Abstract MP055: Early and Long-Term Average Lipid Levels, As Compared With Contemporary Lipid Levels, Are More Strongly Associated with Coronary Artery Calcium in the Framingham Heart Study
Introduction We hypothesized that early and long-term average lipid levels, as compared with contemporary lipid measures, would be more strongly associated with coronary artery calcium (CAC), a marker of atherosclerosis burden, in the Framingham Heart Study (FHS). We sought to further examine the cumulative impact of genetic variants known to alter lipid levels on these associations.
Methods FHS Offspring Cohort participants (n=1156, 44%M, 63±9 yrs) underwent serial fasting lipid profiles [low-density lipoprotein (LDL), high-density lipoprotein (HDL), and triglycerides (TG)] from Exam 1 (1971-1975) to Exam 7 (1998-2001). CAC was measured by multi-detector computed tomography (2002-2005). Lipid genetic risk scores for each lipid subtype were computed from significant single nucleotide polymorphism associations reported by the Global Lipids Genetic Logistic Consortium. Logistic regression assessed the association between early, long term average (∼30 yrs), and contemporary lipid measurements and the 75th percentile of CAC. C statistics assessed the impact of addition of lipid genetic risk score to models.
Results For each lipid subtype, Exam 1 and average lipid measurements were associated with elevated CAC, with the associations stronger than that for Exam 7 measurements (TABLE). When we examined association of genetic risk scores with CAC, only the LDL genetic risk score was associated with CAC (age- and sex-adjusted OR 1.14, 95% CI 1.00-1.29, p=0.04). However, addition of the genetic risk score to the lipid models did not result in a statistically significant increase in the OR (TABLE) or C-statistic (not shown) for any of the lipid measures.
Conclusion Early and long-term average lipid levels are more strongly associated with elevated CAC when compared to contemporary lipid measures, with lipid genetic risk scores adding only marginal information. Our findings suggest that early and long-term average lipid levels should be considered as important information when assessing patient risk for CAD and CV events.
|Model||OR (95% CI)||P value|
|LDL, Exam 1|
|MV + LDL genetic risk score||1.41 (1.23-1.62)||<0.0001|
|LDL, Exam 7|
|MV + lipid treatment (Exam 7)||1.06 (0.93-1.20)||0.41|
|MV + lipid treatment + LDL genetic risk score||1.04 (0.91-1.18)||0.56|
|LDL, Average Exams 1-7|
|MV + lipid treatment (any Exam)||1.24 (1.09-1.42)||0.002|
|MV + lipid treatment + LDL genetic risk score||1.23 (1.07-1.42)||0.004|
|HDL, Exam 1|
|MV + HDL genetic risk score||0.85 (0.75-0.97)||0.01|
|HDL, Exam 7|
|MV + lipid treatment (Exam 7)||0.91 (0.80-1.03)||0.13|
|MV + lipid treatment + HDL genetic risk score||0.92 (0.80-1.05)||0.19|
|HDL, Average Exams 1-7|
|MV + lipid treatment (any Exam)||0.84 (0.74-0.95)||0.007|
|MV + lipid treatment + HDL genetic risk score||0.84 (0.74-0.96)||0.01|
|TG, Exam 1|
|MV + TG genetic risk score||1.36 (1.19-1.56)||<0.0001|
|TG, Exam 7|
|MV + lipid treatment (Exam 7)||1.19 (1.05-1.35)||0.008|
|MV + lipid treatment + TG genetic risk score||1.18 (1.03-1.34)||0.01|
|TG, Average Exams 1-7|
|MV + lipid treatment (any Exam)||1.36 (1.18-1.56)||<0.0001|
|MV + lipid treatment + TG genetic risk score||1.36 (1.18-1.57)||<0.0001|
MV= multivariable: adjusted for age and sex.
- © 2012 by American Heart Association, Inc.