Study of glucose tolerance in consecutive patients harbouring incidental adrenal tumours. Study Group of Incidental Adrenal Adenoma.

Abstract

Clinically silent cortisol hypersecretion has been frequently observed in recent series of adrenal 'incidentalomas'. A significant body of data indicates that a spectrum of cortisol excess exists. Up to 90% of patients with cortisol hypersecretion are glucose intolerant. The aim of this study was (1) to assess glucose tolerance in consecutive patients with 'nonfunctioning' adrenal adenomas, and (2) to study the influence of the status of the hypothalamo-pituitary-adrenal axis (HPAA) on carbohydrate tolerance. Sixty-four consecutive patients with nonfunctioning adrenal adenomas (non-hypersecretory) diagnosed between September 1995 and July 1996 in five hospitals were included in the study. The prevalence of glucose intolerance or diabetes mellitus was determined with an oral glucose tolerance test (oGTT) according to the recommendations of the National Diabetes Data Group. Twenty-three consecutive unselected patients diagnosed with 'nonfunctioning' adrenal adenoma were enrolled in a study to determine the influence of HPAA status on carbohydrate tolerance. In these patients, in addition to basal and post-dexamethasone serum and urinary free cortisol concentrations, assessment of peripheral sensitivity to insulin (estimated during the oGTT according to a previously validated method) and adrenal sensitivity to ACTH (calculated from the CRH test) were performed. Twenty-five patients were considered to have normal glucose tolerance (95% confidence interval (95% CI) for the mean of 2 h glucose, 5.5-6.5 mmol/l); 17 showed glucose intolerance (95% CI 8.5-9.3 mmol/l); and 22 were classified as having diabetes mellitus (95% CI 12.1-14.6 mmol/l), including six patients with previously known diabetes mellitus. These three groups were comparable in age, sex, body mass index, waist-hip ratio and concomitant diseases. Thus, the prevalence of disturbed glucose tolerance was 39/64 (61%), well above the prevalence of NIDDM described in a population of similar age in our area. Among the 23 patients included in the study of HPAA status, the size of the tumour correlated with serum cortisol after dexamethasone (DXM) (r = 0.52, P < 0.01) and 24-h urinary free cortisol (UFC) after DXM (r = 0.55, P < 0.01), and negatively with DHEAs (r = -0.42, P = 0.04). The area under the curve for ACTH after hCRH (AUCacth) correlated negatively with both UFC (r = -0.40, P = 0.04) and serum cortisol after dexamethasone (r = -0.47, P = 0.02). The degree of peripheral sensitivity to insulin (SI) positively correlated with adrenal sensitivity (r = 0.56, P = 0.005). A high prevalence (61%) of disturbed glucose tolerance was found among consecutive patients harbouring incidental 'nonfunctional' adrenal adenomas. Therefore, patients with incidental adrenal tumours should be tested for glucose tolerance. The positive correlation between insulin sensitivity and adrenal sensitivity to ACTH suggests that, in these patients, insulin resistance hampers ACTH action at the level of the adrenal.