[Dexamethasone suppression test using saliva cortisol measurement in affective disorders].

In order to perform the dexamethasone suppression test (DST) utilizing saliva as an alternative to serum, we developed a direct assay method for saliva cortisol by modification of a commercial assay kit for plasma cortisol. The correlation coefficient between saliva and serum cortisol levels was 0.94. Using this method, DST was performed in 124 patients consecutively admitted to one psychiatric ward. One mg of dexamethasone was given at midnight on the previous day, and saliva samples were collected every 4 h between 8 AM and midnight. When the maximum value was used for the discrimination of non-suppressors, the sensitivity, specificity, and diagnostic confidence levels for melancholia vs non-melancholia were 33.3%, 91.3%, and 77.8% at the criterion value of 0.3 microgram/dl, respectively. While the cortisol secretion rate in saliva was calculated from 5 measured values, the sensitivity, specificity, and diagnostic confidence for melancholia were respectively enhanced to 66.7%, 91.3%, and 87.5% at the criterion value of 0.06 microgram/dl/h, and 38.1%, 95.9%, and 88.9% at the criterion value of 0.12 microgram/dl/h. Even when compared to all patients with other psychiatric diagnoses, these indices maintained the rather satisfactory levels of 38.1%, 90.3%, and 44.4%, respectively. These figures are comparable to or even higher than those obtained by DST using serum cortisol, indicating that the measurement of saliva cortisol is sufficiently applicable to clinical situations.