A 52-week, double-blind evaluation of the metabolic effects of aripiprazole and lithium in bipolar I disorder

Abstract

Introduction: Metabolic risk factors, termed metabolic syndrome, which include obesity, diabetes, dyslipidemia, and hypertension, are more common in patients with bipolar disorder than in the general population. Moreover, medications used to treat bipolar disorder carry some risk of worsening metabolic parameters.

Method: The study was conducted at 46 study centers in the United States, although only 31 study centers enrolled patients in the 40-week extension phase. Patients with acute bipolar I mania, manic or mixed (DSM-IV-TR criteria; Young Mania Rating Scale score ≥ 20), who required hospitalization were randomly assigned to double-blind aripiprazole (15-30 mg/d), lithium (900-1500 mg/d), or placebo for 3 weeks. Patients treated with aripiprazole or lithium continued treatment to week 12, after which they could enter a double-blind 40-week extension phase. Patients were enrolled in the 12-week acute treatment phase between April 2004 and July 2006; the first patient entered extension treatment in October 2004, and the last patient completed treatment in May 2007. Changes in metabolic parameters were compared between patients treated with aripiprazole or lithium for up to 52 weeks using last observation carried forward and analysis of covariance. Analysis stratified by baseline body mass index (BMI) was also conducted.

Results: Modest increases in body weight were observed in both groups: +0.97 kg (2.1 lb) for aripiprazole (n = 127) and + 0.74 (1.6 lb) for lithium (n = 136), P = .60. A significant difference in body weight increase was observed only among patients with a BMI < 25: + 2.66 kg (5.9 lb) for aripiprazole (n = 35) and + 0.40 kg (0.9 lb) for lithium (n = 37), P = .02. Mean changes from baseline to week 52 in fasting levels of total cholesterol, high-density lipoprotein cholesterol, low-density lipoprotein cholesterol, plasma glucose, triglycerides, or insulin (last observation carried forward) were small in both aripiprazole and lithium treatment groups; no significant differences were observed. Mean laboratory values were within the normal or borderline range for both treatment groups across all BMI categories.

Conclusion: Comparably modest and similar changes in metabolic parameters were observed in patients with bipolar disorder treated for up to 1 year with either lithium or aripiprazole.

Trial registration: clinicaltrials.gov Identifier: NCT00095511.

Figures

Figure 1.

Mean (SE) Change in Weight From Baseline to End of Week 52 for Total Safety Sample and Stratified by Baseline BMI (last observation carried forward) aNormal/underweight = BMI < 25 kg/m2. bOverweight = BMI ≥ 25 kg/m2 and < 30 kg/m2. cObese = BMI ≥ 30 kg/m2. *Aripiprazole vs lithium: P = .02. Abbreviations: BMI = body mass index, SE = standard error.

Figure 2.

Mean (SE) Change in Fasting Metabolic Measures From Baseline to Week 52 for the Total Safety Sample aLithium: n = 87, aripiprazole: n = 80. bLithium: n = 7, aripiprazole: n = 5. Abbreviations: HDL-C = high-density lipoprotein cholesterol, LDL-C = low-density lipoprotein cholesterol, SE = standard error.

Figure 2.

Mean (SE) Change in Fasting Metabolic Measures From Baseline to Week 52 for the Total Safety Sample aLithium: n = 87, aripiprazole: n = 80. bLithium: n = 7, aripiprazole: n = 5. Abbreviations: HDL-C = high-density lipoprotein cholesterol, LDL-C = low-density lipoprotein cholesterol, SE = standard error.