Body Mass Index, Change in Weight, Body Weight Variability and Outcomes in Type 2 Diabetes Mellitus (from the ACCORD Trial)

Abstract

Weight management is highly recommended to patients with diabetes mellitus. However, this prescription is often characterized by weight fluctuations. It remains unclear the effects of weight fluctuations on outcomes in diabetes mellitus. We used the public use dataset from the Action to Control Cardiovascular Risk in Diabetes (ACCORD) trial to assess the effects of baseline weight, change in weight, and body weight variability (BWV) on outcomes. The ACCORD trial participant's weights were documented annually during the trial. Our primary predictor variables were baseline weight, change in body weight (Initial - final) and BWV defined as average successive variability in weight (average absolute difference between successive values) during the trial. Cox proportional hazards model was used. Out of the 10,251 ACCORD participants, 911(8.9%), 2985(29.1%), and 6355(62%) were normal weight, overweight, and obese. After a mean of 3.5 years of follow-up, 10.2% had the primary outcome (nonfatal MI or nonfatal stroke or CV death), 4.3% had heart failure, 7% died, and 60.7% reported a microvascular complication. BWV was associated with the primary outcome, heart failure, death, and microvascular events in our full models which included BMI [HR (95%CI): 1.25(1.15 to 1.36), 1.59(1.45 to 1.75), 1.74(1.63 to 1.85) and 1.18(1.13 to 1.22), p < 0.0001 respectively). Participants who died were in the quartile that gained the most weight. In this post hoc analysis of ACCORD trial, body weight variability was significantly associated with poor outcomes independent of CVD risk factors and BMI. Our study is consistent with significant risk associated with weight fluctuations in patients with diabetes mellitus.