Pre-clinical Cushing's syndrome: an unexpected frequent cause of poor glycaemic control in obese diabetic patients

OBJECTIVE Autonomous cortisol secretion without clinical stigmata of Cushing's syndrome (CS) has been recently recognized and termed pre-clinical or sub-clinical CS. The common assumption is that CS is an extremely rare cause of uncontrolled diabetes; however, the prevalence of this entity has not been studied. We assessed the prevalence of pre-clinical CS among obese patients with uncontrolled diabetes. PATIENTS AND DESIGN (1) In a retrospective analysis, the medical records of 63 patients with endogenous CS were reviewed. (2) In a cross-sectional study, 90 obese patients (BMI >25 kg/m²) followed in a University Hospital and the local Health Fund endocrine and diabetes clinics, with poorly controlled diabetes (glycosylated haemoglobin >9%), underwent an overnight 1 mg dexamethasone suppression. In patients with non-suppressible cortisol levels (>140 nmol/l), Liddle's 2 and 8 mg dexamethasone suppression tests and imaging studies were performed. MEASUREMENTS The prevalence of poorly controlled diabetes, the major presenting symptom of CS, was assessed in the retrospective analysis. The prevalence of ‘true’ CS and the false positive rate in the overnight dexamethasone suppression test were calculated. The endocrine evaluation of the patients with pre-clinical CS and the effects of surgical cure on glycaemic control are described. RESULTS In the retrospective analysis, 11 (17.5%) had diabetes and 2 (3.2%) lacked the classic physical characteristics of the syndrome. In the cross-sectional study, 4 patients failed to suppress plasma cortisol (<140 nmol/l). In one patient the diagnosis of CS was not confirmed by a standard Liddle’s test and was therefore considered false positive. In the other 3, the diagnosis of CS was confirmed (prevalence of 3.3%, 95% confidence interval 1–9%). In all other patients the overnight cortisol suppression test was normal (cortisol level 47.3 ± 2.5 nmol/l (mean ± SEM)). After surgical treatment of CS, glycaemic control was markedly improved in all 5 patients (2 from retrospective and 3 from cross-sectional studies). CONCLUSIONS The prevalence of pre-clinical Cushing's syndrome in obese patients with poorly controlled diabetes appears to be considerably higher than previously believed. The overnight dexamethasone suppression test proved to be a simple, sensitive and highly specific screening test for Cushing's syndrome despite the presence of obesity and hyperglycaemia.