Progress in Filling the Gaps in Bariatric Surgery

Abstract

THE OBESITY EPIDEMIC IS A MAJOR THREAT TO PUBLIC health. Estimates by the World Health Organization indicate that in 2008 more than 1.4 billion adults or 21% of the world population were overweight, a substantial proportion of whom ( 500 million) were obese, and these numbers continue to increase. Sixty-five percent of the world’s population live in countries where overweight and obesity are linked to more deaths than underweight and malnutrition. For severe obesity, bariatric surgery is the only treatment resulting in significant and sustained weight loss. Nevertheless, only a very small fraction of the eligible candidates undergo surgery. The gaps in the bariatric surgery evidence base contribute to the small fraction of obese patients who undergo surgery. These gaps have included insufficient randomized controlled trials, limits to the generalizability of the reports of bariatric surgical interventions, and uncertainty regarding long-term outcomes. Addressing the various and broad health outcomes resulting from bariatric surgery such as diabetes, hypertension, sleep apnea, psychosocial factors, and many others has proven difficult. In this issue of JAMA, a theme issue on obesity, 3 articles address some of the evidence gaps and add bricks to the wall of data needed to create a solid and respected evidence structure. A prospective cohort study conducted by Adams and colleagues assessed long-term weight loss and health outcomes in severely obese patients undergoing gastric bypass surgery compared with severely obese control patients. There were 418 patients in the Roux-en-Y gastric bypass (RYGB) group compared with 2 control groups; one control group included patients seeking surgery (n=417) and the other included a random sample of severely obese adults (n=321). Although weight loss studies typically are limited by very high dropout rates, a remarkable aspect of the study by Adams et al was the 6-year follow-up rate of 93% in the RYGB group. At 6 years, mean weight loss for RYGB patients was 28% of initial body weight compared with weight gain of 0.2% and 0% in the 2 control groups, respectively. Diabetes remission rates (defined as normal fasting glucose and hemoglobin A1c) were 62% in the RYGB group and 8% and 6% in each of the control groups. Incident cases of diabetes at 6 years were 2% in the RYGB group compared with 17% and 15% in the 2 control groups, respectively. At 6 years of follow-up, the RYGB group showed improvement in all other health outcomes (eg, hypertension, high-density lipoprotein cholesterol) that were examined with the exception of the 36-item Short Form Health Survey mental component summary score. The mortality rate at 6 years was 3% in the RYGB patients compared with 3% in the obese patients evaluated for but who did not undergo surgery and 1% in the other control group of obese adults. Of all deaths, 4 suicides and 2 of the 3 poisonings were in the RYGB cohort. Another important aspect of these findings is that despite the attenuation of weight loss between 2 and 6 years in the RYGB group (shown in Figure 2 of the article), the control of comorbid conditions remained very good. These findings are important because they show in a RYGB cohort and control group with nearly complete follow-up at 6 years that weight loss and associated health benefits following RYGB are durable. The mortality rates in this study were too small to assess statistically, but serve as a reminder of an uncommon but important outcome needing objective monitoring. It is widely held that sleep apnea responds to surgically induced weight loss. In a randomized controlled trial reported in this issue of JAMA, Dixon and colleagues assessed the effect of weight loss with laparoscopic adjustable gastric banding (LAGB) on obstructive sleep apnea (OSA). Sixty obese patients (body mass index of 35-55) with recently diagnosed moderate to severe OSA were randomized and compared with a control group undergoing treatment with a conventional weight loss program. Polysomnography was performed 2 years later in both groups. Patients in the LAGB group lost 27.8 kg (95% CI, 20.9 to 34.7 kg), which was significantly more weight than the 5.1 kg (95% CI, 0.8 to 9.3 kg) lost by patients in the control group (P .001). Both groups experienced reduction in total apneahypopnea index (AHI) at 2 years compared with baseline (25.5 events/hour [95% CI, 14.2 to 36.7 events/hour] in the LAGB group vs 14.0 events/hour [95% CI, 3.3 to 24.6 events/