Panic disorder and response to sertraline: The effect of previous treatment with benzodiazepines

More than 50% of patients who seek psychiatric care for panic disorder have previously received prescriptions for a benzodiazepine (BZ). Research on t he treatment of generalized anxiety suggests that a history of BZ exposure might decrease the efficacy and tolerability of treatment with a serotonerg ic anxiolytic. This study examines the effect of prior BZ treatment on the efficacy and tolerability of sertraline treatment for panic disorder. Data were pooled (N = 705) from four double-blind, placebo-controlled studies of the efficacy of sertra-line for the treatment of panic. Two of the studies were 12-week fixed-dose studies with starting doses of 50 mg, whereas 2 me re flexible-dose studies of 10-week duration with starting doses of 25 mg. The effect of study treatment on the frequency of panic attacks, Clinical G lobal Impressions (CGI) Improvement Scale, and tolerability was examined fo r patients with or without prior BZ treatment. The efficacy of sertraline w as not affected by prior treatment with BZs. The mean endpoint reduction in panic attack frequency was identical in patients with or without prior BZ use: 79% vs. 80% (not significant). A history of good versus poor response to prior BZ treatment did not significantly influence CGI responder rates f or sertraline-treated patients (67% vs. 61%, respectively). Sertraline CGI responder rates were significantly greater than placebo response, which mas 47% for the good-response prior-BZ subgroup (p = 0.007), and 36% for the p oor-response BZ subgroup (p = 0.013). Placebo response was lower in patient s with any prior BZ use by 10%, on an intent-to-treat last-observation-carr ied-forward analysis (p = 0.106) and by 15% on a completer analysis (p = 0. 045). Prior BZ use did not influence either rates of adverse events or disc ontinuation rates within the first 3 weeks in patients treated with either sertraline or placebo. Sertraline is both well-tolerated and has significan t efficacy in patients with panic disorder, including the subset of patient s with panic disorder who have previously been treated with BZs.