Baseline characteristics and early response at week 1 predict treatment outcome in adolescents with bipolar manic or mixed episode treated with olanzapine: Results from a 3-week, randomized, placebo-controlled trial

Le Xiao, Stephen J. Ganocy, Robert L. Findling, Kiki Chang, Melissa P. DelBello, John M. Kane, Mauricio Tohen, Yu Tao Xiang, Christoph U. Correll

Research output: Contribution to journalReview articlepeer-review

Abstract

Background: Early predictors of response and remission in pediatric mania are lacking, requiring further study. Methods: This was a post hoc analysis of a 3-week, randomized, placebo-controlled trial of olanzapine conducted between November 2002 and May 2005 in 161 adolescents aged 13–17 years who were diagnosed with a DSM-IV acute manic or mixed episode of bipolar I disorder. Data from the olanzapine arm were analyzed to investigate the predictive power of early response or early nonresponse (≥25% or < 25% reduction in Young Mania Rating Scale [YMRS] score, respectively) at week 1 for ultimate response or nonresponse (≥ 50% or < 50% reduction in YMRS score, respectively) and for remission (YMRS total score ≤ 12 [standard definition] or ≤ 8 [stringent definition]) at week 3. Correlates of early response and ultimate response were examined in multivariable regression models. Results: By week 1, 69.2% of olanzapine-treated adolescents (n = 104, 2.5–20.0 mg/d) achieved early response, and 49.0% reached ultimate response at week 3. Patients with early response and early nonresponse were similar regarding baseline variables except higher scores for sleep and thought content were found with early response (P< .05) and higher olanzapine doses with early nonresponse (P< .01). At week 3, early response was associated with significantly greater improvements in YMRS, Clinical Global Impressions–Severity of Illness scale (both P< .001), and Overt Aggression Scale scores (P= .024). Adverse events were similar in patients with early response and early nonresponse, except for higher AIMS scores for patients with early nonresponse (P= .036). Early response significantly predicted ultimate response (OR = 5.61, P< .001; sensitivity = 86.3, specificity = 47.2, positive predictive value = 61.1, negative predictive value = 78.1). Significantly more early response than early nonresponse patients achieved ultimate response (61.1% vs 21.9%, P< .001) and remission defined by YMRS score ≤ 12 (45.8% vs 12.5%, P< .001) and YMRS score ≤ 8 (33.3% vs 3.1%, P< .001). In multivariable analyses, among other variables, early response remained an independent correlate of ultimate response and remission. Conclusions: In acute pediatric manic or mixed episodes, early response to olanzapine at week 1 was strongly associated with ultimate response and remission at week 3, while absence of early response predicted the unlikely success of further treatment.

Original languageEnglish (US)
Pages (from-to)e1158-e116
JournalJournal of Clinical Psychiatry
Volume78
Issue number9
DOIs
StatePublished - Nov 1 2017

ASJC Scopus subject areas

  • Psychiatry and Mental health

Fingerprint Dive into the research topics of 'Baseline characteristics and early response at week 1 predict treatment outcome in adolescents with bipolar manic or mixed episode treated with olanzapine: Results from a 3-week, randomized, placebo-controlled trial'. Together they form a unique fingerprint.

Cite this