Umbrella Review: Association Between Antipsychotic Drugs and Metabolic Syndrome Hallmarks in Children and Adolescents

Objective: To summarize the available evidence on metabolic parameters indicating metabolic adverse effects and risk of metabolic syndrome in children and adolescents treated with antipsychotics, following a pre-specified protocol (PROSPERO ID 252336). Method: We searched PubMed, Embase and PsycINFO until May 14, 2021, to identify systematic reviews (SR), meta-analyses (MA) and network meta-analyses (NMA) examining symptoms associated to metabolic syndrome in patients <18 years of age who required treatment with oral antipsychotic drugs. Evidence from quantitative analyses for all outcomes related to anthropometric, glyco-metabolic, and blood pressure parameters (measured from baseline to intervention-end and/or follow-up, in subjects exposed to antipsychotics and placebo) was reported on the basis of their metrics (median difference [medianD], mean difference [MD], standardized mean difference [SMD], odds ratio [OR], risk ratio ([RR]). A qualitative synthesis was also made. A formal quality assessment of the included studies was carried out by using the AMSTAR 2. We also provided a hierarchical stratification of the evidence from meta-analyses based on the class of evidence. Results: A total of 23 articles (13 MA, 4 NMA and 6 SR) were included for review. As compared with placebo, an increase in triglyceride levels was associated with olanzapine (medianD [95% CI]: 37 [12.27, 61.74] mg/dL; MD [95% CI]: 38.57 [21.44, 55.77] mg/dL) and quetiapine (medianD [95% CI]: 21.58 [95% CI]: 4.27, 38.31 mg/dL; MD [95% CI]: 34.87 [20.08, 49.67] mg/dL; SMD [95% CI]: 0.37 [0.06, 0.068]), whereas decreased triglyceride levels were found for lurasidone. Increased total cholesterol level was associated with asenapine (medianD [95% CI]: 9.1 [1.73, 16.44] mg/dL), quetiapine (medianD [95% CI]: 15.60 [7.30, 24.05] mg/dL; olanzapine (MD [95% CI] from 3.67 [1.43, 5.92] mg/dL to 20.47 [13.97, 26.94] mg/dL]; and lurasidone (medianD [95% CI]: 8.94 [1.27, 16.90] mg/dL). Change in glucose levels did not differ among antipsychotics or placebo. Lurasidone, molindone, and ziprasidone were the best tolerated in terms of weight gain. According to the AMSTAR 2 scoring system, 13 (56.5%) reviews were rated as very low quality. According to classes of evidence, most MA were level 4, especially because of their limited total sample size. Conclusion: By collating meta-analyses assessing biochemical markers of metabolic syndrome in antipsychotic-treated children, we conclude that olanzapine should not be the antipsychotic of choice in patients at risk for hypertriglyceridemia or hypercholesterolemia. Aripiprazole and lurasidone appear to be better tolerated in terms of metabolic adverse events. Insufficient meta-analytic data are available to provide a precise risk estimate of metabolic syndrome, and, overall, the quality of evidence is low.