Minimal access surgery compared with medical management for gastro-oesophageal reflux disease: five year follow-up of a randomised controlled trial (REFLUX).

Abstract

To determine the long term clinical effectiveness of laparoscopic fundoplication as an alternative to drug treatment for chronic gastro-oesophageal reflux disease (GORD). Five year follow-up of multicentre, pragmatic randomised trial (with parallel non-randomised preference groups). Initial recruitment in 21 UK hospitals. Responders to annual questionnaires among 810 original participants. At entry, all had had GORD for >12 months. The surgeon chose the type of fundoplication. Medical therapy was reviewed and optimised by a specialist. Subsequent management was at the discretion of the clinician responsible for care, usually in primary care. Primary outcome measure was self reported quality of life score on disease-specific REFLUX questionnaire. Other measures were health status (with SF-36 and EuroQol EQ-5D questionnaires), use of antireflux medication, and complications. By five years, 63% (112/178) of patients randomised to surgery and 13% (24/179) of those randomised to medical management had received a fundoplication (plus 85% (222/261) and 3% (6/192) of those who expressed a preference for surgery and for medical management). Among responders at 5 years, 44% (56/127) of those randomised to surgery were taking antireflux medication versus 82% (98/119) of those randomised to medical management. Differences in the REFLUX score significantly favoured the randomised surgery group (mean difference 8.5 (95% CI 3.9 to 13.1), P<0.001, at five years). SF-36 and EQ-5D scores also favoured surgery, but were not statistically significant at five years. After fundoplication, 3% (12/364) had surgical treatment for a complication and 4% (16) had subsequent reflux-related operations-most often revision of the wrap. Long term rates of dysphagia, flatulence, and inability to vomit were similar in the two randomised groups. After five years, laparoscopic fundoplication continued to provide better relief of GORD symptoms than medical management. Adverse effects of surgery were uncommon and generally observed soon after surgery. A small proportion had re-operations. There was no evidence of long term adverse symptoms caused by surgery. Current Controlled Trials ISRCTN15517081.