Obesity is associated with increased morbidity and mortality. Surgery for morbid obesity is considered when other treatments have failed. A number of procedures are available, but the effects of these surgical procedures compared with medical management and with each other are uncertain.To assess the effects of surgery for morbid obesity.Studies were obtained from computerized searches of multiple electronic bibliographic databases, supplemented with hand searches of selected journals and consultation with experts in obesity research. Date of the most recent searches: December 2004.Randomised controlled trials comparing different surgical procedures, and randomised controlled trials and prospective cohort studies comparing surgery with non-surgical management for morbid obesity.Data were extracted by one reviewer and checked independently by two reviewers. Two reviewers independently assessed trial quality.Twenty-six trials were included. Two randomised controlled trials and three prospective cohort studies compared surgery with non-surgical management, and 21 randomised controlled trials compared different surgical procedures. The quality of most of the trials was poor; just three trials had adequate allocation concealment. A meta-analysis was not possible due to differences in the surgical procedures performed, measures of weight change and length of follow-up. Compared with conventional management, surgery resulted in greater weight loss (21 kg weight loss at eight years versus weight gain), with improvements in quality of life and comorbidities. Some complications of surgery occurred, such as wound infection. Gastric bypass was associated with greater weight loss, better quality of life and fewer revisions, reoperations and/or conversions than gastroplasty, but had more side-effects. Greater weight loss and fewer side-effects and reoperations occurred with adjustable gastric banding than vertical banded gastroplasty, but laparoscopic vertical banded gastroplasty produced more patients with an excellent or good result and fewer late complications than laparoscopic adjustable silicone gastric banding. Vertical banded gastroplasty was associated with greater weight loss but more vomiting than horizontal gastroplasty. Some postoperative deaths occurred in the studies. Weight loss was similar between open and laparoscopic procedures. Fewer serious complications occurred with laparoscopic surgery, although conversion to open surgery was sometimes required. Most studies found that laparoscopic surgery had a longer operative time. But, it resulted in reduced blood loss and quicker recovery.The limited evidence suggests that surgery is more effective than conventional management for weight loss in morbid obesity. The comparative safety and effectiveness of different surgical procedures is unclear.