BACKGROUND: Revisional bariatric operations performed for weight loss failure are frequently associated with inconsistent weight reduction and serious perioperative complications. METHODS: Outcomes of 151 consecutive revisional operations performed by one surgeon for unsatisfactory weight loss were compared to determine whether postoperative weight loss is influenced by the type of primary procedure. Minimum follow-up was 12 months. RESULTS: Primary operations included 14 jejunoileal bypass (JIB): one revised to gastroplasty, 13 to RY gastric bypass; 71 gastroplasty/banding (GP/B): all revised to Roux-en-Y gastric bypass (RYGB); and 66 gastric bypass: 49 revised to distal/malabsorptive RYGB, 12 restapled without malabsorption, and 5 loop bypasses revised to standard RYGB. Perioperative morbidity/mortality rates were 21.8% and 1.3%, respectively. Follow-up at 12 months was 93%. Mean weight/body mass index unit loss after revision of JIB was 90 pounds/17 units versus 113 pounds/16 units after revision of GP/B and 71 pounds/11 units after revision of gastric bypass (P < or = 0.05) with corresponding mean percent of excess weight loss of 51% for JIB, 56% for GP/B, and 48% for gastric bypass. Five of the JIB revisions (38%) lost > or = 50% excess weight loss versus 39 of the GP/B revisions (61%) and 28 of the gastric bypass revisions (48%). Comorbidities improved/resolved in 100% of those who lost > or = 50% of excess weight versus 89% who did not. CONCLUSIONS: Weight loss after revision of pure restrictive operations is significantly better than after revision of operations with malabsorptive components. Improvement of comorbidities in the great majority of patients justifies revision of all types of bariatric operations for unsatisfactory weight loss.