LOW GLUCOSE IS A HIDDEN CARDIOVASCULAR RISK FACTOR IN APPARENTLY HEALTHY YOUNG ADULTS

Abstract

Therapeutic Area: ASCVD/CVD Risk Factors Background: Previously we reported that the impact of fasting insulin on future CVD risk is modified by waist circumference (WC) and fasting glucose. Here, we tested the hypothesis that low-normal fasting glucose (<82 mg/dL) is a hidden CVD risk factor in specific subpopulations of healthy young adults. Methods: This is a retrospective cohort study of CARDIA: Coronary Artery Risk Development in Young Adults. The parent study enrolled 5,114 participants, ages 18-30 at baseline. In this retrospective analysis, the baseline exclusion criteria were CVD, diabetes, pregnancy, hypertriglyceridemia hyperglycemia, low HDL, or fasting <8 hours. The data for 3,292 participants without metabolic syndrome or prediabetes were analyzed using Cox proportional hazard regression in Stata 17.0. The primary outcome was time-to-incident CVD, defined as fatal/nonfatal MI, coronary revascularization, acute coronary syndrome, CHF, stroke, TIA, carotid or peripheral artery disease. Covariates were the 2019 ACC/AHA risk factors for ASCVD: increased WC, family history of premature ASCVD, hypertension, physical inactivity, LDL ≥160, nicotine use (serum cotinine), race and sex. Cox results were reported as hazard ratio (HR) with 95% confidence interval (CI) and Harrell's c-statistic (c). Results: Cox Model 1 included only the ACC/AHA risk factors; c=0.714. Model 2 added insulin, GGT and platelet count to Model 1; c=0.729. Model 3 added categorical glucose (high-normal vs. low-normal) to Model 2: glucose HR=0.9, 95% CI: 0.7, 1.3, p=0.576, c=0.729. Model 4 added a three-way categorical interaction between glucose, insulin, and WC; HR=2.5, 95% CI: 1.2, 5.0, p=0.011 for the low-insulin/low-WC/low-normal glucose group relative to the reference group with high-normal glucose; c=0.740. Model 5 was identical to Model 4, except it excluded individuals with glucose <65: HR=2.4, 95% CI: 1.2, 4.8, p=0.018, c=0.740. Conclusions: At first glance, normal-range fasting glucose appeared as a non-significant factor when assessing future CVD risk in young adults. However, its significance was unveiled when a categorical interaction between glucose, insulin and WC was introduced. Surprisingly, the combination of low-normal glucose (<82 mg/dL), low fasting insulin and low WC more than doubled the risk for future CVD, as compared with high-normal glucose (82-99 mg/dL). Further investigation is warranted.