Islet cell antibodies (ICAs) were determined in a large cohort of white nondiabetic schoolchildren (n = 4287) from a homogenous population in southern Germany. The prevalence of ICA levels ≥5 Juvenile Diabetes Foundation (JDF) U was 1.05% (95% confidence interval 0.8–1.4%). Analysis of HLA-DRβ and -DQβ alleles revealed that the specificities found to be increased in insulin-dependent (type I) diabetic subjects with the same ethnic background were also associated with ICA positivity in the nondiabetic schoolchildren. HLA-DR3 (P < 0.01) and -DR4 (P < 0.01) phenotypes and absence of Asp residue (P < 0.01) at codon 57 of the HLA-DQ β-chain were significantly increased in ICA+ compared with control subjects. High levels of ICAs, which were categorized as either ≥17 or ≥30 JDF U, were found to be associated with amino acids other than Asp at position 57 of the HLA-DQ β-chain. No association of ICA level was found for HLA-DR phenotypes.

Under the influence of high estrogen levels, the suppression of total serum cortisol in the dexamethasone test has often been found to be incomplete. Its measurement for the purpose of excluding Cushing's disease or adrenal tumors in women taking oral contraceptives is, therefore, considered unreliable. This study was designed to compare the reliability of measurements of total cortisol, unbound cortisol and ACTH suppression during chronic hyperestrogenaemia. An overnight suppression test with 2 mg of dexamethasone was performed in 19 women receiving long-term contraceptive treatment (group A) and in 12 controls (group C). Baseline and post-dexamethasone morning levels of ACTH, total serum cortisol and unbound salivary cortisol were determined by RIA. In addition, unbound serum cortisol was measured by equilibrium dialysis. Mean baseline levels of all four parameters were significantly higher in group A. This result points towards the possibility of a direct stimulatory effect of estrogens upon the corticotroph axis which is independent from CBG-mediated increase of total cortisol. Post-dexamethasone values of total and unbound cortisol showed no statistically significant differences, while ACTH suppression in group A was even slightly better than in group C. From these data it is concluded that there is no need for post-dexa routine measurement of ACTH or unbound cortisol under contraceptive treatment since neither one of these parameters provides any additional information in comparison to total cortisol.

No abstract