RESUMEN
OBJECTIVE: This study compared relapse prevention and acceptability of long-acting injectable (LAI) antipsychotics in the maintenance treatment of adults with nonaffective psychoses. METHODS: The authors searched MEDLINE, Embase, PsycINFO, CINAHL, CENTRAL, and online registers for randomized controlled trials published until June 2020. Relative risks and standardized mean differences were pooled using random-effects pairwise and network meta-analysis. The primary outcomes were relapse rate and all-cause discontinuation ("acceptability"). The quality of included studies was rated with the Cochrane Risk of Bias tool, and the certainty of pooled estimates was measured with GRADE (Grading of Recommendations Assessment, Development, and Evaluation). RESULTS: Of 86 eligible trials, 78 (N=11,505) were included in the meta-analysis. Regarding relapse prevention, most of the 12 LAIs included outperformed placebo. The largest point estimates and best rankings of LAIs compared with placebo were found for paliperidone (3-month formulation) and aripiprazole. Moderate to high GRADE certainty for superior relapse prevention compared with placebo was also found for (in descending ranking order) risperidone, pipothiazine, olanzapine, and paliperidone (1-month formulation). In head-to-head comparisons of LAIs, only haloperidol was inferior to aripiprazole, fluphenazine, and paliperidone. For acceptability, most LAIs outperformed placebo, with moderate to high GRADE certainty for (in descending ranking order) zuclopenthixol, aripiprazole, paliperidone (3-month formulation), olanzapine, flupenthixol, fluphenazine, and paliperidone (1-month formulation). In head-to-head comparisons, only LAI aripiprazole had superior acceptability to other LAIs (bromperidol, fluphenazine, paliperidone [1-month formulation], pipothiazine, and risperidone). CONCLUSIONS: LAI formulations of paliperidone (3-month formulation), aripiprazole, olanzapine, and paliperidone (1-month formulation) showed the highest effect sizes and certainty of evidence for both relapse prevention and acceptability. Results from this network meta-analysis should inform frontline clinicians and guidelines.