Is late-night salivary cortisol a better screening test for possible cortisol excess than standard screening tests in obese patients with Type 2 diabetes?

AIM: To compare the performance, in terms of specificity for cortisol excess, of late-night salivary cortisol with 24-hour urine-free cortisol (24hr UFC) and overnight 1mg dexamethasone suppression test (1mg DST) in a group of obese T2DM patients.
METHODS: Forty obese patients with T2DM without clinical features of Cushing's syndrome were recruited. Plasma, urinary and salivary cortisol were measured directly by an enzyme-linked immunosorbent assay using monoclonal antibodies. The specificities of the three tests using various cutoffs were calculated and compared, employing the assumption that none of the patients had hypercortisolism.
RESULTS: The patients had a mean age and BMI of 56 years (range 31-75) and 37 kg/m² (31-56) respectively. All 40 provided late-night salivary cortisol samples. Thirty-eight patients completed all three tests. Two patients only completed two screening tests. The specificities of late-night salivary cortisol (cutoff 10 nmol/L), 24hr UFC (400 nmol) and 1mg DST (50 nmol/L) were 70% (95% CI 53-83%), 90% (76-97%) and 72% (55-85%) respectively. The specificity of late-night salivary cortisol was significantly less than 24 hr UFC (P=0.039) but not 1mg DST (P>0.99).
CONCLUSION: Late-night salivary cortisol has a poor specificity for cortisol excess in obese patients with T2DM with 24 hr UFC showing significantly better specificity in our population.