Fall in plasma testosterone levels in normal male subjects in response to an oral glucose load: comparison of morning and afternoon responses

Standard 50 g oral glucose tolerance tests were carried out on two occasions, one week apart, on six normal male subjects. The tests were performed in randomised order, after a nine hour fast, and were begun at 0900 and 1400 hr respectively. Blood samples were obtained fasting and at 30, 60, 90, 120 and 150 minutes after the glucose load. Estimations included blood glucose and plasma insulin, growth hormone (GH), free fatty acids (FFA) and testosterone. The latter was measured using a radio immunoassay. Fasting testosterone levels were within the normal range on all occasions, although, as expected, mean levels were lower in the afternoon (555 ± 36; 463 ± 64 (S.E.M.) respectively). In ten of twelve tests the level of plasma testosterone fell after the glucose load, rising to near fasting levels in nine of these by the end of the test. This pattern was observed in both morning and afternoon tests in four subjects and in the afternoon tests, in two subjects. The lowest level occurred at 60 or 90 minutes after the glucose load in all but one subject, and tended to occur later in the morning test. The observed fall in plasma testosterone is considered unlikely to be due to stress, for fasting growth hormone levels were low and both growth hormone and free fatty acid levels suppressed in every test. The probable explanation is a suppression of LH release comparable to that of GH, by glucose or by some hormonal/metabolic event triggered by glucose ingestion. Measurements of corresponding plasma LH levels are in progress to test this hypothesis.