The purpose of this study was to delineate the possible endocrine effects of exercise-induced GH secretion. Twelve healthy adult males were studied during short (20 min) and subsequent-prolonged (2 h) physical exercise and recovery period (2 h), both after injection of a long acting somatostatin analog [Sandostatin (ST); 0.1 or 0.05 mg, sc] and after a control saline injection. Additional subjects were studied during rest with similar injections of ST (0.1 mg) and saline (n = 7) or using a lower ST dose (0.01 mg; n = 6). Several venous blood samples were taken during the trials and analyzed for selected hormones, monitoring pituitary, testicular, and adrenal functions. ST injection blocked the serum GH response to short term maximal bicycle ergometer exercise, but not to the following prolonged bicycle exercise. No relationship of the exercise-associated GH increase to the concomitant endocrine responses of the adrenals and testes was observed. Unexpectedly, the higher ST doses (0.1 and 0.05 mg) increased the mean levels of serum testosterone by 18-25% in both exercise (P = 0.0017) and rest trials (P < 0.0001), respectively. ST did not affect the levels of LH, FSH, or cortisol. ST slightly increased serum sex hormone-binding globulin (3%; P = 0.021) and albumin (4%; P = 0.017) concentrations, but not that of free testosterone. Because the testosterone response to somatostatin was fast and without a simultaneous increase in LH, it was consistent with a direct testicular response. The explanation for this novel ST effect remains obscure, but it may be due to modulation of some paracrine mechanisms inhibiting testicular steroidogenesis.