Studies on animals and humans have suggested that dehydroepiandrosterone sulphate (DHEAS) has antiatherogenic effects. It has been hypothesized that insulin may have an atherogenic role and it has been reported recently that, surprisingly, DHEAS levels decreased in normal men and women during the hyperinsulinemic-euglycemic technique. Since a hyperinsulinemia frequently occurs during insulin therapy in patients with insulin dependent diabetes mellitus (IDDM), the present work was undertaken to determine whether DHEAS serum concentrations were decreased in IDDM patients as compared to controls and if so, to discover the possible causes. To this, purpose, out of 805 outpatients afferent to our Diabetes Centre from 1989 to 1992, three groups were selected on the basis of the criteria described below. Known interferences with the DHEAS serum concentrations such as gender (all males), age (aged 20-40 years) and Body Mass Index (BMI < 30) were excluded. Group A (cross-sectional study) was made up of 15 IDDM patients on insulin treatment with good metabolic control (HbA1C < 8%); group B (control study) was made of 18 healthy subjects (these patients were selected also on the basis of their normal oral glucose tolerance test) and group C (longitudinal study) was made up of 7 IDDM patients who had been examined previously and who were on insulin treatment.
Methods: In all three groups serum concentrations of DHEAS, 17 OH progesterone (17 OHP), delta 4 androstenedione (A4) and cortisol (F) were measured. In 10 patients from group A and in 9 patients from group B the ACTH test (9.25 mg IM Synacthen) was administered and the same hormonal pattern was measured after 60 min. In group C the same hormonal evaluation was performed 5 +/- 2.8 months after commencement of insulin therapy.
Results: DHEAS serum concentrations were significantly decreased in group A (median 2.9; range 1.1-5.2 mumol/l) with respect to group B (median 5.7; range 3.0-9.5 mumol/l) (p < 0.0012). However, the serum concentrations of 17 OHP (median 3.9 nm/l; range 2.9-6.9 nm/l and A4 (median 5.2 nm/l; range 1.8-10.2 nm/l) were also significantly reduced, while cortisol levels and the 17 OHP/A4 ratio were comparable to group B. After administration of ACTH, the delta increment in cortisol percentage showed a frank increase (55.1%) in group A with respect to group B (33.1%) (p < 0.01). The rise in DHEAS showed a lower increase in group A (10.2%) with respect to group B (65.5%) even though not statistically significant, while the other hormones showed an overlap between the two groups. In group C the serum concentrations of hormones before insulin therapy did not show any statistical differences with respect to the values in group B. A second evaluation, which was performed during insulin therapy, showed that only the 17 OHP/A4 ratio tended towards higher values with respect to pretherapy values (1.1 and 0.6 respectively; p = 0.07). In conclusion our data confirm low DHEAS levels during chronic insulin administration therapy. The underlying mechanism could be a general aspecific reduction in the activity of P 450 C 21 SCC enzymes in contrast with the specific inhibition of 17.20-lyase obtained during insulin bolus. Whether the low serum concentrations of DHEAS can determine an atherogenic effect of insulin needs further investigation, but the hormone could constitute a new parameter for the follow-up of patients affected by diabetes mellitus.