Severe and anxious depression: Combining definitions of clinical sub-types to identify patients differentially responsive to selective serotonin reuptake inhibitors

Abstract

Patients with severe major depressive disorder are more likely than those with mild/moderate depression to experience the relative benefits of antidepressant therapy versus placebo. Several studies have, unexpectedly, failed to show a similar antidepressant–placebo discrepancy between patients with versus without anxious depression, although patients with anxious depression are more likely to meet criteria for severe depression than those without. The aim of this study was to confirm the absence of treatment moderating effects for anxious depression in a general clinical trial population, and to examine for the presence of treatment moderating effects in severe depression. Patient-level outcome data from all randomized, double-blind, placebo-controlled trials involving the use of the selective serotonin reuptake inhibitor escitalopram for adults with major depressive disorder sponsored by H. Lundbeck A/S or Forest Laboratories were pooled. Studies focusing on patients with a specific axis-I or -III co-morbidity were excluded. Data from five trials were pooled. Anxious depression was not found to serve as a treatment moderator for selective serotonin reuptake inhibitor therapy versus placebo. However, when patients with severe depression were analyzed separately, anxious depression significantly influenced the relative degree of symptom reduction with selective serotonin reuptake inhibitors versus placebo (p=0.0094). In fact, the numbers needed to treat for remission for these two sub-types were the largest and smallest reported to date from analyses of large datasets of antidepressants (22 for severe anxious versus 4 for severe non-anxious depression). Subdividing patients with severe major depressive disorder into those with versus without anxious depression results in the characterization of sub-types that are particularly “responsive” (severe non-anxious) and “unresponsive” (severe anxious) to selective serotonin reuptake inhibitor therapy (relative to placebo). These findings are preliminary, of yet undetermined clinical relevance, and warrant replication and further exploration.