Meta-analysis of Long-term De Novo Acid Reflux-Related Outcomes Following Sleeve Gastrectomy: Evidence Against the Need for Routine Postoperative Endoscopic Surveillance.

Abstract

To evaluate the incidence of long-term de novo acid reflux-related complications following sleeve gastrectomy (SG) to determine whether routine postoperative surveillance endoscopy is necessary. A systematic search of Medline, Embase, CINAHL, CENTRAL, the Web of Science, and bibliographic reference lists was conducted. A proportion meta-analysis model was constructed to quantify the risk of the de novo gastro-oesophageal reflux disease (GORD), oesophagitis, and Barrett's oesophagus (BE) at least 4years after SG. Random-effects modelling was applied to calculate pooled outcome data. Thirty-two observational studies were included reporting a total of 7904 patients who underwent primary SG and were followed up for at least 4years. The median follow-up period was 60months (48-132). Preoperative acid-reflux symptoms existed in 19.1% ± 15.1% of the patients. The risk of development of de novo GORD, oesophagitis, and BE after SG was 24.8% (95% CI 18.6-31.0%), 27.9% (95% CI 17.7-38.1%), and 6.7% (95% CI 3.7-9.7%), respectively. The between-study heterogeneity was significant in all outcome syntheses. It was suspected that several of the included studies have not reported BE and oesophagitis because such events might not have happened in their cohorts. Long-term risk of de novo GORD after SG seems to be comparable with those of the general population which questions the merit of surveillance endoscopy after SG in asymptomatic patients. De novo BE and oesophagitis after SG have not been reported by most of the available studies which may lead to overestimation of the rates of both outcomes in any evidence synthesis. We recommend endoscopic surveillance for symptomatic patients only.