Abstract 17413: Plasma Lipidomic Profiles in Two Large Independent Cohorts Improve Upon Conventional Risk Factors to Predict Cardiovascular Events
Introduction: Secondary prevention in cardiovascular disease (CVD) consumes considerable resources due in part to our limited ability to stratify risk in this group. We used lipidomics to evaluate plasma lipid species as additional markers of CVD risk in secondary prevention.
Methods: Plasma lipids (345 species) were measured using mass spectrometry in a sub-cohort (n=5991) from the Long-Term Intervention with Pravastatin in Ischemic Disease (LIPID) Study. There were 1359 participants who had major CVD events (non-fatal MI, non-fatal stroke and CVD death) in the ~eight-year follow-up period. Cox regression was used to identify lipids associated with CVD events and death. Multivariable risk classification models combining lipid species with conventional clinical risk-factors were developed. C-statistics and net reclassification improvement (NRI) were used to evaluate the ability of plasma lipids to improve upon clinical risk-factors. Results were validated on an independent sub-cohort (3,779) from the Action in Diabetes and Vascular disease: preterAx and diamicroN-MR Controlled Evaluation (ADVANCE) Study.
Results: Sphingolipids, phospholipids, cholesteryl esters and glycerolipids were associated with future CVD events (55 species) and CVD death (95 species). Compared to a base model to predict CVD death (22 risk-factors/medications), the addition of eight lipid species resulted in an increased c-statistic (0.703 to 0.741 (95%CI, 0.739-0.742)), a NRI of 10.3% (95%CI, 9.9%-10.6%) based on a categorical model of <10, 10-15, and >15% 5-year risk, and a continuous NRI of 42.0% (95%CI, 41.4%-42.6%). The addition of lipids to risk-factors also resulted in a significant improvement in the prediction of CVD events. Multivariable models to predict CVD death adding the same eight lipid species to risk factors in the ADVANCE Study showed an increased c-statistic (0.765 to 0.782), a categorical NRI of 10.2% and a continuous NRI of 39.2. Significant improvement in model performance to predict CVD events was also observed.
Conclusion: The improvement in the classification of risk of CVD events and death, above clinical risk-factors, demonstrates the potential of plasma lipid species as biomarkers for CVD risk stratification in secondary prevention.
Author Disclosures: P.A. Mundra: None. C.K. Barlow: None. P.J. Nestel: None. E.H. Barnes: None. A. Kirby: None. P. Thompson: None. D.R. Sullivan: None. Z.H. Alshehry: None. N.A. Mellett: None. K. Huynh: None. M.J. McConville: None. S. Zoungas: None. G.S. Hillis: None. J. Chalmers: None. M. Woodward: None. G. Wong: None. B.A. Kingwell: None. J. Simes: None. A.M. Tonkin: None. P.J. Meikle: Other; Modest; Zora Biosciences.
- © 2016 by American Heart Association, Inc.