Second-generation antipsychotics and adiponectin levels in schizophrenia: A comparative meta-analysis

People with schizophrenia treated with second-generation antipsychotics (SGAs) have lower plasma adiponectin levels, as compared with general population, that may lead to metabolic abnormalities. However, the contribution of different SGAs on adiponectin dysregulation is still unclear. The objective of this systematic review and meta-analysis was to estimate differences in adiponectin levels among people with schizophrenia treated with different SGAs. We systematically searched for observational studies published up to March 2015 in main electronic databases. Different SGAs were included if data on adiponectin were available from at least three different samples involving as a minimum five participants per treatment arm. Standardized mean differences with relevant 95% confidence intervals were generated. I-2 was used to test heterogeneity among studies. Eight studies were included with data suitable for carrying out four different comparisons: Clozapine vs. Olanzapine (including n=877 individuals with schizophrenia); Clozapine vs. Risperidone (n=660); Olanzapine vs. Risperidone (n=738); Quetiapine vs. Risperidone (n=186). There were no differences on adiponectin levels between people taking Clozapine and those taking Olanzapine (p=0.86), but high heterogeneity was detected (I-2=824 Both individuals taking Clozapine (p<0.001; I-2=0%) and those taking Olanzapine (p =0.02; 12=9%), but not subjects treated with Quetiapine (p=0.47; I-2=0%), had adiponectin levels significantly lower than people taking Risperidone. Our findings are consistent with previous evidence showing greater metabolic abnormalities attributable to Clozapine and Olanzapine, as compared with other SGAs. Although mechanisms whereby both these SGAs influence adiponectin remain unexplained, its reduction might mediate relevant abnormalities. Prospective evaluations of long-term effects of different SGAs on adiponectin are needed. (C) 2015 Elsevier B.V. and ECNP. All rights reserved.