Effect of antipsychotics and mood stabilisers on metabolism in bipolar disorder: a network meta-analysis of randomised-controlled trials

Background Antipsychotics and mood stabilisers are gathering attention for the disturbance of metabolism. This network meta -analysis aims to evaluate and rank the metabolic effects of the commonly used antipsychotics and mood stabilisers in treating bipolar disorder (BD). Methods Registries including PubMed, Embase, Cochrane Library, Web of Science, Ovid, and Google Scholar were searched before February 15th, 2024, for randomised controlled trials (RCTs) applying antipsychotics or mood stabilisers for BD treatment. The observed outcomes were twelve metabolic indicators. The data were extracted by two reviewers independently, and con fi rmed by another four reviewers and a corresponding author. The above six reviewers all participated in data analyses. Data extraction was based on PRISMA guidelines, and quality assessment was conducted according to the Cochrane Handbook . Use a random effects model for data pooling. The PROSPERO registration number is CRD42023466669. Findings Together, 5421 records were identi fi ed, and 41 publications with 11,678 complete -trial participants were con fi rmed eligible. After eliminating possible sensitivity, risperidone ranked 1st in elevating fasting serum glucose (SUCRA = 90.7%) and serum insulin (SUCRA = 96.6%). Lurasidone was most likely to elevate HbA1c (SUCRA = 82.1%). Olanzapine ranked 1st in elevating serum TC (SUCRA = 93.3%), TG (SUCRA = 89.6%), and LDL (SUCRA = 94.7%). Lamotrigine ranked 1st in reducing HDL (SUCRA = 82.6%). Amisulpride ranked 1st in elevating body weight (SUCRA = 100.0%). For subgroup analyses, quetiapine is more likely to affect indicators of glucose metabolism among male adult patients with bipolar mania, while long-term lurasidone tended to affect glucose metabolism among female patients with bipolar depression. Among patients under 18, divalproex tended to affect glucose metabolism, with lithium affecting lipid metabolism. In addition, most observed antipsychotics performed higher response and remission rates than placebo, and displayed a similar dropout rate with placebo, while no between -group signi fi cance of rate was observed among mood stabilisers. Interpretation Our fi ndings suggest that overall, antipsychotics are effective in treating BD, while they are also more likely to disturb metabolism than mood stabilisers. Attention should be paid to individual applicability in clinical practice. The results put forward evidence -based information and clinical inspiration for drug compatibility and further research of the BD mechanism. Copyright (c) 2024 The Author(s). Published by Elsevier Ltd. This is an open access article under the CC BY -NC -ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/).