Chinese herbal medicine combined with conventional therapy for gastric precancerous lesions: a systematic review and meta-analysis

Abstract

Background: Chronic atrophic gastritis and gastric precancerous lesions are associated with increased gastric cancer risk. Chinese herbal medicine is widely used as an adjunct to conventional therapy, but the certainty and clinical interpretation of its effects remain unclear. Objective: To evaluate the efficacy and safety of Chinese herbal medicine added to conventional therapy for chronic atrophic gastritis and gastric precancerous lesions. Methods: Eight databases and three trial registries were searched from inception to 31 March 2026 and updated to 16 August 2026. Randomised controlled trials comparing Chinese herbal medicine plus conventional therapy with the same conventional therapy alone and reporting extractable pathological outcomes were included. Random-effects meta-analyses and sensitivity analyses were performed. Risk of bias was assessed using RoB 2 and certainty using GRADE. Results: Sixty-nine randomised trials comprising 6,983 participants were included. Exploratory broad analyses favoured combined treatment for gastric mucosal atrophy (30 studies; SMD -1.63, 95% CI -2.03 to -1.24), intestinal metaplasia (29 studies; SMD -1.48, 95% CI -1.90 to -1.06), and dysplasia/intraepithelial neoplasia (21 studies; SMD -1.25, 95% CI -1.51 to -0.99). Heterogeneity was substantial; prediction intervals crossed the null for atrophy and intestinal metaplasia but not dysplasia/intraepithelial neoplasia. Clinically comparable histopathology sensitivities included only three to four studies and were imprecise. H. pylori clearance/negative conversion among baseline-positive participants favoured combined treatment (15 studies; RR 1.25, 95% CI 1.15 to 1.37), whereas strict test-of-cure sensitivity was imprecise (3 studies; RR 1.19, 95% CI 0.91 to 1.56). There was no clear difference in reported adverse events (14-study evidence set; RR-contributing studies=12; RR 0.93, 95% CI 0.58 to 1.50), which does not establish equivalent safety. Certainty was low for all five principal outcomes. Conclusions: Chinese herbal medicine added to conventional therapy may be associated with lower short-term pathological scores, but the magnitude and clinical meaning remain uncertain because of trial limitations, heterogeneity, and imprecision in clinically comparable analyses. The findings do not establish durable histological regression or gastric cancer prevention and should not be interpreted as a class-wide effect of all Chinese herbal medicines. Registry record: PROSPERO CRD420261442138.