Gastric bypass : positive and negative health effects

Sammanfattning: Obesity is a major health problem worldwide. It is preventable in theory, but in practice no country has been able to stop the development of increasing obesity. The only available treatment with sufficient long-term effect on weight loss and comorbid diseases is surgery. Bariatric surgery has increased over the last decades as several studies demonstrated longterm weight-loss, dramatic effect on diabetes and other comorbid diseases and an effect on mortality. The introduction of laparoscopic surgery and refinements of surgical procedures with an overall reduction of postoperative complications and morbidity has also helped to enhance public perception of bariatric surgery. The aim of this thesis was to evaluate the positive effect of bariatric surgery on diabetes, and other potentially negative effects of surgery. As gastric bypass surgery became increasingly popular, several anecdotal reports suggested that these patients were overrepresented in alcohol treatment programs. In paper I we divided all patients who underwent bariatric surgery between 1980-2006 into two cohorts. One cohort consisted of patients that were operated with gastric bypass and the other consisted of patients operated with gastric banding or vertically banded gastroplasty, so called restrictive procedures. Data on inpatient diagnosis of depression, suicide attempt, alcohol and substance use disorders were retrieved from the National Patient Register. The main finding was that the risk for post-operative alcohol use disorders was elevated in the gastric bypass cohort compared to the restrictive cohort. In paper II we compared patients operated with primary gastric bypass in Sweden between 2001-2010, with an age- and sex-matched control group that was sampled from the general population. The same outcome as in paper I was studied. In a subcohort we also analyzed the prescription pattern of antidepressants, benzodiazepines, hypnotics and sedatives and medication against alcohol dependence, before and after surgery. The main findings were that patients undergoing gastric bypass had a higher risk of preoperative diagnosis of depression, substance abuse and suicide attempts. After surgery the risk for all investigated diagnoses were elevated, including alcohol dependence. Prescriptions of benzodiazepines, hypnotics and sedatives, and medication against alcohol dependence increased after gastric bypass surgery. In paper III we used the Scandinavian Obesity Surgery Registry and the Prescribed Drugs Register to investigate the effect of gastric bypass surgery on diabetes type 2 treatment. 67 % of patients with diabetes treatment before surgery were without diabetes treatment 2 years after surgery and 61% after 7 years. In patients with a short duration of diabetes treatment, the surgery was even more effective. Less than 2 % of patients with prediabetes before surgery were on diabetes treatment during 7-year follow up, which is lower than in a control group from the general population. In 2010 we changed the surgical technique used for gastric bypass, from a technique where the mesentery was divided, to a technique where the mesentery was kept intact. In paper IV we compared the 2 techniques retrospectively regarding postoperative complications, more specifically leakage during the first 30 days and ulceration/stenosis in the anastomosis during the first 6 months. Both leakage and ulceration/stenosis were significantly reduced after the change of technique. In summary we demonstrate that gastric bypass surgery has a dramatic effect on diabetes, and also a diabetes preventive effect in non-diabetic patients. We show that the effect on diabetes is better if the surgery is performed early after diabetes onset. On the other hand, we found that patients seem to have a higher risk for postoperative alcohol abuse disorders after surgery. Finally, our last study on complications indicates that gastric bypass should be performed without division of the mesentery. Gastric bypass is a safe surgical method with good long-term effects, but patients must be informed about potential negative effects and possibly, patients with previous alcohol abuse should be operated with another method.