Abstract 018: Two-year Changes of Adiponectin after Bariatric Surgery and Their Association with Incident Type 2 Diabetes, Cardiovascular Disease, Cancer and Mortality: Results from the Swedish Obese Subjects Study

Abstract

Introduction: Adiponectin has anti-inflammatory, insulin-sensitising and atheroprotective effects in rodents. Although serum adiponectin is uniformly downregulated in obesity, its clinical relevance in humans seems more complex. It is not known whether changes in circulating adiponectin predict type 2 diabetes, cardiovascular disease, cancer and mortality in an obese population. Hypothesis: We hypothesised that adiponectin levels are upregulated substantially after weight loss following bariatric surgery and that pronounced increases of adiponectin should offer better protection for individuals against type 2 diabetes. In addition, findings for type 2 diabetes should be compared to associations with cardiovascular disease, myocardial infarction, stroke, cancer and mortality. Methods: Serum concentrations of total adiponectin were measured in 3,223 participants of the Swedish Obese Subjects (SOS) Study (1,533 in the bariatric surgery group: 229 with gastric bypass, 1056 with vertical banded gastroplasty, 248 with adjustable gastric banding; 1,690 in the control group without surgery) at study baseline and after 2 years. Hazard ratios (HR) and 95% confidence intervals (CI) per 1 standard deviation (SD) of 2-year changes (concentration at year 2 - concentration at baseline) in adiponectin were calculated for incident type 2 diabetes, cardiovascular disease, myocardial infarction, stroke, cancer and mortality in the combined surgery group. Numbers of cases were 93, 122, 78, 55, 82 and 179, respectively. Median follow-up times ranged from 10 years for diabetes up to 16 years for mortality. Results: Mean (SD) levels of adiponectin at baseline were 7,453 (4,150) ng/ml in the combined surgery group and 8,247 (4,846) ng/ml in the control group. During the first 2 years of follow-up, adiponectin levels increased in the surgery group by 4,850 (5,387) ng/ml (parallel to a loss of 24% of body weight) and decreased slightly by 270 (2,650) ng/ml in the control group (parallel to a slight gain of 0.1% body weight). The degree of correlation between changes in adiponectin and weight loss in kg was more pronounced in the surgery groups compared with the control group (p=0.001 for interaction). Two-year increases in adiponectin in the surgery group were associated with decreased risk of type 2 diabetes (HR [95% CI] 0.61 [0.38-0.98], adjusted for baseline data for age, sex, BMI, lipids, blood pressure, alcohol consumption, smoking, anti-hypertensive drugs, glucose, insulin), but not with cardiovascular disease, myocardial infarction, stroke, cancer and mortality (adjusted HR between 0.89 and 1.05). Conclusions: Weight loss after bariatric surgery is paralleled by a substantial increase in circulating adiponectin. The degree of upregulation of adiponectin is associated with protection against future type 2 diabetes, but not with the incidence of cardiovascular outcomes, cancer or mortality.