The aim of this study is to evaluate the muscle strength and basal serum testosterone and urea levels in soccer athletes. Twenty-four soccer players in pre-competitive period had a blood sample collected to have testosterone and urea concentrations analyzed. Subsequently, 1RM tests were applied to the bench press and squat exercises. After data collection, the athletes were randomly divided in two groups submitted to: non-linear periodization program (G1) and non-periodized program (G2), both for 12 weeks. ANOVA for repeated measures showed increase in serum testosterone concentration in G1 (∆ = 3.70 ng/dl; p = 0.0001) and in G2 (∆ = 1.81 ng/dl; p = 0.035) and reduction in urea levels only in G1 (∆ = -3.08mg%; p = 0.0001). G1 presented higher levels of testosterone (∆ = 2.13 ng/dl; p = 0.009) and lower levels of urea (∆ = -1.36mg%; p = 0.026) in the post-test when compared to G2. 1RM tests did not show significant differences. The non-linear training in soccer players was more effective than the non-periodized training in promoting increase in serum testosterone levels and reduction in urea levels.

SummaryStudy aim: To evaluate the effects of two strength training programmes on the explosive force of lower extremities and total testosterone levels in plasma of Brazilian football players. Material and methods: A group of 24 male football players aged 16 -19 years participated in a 12-week training: non-linear periodised (NLP) or non-periodised (NP), 12 subjects each. Before and after training period, all subjects performed countermovement vertical jump (CMJ), horizontal jump and Flegner's power test, and total testosterone concentrations were determined in plasma. Results: Significant, training-induced changes in CMJ were found in both groups; by 7.4% (p<0.05) in the NLP group and by 8% (p<0.001) in NP group. Also in the long jump (SBJ) significant (p<0.05) changes were noted in both groups, by 0.9 and 4.4%, respectively. Serum testosterone concentrations significantly (p<0.05 -0.01) increased post-training in both groups and the increase was significantly (p<0.01) greater in the NLP than in NP group. Conclusion: A larger number of athletes in future studies, as well as other sport categories and levels and a longer training period would be advisable in order to strengthen the causal inferences between the variables studied.