Journal of Molecular Medicine

, Volume 82, Issue 3, pp 182–188

Genetic variations of KCNQ1, KCNH2, SCN5A, KCNE1, and KCNE2 in drug-induced long QT syndrome patients

Authors

  • Aimée D. C. Paulussen
    • Department of PharmacogenomicsJohnson & Johnson Pharmaceutical Research and Development
  • Ronaldus A. H. J. Gilissen
    • Department of PharmacogenomicsJohnson & Johnson Pharmaceutical Research and Development
  • Martin Armstrong
    • Department of PharmacogenomicsJohnson & Johnson Pharmaceutical Research and Development
  • Pieter A. Doevendans
    • Department of CardiologyAcademic Hospital
  • Peter Verhasselt
    • Department of Functional GenomicsJohnson & Johnson Pharmaceutical Research and Development
  • Hubert J. M. Smeets
    • Department of Genetics and Cell BiologyCardiovascular Research Institute Maastricht (CARIM)
  • Eric Schulze-Bahr
    • Department of Cardiology and Angiology, Hospital of the Westfalian Wilhelms-University, Institute for Arteriosclerosis Research University of Münster
  • Wilhelm Haverkamp
    • Department of Cardiology and Angiology, Hospital of the Westfalian Wilhelms-University, Institute for Arteriosclerosis Research University of Münster
  • Günter Breithardt
    • Department of Cardiology and Angiology, Hospital of the Westfalian Wilhelms-University, Institute for Arteriosclerosis Research University of Münster
  • Nadine Cohen
    • Department of PharmacogenomicsJohnson & Johnson Pharmaceutical Research and Development
    • Department of PharmacogenomicsJohnson & Johnson Pharmaceutical Research and Development
Original Article

DOI: 10.1007/s00109-003-0522-z

Cite this article as:
Paulussen, A.D.C., Gilissen, R.A.H.J., Armstrong, M. et al. J Mol Med (2004) 82: 182. doi:10.1007/s00109-003-0522-z

Abstract

Administration of specific drugs may occasionally induce acquired long QT syndrome (aLQTS), a disorder that predisposes to ventricular arrhythmias, typically of the torsade de pointes (TdP) type, and sudden cardiac death. “Forme fruste” mutations in congenital LQTS (cLQTS) genes have been reported repeatedly as the underlying cause of aLQTS, and are therefore considered as an important risk factor. We evaluated the impact of genetic susceptibility for aLQTS through mutations in cLQTS genes. Five cLQTS genes (KCNH2, KCNQ1, SCN5A, KCNE1, KCNE2) were thoroughly screened for genetic variations in 32 drug-induced aLQTS patients with confirmed TdP and 32 healthy individuals. Missense forme frust mutations were identified in four aLQTS patients: D85N in KCNE1 (two cases), T8A in KCNE2, and P347S in KCNH2. Three other missense variations were found both in patients and controls, and are thus unlikely to significantly influence aLQTS susceptibility. In addition, 13 silent and six intronic variations were detected, four of which were found in a single aLQTS patient but not in the controls. We conclude that missense mutations in the examined cLQTS genes explain only a minority of aLQTS cases.

Keywords

Long QT syndromeDrug-inducedArrhythmiaMutation analysis

Abbreviations

cLQTS

Congenital long QT syndrome

aLQTS

Acquired long QT syndrome

TdP

Torsade de pointes

ECG

Electrocardiogram

QTc

Corrected QT

PCR

Polymerase chain reaction

LD

Linkage disequilibrium

Introduction

Congenital long QT syndrome (cLQTS) is a rare, yet in some cases lethal, cardiac disease caused by malfunctioning ion channels in cardiac cells [1]. The syndrome is characterized by symptoms such as syncope and seizures and typically manifests itself with specific, potentially fatal, ventricular arrhythmias of the torsade de pointes (TdP) type. The molecular basis of cLQTS is becoming increasingly understood and mutations in six genes have been identified as causative in cLQTS families: the potassium ion channel genes KCNQ1 (KvLQT1, LQT1), KCNH2 (HERG, LQT2), KCNE1 (LQT5), KCNE2 (LQT6), the sodium channel gene SCN5A (LQT3), and the most recently discovered ANKB (LQT4) gene, encoding a member of a family of versatile membrane adaptors. Currently, over 200 different disease-causing mutations in these genes have been identified in cLQTS families, explaining between 50% and 60% of all cases and thus indicating the benefit of molecular screening in this form of the disease [2, 3, 4].

Acquired long QT syndrome (aLQTS) presents with symptoms that are very similar to those of cLQTS, although the conditions, risk factors and triggers for their occurrence differ. The major cause of aLQTS is drug administration, as indicated by the increasing amount of case reports [5, 6, 7]. Drugs from many different classes with diverse chemical structures have been reported as being capable of prolonging the QT interval and causing TdP, including anti-arrhythmics, antihistamines, anti-fungals and antipsychotics [2]. Most of these drugs have been demonstrated to interact with and block the rapid component of the delayed rectifier current IKr encoded by the KCNH2 gene (http://www.qtdrugs.org). Whilst drug intake is the trigger of acquired LQTS, other factors may also contribute to an increased risk of developing TdP in individual cases, including female gender [8], electrolyte disturbances [9], other heart disease, and underlying mutations or functional polymorphisms in one of the cLQTS disease genes [10]. In the present study, we further explored the genetic component of aLQTS.

Over 20 years ago, Moss et al. [11] suggested that genetic variation may influence the susceptibility to the development of drug-induced TdP. Since then, evidence has been accumulating for the existence of “forme fruste” mutations in aLQTS. Priori et al. [12] reported that the penetrance of disease-causing mutations is very low in many cLQTS families, indicating that many mutation carriers are, under normal circumstances, clinically asymptomatic and have normal QT intervals. On challenge with certain drugs, however, these individuals appear to be more susceptible to cardiac ion channel blockade and QT prolongation. Based on this observation, it is reasonable to assume that individuals may exist within the general population who have not been recognized as cLQTS patients because of small family size and/or low penetrance of the mutation. Indeed, polymorphisms and mutations in the cLQTS genes have been identified as a causal factor in an increasing number of aLQTS case reports (Table 1). This suggests that some aLQTS patients may be identifiable by mutation screening of cLQTS genes, and that for these patients the development of TdP may thus be preventable. Literature reports have mainly focused on individual aLQTS cases, in which a cLQTS gene variant was identified. However, the importance of forme fruste mutations as a risk factor for TdP in larger aLQTS populations remains to be demonstrated. Studies have therefore been initiated to screen cLQTS genes in larger aLQTS patient groups, aiming to determine the frequency of forme fruste mutations and common polymorphisms [13, 14, 15, 16]. Here, we investigated the role of genetic variation in five known cLQTS genes in 32 patients presenting with acquired QT prolongation and TdP as a consequence of drug administration.
Table 1

Overview of mutations and functional polymorphisms in cLQTS genes that have been reported in acquired LQTS patients. n.s. Not specified

Gene

Base pair change

Amino acid change

Drugs

Age (years)

Sex

Additional risk factors

Symptoms

Reference

KCNE1

253G→A

D85N

Sotalol

80

Female

TdP

This study

253G→A

D85N

Quinidine

71

Male

Hypokalaemia

TdP

This study

KCNE2

22A→G

T8A

Amiodarone

12

Male

TdP

This study

22A→G

T8A

Quinidine

n.s.

n.s.

TdP

13

22A→G

T8A

Sulfametoxazole

45

Male

QTc>600 ms

14

25C→G

Q9E

Clarithromycin

76

Female

Hypokalaemia, diabetic, history of stroke

TdP, VF

13

161T→C

M54T

Procainamide

n.s.

n.s.

TdP

14

170T→C

I57T

Oxatomide

n.s.

n.s.

TdP

14

347C→T

A116V

Quinidine, mexiletine

55

Female

History of cardiac arrest

Syncope with TdP

14

KCNH2

1039C→T

P347S

Cisapride, clarithromycin

77

Female

TdP

21, this study

1048C→T

R328C

n.s.

45

Male

TdP

15

2350C→T

R784W

Amiodarone

n.s.

n.s.

TdP

16

KCNQ1

944A→G

Y315C

Cisapride

77

Female

Hypokalaemia

Cardiac arrest

22

1663C→T

R555C

Terfenadine

38

Female

cLQTS family

Sudden death

29

1747C→T

R583C

Dofetilide

n.s.

n.s.

TdP

16

SCN5A

1844G→A

G615E

Quinidine

n.s.

n.s.

TdP

16

1852C→T

L618F

Quinidine

n.s.

n.s.

TdP

16

3748T→C

F1250L

Sotalol

n.s.

n.s.

TdP

16

5474T→C

L1825P

Cisapride

70

Female

Tdp

30

Materials and methods

Patient and control subjects

Thirty-two unrelated aLQTS subjects were recruited from three centres: the Academic Hospital Maastricht, The Netherlands (n=11), the Westfalian Wilhelms-University Münster, Germany (n=20) and the Ottawa Hospital, Ontario, Canada (n=1). The clinical characteristics of the aLQTS patient population included in the study are presented in Table 2. All subjects experienced TdP as a result of drug administration. Torsade de pointes was defined as either non-sustained or sustained ventricular tachycardia showing phasic variation in the electrical polarity of the QRS complex and a “short-long-short” initiating sequence [17]. In addition, all subjects showed a QTc prolongation (QTc>440 ms) and changes in the configuration of the T(-U) wave (i.e., biphasic T waves, beat-to-beat changes in T wave morphology, post-extrasystolic T wave abnormalities such as an increased T-U wave amplitude, or emergence of new large U waves) compared to an electrocardiogram recorded prior to drug exposure implicated in the generation of TdP. In two cases where no baseline and on-drug ECGs could be retrieved from the referring physician, subjects were classified as demonstrating TdP because the qualifying ECG showed the typical TdP pattern. Electrocardiographic intervals were measured from tracings as close as possible to the event. The QT interval was measured from electrocardiogram lead II or the available rhythm strip. For rate-correction, the Bazett equation was applied.
Table 2

Clinical characteristics of the aLQTS patient population. Data are indicated as mean values with the range (min–max). Drugs that have been implicated in QT prolongation and TdP (http://www.qtdrugs.org). Others include cisapride, clarythromycin, ibutilide, risperidone, tamoxifen, and venlafaxine

Total

Female

Male

Number of patients (n)

32

21

11

Age (years)

71 (12–84)

75 (24–82)

70 (12–84)

QTc at rest (ms)

425 (378–490)

422 (378–470)

433 (397–490)

QTc after drug intake (ms)

537 (428–640)

539 (428–640)

533 (456–600)

Trigger drugs (number of cases)

 Sotalol

18

 Amiodarone

8

 Quinidine

3

 Other

6

Relevant concurrent therapy (number of cases)a

11

aEither more than one drug that blocks LQTS related ion channels, or a secondary drug inhibiting the metabolism of the QT prolonging drug

Genotype/allele frequency comparisons were determined with a control group of 32 anonymous, healthy, Caucasians. This group of control subjects was recruited from a large pool of clinical trial volunteers available at the J&J PRD (Belgium). Full IRB approval and subject informed consent was obtained from all participants.

Mutation screening of cLQTS genes

Mutational screening of all individual exons and the intron/exon boundaries of KCNE1, KCNE2, KCNH2, KCNQ1 and SCN5A was performed by PCR amplification and bi-directional direct sequencing using an ABI-Prism-3700 DNA Analyzer (Applied Biosystems, Nieuwerkerk a/d IJssel, The Netherlands). Oligonucleotides (Eurogentec, Seraing, Belgium) and PCR conditions used for amplification have been previously described by Itoh et al. (KCNH2) [18], Splawski et al. (KCNQ1, KCNH2 and KCNE1) [19], Wang et al. (SCN5A) [20], and Abbott et al. (KCNE2) [13]. All sequences were aligned with their respective wild type gene sequences reported in GenBank under accession numbers AF071002 (KCNE1), NM_005136 (KCNE2), U04270 (KCNH2), AF000571 (KCNQ1) and NM_000335 (SCN5A), using the Sequencher software package (Gene Codes Corporation, Ann Arbor, Mich.). Frequencies of known and novel genetic variations were determined in the panels of 32 aLQTS subjects and 32 control subjects.

Results

Table 2 summarizes the clinical characteristics of the aLQTS patient population, including age, sex, pre- and post-dose QTc, the TdP-inducing drug, and eventual concurrent medications. With exception of four patients, all subjects were older than 50 years, among whom 22 (75%) above the age of 65 years. There were approximately twice as many female patients than male patients in this study. The average QTc interval for the aLQTS subjects prior to drug treatment was 425 ms (males=433 ms, females=422 ms) rising to 537 ms post-treatment (males=533 ms, females=539 ms).

The obtained sequences of the five screened cLQTS genes of each of the aLQTS patients were aligned with the wild type sequences deposited in GenBank. Table 3 summarizes the genetic variations detected in the congenital LQTS genes. In Fig. 1, a direct comparison is made between the genetic variations identified in this study and the causal mutations reported in earlier genetic studies of aLQTS patients. The frequency of each of the variations identified in the aLQTS patients was also determined by sequencing in a healthy control group of 32 subjects (Table 3). One aLQTS patient was heterozygous for base pair substitution 22A→G in exon 1 of the KCNE2 gene, leading to a T8A amino acid substitution. This 12-year-old boy had a normal baseline QTc interval (420 ms) and developed TdP after amiodarone treatment. Another subject was heterozygous for base pair substitution 1039 C→T in exon 5 of the KCNH2 gene, resulting in amino acid substitution P347S. The baseline QTc of this patient was normal (440 ms), but was highly prolonged during concomitant drug administration (640 ms). The suspected trigger compound for this subject was cisapride, taken in combination with the antibiotic drug clarithromycin [21]. Besides these two variations that have previously been associated with aLQTS, several other variations were detected in both populations: in total four missense variations were found, including variation D85N in the KCNE1 gene that was detected in two aLQTS subjects but not in any of the control subjects. The three other variations, respectively G38S in KCNE1, K897T in KCNH2, and H558R in SCN5A, were identified in similar frequencies in both study populations. Thirteen silent mutations were detected, including four novel mutations. Three of these silent variations were found only in the aLQTS population (S28S in KCNE1, T377T in KCNQ1 and G628G in KCNQ1), while the others were detected in similar frequencies in both populations. Finally, six intronic variations were present at similar frequencies in both patients and controls, three of which were novel (Table 3). From the data of the individual subjects in Table 3, it can be deduced that partial or complete linkage disequilibrium (LD) exists between identified variations in several genes. For example, I489I and F513F in KCNH2 are in complete LD, and partially with L564L. In KCNQ1, S546S and IVS13+36G→A are in partial LD. The genetic variations IVS9-3C→A and H558R in SCN5A are in almost complete LD, and the variations A29A, E1061E, and D1819D also appear to co-segregate. The possible significance of the LD between these variations is at present not known.
Table 3

Genetic variations detected in aLQTS patients and healthy individuals. Comparisons were made to the database of the “Gene connection for the heart”, assembled and updated by the Working Group on Arrhythmias of the European Society of Cardiology (http://pc4.fsm.it:81/cardmoc/)

Gene

Nucleotide change

Amino acid change

Location of variation within protein

Status

Allele frequency in the aLQTS group (n=32)

aLQTS case reference number carrying the variation (heterozygote//homozygote)

Allele frequency control group (n=32)

KCNE1

84G→A

S28S

N-terminal

Known

0.02

21//–

0

112G→A

G38S

N-terminal

Known

0.28

10,11,13,14,15,16,17,28,30,34,35,38//5,9,21

0.33

253G→A

D85N

C-terminal

Known

0.05

22//32

0

KCNE2

22A→G

T8A

N-terminal

Known

0.02

16//–

0

KCNH2

1039C→T

P347S

PAS/S1

Known

0.02

18//–

0

1467C→T

I489I

S2/S3

Known

0.27

7,8,13,15,17,28,33,36,38//14,31,34,35

0.19

1539C→T

F513F

S3

Known

0.27

7,8,13,15,17,28,33,36,38//14,31,34,35

0.19

1692A→G

L564L

S5

Known

0.45

5,7,8,9,13,15,16,17,18,22,24,25,33,36,42//14,21,28,31,34,35,38

0.36

1956C→T

Y652Y

S6

Known

0.28

5,7,8,13,17,18,19,20,22,24,27,32//12,30,37

0.41

2690A→C

K897T

C-terminal

Known

0.20

9,10,15,16,19,20,25,29,32,36,42//11

0.23

KCNQ1

1131C→G

T377T

S6/C-terminal

Known

0.02

20//–

0

IVS12+14T→C

Known

0.08

7,9,11,14,15//–

0.08

1638G→A

S546S

S6/C-terminal

Known

0.19

8,13,16,20,22,28,30,42//21,29

0.19

IVS13+36G→A

Known

0.33

13,15,16,17,20,22,27,29,30,32,34//8,18,21,28,42

0.33

1884C→T

G628G

S6/C-terminal

Novel

0.02

19//–

0

1986C→T

Y662Y

S6/C-terminal

Known

0.23

9,10,11,12,15,16,21,25,34//7,8,30

0.25

SCN5A

87G→A

A29A

N-terminal

Novel

0.20

7,12,19,21,24,25,28,30,32,34,35,36,42//–

0.28

IVS3-24C→T

Novel

0.03

10,22//–

0.02

IVS9-3C→A

Novel

0.17

11,14,15,25,33,34,35//5,7

0.16

1673A→G

H558R

DIS6/DIIS1

Known

0.34

11,14,15,25,34,35,37//5,7

0.44

IVS16-6C→T

Novel

0.02

42//–

0

3183G→A

E1061E

DIIS6/DIIIS1

Novel

0.14

15,19,24,30,32,35,42//37

0.13

4218G→A

G1406G

DIIIS5/DIIIS6

Novel

0.02

21//–

0.02

IVS24+53T→C

Known

0.05

19,21,32//–

0.05

5457T→C

D1819D

DIVS6/C-terminal

Known

0.33

8,12,15,18,19,22,28,29,30,36,38//21,32,33,37,42

0.38

Fig. 1

Overview of genetic variations identified in aLQTS patients. The genomic structure of each examined cLQTS gene is represented as blocks (exons) and connecting lines (introns). The mutations and polymorphisms identified in the present study are indicated above the genomic structure. Mutations that alter the encoded amino acid are underlined. Causal mutations that have been identified in earlier studies on aLQTS patients are shown below the genomic structure, as well as the position of start and stop codons

Discussion

We investigated the genetic predisposition to the development of aLQTS in a group of 32 patients, all of whom were confirmed by ECG to experience drug-induced TdP. To achieve this aim, we screened the entire coding region of five cLQTS genes in order to identify non-penetrant (or “hidden”) mutation carriers. In our study, female patients were twice as prevalent as male patients, consistent with the observation that female gender may increase the susceptibility of aLQTS [8]. The QTc interval before drug treatment was below a recommended cut-off point for defining QT prolongation (450 and 470 ms for men and women, respectively) [1], but was raised in most cases by more than 100 ms upon drug intake, clearly indicating a severe, drug-induced effect on repolarization.

Individual case reports have shown that patients with drug-induced TdP may carry sporadic mutations in the genes involved in cLQTS [22, 23]. In our group of 32 aLQTS subjects, only one case represented a known forme fruste of cLQTS, carrying missense mutation 1039 C→T (P347S) in KCNH2. This patient had taken the gastro-intestinal drug cisapride in combination with the antibiotic clarithromycin. Cisapride has previously been implicated in aLQTS cases, and is known to be able to block the HERG channel [24]. In addition, both cisapride and clarithromycin are metabolized by the same drug metabolizing enzyme. The metabolism of cisapride might therefore be inhibited by clarithromycin, which can lead to increased circulating plasma levels of cisapride, and thus a higher potency for blockade of the HERG channel [21]. Another patient carried mutation 22 A→G (T8A) in the KCNE2 gene, which has been described previously in a case report of an individual presenting with antibiotic drug-induced cardiac arrhythmia whilst taking bactrim (trimethoprim and sulfmethoxazole), and has been functionally characterized as the causative mutation for the reported case [14]. In our study, the suspected trigger compound was the antiarrhythmic drug amiodarone, indicating that this mutation renders a patient vulnerable to development of TdP by at least two different drugs. Similarly to the T8A variation in KCNE2, the D85N missense polymorphism in the KCNE1 gene was detected in two aLQTS cases but not in controls. The two unrelated patients in the present study carrying this known but rare polymorphism had received different drugs (sotalol and quinidine, respectively). This D85N polymorphism in KCNE1 was more prevalent in aLQTS patients than controls not only in our study but also in at least one other study [25]. Furthermore, functional studies of KVLQT1-MinK (D85N) IKs channels in CHO cells demonstrated that these channels activated slower and deactivated faster than wild-type channels, thereby reducing the level of repolarizing current. In the presence of an IKr drug block, the reduced repolarization reserve caused by the D85N polymorphism may render individuals more susceptible to acquired LQTS [25].

Three other variations that result in altered amino acid sequences were detected in both aLQTS patients and control subjects, and have all been reported previously: G38S in KCNE1, K897T in KCNH2, and H558R in SCN5A [16, 26, 27, 28]. Although allele frequencies of these three variations were similar in both populations, it can not be ruled out that mild alterations to either IKr or INa current may affect the susceptibility to develop early-after depolarizations and TdP when combined with a drug block, as has been described for the D85N polymorphism in KCNE1 [25]. Functional studies are required to establish the contribution of these missense polymorphisms to the QT interval at rest and after challenging with QT-prolonging drugs. Of note, several novel and known silent and intronic variations were detected. Because of their location close to an exon-intron boundary, two of the novel intronic variations in SCN5A may be directly functionally relevant (IVS9-3 C→A and IVS16-6 C→T), as they may potentially induce alternative splicing. The IVS9-3 C→A variation was present in similar frequencies in both the aLQTS and control group, while variant IVS16-6 C→T was detected in a single aLQTS patient only and may therefore possibly represent a novel forme fruste mutation rather than a common polymorphism.

Few other research groups have studied the frequency of genetic polymorphisms and mutations in LQTS genes in larger aLQTS patient groups. The findings of these studies, including this one, are summarized in Table 4. The initial studies of Abbott et al. [13] and Sesti et al. [14] included only the screening of the KCNE2 gene. Chevalier et al. [15] screened four cLQTS genes in 16 aLQTS patients, but did not screen the SCN5A gene. Yang et al. [16] complemented the study of Sesti et al. [14] with the screening of four other cLQTS genes in the same population. For now, the results of these small studies show similar low frequencies of forme fruste mutations in aLQTS patients: between 5–12% of the studied aLQTS populations carry a mutation or functional polymorphism in one of the five investigated cLQTS genes (Table 4). Functional studies on some of these detected forme fruste mutations indicates that they either have no effect on INa function [16] or a relatively mild effect on IK function, presented by overall decreased current densities, [15] or densities reduced at the plateau voltages due to shifts in kinetic properties [29]. Functional rare polymorphisms, such as the T8A variant in KCNE2 or the D85N variant in KCNE1, may also contribute to an increased risk for aLQTS especially because they appear to be more prevalent among aLQTS patients than controls. The significance of the more common coding polymorphisms, as well the non-coding and intronic variations, will need further clarification in the future when very large numbers of aLQTS patients accompanied by appropriate controls can be screened, complemented with functional testing of these variations. Based on these data, one may estimate the implications for molecular screening. In this regard, it is important to recognize that molecular screening of five known cLQTS genes — as was done in the present study — uncovers causal mutations in only 50–60% of the congenital LQTS families. It can therefore be anticipated that additional genes with mutations responsible for cLQTS will be discovered in the future. The recent discovery of a mutation in the ANKB gene was found in cLQTS patients from a large family [4], and demonstrated that not only genes encoding an ion channel (subunit) might be relevant, but also genes involved in the regulation of correct ion channel functioning. However, apart from the members of the single family in which the cLQTS causing ANKB mutation was found, no other cLQTS patients have been reported with mutations in this gene. Since linkage analysis did not reveal significant linkage to this gene in any other cLQTS family analysed, it is currently assumed that the contribution of this particular gene as a causal factor for familial cLQTS is rather limited. Altogether, taking into account that we detect mutations in 5–12% of the aLQTS patients with our current screening approach, it seems reasonable to estimate that up to 1/6 of all aLQTS patients could be carriers of an underlying mutation in a cLQTS gene, and might thus become identifiable by genetic screening in the future.
Table 4

Number of aLQTS patients in whom a disease-associated mutation was identified by genetic screening of cLQTS genes in different studies. n.i. Not investigated

Abbott et al. [13]

Sesti et al. [14]

Yang et al. [16]

Chevalier et al. [15]

This study

Genes screened

 ANKB (Ankyrin-B)

n.i.

n.i.

n.i.

n.i.

 KCNE1 (Mink)

n.i.

0

0

2

 KCNE2 (MiRP1)

1

4

0

1

 KCNH2 (HERG)

n.i.

1

1

1

 KCNQ1 (KvLQT1)

n.i.

1

0

0

 SCN5A (SCN5A)

n.i.

3

n.i.

0

Number of aLQTS patients screened

20

92

16

32

Number of cases explained

1

9

1

4

Percentage of the investigated aLQTS patient population

5%

9%

6%

12%

Acknowledgements

We thank Rosemary Zvonar (Department of Pharmaceutical Services, The Ottawa Hospital, Canada) for providing a patient sample. E. Schulze-Bahr and W. Haverkamp are supported by grants from the Deutsche Forschungsgemeinschaft (grants Schu1082/2–2 and SFB556-A1) and the Fondation Leducq (France). P. Doevendans is supported by the Interuniversitary Cardiology Institute (The Netherlands).

Copyright information

© Springer-Verlag 2004