Systematic review and Meta-analysis of 26 randomized controlled clinical trials of Compound Danshen Dripping Pill for non-proliferating diabetic retinopathy

OBJECTIVE: Diabetic retinopathy (DR) is the retinal consequence of chronic progressive diabetic microvascular leakage and occlusion. Non-proliferating diabetic retinopathy (NPDR) is the early stage of DR. It eventually occurs to some degree in all patients with diabetes mellitus. In recent years, many clinical trials have shown that Compound Danshen Dripping Pill (CDDP) may be associated with the improvement of NPDR symptoms. The aim of this study was to quantitatively summarize the association between CDDP and the therapeutic effects of NPDR. METHODS: It was conducted that a systematic literature search of PubMed, Web of Science, CNKI, VIP and Wanfang Data updated in June 2020 with the following search terms: “diabetic retinopathy” or “retinopathy” or “DR” or “NPDR”, in combination with “Compound Danshen Dripping Pill” or “Salvia miltiorrhiza” or “Danshen”. Risk ratio (RR) and weighted mean difference (WMD) with their 95% confidence interval (CI) was calculated between treatment and control groups. The sensitivity analyses were undertaken by removing each individual study when high heterogeneity appeared. Subgroup analysis, Meta-regression, and publication bias analysis were also conducted. The strength of evidence was evaluated with the Grading of Recommendation, Assessment, Development, and Evaluation (GRADE) method. RESULTS: Twenty-six RCTs involving 2047 subjects were included to conduct a Meta-analysis after screening the studies, extracting the data, and assessing the study quality. The Stata15.0 software was utilized for processing. Meta-analysis indicated that curative effects of treatment group with CDDP was significantly better than control [RR = 0.54, 95% CI (0.40, 0.73); moderate-quality evidence]. In addition, the results showed that CDDP was significantly associated with improving retinal hemorrhages [WMD = −0.62, 95% CI (−0.78, −0.46); low-quality evidence], the vision [WMD = 0.14, and 95% CI (0.09, 0.19), low-quality evidence], fundus fluorescence angiography [RR = 0.37 and 95% CI (0.23, 0.60); low-quality evidence], reduction of retinal microaneurysm [WMD = −3.74 and 95% CI (−4.38, −3.11); moderate-quality evidence], hemangioma volume [WMD = −3.15, 95%CI (−3.45, −2.85); moderate-quality evidence], macular thickness [WMD = −5.52, 95%CI = (−64.27, −48.78); low-quality evidence], mean defect [WMD = −1.65 and 95% CI (−1.95, −1.34); very low-quality evidence], fasting blooding glucose [WMD = −0.95, 95% CI (−1.19, −0.70); low-quality evidence), hemoglobin A1c [WMD = −0.62, 95% CI (−0.93, −0.30); low-quality evidence], high sensitive C reaction protein [WMD = −5.66, 95% CI (−8.01, −3.31); low-quality evidence]. Sensitivity, subgroup, and Meta-regression analyses were also assessed. CONCLUSION: The study demonstrated that CDDP has beneficial clinical effects for treating NPDR and improve the vision. Moreover, it indicated that oral CDDP in NPDR patients led to significant regulation of serum level of fasting blooding glucose, hemoglobin A1c and high sensitive C reaction protein, which was associated with the pathogenesis of NPDR. However, high-quality and large randomized clinical trials will be needed to prove the consequence in future.