Association between follow-up endoscopic inflammation and subsequent esophageal stricture formation after caustic ingestion.

The prognostic accuracy of initial endoscopic findings using the Zargar classification for long-term outcomes remains imperfect. We evaluated whether mucosal inflammation on follow-up endoscopy provides additional prognostic information regarding the risk of esophageal stricture requiring therapeutic intervention. We included 20 patients with corrosive esophagitis who underwent initial endoscopy within 48 hours of caustic ingestion and follow-up endoscopy at Seoul Metropolitan Government Seoul National University Boramae Medical Center between January 1, 2010, and August 30, 2025. The outcome was esophageal stricture requiring balloon dilation or stent insertion. Patients were categorized by the presence or absence of esophageal inflammation on follow-up endoscopy, defined by persistent caustic-related mucosal inflammatory or injury findings, including hyperemia, edema, erosions or ulcerations, friability or bleeding, exudates, clots, or necrotic changes. Firth penalized Cox regression was used to identify predictors of endoscopic intervention. Follow-up endoscopy was performed at a median of 13 (interquartile range, 7-29) days after the initial examination. Seven patients (35%) demonstrated inflammatory findings on follow-up endoscopy, while 13 patients (65%) showed no inflammation. All 4 patients requiring endoscopic intervention had sustained inflammation on follow-up endoscopy, whereas none without inflammation required therapeutic intervention. In multivariable analysis, inflammation on follow-up endoscopy was associated with an increased risk of requiring endoscopic intervention, although the estimate was imprecise, with a wide confidence interval (adjusted hazard ratio, 17.14; 95% confidence interval, 1.67-2308.65; P = .014). Initial Zargar classification grades were not significantly associated with the outcome. Sustained mucosal inflammation on follow-up endoscopy may provide additional prognostic information regarding the risk of esophageal stricture requiring intervention in corrosive esophagitis. These findings warrant validation in larger prospective cohorts.