Abrupt discontinuation compared with a 1-week taper regimen in depressed outpatients treated for 24 weeks with...

The objective of this study was to determine whether the occurrence of discontinuation symptoms was equivalent for abrupt discontinuation versus 1-week taper to desvenlafaxine 25 mg/d after a 24-week treatment with desvenlafaxine 50 mg/d (administered as desvenlafaxine succinate) for major depressive disorder. Adult outpatients with major depressive disorder who completed the 24 weeks of open-label treatment with desvenlafaxine 50 mg/d were randomly assigned to no discontinuation (desvenlafaxine 50 mg/d), taper (desvenlafaxine 25 mg/d), or abrupt discontinuation (placebo) groups for the double-blind (DB) taper phase. The primary end point was Discontinuation-Emergent Signs and Symptoms (DESS) scale total score during the first 2 weeks of the DB phase. The null hypothesis that the absolute difference of greater than 2.5 in DESS scores between taper and abrupt discontinuation groups was tested by calculating the 95% 2-sided confidence interval on the mean difference between the 2 groups. Of the 480 patients enrolled in the open-label phase, 357 (≥1 postrandomization DESS record) were included in the primary analysis. Adjusted mean ± SE DESS scores were 4.1 ± 0.72 for no discontinuation (n = 72), 4.8 ± 0.54 for taper (n = 139), and 5.3 ± 0.52 for abrupt discontinuation (n = 146) groups. The difference in adjusted mean DESS total scores between the abrupt discontinuation and taper groups was 0.50 (95% confidence interval, -0.88 to 1.89) within the prespecified margin (±2.5) for equivalence. The number of patients who discontinued because of adverse events or discontinuation symptoms during the DB period was similar between the taper (2.8%) and abrupt discontinuation (2.1%) groups. These findings indicate that an abrupt discontinuation of desvenlafaxine 50 mg/d produces statistically equivalent DESS scores compared with the 1-week taper using 25 mg/d.