Antipsychotic Drugs and Dysregulated Glucose Homeostasis: A Systematic Review and Meta-Analysis.

Importance: Antipsychotic drug (AP)-induced glucose homeostasis changes are often attributed to AP-induced weight gain. Nevertheless, dysregulated glucose control can occur independently of weight gain. Objective: To examine the association between AP use and glucose homeostasis while considering weight gain propensity, medication type, and treatment duration. Data Sources: MEDLINE, Embase, PsychINFO, CINAHL, the Cochrane Central Register of Controlled Trials (CENTRAL), and Web of Science were searched from inception through February 3, 2025. Study Selection: Blinded randomized clinical trials (RCTs) comparing changes in glucose homeostasis-related parameters between patients with severe mental illness or healthy volunteers assigned to AP or control (placebo or no intervention) groups were included. Studies were limited to English-language human studies without restrictions on study length, AP type, or previous AP exposure. Of 22 773 unique citations, 163 RCTs met inclusion criteria, with 127 studies included in the meta-analysis. Data Extraction and Synthesis: Each article was screened independently by 2 authors using predefined inclusion and exclusion criteria. Data extraction and risk of bias assessment were completed using a standardized spreadsheet. Data were analyzed via random-effects meta-analysis, with subgroup analyses for diagnosis, study length, AP type, age, concomitant medication use, and previous AP exposure. Metaregressions identified covariate effects. Data analysis was completed from October 2023 to February 2025. Main Outcomes and Measures: Primary study outcomes were changes in fasting glucose, fasting insulin, and glycated hemoglobin (HbA1c) following AP treatment. Secondary outcomes included any other glucose metabolism-related parameters including, but not limited to, insulin resistance and hyperglycemia. Results: A total of 35 952 AP-treated patients and 19 010 placebo-treated patients were included in the qualitative synthesis, while 28 975 AP-treated and 15 101 placebo-treated patients were included in the meta-analysis. AP use was associated with significantly increased fasting glucose (mean difference [MD], 0.72 mg/dL; 95% CI, 0.54-1.08 [to convert to millimoles per liter, multiply by 0.0555]; P < .001), fasting insulin (MD, 1.94 μIU/mL; 95% CI, 1.28-2.61 [to convert to picomoles per liter, multiply by 6]; P < .001), glycated hemoglobin (MD, 0.04%; 95% CI, 0.02%-0.05% [to convert to proportion of total hemoglobin, multiply by 0.01]; P < .001), and hyperglycemia (odds ratio, 1.29; 95% CI, 1.04-1.59; P = .02) vs placebo. Findings were corroborated in healthy volunteers. Subgroup analyses suggested that AP type, diagnosis, age, concomitant medication use, and previous AP exposure do not consistently affect dysglycemia risk. In metaregression analyses, AP-associated dysregulations in glucose homeostasis were independent of study length and AP dose. Conclusions and Relevance: In this systematic review and meta-analysis, results indicate that AP exposure significantly disrupts glucose homeostasis independent of exposure time, dose, diagnosis, and weight gain propensity. Increased awareness of AP-induced dysregulations in glucose homeostasis alongside ongoing metabolic monitoring and potential treatment is warranted.