Predictors of Relapse During Fluoxetine Continuation or Maintenance Treatment of Major Depression

Background: The goal was to examinepredictors of relapse during continuation/maintenance treatmentof major depression that had remitted following 12 to 14 weeks offluoxetine therapy.

Method: The study utilizes data collectedin a collaborative clinical trial including patients withDSM-III-R major depression at 5 university-affiliated outpatientpsychiatry clinics. Three hundred ninety-five patients whoremitted with fluoxetine therapy were randomly assigned to 1 of 4treatments: fluoxetine for 14 weeks followed by placebo for 36weeks, fluoxetine for 38 weeks followed by placebo for 12 weeks,fluoxetine for 50 weeks, or placebo for 50 weeks. Coxproportional hazard models were used to identify predictors oftime to relapse.

Results: In addition to the previouslyreported longitudinal pattern of response during acute treatment,neurovegetative symptom pattern was a predictor of fluoxetinebenefit compared with placebo. Greater chronicity predictedpoorer survival, which was not differential by treatment. Themost robust advantage of fluoxetine was seen for patients withendogenous vegetative symptoms, chronic depression, and acutetreatment response characterized by onset in the third week orlater and persistence of response once attained.

Conclusion: Both nonspecific pattern ofresponse and neurovegetative symptoms characteristic of atypicaldepression were predictive of lack of fluoxetine efficacy incontinuation/maintenance treatment. These findings haveimportance for both clinical management and analyses of futuremaintenance trials.