Development and internal validation of a nomogram incorporating C-reactive protein-to-albumin ratio for predicting esophageal stricture after endoscopic submucosal dissection in early esophageal neoplasia.

Background: Endoscopic submucosal dissection (ESD) for early esophageal neoplasia may cause strictures that impair eating and quality of life. Existing models rely on anatomical factors; the value of inflammatory-nutritional markers like the C-reactive protein-to-albumin ratio (CAR) remains unclear. We aimed to develop and internally validate a CAR-based nomogram for post-ESD stricture. Methods: This retrospective single-center study (2015-2024) included consecutive ESD patients with early esophageal neoplasia, excluding those with additional therapy, multiple lesions, comorbidities, recent infections, steroids, or loss to follow-up. Stricture: dysphagia or inability to pass a 9.8-mm scope within 12 months. CAR = preoperative C-reactive protein / albumin. The model was developed on the full cohort and internally validated via 1,000 bootstrap resamples (optimism-corrected). Logistic regression built the nomogram. Performance: area under the curve (AUC), sensitivity/specificity, calibration (Brier), and decision curve analysis (DCA). Results: Of 111 patients, 36 (32.4%) developed stricture. Independent predictors were infiltration depth (< M2 Conclusions: CAR independently predicts post-ESD stricture. The nomogram is a promising preliminary risk-stratification tool, but requires external validation before its routine clinical use.