Abstract
IMPORTANCE: Gastroesophageal reflux disease (GERD) is very common in high-income countries. Nonerosive GERD is increasingly being considered a different entity from erosive GERD, with a more benign disease course but less responsiveness to antireflux medication. Whether nonerosive GERD responds less well than erosive GERD to antireflux surgery is unclear. OBJECTIVE: To assess whether patients with nonerosive GERD retain more reflux symptoms after antireflux surgery than patients with erosive GERD. DESIGN, SETTING, AND PARTICIPANTS: This population-based cohort study included all patients in Finland and Sweden who underwent primary laparoscopic fundoplication for GERD between January 1, 1996, and December 31, 2019. Statistical analysis was conducted from March to April 2024. EXPOSURE: Patients with nonerosive GERD (ie, no erosive esophagitis or Barrett esophagus detected during preoperative endoscopy) were compared with patients with erosive GERD (ie, erosive esophagitis detected during preoperative endoscopy). MAIN OUTCOME AND MEASURES: The main outcome was reflux recurrence, defined as 6 months or more of postoperative antireflux medication or secondary antireflux surgery. Poisson regression provided hazard ratios (HRs) with 95% CIs, adjusted for sex, age, comorbidity, hospital volume of antireflux surgery, calendar year, and country. RESULTS: Of 6194 patients (median age, 53 years [IQR, 42-62 years]; 3310 women [53.4%]) who underwent primary fundoplication, 2700 (43.6%) received a diagnosis of nonerosive GERD, and 3494 (56.4%) received a diagnosis of erosive GERD. During up to 23 years of follow-up (range, 0-23 years; median, 8.8 person-years [IQR, 4.3-13.5 person-years]), the frequency of reflux recurrence was similar among patients with nonerosive GERD (17.1% [461 of 2700]) and those with erosive GERD (17.1% [596 of 3494]). Patients with nonerosive GERD had a similar overall risk of reflux recurrence as patients with erosive GERD (adjusted HR, 0.98; 95% CI, 0.87-1.11) and a similar risk of reflux recurrence when studying recurrence by antireflux medication (HR, 1.04; 95% CI, 0.90-1.21) and secondary antireflux surgery (HR, 0.91; 95% CI, 0.75-1.10) separately. There were no differences in HRs between various follow-up categories after fundoplication or in analyses stratified by the 6 variables included in the multivariable model. CONCLUSIONS AND RELEVANCE: This cohort study of patients who underwent primary laparoscopic fundoplication suggests that the risk of reflux recurrence was similar among patients with nonerosive GERD and those with erosive GERD. This finding is in contrast with evidence showing that nonerosive GERD responds less well to antireflux medication. Thus, the absence of erosive GERD detected by an upper endoscopy may not be used as an argument for abstaining from antireflux surgery.