Donate Help Contact The AHA Sign In Home
American Heart Association
Circulation
Search: search_blue_button Advanced Search
« Previous Article | Table of Contents | Next Article »
Circulation. 2007;116:1128-1136
Published online before print August 20, 2007, doi: 10.1161/CIRCULATIONAHA.107.710780
CLINICAL PERSPECTIVE
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Data Supplement
Right arrow Correction
Right arrow Correction (v117,pe9)
Right arrow All Versions of this Article:
116/10/1128    most recent
CIRCULATIONAHA.107.710780v1
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow Request Permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Newton-Cheh, C.
Right arrow Articles by O’Donnell, C. J.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Newton-Cheh, C.
Right arrow Articles by O’Donnell, C. J.
Right arrowPubmed/NCBI databases
*Gene*GEO Profiles
*HomoloGene*SNP
*UniGene
*Genetics Home Reference
Related Collections
Right arrow Electrophysiology
Right arrow Clinical genetics
Right arrow Epidemiology
Right arrow Electrocardiology

(Circulation. 2007;116:1128-1136.)
© 2007 American Heart Association, Inc.

Genetics

Common Genetic Variation in KCNH2 Is Associated With QT Interval Duration

The Framingham Heart Study

Christopher Newton-Cheh, MD, MPH; Chao-Yu Guo, PhD; Martin G. Larson, ScD; Stacy L. Musone, BA; Aarti Surti, BA; Amy L. Camargo, BA; Jared A. Drake, BA; Emelia J. Benjamin, MD, ScM; Daniel Levy, MD; Ralph B. D’Agostino, Sr, PhD; Joel N. Hirschhorn, MD, PhD*; Christopher J. O’Donnell, MD, MPH*

From the National Heart, Lung and Blood Institute’s Framingham Heart Study, Framingham, Mass (C.N.-C., C.Y.G., M.G.L., E.J.B., D.L., C.J.O.); Broad Institute of Harvard and MIT (C.N.-C., S.M., A.S., A.L.C., J.N.H.), Cambridge, Mass; Cardiology Division (C.N.-C., C.J.O.), Massachusetts General Hospital, Boston; Harvard Medical School (C.N.-C., D.L., J.N.H., C.J.O.), Boston, Mass; Department of Medicine (D.L.), Beth Israel-Deaconess Hospital, Boston, Mass; Evans Department of Medicine and Whitaker Cardiovascular Institute (E.J.B.), Boston University School of Medicine, Boston, Mass; Department of Mathematics and Statistics (C.Y.G., M.G.L., R.B.D.), Boston University, Boston, Mass; National Heart, Lung, and Blood Institute (D.L., C.J.O.), Bethesda, Md; and Children’s Hospital of Boston (J.D., J.N.H.), Boston, Mass.

Correspondence to Christopher Newton-Cheh, MD, MPH, Cardiology Division, Massachusetts General Hospital, 32 Fruit St, GRB847, Boston, MA 02114 (e-mail cnewtoncheh{at}partners.org); or Christopher J. O’Donnell, MD, MPH, NHLBI’s Framingham Heart Study, 73 Mt Wayte Ave, Suite 2, Framingham, MA 01702–5827 (e-mail odonnellc@nhlbi.nih.gov).

Received April 21, 2007; accepted July 10, 2007.


*    Abstract
up arrowTop
 *Abstract
down arrowIntroduction
 down arrowMethods
 down arrowResults
 down arrowDiscussion
 down arrowReferences
 
Background— QT prolongation is associated with increased risk of sudden cardiac death in the general population and in people exposed to QT-prolonging drugs. Mutations in the KCNH2 gene encoding the HERG potassium channel cause 30% of long-QT syndrome, and binding to this channel leads to drug-induced QT prolongation. We tested common KCNH2 variants for association with continuous QT interval duration.

Methods and Results— We selected 17 single nucleotide polymorphisms and rs1805123, a previously associated missense single nucleotide polymorphism, for genotyping in 1730 unrelated men and women from the Framingham Heart Study. rs3807375 genotypes were associated with continuous QT interval duration in men and women (2-df P=0.002), with a dominant model suggested (P=0.0004). An independent sample of 871 Framingham Heart Study men and women replicated the association (1-sided dominant P=0.02). On combined analysis of 2123 subjects, individuals with AA or AG genotypes had a 0.14-SD (SE, 0.04) or 3.9-ms higher age-, sex- and RR-adjusted QT interval compared with GG individuals (P=0.00006). The previously reported association of rs1805123 (K897T) replicated under a dominant (AA/AC, 0.12 SD [SE, 0.07] or 3.1 ms higher versus CC; 1-sided P=0.04) or additive model (0.06 SD [SE, 0.03] or 1.6 ms higher per A allele; 1-sided P=0.01).

Conclusions— Two common genetic variants at the KCNH2 locus are associated with continuous QT interval duration in an unselected community-based sample. Studies to determine the influence of these variants on risk of sudden cardiac death and drug-induced arrhythmias should be considered.


Key Words: arrhythmia • electrocardiography • genetics • ion channels • long-QT syndrome


*    Introduction
up arrowTop
 up arrowAbstract
 *Introduction
down arrowMethods
 down arrowResults
 down arrowDiscussion
 down arrowReferences
 
Sudden cardiac death results in 250 000 to 350 000 deaths in the United States annually.1,2 QT interval prolongation, reflecting abnormal myocardial repolarization time, is a risk factor for sudden cardiac death and cardiovascular events in the general population in most but not all studies3–10 and for sudden infant death syndrome.11 QT interval prolongation is a complex cardiovascular trait with multiple environmental and genetic contributors, including female sex, older age, left ventricular hypertrophy, drug exposure, and genetic factors.12 Congenital long-QT syndrome (LQTS) appears to underlie a fraction of all sudden cardiac death, particularly absent structural heart disease,13,14 as well as a substantial fraction of sudden infant death syndrome.15 QT prolongation with resulting torsade de pointes is a rare but lethal side effect of many noncardiac medications and a major barrier to drug development.16,17 Identification of the genetic determinants of QT interval variation could improve our understanding of the pathophysiology of sudden death and suggest targets for improved pharmacotherapy and risk stratification.18

Clinical Perspective p 1136

Whereas highly penetrant mutations in ion channel genes can result in LQTS, most individual mutations are rare and do not contribute significantly to the population risk of sudden death.19 Whether common genetic variation that influences QT interval duration can reduce repolarization reserve20 and contribute incrementally to increase the risk of arrhythmias is unknown. We and others have shown that continuous QT interval variation across the range of values has a heritability of 35%, indicating that 35% of its variability is attributable to additive genetic factors.12 We recently identified with collaborators common variation in the nitric oxide synthase 1 adaptor protein (NOS1AP) gene that is reproducibly associated with continuous QT interval variation.21,22 Whether such variants of modest genetic effect on intermediate traits contribute significantly to the population burden of cardiovascular disease remains to be seen, but their prevalence in the general population could make even modest effects important.23

We report here our characterization of common genetic variation in KCNH2 (HERG) encoding the {alpha} subunit of the channel underlying the rapidly activating component of the delayed rectifier (IKr) potassium current. Binding to this potassium channel subunit is a major determinant of drug-induced QT prolongation as either the desired therapeutic effect of antiarrhythmic agents or an unintended side effect of noncardiac drugs. Assays for the affinity of binding to the IKr channel are a routine component of pharmaceutical development, but late recognition of arrhythmogenic potential has been a major barrier to the drug pipeline.24 In LQTS families with identifiable genetic mutations, 40% occur in the HERG channel, resulting in LQT2.25,26 Moreover, studies have reported possible association of a missense single nucleotide polymorphism (SNP) in KCNH2 (rs1805123, K897T) with QT interval duration27–30 and as a possible modifier of congenital LQTS.31

We hypothesized that common variants in KCNH2 alter QT interval duration. Therefore, we genotyped a set of common polymorphisms capturing the majority of common variation at the KCNH2 locus and tested them for association with QT interval duration in participants in the Framingham Heart Study (FHS).


*    Methods
up arrowTop
 up arrowAbstract
 up arrowIntroduction
 *Methods
down arrowResults
 down arrowDiscussion
 down arrowReferences
 
Study Participants
The FHS is a prospective epidemiological study established in 1948 to evaluate potential risk factors for cardiovascular disease.32,33 The offspring cohort participants studied in this report are drawn from the 5124 subjects recruited from 1971 to 1975, at which time routine resting ECGs were obtained during the course of a comprehensive examination.34,35 Original cohort participants attending examination cycle 11 from 1968 to 1971 were eligible for study in the validation sample. The FHS participants are predominantly white of European ancestry with no evidence of gross population substructure using genome-wide SNP data.36 Participants surviving to offspring examination cycle 6 or original examination 25 (1996 to 1997) provided blood samples for DNA extraction, and DNA from a set of 1809 unrelated offspring participants was available for genetic analyses in the first stage. Subjects were excluded for known prevalent coronary heart disease (offspring cohort, 1.3%; original cohort, 10.7%), use of cardiac glycosides or quinidine (0.4% and 3.6%, respectively), atrial fibrillation (0.06% and 1.4%), or technically poor ECGs (0.4% and 0.7%), resulting in a primary study sample of 1730 men and women. For genotyping of a missense SNP K897T (rs1805123), a slightly different DNA plate set was available (>80% of samples in common), totaling 1746 after exclusions. For genotyping of 2 SNPs, an additional independent validation set of up to 871 FHS offspring and original cohort men and women in families was genotyped. All study participants gave written informed consent.

Determination of ECT Intervals
Resting 12-lead ECGs were obtained at 25 mm/s and 0.1 mV/mm on strips of lined paper (Hewlett-Packard, Palo Alto, Calif). Digital calipers were used to measure intervals on scanned ECGs as previously described.12 The QT interval was defined as the time (ms) from onset of the QRS to the return of the T wave to baseline, taking care to exclude U waves if present, in leads II, V2, and V5. If a TU complex was present, the T-wave offset was taken to be the nadir of the curve between the T and U waves. The QT interval phenotype studied was the average of the lead II, V2, and V5 residuals from sex-specific models of QT interval regressed on RR interval and age (see below). We have previously shown these measures to have good reproducibility.12

Clinical Characteristics of the Study Sample
The methods of risk factor assessment have been described elsewhere.37 Each examination included comprehensive assessment of cardiovascular disease, ECG, and blood testing. All covariates included in the present study were measured contemporaneously to the ECG.

Linkage Disequilibrium Characterization and SNP Selection
All genotyping was performed on the Sequenom (San Diego, Calif) platform that resolves allele-specific single-base extension products with mass spectrometry (MALDI-TOF) as previously reported.38,39 We constructed linkage disequilibrium (LD) maps by genotyping 53 SNPs in 93 individuals in multigenerational pedigrees from families of European ancestry recruited in Utah (Centre d’Etude du Polymorphisme Humain [CEPH] reference sample, 96 independent chromosomes). A total of 47 SNPs (89%) met our quality control thresholds (call rate ≥85%; Hardy-Weinberg equilibrium; P>0.01; ≤1 mendelian inconsistency), and 37 were polymorphic with minor allele frequency ≥5%, spanning a total genomic distance of 66 kilobases (kb).

As is typical of the genome, the KCNH2 locus shows strong LD, allowing a parsimonious set of SNPs to be selected to capture most of the common genetic variation. Association of a trait with a genetic variant can be detected if the causal genetic variant is genotyped directly (rarely, by chance) or more commonly if it is correlated with a genotyped SNP or combination of SNP alleles (haplotype). We defined blocks of strong LD in which limited historical recombination has occurred, using the ‘solid spine of LD’ method implemented in HaploView 2.03.40 We inferred phase from unphased genotypes using an expectation-maximization algorithm41 implemented in the TagSNPs program.42 To select tag SNPs capturing most of the genetic variation at the KCNH2 locus, we selected a set of SNPs within each block of strong LD that had an r2 (square of the correlation coefficient) ≥0.8 to all haplotypes with frequency ≥5%.

We genotyped all tag SNPs and all SNPs in the gap between blocks (low LD) in 1730 FHS offspring participants. All SNPs genotyped in FHS were in Hardy-Weinberg equilibrium (1-df {chi}2 test P>0.01) with genotyping >85%. Separately, we genotyped SNP rs1805123 (K897T) reported in the literature for association with QT interval duration in 1746 offspring, of whom >80% overlap with the 1730 tested above, using a modified Sequenom assay. We genotyped rs3807375 (because of nominal association in the initial sample) and rs1805123 in an additional 871 FHS subjects in families.

Statistical Analysis
Using SAS version 8.1 (SAS Institute, Cary, NC), we regressed in multivariable linear models QT interval duration on RR interval and age in sex-specific models for each of the 3 leads (II, V2, V5) in the entire offspring cohort with ECG data (2288 men, 2491 women).43 For example, the model for lead II in men was QTresidual=QTms–230–0.133xRRms–0.477xagey and in women was QTresidual=QTms–223–0.159xRRms–0.316xagey.

QT residuals from models for each lead and sex were standardized to a mean of 0 and SD of 1. Sex-specific averages of the standardized QT residuals from each of the 3 leads were analyzed together in sex-pooled analyses (and as a secondary sex-specific analysis). For validation genotyping in the related sample of FHS families, averaged, standardized residuals from cohort- and sex-specific models from the offspring and original cohorts were analyzed.

QT residuals were screened for association with individual SNPs using a 2-df general test44 of the null hypothesis that mean adjusted QT interval does not differ by genotype group (AA, Aa, aa) using linear regression. To exclude the presence of >1 association signal, we used stepwise selection to sequentially add nominally significant SNPs to a model predicting adjusted QT residuals using SAS. We tested for an interaction of sex with SNPs independently associated with QT interval duration. Association testing of 2 SNPs in the validation sample was performed with PROC MIXED to account for familial relationships. We used a 1-sided probability value for the test of the null hypothesis that the specific genetic effect observed in the derivation sample (rs3807375) or the literature (rs1805123) was not supported. Results in the 2 samples were combined through the use of inverse variance weights.45

QT residuals were tested for association with haplotypes ≥5% in blocks of strong LD with the haplo.stats program.46 The haplo.stats program uses an expectation-maximization algorithm41 to develop a posterior probability distribution of haplotype pairs, conditional on observed single SNP genotypes, and regresses adjusted QT interval residuals on haplotypes. A score statistic is used to determine a global probability value for the test that no haplotype is associated with a different mean QT residual (analogous to ANOVA) and haplotype-specific probability values for the test that a given haplotype is not associated with a different mean QT residual compared with all other haplotypes. We considered the global test nominal probability values the primary haplotype test.

To account for multiple tests (multiple SNPs, sex pooled and sex specific), we developed empirical gene-wide probability values for the single SNP association results. Null data sets were created by randomly sampling with replacement sets of genotypes (considered collectively to preserve their correlation) for assignment to each of the individual QT residuals. We conducted all 51 single SNP tests (3 samples and 17 SNPs) on each null data set and recorded the minimum probability value. The distribution of minimum probability values from 1000 repetitions was used to determine an empirical gene-wide probability value corresponding to the best nominal result.

Association testing of SNP rs1805123, reported to be associated with QT duration in the literature, was considered independently because it was selected to test a specific hypothesis rather than to capture common genetic variation.

The authors had full access to and take full responsibility for the integrity of the data. All authors have read and agree to the manuscript as written.


*    Results
up arrowTop
 up arrowAbstract
 up arrowIntroduction
 up arrowMethods
 *Results
down arrowDiscussion
 down arrowReferences
 
Study Sample Characteristics
Our primary study sample included 859 men and 871 women (n=1730 total) with mean ages of 37.1 and 36.7 years, respectively (Table 1). The mean unadjusted QT interval in lead II was 363 ms (SD, 27 ms) in men and 361 ms (SD, 30 ms) in women. RR interval– and age-adjusted QT interval duration in lead II was longer in women (394 ms) compared with men (380 ms), as has been demonstrated previously.9 Additional clinical characteristics are presented in Table 1.


View this table:
[in this window]
[in a new window]

  		TABLE 1. FHS First-Stage Sample Characteristics

LD in Reference Samples and Tag SNP Selection
Of 47 working SNP assays in the CEPH reference samples, 37 SNPs were polymorphic with a minor allele frequency ≥5% (average inter-SNP distance, 1870 bp; online Data Supplement Table I). Two blocks of strong LD, each containing limited haplotype diversity, were found (the Figure). Of the 66-kb genomic span of our markers (14 kb upstream, 33 kb KCNH2 locus, 19 kb downstream), 94% fell within either of the 2 blocks of strong LD. The observed haplotypes ≥5% in the 2 blocks explained 91% and 82% of the 96 chromosomes tested. The 4-kb region of lower LD between the 2 blocks was saturated with SNPs (average inter-SNP distance, 572 bp). In unrelated Framingham offspring participants, we genotyped 4 and 7 tag SNPs in blocks 1 and 2, respectively, and all 6 nonredundant SNPs in the intervening gap of low LD. Using tagger (www.broad.mit.edu/mpg/tagger/; October 2006),47 we found that 97% of the 37 SNPs were captured at an r2 >0.8 with a mean r2 for all SNPs of 0.97. The tag SNPs genotyped in the FHS sample revealed stronger LD in the gap region than in the CEPH reference samples (online Data Supplement Figure). This difference in LD estimates likely stems from the more precise estimate of LD obtained from 3460 Framingham chromosomes compared with 96 CEPH chromosomes.


Figure 1185945
View larger version (44K):
[in this window]
[in a new window]

  		LD map of KCNH2 locus in CEPH reference sample. The exon structure of 3 known isoforms of KCNH2 is represented schematically in 5' to 3' orientation. The relationship of the 37 SNPs passing quality control in the CEPH reference samples to the genomic orientation of the gene is displayed (gray), along with the subset of SNPs that were polymorphic (blue) and the subset that were genotyped in the FHS samples (pink). Two blocks of strong LD (covering 94% of the genomic span of the locus), separated by a 4-kb region of low LD, are shown. In blocks 1 and 2, 3 and 4 haplotypes explain 91% and 82% of all chromosomes, respectively (labeled 1A through 1C, 2A through 2D). The LD between all pairs of SNPs tested is represented at the bottom of the figure; red square indicates significant LD; white square, weak LD; and light blue square, inadequate power to determine LD. Figure prepared with LocusView 2.0 (T. Petryshen, A. Kirby, M. Ainscow, unpublished software, http://www.broad.mit.edu/mpg/locusview/).

Single SNP Association Results
Of the 17 SNPs tested, we observed a minimum nominal 2-df probability value of 0.002 for intronic SNP 15 (rs3807375; minor allele frequency, 38%) in the sex-pooled analysis (Table 2). Table 2 shows the results of association testing for the sex-pooled primary analysis with nominal P<0.05. The results of all 51 single SNP association tests (primary sex pooled and secondary sex specific) are shown in online Data Supplement Table II. Stepwise selection confirmed that only the single independent signal from SNP 15 was present at the locus; correlation among SNPs contributed to multiple nominally significant probability values. A dominant model (AA+AG versus GG) for the minor A allele best fit the observed data (P=0.0004). The genotype at SNP 15 explained 0.9% of the variation in adjusted QT duration. We found similar evidence of association of SNP 15 in men and women separately (2df P=0.04 for both; online Data Supplement Table II) and no evidence of an interaction with sex (interaction P=0.79).


View this table:
[in this window]
[in a new window]

  		TABLE 2. First-Stage Sex-Pooled Single SNP Association Results for Adjusted QT Interval Duration With Nominal P<0.05

Adjustment for Multiple Testing
We determined an empirical distribution of single SNP association probability values, including both the primary sex-pooled and the secondary sex-specific analyses, using null data sets in which the correlation between each individual’s phenotype and genotype had been broken. A nominal P=0.002 as found for SNP 15 was observed in 63 of 1000 null data sets, corresponding to an empiric gene-wide corrected P=0.063.

Multimarker Haplotype Association Results
We tested haplotypes with ≥5% frequency within 2 blocks of strong LD for association with adjusted QT interval duration. We found a nominal global P=0.01 in block 2. The Hap scores for haplotypes 2C and 2D, each containing the SNP 15 (rs3807375) A allele, are both positive (associated with increased QT interval), consistent with the A allelic effect seen on analysis of the single SNP (Tables 3 and 4Down).


View this table:
[in this window]
[in a new window]

  		TABLE 3. Association of Common Haplotypes Within 2 Blocks of Strong LD in FHS With Adjusted QT Interval Duration, Block 1


View this table:
[in this window]
[in a new window]

  		TABLE 4. Association of Common Haplotypes Within 2 Blocks of Strong LD in FHS With Adjusted QT Interval Duration, Block 2

QT Association of rs3807375 in a Second Sample
We genotyped rs3807375 in an additional 871 FHS men and women independent of the original sample. We replicated the association of the SNP with QT residual under a dominant model (1-sided P=0.02). On combined sex-pooled analysis from the 2 samples (n=2123) using inverse variance weights, carriers of at least 1 A allele had a 0.14-SD (SE, 0.04) -higher QT residual (equivalent to 3.9-ms increase) compared with GG individuals (dominant model, P=0.00006). A 37-year-old man with a heart rate of 60 bpm with a GG genotype has a predicted QT interval duration in lead II of 376.3 ms compared with 381.4 ms for a man with either the AA or AG genotype (Table 5). A 37-year-old woman with a heart rate of 60 bpm and a GG genotype has a predicted adjusted QT interval of 392.3 ms compared with a mean of 395.1 ms for a woman with either the AG or AA genotype (Table 5).


View this table:
[in this window]
[in a new window]

  		TABLE 5. Sex-Specific Mean-Adjusted QT by rs3807375 (SNP 15) Genotype in 2123 FHS Men and Women

Genomic Context of rs3807375
SNP 15 (rs3807375) lies in intron 2 in a sequence that is not highly conserved across species. In our reference sample of 96 CEPH chromosomes, SNP 15 was perfectly correlated (r2=1.0; 95% CI, 0.92 to 1.0) with SNP 28 (rs1805121), a highly conserved synonymous coding SNP (L564L) 18 kb downstream in exon 7. In our CEPH reference sample genotyping, SNP 15 also was correlated with other intronic SNPs: SNP 20 (r2=0.86), SNP 19 (r2=0.73), and SNP 4 (r2=0.52). In 3460 independent FHS chromosomes, rs3807375 showed comparable correlation with SNP 28 (r2=0.97), SNP 20 (r2=0.86), and SNP 19 (r2=0.87). In HapMap samples of European ancestry (CEU), additional SNPs show lesser degrees of correlation: SNP 15 is partially correlated with rs11762978 downstream of KCNH2 (r2=0.67) and rs10277237 upstream of KCNH2 (r2=0.52).

QT Association of rs1805123 (K897T)
In addition to the LD-based approach relating common genetic variation at the KCNH2 locus, we also tested rs1805123 (A/C, SNP 30) responsible for a lysine-to-threonine substitution in amino acid 897 (K897T) that has been found to be associated with QT interval in some studies. We successfully genotyped rs1805123 in 2515 FHS men and women. In 2 prior well-powered reports, modest association of rs1805123 was observed under a dominant model28 or an additive model of increased QTc with A allele carriage.29 Under a dominant model, rs1805123 was associated with a 0.12-SD (SE, 0.07) or 3.1-ms-higher adjusted QT interval in AA or AC individuals compared with CC individuals (1-sided P=0.04). Under an additive model, rs1805123 was associated with a 0.06-SD (SE 0.03) or 1.6-ms-higher adjusted QT interval per A allele copy (1-sided P=0.01). Discrimination between a dominant or an additive model requires a much larger sample size. Our findings thus confirm a modest effect of rs1805123 (K897T) on QT interval duration.


*    Discussion
up arrowTop
 up arrowAbstract
 up arrowIntroduction
 up arrowMethods
 up arrowResults
 *Discussion
down arrowReferences
 
Principal Findings
We defined a parsimonious set of tag SNPs that comprehensively capture common genetic variation at the KCNH2 locus encoding the HERG potassium channel {alpha} subunit. We identified a novel association of a common variant with myocardial repolarization and replicated the previously reported association of a second independent missense variant. We found evidence of association of SNP 15 (rs3807375) with age-, sex-, and RR-adjusted QT interval duration under a dominant model in our large, community-based sample (nominal P=0.00006). We also extend previous work reported in other large community-based samples in our finding of modest supportive evidence for the association of SNP 30 (rs1805123, K897T) with QT interval duration (P=0.01).28,29

rs3807375 Association With QT Interval
We show for the first time that rs3807375, a SNP in intron 2 with a minor allele frequency of 38%, is associated with QT interval duration (P=0.00006). Probability values in genome-wide surveys of common variant–common trait associations likely need to be <10–7 to have a substantial probability of not being false-positives because 10 000 000 common variants48 exist and a minority of them are likely to be functional. However, in a candidate gene such as KCNH2, which is among the top candidates to influence QT interval duration, we have previously estimated the posterior probability of a true-positive result to be {approx}65% for an observed P<10–4.49

In a 2-stage survey of common genetic variants in potassium channels in a large, community-based sample of 3966 Germans from the KORA (Cooperative Health Research in the Region of Augsburg) project, Pfeufer et al29 reported association of 2 variants in KCNH2 with continuous QT interval duration. They found association of rs3815459, which tags a 23% haplotype (2C in our study, h3 in the KORA article), and SNP 30 (rs1805123), which tags a 23% haplotype (2B in our study, h2 in KORA).50 In our study, the minor alleles of SNP 15 (rs3807375) and its perfect proxy 28 (rs1805121) are found on haplotype 2C/h3 and a related 11% haplotype 2D/h4. The association in the KORA study with SNPs on haplotype 2C but not SNP 15 or other correlated SNPs (with minor alleles on both haplotypes 2C and 2D) could stem from the limited power to detect modest effects in a screening stage of 689 subjects or a chance difference. Our observation that SNP 15 shows stronger association with QT than either haplotype has effectively fine mapped the prior association by Pfeufer et al. We are not aware of another study that has reported on the allele that we have identified and QT interval duration.

LD in the human genome allows identification of a functional allele without directly genotyping it through correlation with genotyped variants, but LD makes distinguishing the functional variant from all correlated and equally associated variants difficult. SNP 15 (rs3807375) is perfectly correlated with a synonymous coding SNP 28 (rs1805121) and is partially correlated with other intronic or nongenic variants. Coding SNPs in general are under significant selective constraint, as we and others have shown.50 Some synonymous coding SNPs have been found to interrupt exonic splicing enhancers51 and are generally under selective constraint.52 SNP 28 (rs1805121) does not fall in any of 238 exonic splicing enhancers identified by Fairbrother et al (http://genes.mit.edu/burgelab/rescue-ese/).51 However, deep resequencing is required to fully define the set of all candidate alleles, with much larger sample sizes needed to distinguish among the several correlated alleles at the locus. Genotyping both SNPs 15 and 28 in individuals of African ancestry could allow improved resolution of the signal of association at the locus because Yoruban HapMap samples show substantially less LD at the locus.

rs1805123/K897T Association With QT Interval Duration
We found modest statistical evidence in support of the association of rs1805123 (A/C), encoding a lysine-to-threonine change at amino acid 897 (K897T) with QT interval duration. This effect is distinct from the rs3807375 effect because the variants show minimal correlation (r2=0.18). Several studies have examined the association of this missense SNP in exon 12 in relation to QT interval duration. Pietila et al27 reported a study of 226 men and 187 women from Finland and found an increased QTc to be associated with C allele carriage: AA<(AC+CC). Bezzina et al28 reported a study of 1030 men and women from Augsburg, Germany, and found a higher QTc to be associated with A allele carriage: CC<(AC+AA). Pfeufer et al29 reported a separate sample of 3966 middle-aged men and women from Augsburg and found an increasing QTc with increasing copies of the major A allele (CC<AC<AA). Most recently, Gouas et al30 reported a study of 200 men and 200 women sampled from the extremes of the QTc distribution of a community-based sample in France and found an enrichment of major allele homozygotes in the high-QTc group: (CC+AC)<AA. Although our findings are marginally statistically significant in isolation, when they are considered together with the other published findings in large samples, consistent association exists of increased QT interval duration with genotypes containing the major A allele.

An interesting report by Crotti et al31 demonstrated in cellular assays a modifier effect of the K897T polymorphism on an LQTS-causing mutation in KCNH2 found in an Italian family. The authors hypothesized that the more modest expression observed in the other LQTS-mutation carriers in the family resulted from the absence of the 897T allele, but limited numbers of individuals in each genotype class preclude definitive statements. Cellular assays of the K897T polymorphism, which resides in the carboxy terminus of the HERG channel, have suggested reduced channel expression,53 activation at more negative potentials for the minor 897T allele28,54 with hastened28 activation, and both hastened28,54 and delayed53 inactivation in different assays. Differences in assay methods could underlie apparently conflicting results. Taken in aggregate, both human genetic and cellular electrophysiological data support a functional effect of the K897T polymorphism on myocardial repolarization.

Study Strengths and Limitations
Strengths of our study include the large community-based sample ascertained independently of ECG or other clinical characteristics, the uniformly and reproducibly measured QT interval, and the comprehensive LD-based approach to characterizing common genetic variation at the KCNH2 locus. Limitations clearly apply. Our study sample is of European ancestry, and the relevance of the 2 variants described to populations of different continental ancestry, in whom substantial allele frequency differences are observed in HapMap, remain to be defined. Our study assayed only common variation at this locus and cannot address the role of rare variation at KCNH2 on myocardial repolarization, already well established for mendelian syndromes. Moreover, detecting more modest effects than observed here requires much larger sample sizes. Lastly, use of QT-prolonging medications in the early 1970s was sufficiently uncommon to preclude tests for interaction of KCNH2 variants with such medication use.

Clinical Implications and Future Directions
The identification of 2 common modest QT-modifying alleles in KCNH2 could ultimately have important implications for clinical care. HERG binding is the dominant known mechanism for drug-induced QT prolongation and resultant arrhythmias. Whether polymorphisms in KCNH2 contribute to interindividual variation in susceptibility to this disastrous complication of pharmacotherapy needs to be tested in drug-exposed patient collections. The identification of host factors that contribute to liability to cardiotoxic effects of common medications could improve personalized risk assessment and rational assignment to drug therapy through the development of risk prediction models incorporating both clinical and genetic factors. Ultimately, identifying the full set of variants that modulate an intermediate trait such as QT interval duration will describe a superset of variants to be tested in the general population, in high-risk subsets (eg, heart failure, after myocardial infarction, congenital LQTS), and in those exposed to QT-prolonging drugs to define their impact on clinical events.18 Although modest genetic risk factors such as those found in KCNH2 may ultimately be found to lack clinical discrimination, they can still point the way to a better understanding of myocardial repolarization in the general population.


*    Acknowledgments
 
This study was conducted in the Framingham Heart Study of the National Heart, Lung, and Blood Institute of the National Institutes of Health and Boston University School of Medicine.

Sources of Funding

This work was supported by the National Heart, Lung, and Blood Institute’s Framingham Heart Study (contract No. N01-HC-25195) and CardioGenomics Program for Genomic Applications (HL66582). Dr Newton-Cheh has been supported by an award from the GlaxoSmithKline Competitive Grants Award Program for Young Investigators, National Institutes of Health grants T32HL07575 and K23HL080025, and a Doris Duke Charitable Foundation Clinical Scientist Development Award. ECG measurements were supported in part by an unrestricted grant from Pfizer, Inc.

Disclosures

None.


*    References
up arrowTop
 up arrowAbstract
 up arrowIntroduction
 up arrowMethods
 up arrowResults
 up arrowDiscussion
 *References
 

  1. Rosamond W, Flegal K, Friday G, Furie K, Go A, Greenlund K, Haase N, Ho M, Howard V, Kissela B, Kittner S, Lloyd-Jones D, McDermott M, Meigs J, Moy C, Nichol G, O’Donnell CJ, Roger V, Rumsfeld J, Sorlie P, Steinberger J, Thom T, Wasserthiel-Smoller S, Hong Y. Heart disease and stroke statistics: 2007 update: a report from the American Heart Association Statistics Committee and Stroke Statistics Subcommittee. Circulation. 2007; 115: e69–e171.[Free Full Text]
  2. Fox CS, Evans JC, Larson MG, Kannel WB, Levy D. Temporal trends in coronary heart disease mortality and sudden cardiac death from 1950 to 1999: the Framingham Heart Study. Circulation. 2004; 110: 522–527.[Abstract/Free Full Text]
  3. Algra A, Tijssen JG, Roelandt JR, Pool J, Lubsen J. QTc prolongation measured by standard 12-lead electrocardiography is an independent risk factor for sudden death due to cardiac arrest. Circulation. 1991; 83: 1888–1894.[Abstract/Free Full Text]
  4. Schouten EG, Dekker JM, Meppelink P, Kok FJ, Vandenbroucke JP, Pool J. QT interval prolongation predicts cardiovascular mortality in an apparently healthy population. Circulation. 1991; 84: 1516–1523.[Abstract/Free Full Text]
  5. Goldberg RJ, Bengtson J, Chen ZY, Anderson KM, Locati E, Levy D. Duration of the QT interval and total and cardiovascular mortality in healthy persons (the Framingham Heart Study experience). Am J Cardiol. 1991; 67: 55–58.[CrossRef][Medline] [Order article via Infotrieve]
  6. Dekker JM, Schouten EG, Klootwijk P, Pool J, Kromhout D. Association between QT interval and coronary heart disease in middle-aged and elderly men: the Zutphen Study. Circulation. 1994; 90: 779–785.[Abstract/Free Full Text]
  7. Siscovick DS, Raghunathan TE, Rautaharju P, Psaty BM, Cobb LA, Wagner EH. Clinically silent electrocardiographic abnormalities and risk of primary cardiac arrest among hypertensive patients. Circulation. 1996; 94: 1329–1333.[Abstract/Free Full Text]
  8. Karjalainen J, Reunanen A, Ristola P, Viitasalo M. QT interval as a cardiac risk factor in a middle aged population. Heart. 1997; 77: 543–548.[Abstract/Free Full Text]
  9. Dekker JM, Crow RS, Hannan PJ, Schouten EG, Folsom AR. Heart rate-corrected QT interval prolongation predicts risk of coronary heart disease in black and white middle-aged men and women: the ARIC study. J Am Coll Cardiol. 2004; 43: 565–571.[Abstract/Free Full Text]
  10. Straus SM, Kors JA, De Bruin ML, van der Hooft CS, Hofman A, Heeringa J, Deckers JW, Kingma JH, Sturkenboom MC, Stricker BH, Witteman JC. Prolonged QTc interval and risk of sudden cardiac death in a population of older adults. J Am Coll Cardiol. 2006; 47: 362–367.[Abstract/Free Full Text]
  11. Schwartz PJ, Stramba-Badiale M, Segantini A, Austoni P, Bosi G, Giorgetti R, Grancini F, Marni ED, Perticone F, Rosti D, Salice P. Prolongation of the QT interval and the sudden infant death syndrome. N Engl J Med. 1998; 338: 1709–1714.[Abstract/Free Full Text]
  12. Newton-Cheh C, Larson MG, Corey DC, Benjamin EJ, Herbert AG, Levy D, D’Agostino RB, O’Donnell CJ. QT interval is a heritable quantitative trait with evidence of linkage to chromosome 3 in a genome-wide linkage analysis: the Framingham Heart Study. Heart Rhythm. 2005; 2: 277–284.[CrossRef][Medline] [Order article via Infotrieve]
  13. Chugh SS, Senashova O, Watts A, Tran PT, Zhou Z, Gong Q, Titus JL, Hayflick SJ. Postmortem molecular screening in unexplained sudden death. J Am Coll Cardiol. 2004; 43: 1625–1629.[Abstract/Free Full Text]
  14. Tester DJ, Ackerman MJ. Postmortem long QT syndrome genetic testing for sudden unexplained death in the young. J Am Coll Cardiol. 2007; 49: 240–246.[Abstract/Free Full Text]
  15. Arnestad M, Crotti L, Rognum TO, Insolia R, Pedrazzini M, Ferrandi C, Vege A, Wang DW, Rhodes TE, George AL Jr, Schwartz PJ. Prevalence of long-QT syndrome gene variants in sudden infant death syndrome. Circulation. 2007; 115: 361–367.[Abstract/Free Full Text]
  16. Roden DM. Drug-induced prolongation of the QT interval. N Engl J Med. 2004; 350: 1013–1022.[Free Full Text]
  17. Straus SM, Sturkenboom MC, Bleumink GS, Dieleman JP, van der Lei J, de Graeff PA, Kingma JH, Stricker BH. Non-cardiac QTc-prolonging drugs and the risk of sudden cardiac death. Eur Heart J. 2005; 26: 2007–2012.[Abstract/Free Full Text]
  18. Newton-Cheh C, Shah R. Genetic determinants of QT interval variation and sudden cardiac death. Curr Opin Genet Dev. 2007; 17: 213–221.[CrossRef][Medline] [Order article via Infotrieve]
  19. Reich DE, Lander ES. On the allelic spectrum of human disease. Trends Genet. 2001; 17: 502–510.[CrossRef][Medline] [Order article via Infotrieve]
  20. Roden DM. Taking the "idio" out of "idiosyncratic": predicting torsades de pointes. Pacing Clin Electrophysiol. 1998; 21: 1029–1034.[CrossRef][Medline] [Order article via Infotrieve]
  21. Arking DE, Pfeufer A, Post W, Kao WH, Newton-Cheh C, Ikeda M, West K, Kashuk C, Akyol M, Perz S, Jalilzadeh S, Illig T, Gieger C, Guo CY, Larson MG, Wichmann HE, Marban E, O’Donnell CJ, Hirschhorn JN, Kaab S, Spooner PM, Meitinger T, Chakravarti A. A common genetic variant in the NOS1 regulator NOS1AP modulates cardiac repolarization. Nat Genet. 2006; 38: 644–651.[CrossRef][Medline] [Order article via Infotrieve]
  22. Aarnoudse AJ, Newton-Cheh C, de Bakker PI, Straus SM, Kors JA, Hofman A, Uitterlinden AG, Witteman JC, Stricker BH. Common NOS1AP variants are associated with a prolonged QTc interval in the Rotterdam Study. Circulation. 2007; 116: 10–16.[Abstract/Free Full Text]
  23. Altshuler D, Hirschhorn JN, Klannemark M, Lindgren CM, Vohl MC, Nemesh J, Lane CR, Schaffner SF, Bolk S, Brewer C, Tuomi T, Gaudet D, Hudson TJ, Daly M, Groop L, Lander ES. The common PPARgamma Pro12Ala polymorphism is associated with decreased risk of type 2 diabetes. Nat Genet. 2000; 26: 76–80.[CrossRef][Medline] [Order article via Infotrieve]
  24. Finlayson K, Witchel HJ, McCulloch J, Sharkey J. Acquired QT interval prolongation and HERG: implications for drug discovery and development. Eur J Pharmacol. 2004; 500: 129–142.[CrossRef][Medline] [Order article via Infotrieve]
  25. Splawski I, Shen J, Timothy KW, Lehmann MH, Priori S, Robinson JL, Moss AJ, Schwartz PJ, Towbin JA, Vincent GM, Keating MT. Spectrum of mutations in long-QT syndrome genes: KVLQT1, HERG, SCN5A, KCNE1, and KCNE2. Circulation. 2000; 102: 1178–1185.[Abstract/Free Full Text]
  26. Napolitano C, Priori SG, Schwartz PJ, Bloise R, Ronchetti E, Nastoli J, Bottelli G, Cerrone M, Leonardi S. Genetic testing in the long QT syndrome: development and validation of an efficient approach to genotyping in clinical practice. JAMA. 2005; 294: 2975–2980.[Abstract/Free Full Text]
  27. Pietila E, Fodstad H, Niskasaari E, Laitinen PP, Swan H, Savolainen M, Kesaniemi YA, Kontula K, Huikuri HV. Association between HERG K897T polymorphism and QT interval in middle-aged Finnish women. J Am Coll Cardiol. 2002; 40: 511–514.[Abstract/Free Full Text]
  28. Bezzina CR, Verkerk AO, Busjahn A, Jeron A, Erdmann J, Koopmann TT, Bhuiyan ZA, Wilders R, Mannens MM, Tan HL, Luft FC, Schunkert H, Wilde AA. A common polymorphism in KCNH2 (HERG) hastens cardiac repolarization. Cardiovasc Res. 2003; 59: 27–36.[Abstract/Free Full Text]
  29. Pfeufer A, Jalilzadeh S, Perz S, Mueller JC, Hinterseer M, Illig T, Akyol M, Huth C, Schopfer-Wendels A, Kuch B, Steinbeck G, Holle R, Nabauer M, Wichmann HE, Meitinger T, Kaab S. Common variants in myocardial ion channel genes modify the QT interval in the general population: results from the KORA study. Circ Res. 2005; 96: 693–701.[Abstract/Free Full Text]
  30. Gouas L, Nicaud V, Berthet M, Forhan A, Tiret L, Balkau B, Guicheney P. Association of KCNQ1, KCNE1, KCNH2 and SCN5A polymorphisms with QTc interval length in a healthy population. Eur J Hum Genet. 2005; 13: 1213–1222.[CrossRef][Medline] [Order article via Infotrieve]
  31. Crotti L, Lundquist AL, Insolia R, Pedrazzini M, Ferrandi C, De Ferrari GM, Vicentini A, Yang P, Roden DM, George AL Jr, Schwartz PJ. KCNH2-K897T is a genetic modifier of latent congenital long-QT syndrome. Circulation. 2005; 112: 1251–1258.[Abstract/Free Full Text]
  32. Dawber T, Meadors GF, Moore FE. Epidemiologic approaches to heart disease: the Framingham study. Am J Public Health. 1951; 41: 279–286.[Free Full Text]
  33. Dawber T, Kannel WB, Lyell LP. An approach to longitudinal studies in a community: the Framingham Heart Study. Ann N Y Acad Sci. 1963; 107: 539–556.[Medline] [Order article via Infotrieve]
  34. Feinleib M, Kannel WB, Garrison RJ. The Framingham Offspring Study: design and preliminary data. Prev Med. 1975; 4: 518–525.[CrossRef][Medline] [Order article via Infotrieve]
  35. Kannel WB, Feinleib M, McNamara PM. An investigation of coronary heart disease in families: the Framingham Offspring Study. Am J Epidemiol. 1979; 110: 281–290.[Abstract/Free Full Text]
  36. Wilk JB, Manning AK, Dupuis J, Cupples LA, Larson MG, Newton-Cheh C, Demissie S, DeStefano AL, Hwang SJ, Liu C, Yang Q, Lunetta KL. No evidence of major population substructure in the Framingham Heart Study. Genet Epidemiol. 2005; 29: 286.
  37. Cupples LA, D’Agostino R., Garrison, RJ, eds. Some Risk Factors Related to the Annual Incidence of Cardiovascular Disease and Death Using Pooled Repeated Biennial Measurements: Framingham Study, 30-Year Follow-Up. Bethesda, Md: National Heart, Lung, and Blood Institute; 1987. DHHS publication No. NIH87–2703.
  38. Buetow KH, Edmonson M, MacDonald R, Clifford R, Yip P, Kelley J, Little DP, Strausberg R, Koester H, Cantor CR, Braun A. High-throughput development and characterization of a genomewide collection of gene-based single nucleotide polymorphism markers by chip-based matrix-assisted laser desorption/ionization time-of-flight mass spectrometry. Proc Natl Acad Sci U S A. 2001; 98: 581–584.[Abstract/Free Full Text]
  39. Gabriel SB, Schaffner SF, Nguyen H, Moore JM, Roy J, Blumenstiel B, Higgins J, DeFelice M, Lochner A, Faggart M, Liu-Cordero SN, Rotimi C, Adeyemo A, Cooper R, Ward R, Lander ES, Daly MJ, Altshuler D. The structure of haplotype blocks in the human genome. Science. 2002; 296: 2225–2229.[Abstract/Free Full Text]
  40. Barrett JC, Fry B, Maller J, Daly MJ. Haploview: analysis and visualization of LD and haplotype maps. Bioinformatics. 2005; 21: 263–265.[Abstract/Free Full Text]
  41. Excoffier L, Slatkin M. Maximum-likelihood estimation of molecular haplotype frequencies in a diploid population. Mol Biol Evol. 1995; 12: 921–927.[Abstract]
  42. Stram DO, Haiman CA, Hirschhorn JN, Altshuler D, Kolonel LN, Henderson BE, Pike MC. Choosing haplotype-tagging SNPS based on unphased genotype data using a preliminary sample of unrelated subjects with an example from the Multiethnic Cohort Study. Hum Hered. 2003; 55: 27–36.[CrossRef][Medline] [Order article via Infotrieve]
  43. Sagie A, Larson MG, Goldberg RJ, Bengtson JR, Levy D. An improved method for adjusting the QT interval for heart rate (the Framingham Heart Study). Am J Cardiol. 1992; 70: 797–801.[CrossRef][Medline] [Order article via Infotrieve]
  44. Lettre G, Lange C, Hirschhorn JN. Genetic model testing and statistical power in population-based association studies of quantitative traits. Genet Epidemiol. 2007; 31: 358–362.[CrossRef][Medline] [Order article via Infotrieve]
  45. SAS Institute. SAS/STAT User’s Guide, Version 8.1. Cary, NC: SAS Institute; 2000.
  46. Schaid DJ, Rowland CM, Tines DE, Jacobson RM, Poland GA. Score tests for association between traits and haplotypes when linkage phase is ambiguous. Am J Hum Genet. 2002; 70: 425–434.[CrossRef][Medline] [Order article via Infotrieve]
  47. de Bakker PI, Yelensky R, Pe’er I, Gabriel SB, Daly MJ, Altshuler D. Efficiency and power in genetic association studies. Nat Genet. 2005; 37: 1217–1223.[CrossRef][Medline] [Order article via Infotrieve]
  48. Kruglyak L, Nickerson DA. Variation is the spice of life. Nat Genet. 2001; 27: 234–236.[CrossRef][Medline] [Order article via Infotrieve]
  49. Newton-Cheh C, Hirschhorn JN. Genetic association studies of complex traits: design and analysis issues. Mutat Res. 2005; 573: 54–69.[Medline] [Order article via Infotrieve]
  50. Drake JA, Bird C, Nemesh J, Thomas DJ, Newton-Cheh C, Reymond A, Excoffier L, Attar H, Antonarakis SE, Dermitzakis ET, Hirschhorn JN. Conserved noncoding sequences are selectively constrained and not mutation cold spots. Nat Genet. 2006; 38: 223–227.[CrossRef][Medline] [Order article via Infotrieve]
  51. Fairbrother WG, Yeh RF, Sharp PA, Burge CB. Predictive identification of exonic splicing enhancers in human genes. Science. 2002; 297: 1007–1013.[Abstract/Free Full Text]
  52. Carlini DB, Genut JE. Synonymous SNPs provide evidence for selective constraint on human exonic splicing enhancers. J Mol Evol. 2006; 62: 89–98.[CrossRef][Medline] [Order article via Infotrieve]
  53. Paavonen KJ, Chapman H, Laitinen PJ, Fodstad H, Piippo K, Swan H, Toivonen L, Viitasalo M, Kontula K, Pasternack M. Functional characterization of the common amino acid 897 polymorphism of the cardiac potassium channel KCNH2 (HERG). Cardiovasc Res. 2003; 59: 603–611.[Abstract/Free Full Text]
  54. Anson BD, Ackerman MJ, Tester DJ, Will ML, Delisle BP, Anderson CL, January CT. Molecular and functional characterization of common polymorphisms in HERG (KCNH2) potassium channels. Am J Physiol Heart Circ Physiol. 2004; 286: H2434–H2441.[Abstract/Free Full Text]
 

CLINICAL PERSPECTIVE

Disordered myocardial repolarization as manifest in electrocardiographic QT interval prolongation is a consistent risk factor for sudden cardiac death. As a consequence of noncardiac medications, QT prolongation and arrhythmias, in part a result of binding to the HERG channel, are also a major impediment to bringing common drugs to market safely. We tested common genetic variants in the KCNH2 gene encoding the {alpha} subunit of the HERG channel and observed association of a novel intronic single nucleotide polymorphism and a previously reported missense single nucleotide polymorphism with continuous QT interval duration in the Framingham Heart Study. The variants contribute to modest changes in QT interval, 3.9- and 3.1-ms increases; thus, their relevance to clinical disease is currently unclear. In trials of drugs that prolong the QT interval on resting ECG, increases of 5 to 10 ms in the group mean have been associated with increased risk of arrhythmia. These group means on resting ECGs reflect some patients with little QT change and some with substantial increases who likely bear a large fraction of risk. Future tests of common variants of modest effect such as those in KCNH2 or the recently identified NOS1AP myocardial repolarization gene ({approx}7-ms increase) in individuals at high risk for arrhythmias (eg, heart failure, status post myocardial infarction) and those exposed to QT-prolonging drugs are needed to determine whether these small changes in group means hide substantially increased risk for certain individuals when combined with other repolarization perturbations.


*    Footnotes
 
Guest Editor for this article was Michael E. Cain, MD.

The online Data Supplement, consisting of tables and a figure, can be found with this article at http://circ.ahajournals.org/cgi/content/full/CIRCULATIONAHA.107.710780/DC1.

*Dr Hirschhorn and Dr O’Donnell contributed equally. Back

Presented in part at the 45th Annual Conference on Cardiovascular Disease Epidemiology and Prevention of the American Heart Association, Washington, DC, May 2005.




This article has been cited by other articles:


Home page
 Eur Heart JHome page
P. T. Ellinor and D. J. Milan
Polymorphisms and atrial fibrillation: sorting the wheat from the chaff
Eur. Heart J., April 1, 2008; 29(7): 843 - 845.
[Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Data Supplement
Right arrow Correction
Right arrow Correction (v117,pe9)
Right arrow All Versions of this Article:
116/10/1128    most recent
CIRCULATIONAHA.107.710780v1
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow Request Permissions
Citing Articles
Right arrow Citing Articles via HighWire