Contributing Factors to Gastric Ulcer Healing After Endoscopic Submucosal Dissection Including the Promoting Effect of Rebamipide

Abstract

The healing process for artificial ulcers resulting from endoscopic submucosal dissection (ESD) for gastric cancer is not understood. To clarify factors that promote healing and the additional healing effects of rebamipide, we conducted a randomized controlled trial to compare proton pump inhibitor (PPI) and combination PPI plus rebamipide treatments. One hundred and seventy patients with early gastric cancers that had undergone ESD were enrolled. Follow-up endoscopy was scheduled at 4-6 weeks after ESD. We assessed marginal healing and basal healing independently by endoscopic observation. Marginal healing was determined by a regenerating epithelium and/or converging folds around the periphery of the ulcer. Basal healing was declared when the ulcer was covered by white coat thinning such that basal granulation could be seen. The sizes of the artificial ulcers were divided into normal size (area <1,200 mm(2)) or large size (area ≥ 1,200 mm(2)). Initial ulcer size and duration after ESD were significantly correlated with both marginal and basal healing rates by univariate analysis. The marginal healing rate of antral lesions was higher than that of body lesions. Multivariate analysis showed a large-sized ulcer was the only significant predictor of delayed healing, with delayed healing defined as no observed marginal or basal healing (p < 0.0001). Subgroup analysis for the effect of rebamipide on large-sized ulcers showed a significantly higher rate of basal healing in the combination PPI and rebamipide group (p = 0.015). Healing in ESD-induced ulcers was dependent on the initial ulcer size. In large-sized ulcers, rebamipide promotes basal healing.