The dexamethasone suppression test: a study in a normal population.

One hundred healthy, non-depressed volunteers were given a standard dexamethasone suppression test (DST) to determine the appropriate criterion values of plasma cortisol to define suppression or nonsuppression. By radioimmunoassay (RIA) of cortisol, the criterion value for 5% nonsuppression was plasma cortisol greater than 187 nmol/l, and for suppression less than 153 nmol/l, with an indeterminate range between these values. Use of the widely accepted pre-determined criterion value of 138 nmol/l gave a significantly greater frequency of nonsuppression. Values of cortisol measured by two RIAs in a subset of 43 volunteers were not equivalent. With the experimentally determined criterion value, no significant differences between nonsuppressors and suppressors were found for any measured physical or psychological parameters. Women taking oral contraceptives had significantly higher plasma cortisol pre-dexamethasone and post-dexamethasone. Their exclusion did not alter the calculated criterion value for the remainder, but their separately estimated criterion value was significantly higher. Caution should be exercised when classifying the DST status of women on oral contraceptives, particularly when values are at the lower end of the nonsuppressor range. Determination of a separate normal range for them may be warranted.