Abstract P018: Overweight Is Not A Cardiovascular Risk Factor For Insulin-sensitive Individuals: CARDIA 30-year Follow Up

Abstract

Introduction: Overweight and obeseare widely reported risk factors for atherosclerotic cardiovascular disease(ASCVD). However, their true impact ondisease risk may be complicated by interactions between measures of adiposityand insulin resistance. Hypothesis: Due to effect modification, the degree towhich overweight or obese increases ASCVD risk depends on fasting insulinstatus. Specific Aim: Determine the effects of overweight and obeseon incident ASCVD by estimating Cox hazard ratios for BMI categories, includingpotential interactions with fasting insulin. Methods: Cox proportional hazards regression models weredeveloped using 30-year follow up datafrom the Coronary Artery Risk Development in Young Adults (CARDIA) study. After excluding participants with baselinediabetes or cardiovascular disease, or those fasting <8 hours or pregnant, atotal of 4,555 subjects were analyzed. The primary outcome was incident ASCVD, defined as any fatal ornon-fatal MI, stroke, TIA, coronary revascularization, non-MI acute coronarysyndrome, CHF, carotid or peripheral artery disease. Cox models were developed, adjusting forcanonical ACC/AHA risk factors and the potential interaction between BMI and fastinginsulin. The reported effect size ishazard ratio (HR) with 95% confidence intervals (CI) and p-values. Results: Cox model 1 adjusted for family history ofpremature ASCVD, metabolic syndrome criteria, LDL-C >160 mg/dL, nicotine use(serum cotinine), chronic kidney disease (eGFR<60), history of earlymenopause, fitness level by treadmill, race, fasting insulin and BMI category. In model 1, both overweight (HR 1.46, 95% CI:1.07, 1.99; p=0.0158) and obese (HR 2.21, 95% CI: 1.52, 3.21; p<0.0001) appearedto be independently associated with incident ASCVD. Cox model 2 included an interacting categoricalvariable that included both BMI andfasting insulin. In model 2, overweightwas not a risk factor for CARDIA participants with fasting insulin values inthe lower half (<9.0 μIU/mL): HR 1.36, 95% CI: 0.81,2.30, p=0.2482. However, overweight wasa significant risk factor for those with fasting insulin in the upperhalf: HR 2.21, 95% CI: 1.51, 3.24, p<0.0001. This result was cross-checked and confirmed using otherwise identical modelsstratified by fasting insulin level. Thehazard ratios were 1.34 (95% CI: 0.79, 2.27, p=0.2818) for normoinsulinemic,overweight individuals and 1.51 (95% CI 1.03, 2.21, p=0.0367) forhyperinsulinemic, overweight individuals, as compared with those with BMI<25. In contrast to overweight, obese was anindependent risk factor for ASCVD regardless of fasting insulin status. Conclusion: Overweight was not an ASCVD risk factor forCARDIA participants with low fasting insulin. If confirmed, these findings could inform tailored weight-lossrecommendations for overweight individuals.