTG trained only in the morning and performed tests in the morning and in the evening. However, the subjects of the ETG trained only in the evening and were also tested in the morning and in the evening. The morning and evening test sessions were scheduled at the same time of the day as the morning and evening training sessions.

Different precautions were taken before each experiment or test session. Instructions about sleep and diet were given to the subjects. In that, the subjects were asked to keep their usual sleeping habits with a minimum of 6 hours of sleep taken on the night preceding each test session. During the period of investigation, they were prohibited from consuming food, beverages, or any known stimuli (e.g., caffeine) or depressants (e.g, alcohol) that would possibly enhance or compromise alertness. Throughout the experimental period, the participants were requested to maintain their habitual physical activity and to avoid strenuous activity during the 24 hours before the test sessions. It was easy to control compliance with these directions because the participants were students who had exactly the same daily schedules in our institution. The subjects who were tested at 07:00 hours were requested to come to the laboratory at 06:00 hours in a fasting state. Each subject was permitted to drink only 1 glass (150–200 ml) of water before exercise to standardize the level of hydration and to avoid the effects of postprandial thermogenesis. After the morning session, free food intake was allowed. All the subjects had the same standard isocaloric meal at 12:00, which ended at least 5 hours before the tests at 17:00 hours. After that meal, only 1 glass (15–20 cl) of water was allowed ad libitum. The diet was identical for all participants, before each test session. The overall daily energy intake goal was set at 10.5 MJ (2,500 kcal) per capita per day. The laboratory temperature was maintained constant at 20.4 ± 1.1°C during the entire experimental period to minimize the effects of ambient temperature changes on the way the muscular system works (³).

At the beginning of each experiment, the body mass was measured using an electronic scale (Tanita, Tokyo, Japan). After 10 minutes of seated rest, the oral temperature was measured with a digital clinical thermometer (Omron®, Paris, France; accuracy ± 0.05°C) inserted sublingually for at least 3 minutes.

The CG was included to determine whether the increase of performance results from the training program or from the physical activity at the university.

Training Protocol

The participants assigned to the MTG and the ETG trained in 3 sessions per week for 8 weeks with training times being 07:00–08:00 hours for the MTG and 17:00–18:00 hours for the ETG. They had at least 48 hours of rest between 2 successive sessions. The training period was planned as lower-extremity periodized programs using a linear method of progression and focusing on the knee extensors and flexors muscles of both legs concomitantly using 3 devices: leg extension, leg curl, and squat. All the training sessions were supervised. One-repetition maximum (1RM) values for all exercises were obtained during the first training session and adjusted after 4 weeks of training. Hypertrophic type of training (60–70% of 1RM, 6 repetitions per set, 6 sets, 2 minutes of rest between sets) was performed during the first 2 weeks of the preparatory training period and strength type of training (100–120% of 1RM, 3–6 repetitions per set, 3–5 sets, 7–9 minutes of rest between sets) was performed during the last 6 weeks (Table 1). The weights were progressively increased to maintain this range of repetitions per set (110–120% during weeks 6–8).

Table 1: Basic variables of the resistance training protocol for both MTG and ETG.*

Data Collection

The P_peak, P_mean, SJ, and CMJ of the legs were determined by the Wingate and the vertical jump tests. The subjects were verbally encouraged throughout all tests to avoid pacing and to perform their maximal effort.

Wingate Test

The Wingate test was conducted for 30 seconds of maximal sprint on a cycle ergometer against constant resistance related to body mass (0.087 kg·kg⁻¹ body mass) using the optimization tables of Bar-Or (²). The tests were conducted on a calibrated friction-loaded cycle ergometer (Monark 894^E, Stockholm, Sweden) equipped with an electronic revolution counter and a recorder fitted onto the wheel and an option of setting a series of resisting forces before the beginning of an exercise. Seat height was adjusted to each participant's satisfaction. This seat height was recorded and kept the same for each participant in all tests, and the subject's feet were firmly held onto the pedals by means of toe clips. The participants remained in a sitting position during the test. Before starting, they were asked to assume the ready position, to maintain the pedaling rate close to 60 rpm and await the start signal announced by a countdown given by the experimenter. At the start signal, they were instructed to pedal as fast as they could for 30 seconds. The subjects were given vigorous verbal encouragement to reach the maximal pedaling rate as quickly as possible during the test.

The P_peak is the highest mechanical power elicited during the 30 seconds. This index was taken as the highest power produced during any 3- to 7-second part of the test. The P_mean was calculated as the average power during the test duration. The characteristics of the ergometer allowed the calculation of P_peak and P_mean as follows: P (W) = (F [kg] × 9.81 × V [rev·min⁻¹]/60) × 6.12 m = F (kg) × V (rev·min⁻¹).

Squat Jump and Countermovement Jump Tests

The SJ consists of a maximal jump from a flexed position with the hands on the hips, whereas the CMJ is a leg flexion from the standing position immediately followed by a maximal jump with the hands off the hips. These jump tests were monitored with the Takei Kiki Kogyo vertical jump meter (T.K.K.5406), which recorded the jump height. The participants performed 2 trials of each test with a recovery of 15 seconds between 2 successive jumps, and the best one was recorded.

Statistical Analyses

The data were analyzed using multivariate analysis of variance (MANOVA). The effects of group, time of the day and pre/posttraining were verified by a 3-way analysis of variance with repeated measures (3 [training group] × 2 [training] × 2 [time of day]). When appropriate, significant differences among means were tested using the least significant different post hoc test. The data are expressed as mean ± SD in the text and in the figures. A probability level of 0.05 was selected as the criterion for statistical significance. Statistical power was determined to be 0.80 for the sample size used at the 0.05 α level. Effect sizes were calculated as partial eta-squared η_p² to estimate the meaningfulness of significant findings. The test-retest reliability was expressed by intraclass correlation coefficients (ICCs), SEM, and the minimal differences (MDs) needed to be considered real. All statistical tests were processed using STATISTICA Software (StatSoft, France).

Results

Strength

Table 2 presents the results of the 1RM measured during leg extension, leg curl, and squat in the morning and in the evening, before and after training, for all groups.

Table 2: Subjects 1RM (mean ± SD) at 07:00 and 17:00 hours before and after training in the MTG (n = 10), ETG (n = 10), and CG (n = 10).*

Leg Extension

The ICC and SEM for 1RM in leg extension showed high reliability (ICC >0.86 and absolute SEM <3.1 kg). The observed differences of measurement between first and second evaluations were without exception lower than the calculated MD at a level of 8%.

There was a significant main effect for groups (F(2.27) = 3.56, p < 0.05; η_p² = 0.2), pre/posttraining (F(1.27) = 160.94, p < 0.001; η_p² = 0.85) and time of the day (F(1.27) = 44.03, p < 0.001; η_p² = 0.61). In contrast, there was no main effect for groups × pre/posttraining × time-of-day interaction (F(2.27) = 2.33, p > 0.05; η_p² = 0.14).

The post hoc test revealed that, before training, all groups behaved similarly. 1RM improved between the morning and the evening (p < 0.05 for the MTG and ETG and p < 0.01 for the CG) (amplitude: 3.62, 3.52, and 4.17%, for MTG, ETG, and CG, respectively). In the posttraining, this intraday variation disappeared in the MTG and persisted in the ETG (p < 0.01) and in the CG (p < 0.05). Moreover, the MTG and the ETG improved their 1RM in the morning (p < 0.001) and in the evening (p < 0.001). However, the improvement was larger in the morning than in the evening (16.5 ± 4.74 vs. 13.5 ± 4.12 kg, respectively) for the MTG, but gains were lower in the morning than in the evening (12 ± 6.32 vs. 13 ± 4.22 kg, respectively) for the ETG.

Leg Curl

High reliability was observed for 1RM in leg curl (ICC > 0.86 and absolute SEM < 3.3 kg). The observed differences of measurement between first and second evaluations were without exception lower than the calculated MD at a level of 9%.

There was a significant main effect for groups (F(2.27) = 6.42, p < 0.01; η_p² = 0.32), pre/posttraining (F(1.27) = 188.4, p < 0.001; η_p² = 0.87) and time of the day (F(1.27) = 20.16, p < 0.001; η_p² = 0.42). Moreover, there was a significant main effect for groups × pre/posttraining × time-of-day interaction (F(2.27) = 4.5, p < 0.05; η_p² = 0.25).

The post hoc test revealed that before training, 1RM increased significantly from morning to evening (p < 0.001 for the MTG and p < 0.01 for ETG and the CG) (amplitude: 4.17, 3.75, and 3.56%, for MTG, ETG, and CG, respectively). This diurnal fluctuation disappeared after training in the MTG and persisted in the ETG and the CG (p < 0.001). When we consider the effect of training, the MTG and the ETG improved their 1RM in the morning (p < 0.001) and in the evening (p < 0.001). However, the improvement was greater in the morning than in the evening (15.5 ± 2.84 vs. 12 ± 5.37 kg, respectively) for the MTG and in the evening than in the morning (14 ± 4.59 vs. 13 ± 5.37 kg, respectively) for the ETG.

Squat

The reliability of 1RM in squat was high (ICC ≥ 0.86) and absolute SEM showed low values (≤3.3 kg). The differences between the first and second evaluations were lower than the calculated MD at a level of 9%.

There was a significant main effect for groups (F(2.27) = 4.79, p < 0.05; η_p² = 0.26), pre/posttraining (F(1.27) = 302.17, p < 0.001; η_p² = 0.91) and time of the day (F(1.27) = 27.98, p < 0.001; η_p² = 0.5). Moreover, there was a significant main effect for groups × pre/posttraining × time-of-day interaction (F(2.27) = 7.83, p < 0.01; η_p² = 0.36).

The post hoc test revealed that before training, in all groups, 1RM recorded at 17:00 hours was higher than those recorded at 07:00 hours, with amplitudes of 3.45 ± 3.7, 4.38 ± 3.87, and 2.22 ± 3.66% for MTG, ETG, and CG respectively. After training, this diurnal variation disappeared in the MTG. Moreover, the MTG and the ETG improved their 1RM in the morning (p < 0.001) and in the evening (p < 0.001). However, the improvement was greater in the morning than in the evening (15.5 ± 2.84 vs. 12 ± 4.22 kg, respectively) for the MTG and in the evening than in the morning (13.5 ± 3.37 vs. 12.5 ± 3.54 kg, respectively) for the ETG.

Oral Temperature

Figure 1 presents the results of oral temperature obtained in the morning and in the evening, before and after training.

Figure 1: Changes in oral temperature (mean ± SD) in the morning and the evening tests before and after training for all groups. **, ***Significant differences between the time points at the level of p < 0.01 and p < 0.001, respectively.

In all the groups, a significant diurnal variation was found at rest in oral temperature before (p < 0.001) and after (p < 0.01) training. In fact, the oral temperature was significantly lower during morning trials.

Wingate Test

Figures 2 and 3 present the results of the Wingate test variables calculated in the morning and in the evening, before and after training, for all groups (P_peak and P_mean, respectively).

Figure 2: Changes in peak power (mean ± SD) at 07:00 and 17:00 hours before and after training in the morning training group (MTG, n = 10), evening training group (ETG, (n = 10), and control group (CG, n = 10). M pre = morning pretraining tests; E pre = evening pretraining tests; M post = morning posttraining tests; E post = evening posttraining tests. **, ***Significant differences between the time points at the level of p < 0.01 and p < 0.001, respectively. +++Significant differences from pre to post at the level of p < 0.001.

Figure 3: Changes in mean power (mean ± SD) at 07:00 and 17:00 hours before and after training in the morning training group (MTG, n = 10), evening training group (ETG, n = 10), and control group (CG, n = 10). M pre = morning pretraining tests; E pre = evening pretraining tests; M post = morning posttraining tests; E post = evening posttraining tests. **, ***Significant differences between the time points at the level of p < 0.01 and p < 0.001, respectively. +++Significant differences from pre to post at the level of p < 0.001.

Peak Power

ICC and SEM for P_peak showed very high reliability (ICC > 0.99 and absolute SEM < 0.05 W·kg⁻¹). The observed differences of measurement between first and second evaluations were without exception lower than the calculated MD at a level of 5%.

For P_peak, there was a significant (F(1.27) = 64.33, p < 0.001) time-of-day effect. There was also a significant (F(1.27) = 107.67, p < 0.001) pre/posttraining effect. The group effect was not significant (F(2.27) = 1.59, p > 0.05). However, there was a significant group × pre/posttraining × time-of-day interaction (F(2.27) = 57.87, p < 0.001). The effect size η_p² were at 0.7, 0.8, 0.1, and 0.81 for time of the day, pre/posttraining, group, and interaction among the 3, respectively.

The post hoc test revealed that, before training, all groups behaved similarly. P_peak improved between the morning and the evening (p < 0.001) (amplitude: 10.75, 5.9, and 6.38%, for ETG, MTG, and CG, respectively). In the posttraining, the ETG showed an improvement of P_peak (p < 0.001) in the evening compared with the morning test value. In contrast, the MTG showed an improvement (p < 0.01) in the morning test compared with the evening tests. When we consider the effect of training, the MTG improved their P_peak in the morning (p < 0.001) and in the evening (p < 0.001) tests. Although the absolute increase was larger in the morning (1.67 ± 0.42 W·kg⁻¹) than in the evening (0.73 ± 0.44 W·kg⁻¹). The ETG did not improve their P_peak in the morning, whereas they improved this in the evening (p < 0.001).

Mean Power

ICC and SEM for P_mean showed very high reliability (ICC > 0.95 and absolute SEM < 0.12 W·kg⁻¹). The observed differences of measurement between first and second evaluations were without exception lower than the calculated MD at a level of 5%.

For P_mean, there was a significant effect of both time of the day (F(1.27) = 55.81, p < 0.001) and pre/posttraining (F(1.27) = 33.65, p < 0.001). The group effect was not significant (F(2.27) = 1.23, p > 0.05). However, there was a significant group × pre/posttraining × time of day of test interaction (F(2.27) = 20.95, p < 0.001). The effect size η_p² were at 0.6, 0.42, 0.001, and 0.43 for time of day, pre/posttraining, group, and interaction among the 3, respectively.

The post hoc test revealed that before training, in all groups, P_mean increased significantly from morning to evening (p < 0.01) (amplitude: 5.49, 5.88, and 5.57%, for ETG, MTG, and CG, respectively). After training, this intradaily variation disappeared in the MTG. However, the ETG showed an improvement (p < 0.001) in the evening compared with the morning tests. When we consider the training effect, the MTG improved their P_mean in the morning (p < 0.001) and in the evening (p < 0.001) trials. The improvement was greater in the morning than in the evening (1.13 ± 0.27 W·kg⁻¹ in the morning vs. 0.53 ± 0.18 W·kg⁻¹ in the evening). However, the ETG improved their P_mean only in the evening (p < 0.001).

Vertical Jump Tests

Mean SJ and CMJ in the morning and evening, before and after training, for all groups are presented in Figures 4 and 5, respectively.

Figure 4: Changes in squat jump (mean ± SD) at 07:00 and 17:00 hours before and after training in the morning training group (MTG, n = 10), evening training group (ETG, n = 10), and control group (CG, n = 10). M pre = morning pretraining tests; E pre = evening pretraining tests; M post = morning posttraining tests; E post = evening posttraining tests. ***Significant differences between the time points at the level of p < 0.001. +++Significant differences from pre to post at the level of p < 0.001.

Figure 5: Changes in countermovement jump (CMJ; mean ± SD) at 07:00 and 17:00 hours before and after training in the morning training group (MTG, n = 10), evening training group (ETG, n = 10), and control group (CG, n = 10). M pre = morning pretraining tests; E pre = evening pretraining tests; M post = morning posttraining tests; E post = evening posttraining tests. ***Significant differences between the time points at the level of p < 0.001. +++Significant differences from pre to post at the level of p < 0.001.

Squat Jump Test

ICC and SEM for SJ showed very high reliability (ICC > 0.93 and absolute SEM < 1.67 cm). The observed measurement differences between first and second evaluations were without exception lower than the calculated MD at a level of 5%.

For SJ, there was a significant effect of both time of the- day (F(1.27) = 1,453.18, p < 0.001) and pre/posttraining (F(1.27) = 657.41, p < 0.001). The group effect was not significant (F(2.27) = 0.52, p > 0.05). However, there was a significant group × pre/posttraining × time-of-day interaction (F(2.27) = 311.14, p < 0.001). The effect size η_p² were at 0.96, 0.98, 0.04, and 0.96 for time of day, pre/posttraining, group, and interaction among the 3, respectively.

Before training, in all groups, the post hoc test revealed that SJ increased significantly between the morning and the evening (p < 0.001) (amplitude: 12.25, 17.55, and 15.42%, for ETG, MTG, and CG, respectively). After training, the ETG showed an improvement in SJ (p < 0.001) in the evening compared with that in the morning. On the other hand, there was no difference between morning and evening posttraining values for the MTG. When we consider the effect of training, the MTG improved their SJ in the morning (p < 0.001) and in the evening (p < 0.001) tests. The absolute increase was higher in the morning than in the evening (15.3 ± 2.31 cm in the morning vs. 7.8 ± 2.14 cm in the evening). The ETG did not improve their SJ in the morning, whereas they improved this in the evening (p < 0.001).

Countermovement Jump

The ICC and SEM for CMJ showed very high reliability (ICC > 0.97 and absolute SEM < 0.96 cm). The observed differences of measurement between first and second evaluations were without exception lower than the calculated MD at a level of 5%.

For CMJ, there was a significant effect of both time of the day (F_(1.27) = 1146.69, p < 0.001) and pre/posttraining (F(1.27) = 728.52, p < 0.001). The group effect was not significant (F(2.27) = 1.72, p > 0.05). However, there was a significant group × pre/posttraining × time-of-day interaction (F(2.27) = 534.78, p < 0.001). The effect size η_p² were at 0.96, 0.97, 0.11, and 0.97 for time of day, pre/posttraining, group, and interaction among the 3, respectively.

In the pretraining tests, the post hoc test revealed that all the groups behaved similarly; CMJ improved between the morning and evening (p < 0.001) (amplitude: 11.49, 15.99, and 15.61%, for ETG, MTG, and CG, respectively). In the posttraining, the ETG showed an improvement (p < 0.001) in the evening compared with that in the morning. In contrast, the MTG showed an enhancement in CMJ (p < 0.01) in the morning compared with that in the evening. Regarding the effect of training, the MTG improved their CMJ in the morning (p < 0.001) and in the evening (p < 0.001) tests. The absolute increase was higher in the morning than in the evening (18.1 ± 2.33 vs. 9.9 ± 1.79 cm, respectively). The ETG did not improve their CMJ in the morning, whereas they improved it in the evening (p < 0.001).

Discussion

The major result of this study was that adaptation to strength training is associated with greater improvements in anaerobic performances for the MTG and ETG at the training time. Moreover, strength training performed in the morning hours can improve typically poor morning performances to the same or even higher level as their normal daily peaks typically observed in the evening.

Before training, anaerobic performances (Wingate test, SJ, CMJ, and 1RM) were significantly higher in the evening compared with those of the morning test sessions with an acrophase between 17:00 and 18:00 hours and amplitudes amounting to between 5.9 and 17.55%. These findings confirm those reported in previous studies (^{4,13,16,19,21-23}), which demonstrated significant diurnal fluctuations during various anaerobic tasks. On the basis of a large number of published scientific studies, the diurnal variation in maximal short-term performance has been linked to variation in muscle contractile properties, rather than to variation in central nervous command (^4,16). The effect of intraday variation on muscle contractile properties could be attributed, in part, to the intracellular variation in the muscle, for example, the circadian rhythm in inorganic phosphate concentration (^9,12) and, in part, to the circadian rhythm in central temperature, which could influence calcium release by the sarcoplasmic reticulum (^12,16). In line with the findings in the literature, our results showed increases in oral temperature over the course of the day. Furthermore, the gain observed in our study is in accordance with the amplitude (peak-to-trough variation) found in other studies (^21,23). Thus, this result shows that the times of the day of the 2 test sessions (07:00 and 17:00 hours) approximate, respectively, the bathyphase (trough time) and the acrophase (peak time) of the circadian rhythm of the oral temperature for the selected subjects. However, other factors could be evoked to explain the diurnal variation in anaerobic performances (e.g., Wingate test). Indeed, the Wingate test is dependent on energy production derived from both aerobic and anaerobic metabolisms (²¹). Recent studies have reported that during cycling exercises, there is a higher aerobic participation in energy production in the evening than in the morning (²¹).

In line with the findings in the literature, the 2 groups showed improved performances after 8 weeks of lower-extremity strength training performed 3 times per week. It is well known, for example, that strength training can enhance jump performance and the anaerobic power developed on a cycle ergometer, and numerous studies have reported that training 2 or 3 times per week for 5 or 6 weeks is sufficient to improve maximal anaerobic performances (^6,22). Some factors could explain this increase. Mathematically, power is the product of force multiplied by velocity (Power = Force × Velocity). By increasing the force or the velocity, an improvement in power output should occur (²²). Previous studies have found a positive relationship between force and anaerobic power in a cycle ergometer (²⁰). Thus, the cycle ergometer test is a very good index of maximal strength in movement. Moreover, the SJ and the CMJ principally activate the series elastic component, with an active part (crossbridges of fiber types) and a passive part (tendons), which contributes greatly to performance and expresses the force and maximal power that can be reached during a very brief exercise (²²). It must be pointed out that the exercises performed during the 8 weeks in this study were not similar to those performed during the power tests. However, an increase in anaerobic power output in the Wingate test (²²) and in the SJ and CMJ tests (^11,17) has been observed after several weeks of regular strength training.

A major finding of this study was that, after 8 weeks of regular strength training, both the ETG and the CG showed significantly lower P_peak, P_mean, SJ, CMJ, and 1RM values in the morning than in the evening. However, the P_peak, P_mean, SJ, CMJ, and 1RM of the MTG measured at 07:00 and 17:00 hours did not differ. Thus, strength training at a specific time of the day increases performances specifically at this time and modifies the diurnal variations of anaerobic performances in the MTG. These findings are consistent with those of some previous research studies that reported significant temporal specificity after strength training (^17-19,22). In contrast to these data, Blonc et al. (⁶) did not report any significant effect of training at a specific hour on SJ and CMJ performances after 5 weeks of training (sprints, jumps, and other exercises). The authors suggested that the passive warm-up effect of the environment may contribute to the apparent no effect of time of the day on either performance or training benefit, which could account for the discrepancies between our results and the results of Blonc et al.

On the other hand, Souissi et al. (²²) suggested that, after 6 weeks of strength training, the participants in the MTG improved their performance in the morning and in the evening, although the absolute increase was larger in the morning and the participants in the ETG improved their performance only in the evening. They concluded that training twice a week at a specific time of the day increases the peak torque and the peak anaerobic power specifically at this hour. More recently, Sedliak et al. (^17-19) suggested that a significant diurnal difference in power during loaded SJs, maximum voluntary isometric strength during unilateral knee extensions, and half-squat 1RM between the 07:00- and 17:00-hour tests decreased, after 10 weeks of regular strength training, in the MTG but not in the ETG or in the CG. The present results are in line with those of Sedliak et al. (¹⁸) and Souissi et al. (²²). However, the adaptations to strength training of the MTG in this study were comparable with those of Souissi et al. and somewhat more pronounced than those of Sedliak et al. In our opinion, it is possible that the preparatory period in the study of Sedliak et al. (¹⁸) before the time-of-day–specific training may discount neural adaptations. Therefore, as the training proceeds with a larger contribution of peripheral factors, the time-specific training effects in the study of Sedliak et al. (¹⁸) might be less pronounced.

The results of this study suggested that adaptations to strength training are greater at the time of the day at which training was performed than at other times. However, it is difficult to explain why greater increases seem to occur at the hour during which strength training is regularly performed, because the training effect represented by the increase in P_peak, P_mean, SJ, CMJ, and 1RM is the sum of many different physiological processes.

Diurnal variations in response to exercise have been shown by body temperature because it relates to metabolism and by-products of the neurohormonal system, such as cortisol (⁷) and catecholamine (²⁴). Bird and Tarpenning (⁵) suggested that plasma concentrations of testosterone and cortisol were higher after evening strength training than after morning strength training. Sedliak et al. (¹⁹) observed that the MTG showed a significant decrease in the serum cortisol concentrations, after 10 weeks of strength training, in the morning, whereas both ETG and CG showed persisting higher morning serum cortisol concentrations after training. Recently, Sedliak et al. (¹⁸) failed to show any adaptation to strength training at a specific time of the day, in electromyography of the knee extensors in men during unilateral isometric knee extension peak torque. They concluded that peripheral rather than neural adaptations are the main source of temporal specificity in strength training. More recently, Sedliak et al. (¹⁷) observed that the magnitude of muscular hypertrophy (quadriceps femoris cross-sectional areas and volume) did not differ between morning and evening training.

Among the factors frequently presented in the literature to explain the diurnal variation in muscular power, some authors have postulated the hypothesis of a causal link between core temperature and muscular power fluctuation (^4,13). In agreement, our results showed a significant diurnal increase in the oral temperature at rest before and after training in the MTG and ETG. However, the results of this study showed that, in the MTG, diurnal fluctuations in anaerobic performances disappeared after the training program. These results suggested that the diurnal fluctuation in oral temperature is not the only explanation of the time-of-day effects on anaerobic performances. This observation was in accordance with data of Martin et al. (¹²), who observed that the diurnal fluctuations in muscle strength persisted despite an artificial heating of the adductor pollicis muscle by 5°C in the morning.

Practical Applications

In summary, our results suggested that adaptation to strength training is greater at the time of the day during which training was performed than at other times. Thus, strength and power athletes required to compete at a certain time of the day (e.g., morning qualifications) may be advised to coincide physical preparation with the time of the day at which one's critical performance is programmed. This is an important because it implies that all training sessions cannot be programmed throughout the day without taking into consideration the time of the day of the competitive event. Further research is needed to confirm the present results and to reveal the possible central or peripheral origin, or both, of the diurnal fluctuation in strength training effects.

Acknowledgments

The authors wish to express their sincere gratitude to all the participants in this study for their maximal effort and cooperation.