How efficacious are traditional Chinese medicine injections in treating angina pectoris? A network meta-analysis of...

Over 50 million adults in China suffer from angina pectoris, which are often treated with traditional Chinese medicine injections (TCMIs). However, the efficacies of TCMIs and conventional drugs as determined by randomized controlled trials (RCTs) were not rigorously compared with one another by network meta-analysis (NMA). This PRISMA-compliant NMA aimed to compare the efficacy and assess the evidence strengths of 24 TCMIs in treating adults with angina pectoris of RCTs.

Following the protocol (PROSPERO registration number CRD42018117720), the RCTs that compared any TCMI with another TCMI or conventional drug on outcome measures including symptomatic and electrocardiography improvements were included. The quality of included RCTs was assessed with the Cochrane's risk of bias 2 tool. Frequentist statistical analyses were performed, including NMA, pairwise meta-analysis (PMA), subgroup analysis, sensitivity analysis, meta-regression, and publication bias analysis. The certainty of evidence was assessed with the GRADE approach.

Totally, 556 eligible RCTs with 57015 participants were identified while the quality of all but five included RCTs was poor. The significant efficacy estimates and insignificant heterogeneity assessment from PMA and NMA indicated that nearly all TCMIs were more efficacious than conventional treatments for angina pectoris. Adequate subgroup and sensitivity analyses found the robust and consistent results. However, the evidence strengths of meta-analyses were assessed as very low to low due to the high risk of RCTs. The comprehensive efficacy estimates suggested that 4 TCMIs (HH, Honghua injection; HHH, Honghua Huangsesu injection; GLP, Gualoupi injection; and SM, Shenmai injection) was the best anti-anginal drugs for adults with angina pectoris.

TCMIs appear to be efficacious for angina pectoris, although evidence evaluation of high-quality RCTs of TCMIs would be necessary. In particular, randomization and blinding procedures of the RCTs should be explicated to meet the CONSORT requirements.