The shorter CAG repeat of the AR gene is associated with more severe CAD, which suggests a role for the sensitivity to androgens in the increased frequency of CAD in males. In addition, a protective role of endogenous oestrogen, which is higher in the longAR subgroup, can contribute to the observed difference.

Acute stroke patients with laboratory findings compatible with pre-existing hypothyroidism on admission, appear to have better clinical presentation and outcome; we speculate that a reduced response to stress and previous TIA's, possibly related to endogenous 'preconditioning', may contribute to this phenomenon.

Reduced bone mass occurring with increased frequency in diabetes mellitus has been attributed to poor blood glucose control but the pathogenetic mechanisms remain unknown. To evaluate the role of calcium metabolism, 59 patients with diabetes and normal renal function (22 Type 1, 37 Type 2) were studied. In all patients plasma calcium (Ca), serum phosphate (PO4), serum parathyroid hormone (PTH), and 24-h urinary calcium (uCa) were determined under both poor and improved control (for at least 7 days) as ascertained by four blood glucose determinations daily. Improvement of blood glucose control (p = 0.001) was associated with reduction of uCa both in Type 1 (6.9 +/- 1 vs 4.9 +/- 0.9 mmol day-1, mean +/- SEM, p = 0.02) and in Type 2 patients (4.2 +/- 0.4 vs 3.2 +/- 0.4 mmol day-1, mean +/- SEM, p = 0.002). Considerably more Type 1 patients (10 out of 22) had PTH values below the detection limit (1.5 pmol l-1) during poor than during improved control (2 out of 22). Comparison between the two types of diabetes showed that in Type 1 under poor control, Ca and PTH were lower (p = 0.03), while uCa was higher (p = 0.003), and after improved control, only uCa continued to be higher (p = 0.035). These findings suggest that increased uCa excretion in association with 'functional hypoparathyroidism' (especially in Type 1 diabetes) is observed during poor blood glucose control, and may be one of the factors leading to reduced bone mass in diabetes mellitus.

On the basis of our data in a large group of clinically euadrenal subjects, we suggest that following LDDST cortisol concentrations should become undetectable with the currently used radioimmunoassays. In patients with adrenal incidentalomas, application of the LDDST confirmed the presence of incomplete suppression of cortisol in the majority of patients. We suggest that the LDDST is a sensitive index of autonomous cortisol production in patients with adrenal incidentalomas; following this test a grading of subtle glucocorticoid excess may be obtained but future studies correlating biochemical, clinical and epidemiological data are required, in order to develop widely agreed cut-off levels of clinically significant glucocorticoid excess in these patients.

We conclude that there is a higher prevalence of follicular carcinoma in our country probably due to a moderate degree of iodine deficiency still existing in Greece. Age and extent of disease at diagnosis were important prognostic factors affecting morbidity and mortality, whereas sex, tumour features and histological type were of minor importance. All these prognostic factors and their relative importance should be taken in consideration in the management of this disease.