Differential effects of venlafaxine in the treatment of major depressive disorder according to baseline severity.

In this meta-analysis, we compare the relative efficacy of venlafaxine to selective serotonin reuptake inhibitors (SSRIs) in patients with major depressive disorder classified according to baseline disease severity. Data from 31 double-blind randomised clinical trials comparing venlafaxine and SSRIs (intent-to-treat n = 6,492) were pooled. For this secondary analysis, patients were stratified into groups based on baseline HAM-D(17) total score (>or=30, <30, >or=25, and <25). Remission rates (HAM-D(17) < 8) were analyzed for each subgroup using Fisher's exact test to compare treatment effects between venlafaxine and SSRIs; last observation carried forward (LOCF) and observed cases (OC) data were analyzed. The number needed to treat (NNT) to benefit was determined for each analysis. Statistically significant remission rate differences, favoring venlafaxine, were seen in LOCF and OC analyses for each subgroup. In patients with baseline HAM-D(17) < 25 (n = 3,928) the differences were (LOCF) 7.3 [P < 0.001; NNT = 14] and (OC) 6.2 [P = 0.003; NNT = 16], and in patients with baseline HAM-D(17) >or= 25 (n = 2,564) were (LOCF) 5.7 [P = 0.002; NNT = 17] and (OC) 6.7 [P = 0.009; NNT = 15]. In patients with baseline HAM-D(17) < 30 (n = 5,836) the differences were (LOCF) 6.4 [P < 0.001; NNT = 16] and (OC) 5.5 [P = 0.001; NNT = 18], and in patients with baseline HAM-D(17) >or= 30 (n = 656) were (LOCF) 8.9 [P = 0.015; NNT = 11] and (OC) 14.8 [P = 0.003; NNT = 7]. In conclusion, these analyses demonstrate that venlafaxine may be superior to SSRIs in achieving remission in both mild/moderate and severely depressed patients. The greater difference in remission rates among patients with baseline HAM-D(17) >or= 30 suggests a more pronounced clinical benefit that may be achieved with venlafaxine in severely depressed patients.