Abstract 12: Associations Between Long-Term Body Mass Index Variability and Adverse Cardiovascular Outcomes: Million Veteran Program Study

Abstract

Objective: Body mass index (BMI) is a commonly used estimate of obesity that is a complex multifactorial trait resulting from genetic and lifestyle factors. Weight gain and loss could correlate with adverse biological processes. Prior studies have reported that increased BMI variability predicts future adverse cardiovascular events. However, there have been no prior studies that have investigated this association in patients without cancer or preexisting cardiovascular disease among multiple ethnicities. Using the Million Veteran Program (MVP), a mega-biobank with longitudinal health record data from a large US healthcare system, we evaluated the association between BMI variability and incident cardiovascular events in an ethnically diverse longitudinal cohort (2011-2019). Methods: We studied 92,363 adults free of cardiovascular disease and cancer at baseline. BMI was longitudinally assessed from the baseline visit. BMI variability was assessed using the standard deviation (SD) and coefficient of variation (CV) of each participant. Cardiovascular disease (CVD) risk factors including age, sex, systolic blood pressure, total cholesterol, high-density lipoprotein (HDL) cholesterol, smoking status, diabetes status and statin use were assessed as covariates. We followed patients for the composite CVD events of incident non-fatal myocardial infarction, acute ischemic stroke, and cardiovascular death. Cox proportional hazards modeling was used to examine the association between BMI variability and incident CVD events after adjustment for dichotomous and continuous measures. We also assessed if results were affected sensitive to adjustment for the BMI polygenic score. Results: The cohort (88% men) included 60,180 non-Hispanic White (NHW), 22,488 non-Hispanic Black (NHB), and 9,695 Hispanic (HIS) participants with mean age 56.8 years. Patients were followed from 2011 to 2019. We observed 4,855 composite CVD events. The CV of BMI was associated with incident composite CVD across all groups, but to a stronger degree for Hispanic participants. Among the NHW, each SD increase of CV (indicative of greater BMI variability) was associated with a 15% rise in the risk of CVD events (i.e HR 1.15, 95%CI, 1.11-1.19 after adjustment of covariates. Among the NHB and HIS, results were similar: HR, 1.16 (95%CI, 1.09 - 1.22) and HR, 1.24 (95%CI, 1.12-1.38), respectively. In sensitivity analyses, we found that BMI variability is similarly predictive of the composite outcome after adjustment for BMI polygenic score. Conclusions and Relevance: Increased BMI variability is an independent risk factor for future CVD events.