Open Access Highly Accessed Open Badges Research article

The contribution of a 9p21.3 variant, a KIF6 variant, and C-reactive protein to predicting risk of myocardial infarction in a prospective study

Dov Shiffman1*, Ellen S O'Meara2, Charles M Rowland1, Judy Z Louie1, Mary Cushman3, Russell P Tracy4, James J Devlin1 and Bruce M Psaty5

Author Affiliations

1 Celera, 1401 Harbor Bay Parkway, Alameda, CA 94502 USA

2 Group Health Research Institute, Seattle, WA 98101 USA

3 Department of Medicine and Pathology, University of Vermont, Colchester Research Facility, Colchester VT 05446 USA

4 Departments of Pathology and Biochemistry, College of Medicine, University of Vermont, Burlington, VT 05405-0068 USA

5 Departments of Medicine, Epidemiology, and Health Services, University of Washington, Seattle, WA 98115 USA

For all author emails, please log on.

BMC Cardiovascular Disorders 2011, 11:10  doi:10.1186/1471-2261-11-10

Published: 15 March 2011



Genetic risk factors might improve prediction of coronary events. Several variants at chromosome 9p21.3 have been widely reported to be associated with coronary heart disease (CHD) in prospective and case-control studies. A variant of KIF6 (719Arg) has also been reported to be associated with increased risk of CHD in large prospective studies, but not in case-control studies. We asked whether the addition of genetic information (the 9p21.3 or KIF6 variants) or a well-established non-genetic risk factor (C-reactive protein [CRP]) can improve risk prediction by the Framingham Risk Score (FRS) in the Cardiovascular Health Study (CHS)--a prospective observational study of risk factors for cardiovascular disease among > 5,000 participants aged 65 or older.


Improvement of risk prediction was assessed by change in the area under the receiver-operator characteristic curve (AUC) and by net reclassification improvement (NRI).


Among white participants the FRS was improved by addition of KIF6 719Arg carrier status among men as assessed by the AUC (from 0.581 to 0.596, P = 0.03) but not by NRI (NRI = 0.027, P = 0.32). Adding both CRP and 719Arg carrier status to the FRS improved risk prediction by the AUC (0.608, P = 0.02) and NRI (0.093, P = 0.008) in men, but not women (P ≥ 0.24).


While none of these risk markers individually or in combination improved risk prediction among women, a combination of KIF6 719Arg carrier status and CRP levels modestly improved risk prediction among white men; although this improvement is not significant after multiple-testing correction. These observations should be investigated in other prospective studies.