Symptoms predicting remission after divalproex augmentation with olanzapine in partially nonresponsive patients experiencing mixed bipolar I episode: a post-hoc analysis of a randomized controlled study

Abstract

Background: Rating scale items in a 6-week clinical trial of olanzapine versus placebo augmentation in patients with mixed bipolar disorder partially nonresponsive to ≥14 days of divalproex monotherapy were analyzed to characterize symptom patterns that could predict remission. At baseline, the two treatment groups were similar.

Findings: Factor analysis with Varimax rotation was performed post hoc on baseline items of the 21-Item Hamilton Depression Rating Scale (HDRS-21) and Young Mania Rating Scale (YMRS). Backwards-elimination logistic regression ascertained factors predictive of protocol-defined endpoint remission (HDRS-21 score ≤ 8 and YMRS score ≤ 12) with subsequent determination of optimally predictive factor score cutoffs.Factors for Psychomotor activity (YMRS items for elevated mood, increased motor activity, and increased speech and HDRS-21 agitation item) and Guilt/Suicidality (HDRS-21 items for guilt and suicidality) significantly predicted endpoint remission in the divalproex+olanzapine group. No factor predicted remission in the divalproex+placebo group. Patients in the divalproex+olanzapine group with high pre-augmentation psychomotor activity (scores ≥10) were more likely to remit compared to those with lower psychomotor activity (odds ratio [OR] = 3.09, 95% confidence interval [CI] = 1.22-7.79), and patients with marginally high Guilt/Suicidality (scores ≥2) were less likely to remit than those with lower scores (OR = 0.37, 95% CI = 0.13-1.03). Remission rates for divalproex+placebo vs. divalproex+olanzapine patients with high psychomotor activity scores were 22% vs. 45% (p = 0.08) and 33% vs. 48% (p = 0.29) for patients with low Guilt/Suicidality scores.

Conclusions: Patients who were partially nonresponsive to divalproex treatment with remaining high vs. low psychomotor activity levels or minimal vs. greater guilt/suicidality symptoms were more likely to remit with olanzapine augmentation.

Trial registration: ClinicalTrials.gov; http://clinicaltrials.gov/ct2/show/NCT00402324?term=NCT00402324&rank=1, Identifier: NCT00402324.

Figure 1

Total HDRS-21 (A and C) and YMRS (B and D) scores in sub-categorized patients. Patients were sub-categorized by cut-off scores for baseline Psychomotor activity (Factor 2) and Guilt/suicidality (Factor 8) during olanzapine augmentation of divalproex-partially nonresponsive patients. HDRS-21: 21-Item Hamilton Depression Rating Scale; YMRS: Young Mania Rating Scale.