WEIGHT-LOSS AND DIETARY-INTAKE AFTER VERTICAL BANDED GASTROPLASTY ANDROUX-EN-Y GASTRIC BYPASS

Objective The purpose of this study was to learn whether preoperative eating habits can be used to predict outcome after vertical banded gas troplasty (VBG) and Roux-en-Y gastric bypass (RYGB). Background Summar y Several independent randomized and sequential studies have reported significantly greater weight loss after RYGB in comparison with VBG. A lthough the mechanism responsible for weight loss after both procedure s is restriction of intake rather than malabsorption, the relationship s between calorie intake, food preferences, and postoperative weight l oss are not well defined. Methods During the past 5 years, 138 patient s were prospectively selected for either VBG or RYGB, based on their p reoperative eating habits. All patients were screened by a dietitian w ho determined total calorie intake and diet composition before recomme nding VBG or RYGB. Thirty patients were selected for VBG; the remainin g 108 patients were classified as ''sweets eaters'' or ''snackers'' an d had RYGB. Detailed recall diet histories also were performed at each postoperative visit. Results Early morbidity rate was zero after VBG versus 3% after RYGB. There were no deaths. Mean follow-up was 39 +/- 11 months after VBG and 38 +/- 14 months after RYGB. Mean weight loss peaked at 74 +/- 23 lb at 12 months after VBG and 99 +/- 24 lb at 16 m onths after RYGB (p less than or equal to 0.001). Twelve of 30 VBG pat ients lost greater than or equal to 50% of their excess weight versus 100 of 108 RYGB patients (p less than or equal to 0.0001). Milk/ice cr eam intake was significantly greater postoperatively in patients who u nderwent VBG versus patients who underwent RYGB after 6 months (p less than or equal to 0.003), whereas solid sweets intake was significantl y greater after VBG during the first 18 months postoperatively (p less than or equal to 0.004). Revision of VBG was performed in 6 of 30 pat ients (20%) for complications or poor weight loss, whereas only 2 of 1 08 patients who underwent RYGB required surgical revisions (p less tha n or equal to 0.001). Conclusions These data show that VBG adversely a lters postoperative eating behavior toward soft, high-calorie foods, r esulting in problematic postoperative weight loss. Conversely, RYGB pa tients had significantly greater weight loss despite inferior preopera tive eating habits. The high rate of surgical revision in conjunction with inconsistent postoperative weight loss has led us to no longer re commend VBG as treatment for morbid obesity.