Abstract: Objective To preliminarily evaluate the safety of bromhexine in domestic application. Methods CNKI, VIP and WanFang database were retrieved, and randomized controlled trials involving bromhexine were included. Meta-analysis for adverse reaction rates of bromhexine was performed by using RevMan5.1. In addition, subgroup analysis including different formulations (injections and tablets), different indications (pneumonia and bronchitis) and different controlling measures (conventional measures and other medications) were performed according to PRISAM guidelines. Results Ten RCTs which involved 1987 patients met the inclusion criteria. The experimental group showed higher adverse reaction incidence than controlling group[OR=4.42, 95%CI(2.55, 7.66), P ＜0.00001], the difference was statistically significant. This difference also exists in each subgroup analysis: injection subgroup (OR=4.41, 95%CI[2.35, 8.28]); tablets subgroup(OR=4.44, 95%CI[1.44, 13.68]); pneumonia subgroup(OR=4.41, 95%CI[3.25, 8.28], P＜0.00001); bronchitis subgroup(OR=4.1, 95%CI[1.22, 13.79], P =0.02); conventional measures subgroup(OR=5.92, 95%CI[2.71, 12.91] P ＜0.00001); other medications subgroup(OR=2.96, 95%CI[1.34, 6.53], P=0.007). Conclusion Bromhexine treatment exhibits higher adverse reaction incidence than other treatment, and this risk should be concerned in clinical use, but the Results need more high-quality randomized controlled trials to testify.

Key words: Kye words: bromhexine, adverse reaction, Meta-analysis, subgroup analysis