Metabolic Effects of 7 Antipsychotics on Patients With Schizophrenia: A Short-Term, Randomized, Open-Label, Multicenter, Pharmacologic Trial

Objective: To compare longitudinal metabolic effects of 7 antipsychotics, including body mass index (BMI), waist circumference (WC), blood pressure (BP), glucose, triglycerides (TG), high-density lipoprotein cholesterol (HDL-C), and low-density lipoprotein cholesterol (LDL-C); to investigate risk factors for metabolic syndrome (MetS); and to make recommendations on frequency and timing of monitoring metabolic measurements. Methods: This randomized, open-label, pharmacologic trial was conducted among patients with schizophrenia (DSM-IV) in 32 hospitals across China. Patients were randomly assigned to 7 groups and assessed at baseline, 2, 4, and 6 weeks. Linear mixed-effect models were used to assess changes of metabolic measures over time. Multivariable logistic regression analysis was performed to investigate the risk factors for MetS. Results: In total, 2,550 (718 drug-naive) of 2,774 patients finished the study between July 6, 2010, and November 30, 2011. We found significant (P<.05) changes for BMI, WC, TG, and LDL-C, with TG and LDL-C reaching a plateau. Interactions between baseline metabolic condition and changes over time were observed for BMI (chi(2) = 43.11, P<.001), WC (chi(2) = 36.34, P<.001), systolic BP (chi(2) = 11.92, P=.002), glucose (chi(2) = 6.09, P=.01), and TG (chi(2) = 6.01, P=.01). Antipsychotics generally had greater adverse effects on patients who were initially screened as metabolically normal. After controlling for other associated factors, we found that antipsychotics resulted in differing risk for incident MetS, with a similar pattern to findings in other populations: olanzapine (odds ratio [OR] = 3.36, P<.001) > quetiapine (OR = 3.29, P<.001) > perphenazine (OR = 2.73, P=.007) > risperidone (OR = 2.21, P=.02) > aripiprazole (OR = 1.74, P=.15) approximate to haloperidol (OR = 1.75, P=.22) approximate to ziprasidone (OR = 1, reference). Conclusions: Metabolic traits should be monitored frequently in early stages of antipsychotic treatment due to rapid and substantial changes. Clinicians should not assume low risk for patients with normal metabolic parameters at baseline.