Single low doses (0.25 mg) of dexamethasone were given at 11 p.m., 2, 5 and 8 a.m. on separate days to five normal subjects. The concentrations of cortisol in plasma on the next day were significantly decreased compared to results after placebo administration, and cortisol suppression was maximal after dexamethasone had been given at 8 a.m. Our findings suggests that the postulated phase-advance of circadian rhythms is not a major cause of cortisol non-suppression in depressives given dexamethasone.The utility of the dexamethasone suppression test (DST) in psychiatry has been the subject of numerous studies. Generally, 1 mg dexamethasone was given at 11 p.m. to patients and cortisol concentrations in plasma during the next day were compared to those measured in matched normal controls. Through such protocols, the known overproduction of cortisol during major depressive episodes (Doig et al. 1966;Conroy et al. 1968;Fullerton et al. 1968) has been confirmed in subgroups of patients.Circadian rhythms of cortisol sercretion might be abnormal, i.e. phase-advanced by a few hours, during major depressive episodes (Jarrett et al. 1983;Sachar et al. 1973), though other authors did not confirm this observation (Claustat et al. 1984). If there exists a phase-advance of cortisol rhythms in depressives, then one could conclude that administering dexamethasone at 11 p.m. to depressives would correspond to giving the drug between 1 and 3 a.m. to normal subjects. We therefore studied the influence on cortisol secretion of the time of dexamethasone administration in normal subjects, to establish whether chronobiological changes in the sensitivity of the hypothalamic-pituitaryadrenal axis (HHAA) might lead to a false positive nonsuppression.
Materials and methodsSeven normal subjects of 25-38 years of age participated. The study was conducted in May and June. A pilot study in two subjects revealed complete suppression after 1 mg dexamethasone (Millicortene R, Ciba) given at 2 or 5 a.m.Offprint requests to: P. Schulz (with the exception of the 8 a.m. samples drawn after the 5 a.m. administration). Partial suppression was obtained after 0.25 mg given at 11 p.m. or 3 a.m. to the same two subjects. Therefore, 0.25 mg dexamethasone was administered orally to the other five subjects, at 11 p.m., 2 a.m., 5 a.m. and 8 a.m., using a cross-over latin square design. An identical placebo was administered once at 11 p.m. Each drug or placebo administration was separated by at least 5 days. According to Yerevanian et al. (1985), weekly administration of dexamethasone to normal subjects does not lead to false positive non-suppression. Venous blood samples were drawn through an indwelling catheter (BectonDickinson), using no heparin, every 2 h from 8 a.m. to 8 p.m. after dexamethasone and plasma was stored at -20 ° C pending analysis. The concentration of cortisol in plasma was measured using a commercial kit (Bio-M6rieux) based on the competition for transcortin. There was an excellent concordance between cortisolaemia measured with this assay a...