Effect of glycemic control on the overnight dexamethasone suppression test in patients with diabetes mellitus.

Because many of the clinical features associated with Cushing's syndrome are frequently found in patients with diabetes mellitus, diabetic patients are often evaluated for Cushing's syndrome. The initial test for Cushing's syndrome is the 1 mg overnight dexamethasone suppression test (DST), but its value as a screening test in diabetic subjects, especially those with poor glycemic control, has been questioned. To address this issue, an overnight DST was administered to 100 subjects with diabetes. Only 7 patients failed to suppress their plasma cortisol to less than 140 nmol/L (5.0 micrograms/dL), achieving a specificity of 93%. There was no relation between acute glycemic control (as measured by the mean of 4 serum glucose values obtained before receiving dexamethasone) or chronic glycemic control (as measured by glycohemoglobin) and false positive responses to the 1 mg overnight DST. The mean of the measures of acute glycemic control of the 7 subjects who had false positive results, 14.4 +/- 2.8 mmol/L, was not significantly different than that of the 93 subjects with normal responses, 13.2 +/- 3.3 mmol/L. Similarly, the mean glycohemoglobin of the subjects with false positive results, 12.8 +/- 2.4%, was not significantly different than that of the subjects with normal responses, 12.9 +/- 2.5%. There was no correlation between plasma cortisol after dexamethasone and glycohemoglobin (r = 0.05), and only a weak correlation with the mean serum glucose (r = 0.21). We conclude that the 1 mg overnight DST is a valid screening test for Cushing's syndrome in patients with diabetes, regardless of glycemic control.