Preoperative Hill grade versus hiatal hernia size: which drives the decision for Dor fundoplication during sleeve gastrectomy? A retrospective cohort study.

BACKGROUND: Reflux risk after sleeve gastrectomy is closely linked to gastroesophageal junction anatomy, but the relative influence of preoperative Hill grade vs. hiatal hernia size on the decision to add concurrent Dor fundoplication remains unclear. Identifying the dominant driver may improve patient selection for an adjunctive anti-reflux procedure. OBJECTIVE: To compare the relative contributions of Hill grade and hiatal hernia size to the decision for Dor fundoplication during sleeve gastrectomy, and to evaluate whether Dor fundoplication is associated with improved reflux outcomes at 12 months. METHODS: This single-centre retrospective cohort study included adult patients who underwent primary sleeve gastrectomy between January 2023 and June 2025. Follow-up was completed and the database locked in June 2026. Patients were divided into Dor and non-Dor groups according to whether concurrent Dor fundoplication was performed. The primary analyses examined factors associated with Dor selection and treatment failure for reflux at 12 months. Secondary outcomes included de novo GERD, persistent GERD, PPI discontinuation, endoscopic and pH-monitoring evidence of reflux, safety, weight loss, and diabetes remission. To minimize selection bias inherent in surgeon-driven treatment decisions, we applied propensity score overlap weighting to create a pseudo-population with balanced baseline covariates between groups, simulating the conditions of a randomized trial. Multivariable logistic regression, sequential model comparisons, decision curve analysis, and propensity score overlap weighting were used to control confounding. RESULTS: Among 316 patients (126 Dor, 190 non-Dor), high-risk Hill grade was the strongest independent factor for Dor selection (adjusted OR 3.48, 95% CI 2.02-6.00). Hiatal hernia size showed a weaker association (adjusted OR 1.42 per 1-cm increase, 95% CI 1.10-1.85). Model comparisons confirmed a greater incremental contribution from Hill grade than from hernia size. After overlap weighting, Dor was associated with a lower risk of treatment failure for reflux (adjusted RR 0.63, 95% CI 0.43-0.91), with consistent reductions in de novo GERD, persistent GERD, and regular PPI use, and a higher PPI discontinuation rate. Operative time was longer in the Dor group, but hospital stay, major complications, weight loss, and diabetes remission did not differ significantly. Sensitivity analyses supported the main findings. CONCLUSION: In patients undergoing sleeve gastrectomy, preoperative Hill grade drives the decision for concurrent Dor fundoplication more strongly than hiatal hernia size. After adjustment for clinical comparability, Dor fundoplication was associated with a lower risk of treatment failure for reflux at 12 months, without an apparent safety or weight-loss penalty.