Adjunctive metformin for antipsychotic-related hyperprolactinemia: A meta-analysis of randomized controlled trials

Wei Zheng, Xin Hu Yang, Dong Bin Cai, Gabor S. Ungvari, Chee H. Ng, Nan Wang, Yu Ping Ning, Yu-Tao Xiang

    Research output: Contribution to journalArticle

    4 Citations (Scopus)

    Abstract

    Hyperprolactinemia is a common and severe antipsychotic-induced adverse drug reaction. This meta-analysis of randomized controlled trials systematically examined the efficacy and safety of adjunctive metformin for antipsychotic-related hyperprolactinemia in schizophrenia patients. Two independent investigators searched, extracted, and synthesized data. Weighted mean differences and risk ratios with their 95% confidence intervals were calculated using random effect model. Four randomized controlled trials (n=509) comparing adjunctive metformin (n=253) with the control groups (n=256), lasting 22.7 weeks of treatment, were included in the meta-analysis. The metformin group had significantly lower serum prolactin level at endpoint (four randomized controlled trials, n=501; weighted mean difference: -'6.87 ug/L (95% confidence interval: -'13.24 to -'0.51), p=0.03; I2=80%) with "moderate quality" based on the grading of recommendations assessment, development, and evaluation system. In patients with menstrual disturbances, the rate of menstruation resumption was 66.7% in the metformin group and 4.8% in the control group. Adverse drug reactions and all-cause discontinuation (three randomized controlled trials, n=339, risk ratio: 0.76 (95% confidence interval: 0.29, 1.97), p=0.57; I2= 0%) were similar between the two groups. Adjunctive metformin appears to be effective and safe for reducing antipsychotic-induced hyperprolactinemia and prolactin-related symptoms in schizophrenia patients. Higher quality randomized controlled trials with a large sample size are warranted to confirm these findings.

    Original languageEnglish
    Pages (from-to)625-631
    Number of pages7
    JournalJournal of Psychopharmacology
    Volume31
    Issue number5
    DOIs
    Publication statusPublished - 1 May 2017

      Fingerprint

    Cite this