Predictors of endoscopic management failure in benign non-caustic esophageal strictures: a retrospective cohort study

Benign Non-caustic Esophageal Strictures (BNES) often require repeated endoscopic dilations, yet a substantial proportion of patients remain refractory and ultimately require esophagectomy. Identifying predictors of endoscopic failure is essential for risk stratification and timely referral to alternative therapies. This study aimed to identify factors associated with endoscopic treatment failure leading to esophagectomy in patients with BNES.

This retrospective cohort study included 51 patients with BNES treated at a single tertiary center from January 2004 to March 2026. Patients with caustic, malignant, or post-esophagectomy anastomotic strictures were excluded. Treatment consisted of balloon or bougie dilation. Endoscopic failure was defined as the need for esophagectomy due to persistent or recurrent dysphagia despite treatment. Multivariable logistic regression was performed to identify independent predictors of failure.

Of 51 patients, 18 (35.3%) required esophagectomy. Failure was associated with non-GERD etiology (p < 0.001), absence of hiatal hernia (p = 0.001), segmental strictures (p = 0.016), and a higher number of dilations (p = 0.026). In multivariable analysis, non-GERD etiology (OR = 7.45, p = 0.034) and absence of hiatal hernia (OR = 8.28, p = 0.016) were independent predictors. In a secondary model, absence of hiatal hernia (OR = 17.02, p = 0.011) and number of dilations (OR = 1.08 per session, p = 0.028) remained significant. Among endoscopically managed patients, 33.3% had persistent symptoms.

One-third of BNES patients required esophagectomy. Non-GERD etiology, absence of hiatal hernia, and increasing dilation burden independently predict endoscopic failure and may support earlier consideration of surgical management.