Post-dexamethasone cortisol as a predictor for the efficacy of electroconvulsive therapy in depressed inpatients Miljana Vukadin a , Tom K. Birkenhäger a, * , André I. Wierdsma a , Theo H.N. Groenland b , Walter W. van den Broek a a Department of Psychiatry, Erasmus Medical Center, PO Box 2040, 3000CA Rotterdam, The Netherlands b Department of Anaesthesiology, Erasmus Medical Center, Rotterdam, The Netherlands article info Article history: Received 13 December 2010 Received in revised form 23 February 2011 Accepted 20 March 2011 Keywords: Major depression Electroconvulsive therapy Dexamethasone suppression test Prediction abstract Background: Although several variables have been studied as a possible predictor for the efficacy of ECT, results regarding hypercortisolism have been inconsistent. This prospective study evaluates the relation between pre-treatment cortisol levels and the efficacy of ECT in a population of drug-free inpatients with severe major depression. Methods: At the inpatient depression unit, 18 patients meeting the DSM-IV criteria for depressive disorder, and with scores of at least 18 on the 17-item Hamilton Rating Scale for Depression (HAM-D), were treated with bilateral ECT twice weekly. The HAM-D evaluated depression severity and was per- formed within 3 days prior to ECT, weekly during the course of ECT, and within 3 days after the last treatment. The outcome criterion was defined a priori as the change on the HAM-D score. Salivary cortisol was assessed within 3 days prior to ECT at two time points, followed by 0.5 mg dexamethasone ingestion. The following day, salivary cortisol was again assessed at two time points. The generalized linear model was used to assess the relation between salivary cortisol levels and reduction in HAM-D score as continuous variables. Results: Higher levels of salivary cortisol at 9 AM after 0.5 mg dexamethasone ingestion are associated with a greater reduction in HAM-D score (B ¼0.279, 95% CI: 0.557 to 0.01, s.e. ¼ 0.13, p ¼ 0.049; R square ¼ 0.23; adjusted R square ¼ 0.13). Conclusion: This study suggests that higher levels of post-dexamethasone salivary cortisol at 9 AM are predictive of ECT efficacy. Ó 2011 Elsevier Ltd. All rights reserved. 1. Introduction Electroconvulsive therapy (ECT) is considered to be the most effective treatment for patients with major depression, with a reported response rate of 60e90% (UK ECT ReviewGroup, 2003). Several clinical variables have been studied as possible predictors for the efficacy of ECT. Although identifying predictors for response has proven difficult, the most convincing clinical predictors are the presence of delusions (Petrides et al., 2001; Birkenhäger et al., 2003), psychomotor retardation (Hickie et al., 1996) and pharma- cotherapy failure prior to ECT (Heijnen et al., 2010). However, data on potential biological predictors for the efficacy of ECT are scarce. Hyperactivation of the hypothalamicepituitarye adrenal (HPA) axis, as indicated by hypercortisolism and non- suppression of cortisol on the dexamethasone suppression test (DST), has been reported in many studies of depression. It occurs most frequently among inpatients with more severe psychotic or melancholic depression (Nelson and Davis, 1997; Fink, 2005). Hyperactivation of the HPA-axis could be a predictor for ECT response, although in a meta-analysis no clear predictive value of the DST for response to either antidepressants or ECT was found (Ribeiro et al., 1993) This apparent lack of predictive value of the DST might be due to studies including heterogeneous populations of depressed patients and, with regard to ECT, the use of low-to-moderate dose unilateral ECT, or administering ECT with benzodiazepines as concomitant medication, both of which may result in limited antidepressant efficacy. Less effective forms of ECT may result in a relatively large proportion of ‘placebo’ responders and a relatively small proportion of ‘true’ responders, leading to a relatively high efficacy in less severely depressed patients without hypercortisolism. * Corresponding author. Tel.: þ31 10 7040139; fax: þ31 10 4633217. E-mail address: t.birkenhager@erasmusmc.nl (T.K. Birkenhäger). Contents lists available at ScienceDirect Journal of Psychiatric Research journal homepage: www.elsevier.com/locate/psychires 0022-3956/$ e see front matter Ó 2011 Elsevier Ltd. All rights reserved. doi:10.1016/j.jpsychires.2011.03.012 Journal of Psychiatric Research 45 (2011) 1165e1169