BACKGROUND: The multicentre randomised trial YOMEGA (NCT02139813) comparing the one anastomosis gastric bypass (OAGB) with the Roux-en-Y gastric bypass (RYGB) confirmed the non-inferiority of OAGB on weight loss outcomes at 24 months. We aimed to report weight loss, metabolic, and safety outcomes at 5 years. METHODS: YOMEGA is a prospective, open-label, non-inferiority, randomised trial conducted at nine centres in France. Inclusion criteria were BMI of 40 kg/m or more, or 35 kg/m or more with comorbidities. Key exclusion criteria were severe gastro-oesophageal reflux disease or Barrett's oesophagus and previous bariatric surgery. Patients were randomly assigned (1 :1) to OAGB (one gastrojejunal anastomosis with a 200 cm biliopancreatic limb) or RYGB (with a 150 cm alimentary limb and a 50 cm biliary limb), stratified by centre, with blocks of variable size. The primary endpoint of this extension study was percentage excess BMI loss and was analysed in the per-protocol population, including patients with data who were operated on with the technique randomly assigned to them and excluding patients with major deviations from the protocol during the follow-up (change of surgical technique, death, or withdrawal of consent). Non-inferiority was concluded for the primary endpoint if the upper bound of the CI was less than the non-inferiority limit (7 percentage points). YOMEGA is registered with ClinicalTrials.gov, NCT02139813, and the 5-year follow-up of YOMEGA is registered with ClinicalTrials.gov, NCT05549271. FINDINGS: Between May 13, 2014, and March 2, 2016, 253 patients were randomly assigned to OAGB (n=129) or RYGB (n=124), and from these patients 114 in the OAGB group and 118 in the RYGB group were included in the per-protocol analysis. In the per-protocol population, at baseline, mean age was 43·0 years (SD 10·8), mean BMI was 44·0 kg/m (5·6), 54 (23%) patients were male and 178 (77%) were female; 55 (27%) of 207 patients had type 2 diabetes. After 5 years, mean percentage excess BMI loss was -75·6% (SD 28·1) in the OAGB group versus -71·4% (SD 29·8) in the RYGB group, confirming non-inferiority (mean difference -4·1% [90% CI -12·0 to 3·7], p=0·0099). Remission of type 2 diabetes was similar in both groups. Nutritional status did not differ; the most common adverse event was clinical gastro-oesophageal reflux disease, occurring in 27 (41%) of 66 patients in the OAGB group versus 14 (18%) of 76 patients in the RYGB group (p=0·0030). Among serious adverse events, ten (8%) of 127 patients converted from OAGB to RYGB. 171 (68%) of 253 patients were followed up. INTERPRETATION: OAGB was not inferior to RYGB regarding percentage excess BMI loss at 5 years with similar metabolic outcomes. The high rate of clinical gastro-oesophageal reflux disease after OAGB raises questions about its long-term consequences, which need to be further investigated. FUNDING: Medtronic.