Correspondence

Genomewide Association Studies of Stroke

N Engl J Med 2009; 361:722August 13, 2009

Article

To the Editor:

The study by Ikram et al. (April 23 issue)1 was designed to provide information about the risk factors and prevalence of cardiovascular diseases, such as coronary heart disease and stroke. Risk factors and pathophysiological features of coronary heart disease and stroke are similar, and the two conditions frequently coexist. In one prospective study, coronary heart disease was present in 53% of patients with atherothrombotic stroke, and atherothrombotic stroke was present in 32% of patients with coronary heart disease.2 The basic assumption of the Cox proportional-hazards model is that the censoring of data from a study population occurs independently of events of interest. However, if the authors censored data from patients with coronary heart disease, such a process would probably result in a preferential censoring of data from patients who might otherwise have gone on to have a stroke. Thus, such censoring would weaken the power of the study to detect loci conferring a risk of both coronary heart disease and stroke. Ikram et al. do not describe the characteristics of patients for whom data were censored, nor did they detect the locus at 9p213-5 that has previously been associated with a risk of both coronary heart disease and stroke.

Jongoh Kim, M.D.
Young Kwang Chae, M.D.
Albert Einstein Medical Center, Philadelphia, PA
dr.jongoh.kim@gmail.com

5 References

1. 1

   Ikram MA, Seshadri S, Bis JC, et al. Genomewide association studies of stroke. N Engl J Med 2009;360:1718-1728
   Full Text | Web of Science | Medline

2. 2

   Aronow WS, Ahn C. Prevalence of coexistence of coronary artery disease, peripheral arterial disease, and atherothrombotic brain infarction in men and women > or = 62 years of age. Am J Cardiol 1994;74:64-65
   CrossRef | Web of Science | Medline

3. 3

   McPherson R, Pertsemlidis A, Kavaslar N, et al. A common allele on chromosome 9 associated with coronary heart disease. Science 2007;316:1488-1491
   CrossRef | Web of Science | Medline

4. 4

   Matarin M, Brown WM, Singleton A, Hardy JA, Meschia JF. Whole genome analyses suggest ischemic stroke and heart disease share an association with polymorphisms on chromosome 9p21. Stroke 2008;39:1586-1589
   CrossRef | Web of Science | Medline

5. 5

   Luke MM, O'Meara ES, Rowland CM, et al. Gene variants associated with ischemic stroke: the Cardiovascular Health Study. Stroke 2009;40:363-368
   CrossRef | Web of Science | Medline

Author/Editor Response

We continued to follow subjects in whom coronary artery disease developed. Data from all subjects in our study (except those who died of stroke) were censored only at the time of death, including death from coronary disease, and at their last evaluation if they were lost to follow-up. All genetic studies of stroke have this potential survivor bias, although studies of younger subjects might be less susceptible to it. Of the four cohorts we analyzed, the Atherosclerosis Risk in Communities (ARIC) cohort was the youngest, and the Cardiovascular Health Study (CHS) excluded subjects who had clinical coronary heart disease at baseline. Sensitivity analyses showed that the exclusion of data from either of these studies did not alter our overall results. For single-nucleotide polymorphism (SNP) rs1537378 (the SNP that was previously reported to be most strongly associated with ischemic stroke at chromosome 9p21),1-5 we observed a weak but nonsignificant association in the expected direction (hazard ratio, 1.06; P=0.22 for all studies; hazard ratio, 1.08; P=0.67 with the exclusion of ARIC; hazard ratio, 1.07; P=0.75 with the exclusion of CHS). After reviewing our data for all eight SNPs that have been associated with ischemic stroke at chromosome 9p21, we observed that only one, rs7044859, showed a significant association (hazard ratio, 1.13; P=0.008). This association was rendered nonsignificant after correction for multiple testing (P=0.06).

M. Arfan Ikram, M.D., Ph.D.
Erasmus Medical Center, Rotterdam, the Netherlands

Sudha Seshadri, M.D.
Boston University School of Medicine, Boston, MA
suseshad@bu.edu

Joshua C. Bis, Ph.D.
University of Washington, Seattle, WA

for the Cohorts in Heart and Aging Research Consortium

5 References

1. 1

   Gschwendtner A, Bevan S, Cole JW, et al. Sequence variants on chromosome 9p21.3 confer risk for atherosclerotic stroke. Ann Neurol 2009;65:531-539
   CrossRef | Web of Science | Medline

2. 2

   Lemmens R, Abboud S, Robberecht W, et al. Variant on 9p21 strongly associates with coronary heart disease, but lacks association with common stroke. Eur J Hum Genet 2009 March 25 (Epub ahead of print).
   

3. 3

   Di Castelnuovo A, Pezzini A, Latella MC, Lichy C, Iacoviello L. Polymorphisms in chromosome 9 and risk of ischemic stroke in two European white populations, and a meta-analysis. J Thromb Haemost 2009;7:365-367
   CrossRef | Web of Science | Medline

4. 4

   Zee RY, Ridker PM. Two common gene variants on chromosome 9 and risk of atherothrombosis. Stroke 2007;38:e111-e111
   CrossRef | Web of Science | Medline

5. 5

   Karvanen J, Silander K, Kee F, et al. The impact of newly identified loci on coronary heart disease, stroke and total mortality in the MORGAM prospective cohorts. Genet Epidemiol 2009;33:237-246
   CrossRef | Web of Science | Medline