Midnight serum cortisol as a marker of increased cardiovascular risk in patients with a clinically inapparent adrenal adenoma

Objective: There is scant information on the morbidity associated with subclinical Cushing ' s syndrome in patients with a clinically inapparent adrenal adenoma. In the present study, we have determined the prevalence of alterations of the hypothalamic -pituitary -adrenal axis in such patients and examined whether any correlation between endocrine data and the clinical phenotype exists. Design and methods: A multi-institutional retrospective study was carried out on 210 patients (135 women and 75 men aged 19-81 years) with an adrenal adenoma detected serendipitously between 1996 and 2000 in four referral centers in Italy. Results: Hypertension was observed in 53.8%, obesity in 21.4%, and hyperglycemia in 22.4% of patients. The 4 7 patients with midnight serum cortisol > 5.4 mu g/dl, a value corresponding to the 97th centile of 100 controls, were older and displayed greater fasting glucose (12 0.4 +/- 52.2 mg/dl vs 105.1 +/- 39.2mg/dI, P=0.04) and systolic blood pressure (148.3 +/- 14.6 mmHg vs 136.4 +/- 1-16.2 mmHg, P = 0.0009) than the 113 patients with normal cortisol levels. The difference in systolic blood pressure remained statistically significant (P = 0.009) when age was used as a covariate. The percentage of hypertensive patients undergoing treatment was not different between the two groups (90.5 and 97.19%) but the percentage of patients with controlled hypertension was significantly lower among the hypercortisolemic patients (12.5 vs 32.4%, P = 0.04). Glycated haemoglobin (HbA1c) levels were higher in the hypercortisolemic diabetic patients (8.9 +/- 1.1% vs 7.1 +/- 1.3%, P = 0.005). Conclusions: Elevated midnight cortisol concentration is a reliable test to select a subgroup of patients with a clinically inapparent adrenal adenoma with an adverse cardiovascular risk profile.