Pharmacopsychiatry 2009; 42 - A59
DOI: 10.1055/s-0029-1240131

Clinical characteristics and treatment outcome in a representative sample of depressed inpatients – Findings from the Munich Antidepressant Response Signature project

JM Hennings 1, T Owashi 2, EB Binder 1, S Horstmann 1, A Menke 1, S Kloiber 1, T Messer 3, T Pollmächer 4, T Nickel 1, A Sonntag 1, M Uhr 1, M Ising 1, F Holsboer 1, S Lucae 1
  • 1Max Planck Institute of Psychiatry, Munich, Germany
  • 2Department of Psychiatry, Showa University Fujigaoka Hospital, Yokohama, Japan
  • 3Bezirkskrankenhaus Augsburg, Augsburg, Germany
  • 4Klinikum Ingolstadt, Ingolstadt, Germany

We report the characteristics of a representative sample of depressed inpatients participating in the Munich Antidepressant Response Signature (MARS) project. 842 inpatients admitted to a psychiatric hospital for treatment of a major depressive episode, recurrent or bipolar depression were thoroughly characterized with respect to demographic factors, clinical history, and the degree of HPA-axis dysregulation evaluated by means of combined dex/CRH tests, and the predictive value of these factors for treatment outcome is investigated. 80.8% of patients responded to treatment and 57.9% reached remission at discharge after a mean treatment period of 11.8 weeks. Regression analysis identified early partial response (within 2 weeks) as the most important positive predictor for achieving remission. Previous ineffective treatment trials in the current episode and presence of a migration background are potent negative predictors for treatment outcome. In addition, remitters were characterized by a more pronounced normalization of an initially dysregulated HPA-axis. We could show that a large majority of inpatients suffering from depression benefits from antidepressant treatment during hospitalization. However, a considerable number of patients failed to achieve remission. We demonstrated that this subgroup can be characterized by a set of demographic, clinical and neuroendocrine variables allowing to predict unfavorable outcome at an early stage of treatment.