Abstract 18516: Common Atrial Fibrillation Susceptibility Alleles Enhance Clinical Risk Prediction Model for Postoperative Atrial fibrillation
Introduction: Atrial fibrillation (AF) is a common adverse event following coronary artery bypass graft (CABG) surgery. The aim of this study was to evaluate whether single nucleotide polymorphisms (SNPs) conferring AF susceptibility identified by genome wide association (GWA) studies improve the traditional clinical risk prediction model for postoperative AF.
Methods: Our study population consisted of 829 prospectively enrolled patients in the Vanderbilt Cardiac Surgery Registry (VCSR) undergoing CABG without concurrent valve surgery. Postoperative AF (12 months post surgery) was evaluated by review of medical records, ECGs and Holter monitors. Clinical variables in the model included age at the time of CABG, race, gender, history of AF, diabetes, hypertension, heart failure, CAD, LVEF, PR interval and use of beta-blockers, ACE inhibitors, statins, aspirin, and COX inhibitors. Thirteen AF-associated SNPs (rs13376333 and rs6666258 at 1q21; rs3903239 at 1q24; rs2200733, rs10033464 and rs6817105 at 4q25; rs3807989 at 7q31; rs10821415 at 9q22; rs10824026 at 10q22; rs1152591 at 14q23; rs7164883 at 15q24; rs2106261 and rs7193343 at 16q22) were genotyped using the Sequenom platform.
Results: Subjects with two or more missing variables were excluded from the analysis (final cohort 444 subjects). SNP rs10033464 was not in Hardy-Weinberg equilibrium and hence removed from the analysis. Using multiple logistic regression, a model containing 12 SNPs, and all clinical covariates was fitted and found to be highly significant (P<0.001). A likelihood ratio test comparing the complete model (clinical and genetic covariates) to a reduced model containing only clinical variables was also highly significant (P<0.001), indicating an improvement in model fit with the inclusion of AF susceptibility alleles. The area under the receiver operator characteristic curve significantly improved from 0.759 (clinical variables only) to 0.816 (combined), (P<0.001).
Conclusion: Our data show that AF susceptibility alleles enhance the clinical risk prediction model for the development of AF after cardiac surgery. Pre-procedural genotyping of AF susceptibility alleles may improve risk stratification in patients undergoing CABG surgery.
- © 2012 by American Heart Association, Inc.