Abstract 11807: Phenotyping Family History of Coronary Heart Disease to Inform Risk Assessment for Cardiovascular Events in MESA

Abstract

Introduction: The relationships between various family history (FH) of coronary heart disease (CHD) phenotypes and cardiovascular disease (CVD) events have not been fully explored. Hypothesis: The association between FH and CVD will differ by, 1) FH at any age vs. premature FH in a parent, sibling, or child, and 2) familial risk score strata (weak, moderate, or strong; see Table for details). Methods: We ascertained CVD events (CHD, angina, stroke, peripheral arterial disease [PAD], and heart failure [HF]) over a median of 10.1 years. Multivariable Cox models examined the association of FH with CVD events (individual and composite). The C-statistic assessed the prognostic impact of FH for predicting CVD when added to traditional RFs. Results: Of 6192 adults [mean (±SD) age 62±10 years, 48% men, 40% white] with information on FH, 36% reported any FH. Those with strong familial risk (n=1240) were more likely to be diabetic, overweight, smoke, or take medication for hypertension and dyslipidemia. Overall, 741 CVD events occurred. Compared to no FH, any FH was associated with incident CHD, angina, and composite CVD in adjusted models [hazard ratio (95% CI)]: 1.4 (1.1-1.8), 1.6 (1.2-2.1), and 1.3 (1.1-1.5), respectively]. Nearly identical results were obtained for premature FH ( Table ). Compared to weak familial risk, strong risk was also associated with CHD, angina and CVD. However, there was no association between FH of any type and stroke, PAD, or HF. FH status (any or premature) and familial risk score improved discrimination of CVD (p =0.0005, p=0.005, and p=0.001, respectively). Familial risk score provided marginal discrimination over and above premature FH status (C statistic from 0.742 to 0.744, p=0.051). Conclusion: The association between FH and CVD risk differs depending on the number of affected family members, suggesting the value of detailed FH information to refine risk assessment. The predictive value of FH of CHD for non-coronary events is limited.