Long Qt Syndrome 1

A number sign (#) is used with this entry because long QT syndrome-1 (LQT1) is caused by heterozygous mutation in the KQT-like voltage-gated potassium channel-1 gene (KCNQ1; 607542) on chromosome 11p15.

Digenic inheritance has also been reported; see MOLECULAR GENETICS.

Description

Congenital long QT syndrome is electrocardiographically characterized by a prolonged QT interval and polymorphic ventricular arrhythmias (torsade de pointes). These cardiac arrhythmias may result in recurrent syncope, seizure, or sudden death (Jongbloed et al., 1999).

A form of torsade de pointes in which the first beat has a short coupling interval has been described (613600).

Genetic Heterogeneity of Long QT Syndrome

Other forms of LQT syndrome (LQTS) are LQT2 (613688), caused by mutation in the KCNH2 gene (152427); LQT3 (603830), caused by mutation in the SCN5A gene (600163); LQT4 (see 600919), caused by mutation in the ANK2 gene (106410); LQT5 (613695), caused by mutation in the KCNE1 gene (176261); LQT6 (613693), caused by mutation in the KCNE2 gene (603796); LQT7 (Andersen cardiodysrhythmic periodic paralysis, 170390), caused by mutation in the KCNJ2 gene (600681); LQT8 (618447), caused by mutation in the CACNA1C gene (114205); LQT9 (611818), caused by mutation in the CAV3 gene (601253); LQT10 (611819), caused by mutation in the SCN4B gene (608256); LQT11 (611820), caused by mutation in the AKAP9 gene (604001); LQT12 (612955), caused by mutation in the SNTA1 gene (601017); LQT13 (613485), caused by mutation in the KCNJ5 gene (600734); LQT14 (616247), caused by mutation in the CALM1 gene (114180), and LQT15 (616249), caused by mutation in the CALM2 gene (114182).

Approximately 10% of LQTS patients in whom a mutation is identified in one ion channel gene carry a second mutation in the same gene or in another ion channel gene (Tester et al., 2005).

Clinical Features

Ward (1964) observed syncope due to ventricular fibrillation in a brother and sister whose resting electrocardiogram showed abnormal prolongation of the QT interval. The mother, although asymptomatic, had a prolonged QT interval also. Her sister had attacks of syncope and died in one of these at the age of 30 years. Deafness was not a feature, making this disorder distinct from the recessively inherited syndrome described by Jervell and Lange-Nielsen (JLNS; see 220400). Similar families with involvement of multiple generations were reported by Romano et al. (1963), Romano (1965), Barlow et al. (1964), and Garza et al. (1970). Hashiba (1978) concluded that in Japan women are more severely affected than men. (As indicated later, Moss et al. (1991) found that the proband was female in 69% of multiplex families and on the average was younger than other affected members.) Gamstorp et al. (1964) reported a family with prolonged QT interval and cardiac arrhythmias without deafness; affected members were hypokalemic and benefited from administration of potassium.

Vincent (1986) found that the resting heart rate was significantly slower in newborns and children under age 3 with WRS but not in older children and adults. He interpreted the data as consistent with right-sided sympathetic deficiency manifested by a slower heart rate in early life, when sympathetic tone is high and contributes to resting heart rate, but not in older persons in whom resting heart rate is predominantly under parasympathetic control. Bonduelle (1993) suggested that death in utero is an expression of the Ward-Romano syndrome in some families.

Moss et al. (1991) prospectively investigated the clinical characteristics and long-term course of 3,343 individuals from 328 families in which one or more members were identified as affected with LQT. The 328 probands were younger at first contact (age 21 +/- 15 years) and more likely to be female (69%), and had a higher frequency of preenrollment syncope or cardiac arrest with resuscitation (80%), congenital deafness (7%), a resting heart rate less than 60 beats/min (31%), and a history of ventricular tachyarrhythmia (47%) than other affected and unaffected family members. Arrhythmogenic syncope often occurred in association with acute physical, emotional, or auditory arousal. The syncopal episodes were frequently misinterpreted as a seizure disorder. By age 12 years, 50% of the probands had experienced at least one syncopal episode or death.

Gohl et al. (1991) tested the hypothesis of sympathetic imbalance by a scintigraphic display of efferent cardiac sympathetic innervation using I-123-MIBG, an analog of norepinephrine and guanethidine. Single photon emission computed tomography (SPECT) was the method of scanning. All scans of the healthy volunteers showed a uniform tracer uptake with sometimes slightly decreased activity in the apex. All 5 patients with prolonged QT and all who had suffered from at least one episode of torsade de pointes, ventricular fibrillation, or syncope had reduced or abolished MIBG uptakes in the inferior and inferior septal parts of the left ventricle. They referred to this as congenital myocardial sympathetic dysinnervation (CMSD). One woman without symptoms or QT prolongation showed an abnormal MIBG SPECT similar to that of her daughter, who did have LQT with symptoms. One male without LQT who had suffered from ventricular fibrillation showed CMSD similar to that of his father, who had LQT but no symptoms. All members of the families with normal MIBG SPECTs had neither LQT nor symptoms.

Pacia et al. (1994) reported 2 cases of LQT presenting as epilepsy and found 8 other cases in the literature.

Vincent et al. (1992) obtained medical histories and electrocardiograms from 199 members of families with LQT. Carriers of the LQT gene (83 subjects) and noncarriers (116 subjects) were distinguished by genetic linkage analysis. A history of syncope was obtained in 52 of the carriers of the long QT gene (63%), and 4 (5%) had a history of aborted sudden death. The QT intervals corrected for heart rate in gene carriers ranged from 0.41 to 0.59 seconds (mean, 0.49). The values for noncarriers ranged from 0.38 to 0.47 seconds (mean, 0.42). Although the QT intervals were, on the average, longer in carriers, there was substantial overlap in the 2 groups. The use of a directed QT interval above 0.44 seconds as a diagnostic criterion resulted in 22 misclassifications among the 199 family members (11%). A corrected QT interval of 0.47 seconds or longer in males and 0.48 seconds or longer in females was completely predictive but resulted in false-negative diagnoses in 40% of the males and 20% of the females. Vincent et al. (1992) concluded that the QT interval cannot be used as the basis of accurate diagnosis and that, whenever possible, DNA markers should be used to obtain a reliable diagnosis.

Ohkuchi et al. (1999) described a fetus who exhibited transient (at most 30 seconds long), repeated episodes of tachyarrhythmia (240 beats per minute). The infant was born at 36 weeks' gestation and showed a markedly prolonged QT interval and transient, repeated episodes of polymorphic ventricular tachycardia. Retrospective analysis of the videotape showing fetal cardiac movement showed that atrioventricular dissociation was present prenatally and thus, that fetal tachyarrhythmia was due to ventricular tachycardia.

An excess of females with long QT syndrome is well recognized. The QT interval is longer in females, even in LQTS, which may bias the diagnostic rate in this group. To investigate the possible age- and sex-related differences in phenotype in carriers of mutations in LQTS genes (KVLQT1 (KCNQ1, 607542); HERG (KCNH2, 152427), and SCN5A, 600163), Locati et al. (1998) analyzed data from 479 probands (335 females and 144 males) referred to the International LQTS Registry. The first cardiac event (defined as syncope, nonfatal cardiac arrest, or sudden unexplained death before the age of 40) occurred significantly earlier in males. In 69 KVLQT1 mutation carriers this effect was more marked, with all first cardiac events occurring before puberty in males. A persisting cumulative risk was demonstrated beyond puberty in females. Locati et al. (1998) suggested that this apparent age- and sex-related phenomenon placed young male gene carriers in a high-risk category and that all female gene carriers should be considered for long-term prophylactic therapy.

Imboden et al. (2006) investigated the distribution of mutant alleles for the long-QT syndrome in 484 nuclear families with type I disease (LQT1 due to mutation in the KCNQ1 gene) and 269 nuclear families with type II disease (LQT2 (613688) due to mutation in the KCNH2 gene; 152427). In offspring of the female carriers of LQT1 or male and female carriers of LQT2, classic mendelian inheritance ratios were not observed. Among the 1,534 descendants, the proportion of genetically affected offspring was significantly greater than that expected according to mendelian inheritance: 870 were carriers of a mutation (57%), and 664 were noncarriers (43%) (P less than 0.001). Among the 870 carriers, the allele for the long-QT syndrome was transmitted more often to female offspring (476; 55%) than to male offspring (394; 45%) (P = 0.005). Increased maternal transmission of the long QT syndrome to daughters was also observed, possibly contributing to the excess of female patients with autosomal dominant long QT syndrome.

Priori et al. (1999) identified 9 families, each with a 'sporadic' case of LQTS, i.e., only the proband was diagnosed clinically as being affected by LQTS. Six probands were symptomatic for syncope, 2 were asymptomatic with QT prolongation found on routine examination, and 1 was asymptomatic but showed QT prolongation when examined following her brother's sudden death while swimming. Five had mutations in HERG (4 missense, 1 nonsense) and 4 had missense mutations in KCNQ1. Four of the mutations were de novo; in the remaining families at least 1 silent gene carrier was found, allowing estimation of penetrance at 25%. This contrasted greatly with the prevailing view that LQTS gene mutations may have penetrances of 90% or more. This study highlighted the importance of detecting such silent gene carriers since they are at risk of developing torsade de pointes if exposed to drugs that block potassium channels. Further, the authors stated, carrier status cannot be reliably excluded on clinical grounds alone.

In 108 first-degree relatives of 26 patients with the sudden infant death syndrome (SIDS), Kukolich et al. (1977) found normal QT intervals in all. Thus, they were unable to confirm the notion that the Ward-Romano syndrome is the basis for a large proportion of cases of SIDS. On the other hand, Schwartz et al. (1998) maintained that a relationship exists between prolongation of the QT interval and the sudden infant death syndrome. The conclusions of this study and the recommendations based thereon were the target of multiple criticisms, as reviewed elsewhere (272120).

Clinical Management

Beta-adrenergic blockade using propranolol may prevent ventricular dysrhythmia (Gale et al., 1970).

Mitsutake et al. (1981) found that the Valsalva maneuver lengthened the QT interval more in patients with this disorder than in controls, and could lead to T-wave alternans and short runs of ventricular tachycardia in patients having attacks. Propranolol suppressed this effect of the Valsalva maneuver, which, therefore, can be used to evaluate the risk of ventricular tachyarrhythmia and the efficacy of drug treatment. In a nonfamilial case, Moss and McDonald (1971) observed benefit from sympathetic denervation of the heart. Stellectomy may also have value (Moss and Schwartz, 1979).

The usual cause of syncope and sudden death in the Ward-Romano syndrome, as well as in acquired forms of prolonged QT, is a specific arrhythmia known as polymorphous ventricular tachycardia, or as torsade de pointes (meaning 'turning of the points,' an allusion to the alternately positive and negative major QRS complex). Secondary torsade de pointes is produced by various drugs and by intracranial disease such as subarachnoid hemorrhage. Stimulation of the left stellate ganglion causes QT prolongation, and ablation causes QT abbreviation. These procedures applied to the right stellate ganglion have opposite effects. Left stellate ganglion block or ablation has been used in the treatment of the long QT syndrome. The automatic implantable defibrillator has usefulness in patients with frequent ventricular arrhythmia from the long QT syndrome.

Di Segni et al. (1980) reported a case of congenital LQT in a patient in whom recurrent torsades were noted on the third day after birth. The only effective drug treatment was continuous isoproterenol infusion. A permanent pacemaker with epicardial leads was implanted at 19 days of age. Pacing decreased the QT interval, and all arrhythmias were gradually suppressed. Klein et al. (1996) reported that the child thrived and was symptom free during the next 12 years; however, attempts to decrease the rate to levels under 110 beats/minute always resulted in immediate prolongation of the QT interval and subsequent emergence of torsades. Furthermore, the prolonged rapid-rate pacing led to progressive left ventricular dilatation and diffuse hypokinesia. For this reason, orthotopic cardiac transplantation was required at 12 years of age.

Shimizu et al. (1998) studied 6 patients with known KVLQT1 mutations. Five had a history of stress-induced syncope and one had experienced occasional palpitations. Four had documented episodes of torsade de pointes. Eight control patients with Wolff-Parkinson-White syndrome were selected. Intravenous infusion of epinephrine resulted in QT prolongation, early after-depolarization phenomena, and ventricular premature complexes in LQT1 patients but in not controls. Co-infusion of the drug nicorandil improved these repolarization abnormalities. The authors commented that nicorandil increases outward cellular potassium current through an ion channel distinct from the KVLQT1 channel and may have a role in the future treatment of patients with LQT1. It was unclear whether nicorandil offered any benefit over conventional beta-blocker therapy.

In the case of the forms of LQT defined at the molecular level, Wang et al. (1996) noted that the information may be useful in devising therapy. For example, the LQT3 form due to the sodium channel abnormality caused by mutations in the SCN5A gene (600163) can be treated with sodium channel blockers, whereas the LQT1 and LQT2 forms caused by mutations in the KCNH2 (152427) and KVLQT1 genes, respectively, should respond to drugs that open potassium channels.

In a retrospective study, Itoh et al. (2001) found that patients carrying mutations in the KCNQ1 gene responded better to beta-adrenergic blocking agents than did those with mutations in the KCNH2 gene (12 of 13 vs 1 of 5; p = 0.0077, Fisher exact test). The authors stated that this is a good example of the power of genetic diagnosis to direct the selection of appropriate therapy for patients with diseases of heterogeneous genetic etiology.

Miller et al. (2001) assessed the value of screening ECG for long QT syndrome in the family with LQT1 originally studied by Ackerman et al. (1998), in which there were 10 carriers of the F339del mutation (607542.0018) and 13 noncarriers. Using a QTc of greater than or equal to 460 ms as a diagnostic cutoff, the positive and negative predictive values for identifying at-risk individuals were 100%. Despite this, the computer-generated ECG diagnostic interpretation erroneously classified 6 of 23 family members, and half of the mutation-positive family members received the diagnostic interpretation 'normal ECG.' Miller et al. (2001) concluded that reliance on computer-generated ECG diagnostic interpretation alone will fail to identify many at-risk family members.

Brink et al. (2005) studied a South African LQTS founder population (SA-A341V) of Afrikaner origin (de Jager et al., 1996) in which there were 166 carriers of the A341V mutation in the KCNQ1 gene (607542.0010; see MOLECULAR GENETICS). Functional analysis revealed that the A341V mutant reduced the magnitude of wildtype channel repolarizing current I(Ks) by approximately 50%. In the South African cohort, patients with a resting HR greater than 73 bpm were at higher risk for cardiac events, whereas among patients with a QTc less than 500 ms, there was a linear correlation between risk of cardiac events and HR. Brink et al. (2005) concluded that HR plays a significant modulating role on the risk for cardiac events and that this arrhythmogenic role is accentuated in the presence of moderate, but not excessive, QT prolongation. Brink et al. (2005) proposed that the activation of I(Ks) during increased heart rate (HR), essential for QT interval adaptation during tachycardia, partly explains the high efficacy of beta-blocker therapy in LQT1 patients, since beta-blocking agents act not only on triggers but on the substrate by modifying HR. The long-term combined incidence of cardiac arrest and sudden death among LQT1 patients taking beta-blockers had been reported to be approximately 1% (Vincent et al., 2009; Priori et al., 2004).

Mapping

In a family estimated to contain more than 400 affected persons, Keating et al. (1991) observed no recombination between the LQT phenotype and the HRAS gene (190020), which is located on the short arm of chromosome 11. As rationale for HRAS as a candidate gene for the site of the mutation, Keating et al. (1991) pointed to the similarity of ras proteins to G proteins, which regulate myocardial and cardiac pacemaker ion channels. Physiologic data showed that p21 ras protein and GAP (guanosine triphosphatase-activating protein; 139150) regulate cardiac potassium channels. In a further study of 6 additional unrelated families with LQT, Keating et al. (1991) again found 'complete linkage' to HRAS; the lod score in these families was 5.25 at a recombination fraction of 0.0. By complete sequencing of the HRAS gene, Keating (1993) probably excluded this gene as the site of the mutation. Roy et al. (1994) presented evidence indicating that HRAS is not in the region containing the LQT gene and that LQT is more centromeric in 11p15.5 than previously thought.

The existence of more than one genetic form of LQT was indicated by the studies of Kerem et al. (1992), which excluded linkage with HRAS1 in a very large affected Jewish family originating from the island of Jerba near Tunis and later residing in Israel; see Benhorin et al. (1993). Curran et al. (1993) likewise found indications of genetic heterogeneity; linkage to HRAS1 and MUC2 (158370) was excluded in 2 kindreds. Genetic heterogeneity was established unequivocally by Jiang et al. (1994), who found linkage to a chromosome 7 marker, D7S483, in 9 families with a combined lod score of 19.41, and to a chromosome 3 marker, D3S1100, in 3 families with a combined lod score of 6.72. These findings localized major LQT genes to 7q35-q36 (LQT2; 613688) and 3p24-p21 (LQT3; 603830). However, in 3 families linkage to loci on chromosomes 3, 7, and 11 were excluded, indicating additional heterogeneity. In Taiwan, Ko et al. (1994) excluded linkage to 11p15.5 markers in a Chinese family. In the course of studies of 13 Japanese families, Tanaka et al. (1994) encountered evidence of genetic heterogeneity, making the 11p15.5 markers unsuitable for genetic diagnosis in most cases. In 15 of 23 families, Towbin et al. (1994) found linkage to HRAS1 on 11p15.5; of the remaining 8 families with negative lod scores, 4 were definitively excluded from linkage with HRAS1.

Dausse et al. (1995) readjusted the localization of the LQT1 gene on 11p15 by use of more extensive markers.

Molecular Genetics

In affected members of 16 families with long QT syndrome, Wang et al. (1996) identified heterozygous mutations in the KVLQT1 gene, including a 3-bp deletion (607542.0001) and 10 different missense mutations (607542.0002-607542.0011).

De Jager et al. (1996) restudied 4 South African families of northern European Afrikaner origin (designated as pedigrees 161, 162, 163, and 164) previously reported by Wang et al. (1996), in which affected individuals were heterozygous for an A341V substitution in the KCNA1 gene (607542.0010). Heterozygosity for the same mutation was identified in affected members of a fifth Afrikaner family with LQT (pedigree 166), and haplotype reconstruction revealed that all 5 families shared a common haplotype, indicating a founder effect. Clinical analysis of the 2 largest pedigrees showed that family 162 had an earlier age of onset of symptoms, fewer asymptomatic carriers, and more syncopal episodes per person, as well as significantly longer QTc interval range and mean in both carriers and noncarriers of the mutation, compared to family 161. De Jager et al. (1996) suggested that these differences in the spectrum of clinical symptoms might reflect the influence of different modulating environmental or genetic backgrounds on expression of the same mutant allele.

Modifier Effects of Variation in the AKAP9 Gene

In 349 members of a South African founder population of Afrikaner origin with long QT syndrome (LQT1; 192500), 168 of whom carried an identical-by-descent A341V mutation in the KCNQ1 gene (607542.0010), de Villiers et al. (2014) genotyped 4 SNPs in the AKAP9 gene (604001) and found statistically significant associations between certain alleles, genotypes, and haplotypes and phenotypic traits such as QTc interval length, risk of cardiac events, and/or disease severity. De Villiers et al. (2014) stated that these results clearly demonstrated that AKAP9 contributes to LQTS phenotypic variability; however, the authors noted that because these SNPs are located in intronic regions of the gene, functional or regulatory variants in linkage disequilibrium with the SNPs were likely to be responsible for the modifying effects.

Russell et al. (1996) used SSCP analysis to screen 2 large and 9 small LQT families for mutations of the KVLQT1 potassium channel gene. They identified a mutation (607542.0012) in the KVLQT1 gene in 2 unrelated families and, in a third family, another mutation (607542.0010) that resulted in the spontaneous occurrence of LQT in monozygotic twin offspring of unaffected parents.

A comprehensive review of the genetic and molecular basis of long QT syndromes was given by Priori et al. (1999, 1999).

Splawski et al. (2000) screened 262 unrelated individuals with LQT syndrome for mutations in the 5 defined genes (KCNQ1; KCNH2; SCN5A; KCNE1, 176261; KCNE2, 603796) and identified mutations in 177 individuals (68%). KCNQ1 and KCNH2 accounted for 87% of mutations (42% and 45%, respectively), and SCN5A, KCNE1, and KCNE2 for the remaining 13% (8%, 3%, and 2%, respectively).

From a cohort of 2,008 healthy individuals, Gouas et al. (2005) analyzed a group of 200 individuals with the shortest QTc intervals and a group of 198 with the longest QTc intervals, comparing the allele, genotype, and haplotype frequencies of polymorphisms in cardiac ion channel genes (10 SNPs in KCNQ1, 2 in KCNE1, 4 in SCN5A, and 1 in KCNH2) between the 2 groups. Based on observed differences, Gouas et al. (2005) suggested that genetic determinants located in these genes influence QTc length in healthy individuals and may represent risk factors for arrhythmias or cardiac sudden death in patients with cardiovascular disease.

Napolitano et al. (2005) screened the KCNQ1, KCNH2, SCN5A, KCNE1, and KCNE2 genes in 430 consecutive patients with LQT syndrome and identified 235 different mutations in 310 (72%) of the patients, 49% of whom had mutations in KCNQ1, 39% in KCNH2, 10% in SCN5A, 1.7% in KCNE1, and 0.7% in KCNE2. Fourteen (4.5%) of the patients carried more than 1 mutation in a gene. Fifty-eight percent of probands carried nonprivate mutations in 64 codons of the KCNQ1, KCNH2, and SCN5A genes; screening in a prospective cohort of 75 probands confirmed the occurrence of mutations at these codons (52%).

Arbour et al. (2008) identified a missense mutation (607542.0040) in the KCNQ1 gene causing long QT syndrome-1 among a First Nations community of northern British Columbia.

Johnson et al. (2008) sought the history of documented atrial fibrillation (AF) in 2 independent cohorts of LQT patients known to carry mutations in LQTS genes. Overall, early-onset AF was documented in 8 (1.7%) of 457 patients: 5 (2.4%) of 211 patients with LQT1 had documented AF (ATFB3; 607554), compared to 0 of 174 patients with LQT2, 1 (1.7%) of 59 patients with LQT3, 1 of 1 patient with Andersen-Tawil syndrome (170390), and 1 (2.9%) of 34 patients with multiple mutations. Johnson et al. (2008) noted that compared to the background prevalence of 0.1%, early-onset AF was observed in almost 2% of patients with mutation-positive LQTS, and concluded that AF should be viewed as an uncommon but possible LQT-related dysrhythmia.

Acquired Long QT Syndrome

In a patient who developed QT prolongation and torsade de pointes while taking the drug dofetilide, Yang et al. (2002) identified heterozygosity for a missense mutation in the KCNQ1 gene (R583C; 607542.0031). In vitro expression studies of the mutant protein confirmed a significant reduction in potassium currents, suggesting that the R583C mutation was responsible for the patient's response to dofetilide.

Digenic Inheritance

Berthet et al. (1999) studied a large Belgian family with LQTS in which both parents of 3 affected sisters had long QT intervals and family histories of sudden death. Haplotype analysis using microsatellite markers revealed linkage to LQT1 in the father and 2 severely affected daughters and linkage to LQT2 in the mother, the same 2 daughters, another more mildly affected daughter, and a grandson. In the 2 most severely affected sisters, who required multiple medications, cardiac sympathectomy, and pacemaker implantation for control of symptoms, Berthet et al. (1999) identified biallelic digenic mutations: a missense mutation in the KCNQ1 gene (A341E; 607542.0009) and a splice site mutation in the KCNH2 gene (2592+1G-A; 152427.0019). The father, 2 of his brothers, and a niece were all heterozygous for the A341E mutation in KCNQ1; the mother, the more mildly affected sister, and the grandson were heterozygous for the splice site mutation in KCNH2. Neither mutation was found in 2 unaffected sibs or in other unaffected members of the family. Berthet et al. (1999) stated that this was the first description of double heterozygosity in long QT syndrome.

Tester et al. (2005) analyzed 5 LQTS-associated cardiac channel genes in 541 consecutive unrelated patients with LQT syndrome (average QTc, 482 ms). In 272 (50%) patients, they identified 211 different pathogenic mutations, including 88 in KCNQ1, 89 in KCNH2, 32 in SCN5A, and 1 each in KCNE1 and KCNE2. Mutations considered pathogenic were absent in more than 1,400 reference alleles. Among the mutation-positive patients, 29 (11%) had 2 LQTS-causing mutations, of which 16 (8%) were in 2 different LQTS genes (biallelic digenic). Tester et al. (2005) noted that patients with multiple mutations were younger at diagnosis, but they did not discern any genotype/phenotype correlations associated with location or type of mutation.

In 44 unrelated patients with LQT syndrome, Millat et al. (2006) used DHLP chromatography to analyze the KCNQ1, KCNH2, SCN5A, KCNE1, and KCNE2 genes for mutations and SNPs. Most of the patients (84%) showed a complex molecular pattern, with an identified mutation associated with 1 or more SNPs located in several LQTS genes; 4 of the patients also had a second mutation in a different LQTS gene (biallelic digenic inheritance; see 607542.0038, 607542.0039, 152427.0023, and 600163.0007). Millat et al. (2006) suggested that because double heterozygosity appears to be more common than expected, molecular diagnosis should be performed on all LQTS-related genes, even after a single mutation has been identified.

Associations Pending Confirmation

For a discussion of a possible association between LQT and mutation in the KCNE3 gene, see 604433.

Genotype/Phenotype Correlations

In a large collaborative study, Zareba et al. (1998) demonstrated that the genotype of the long QT syndrome influences the clinical course. The risk of cardiac events (syncope, aborted cardiac arrest, or sudden death) was significantly higher among subjects with mutations at the LQT1 or LQT2 locus than among those with mutations at the LQT3 locus. Although the cumulative mortality was similar regardless of the genotype, the percentage of cardiac events that were lethal was significantly higher in families with mutations at the LQT3 locus. In this large study, 112 patients had mutations at the LQT1 locus, 72 at the LQT2 locus, and 62 at the LQT3 locus. Thus, paradoxically, cardiac events were less frequent in LQT3 but more likely to be lethal; the likelihood of dying during a cardiac event was 20% in families with an LQT3 mutation and 4% with either an LQT1 or an LQT2 mutation.

Kimbrough et al. (2001) reported the findings of a study of 211 LQT syndrome probands and 791 first-degree relatives. They found that the clinical severity profile of the disease in the proband was not a useful indicator of disease status in parents or sibs.

Priori et al. (2003) stratified risk according to genotype, in conjunction with other clinical variables such as sex and QT interval length, in 647 patients from 193 consecutively genotyped families with LQTS, of whom 386 carried a mutation at the LQT1 locus, 206 a mutation at the LQT2 locus, and 55 a mutation at the LQT3 locus. The cumulative probability of a first cardiac event, defined as the occurrence of syncope, cardiac arrest, or sudden death before the age of 40 years and before the initiation of therapy, was determined according to genotype, sex, and the QT interval corrected for heart rate (QTc). Within each genotype, Priori et al. (2003) also assessed risk in the 4 categories derived from the combination of sex and QTc (less than 500 ms and 500 ms or more). They found that the incidence of a first cardiac event before the age of 40 years and before the initiation of therapy was lower among patients with a mutation at the LQT1 locus (30%) than among those with a mutation at the LQT2 or LQT3 loci (46% and 42%, respectively). Multivariate analysis showed that the genetic locus and the QTc, but not sex, were independent predictors of risk. The QTc was an independent predictor of risk among patients with a mutation at either the LQT1 or the LQT2 locus but not among those with a mutation at the LQT3 locus. Among patients with a mutation at the LQT3 locus, sex was an independent predictor of events, i.e., male patients became symptomatic much earlier than female patients even when their QTc was below 500 ms; the authors noted, however, that caution was required in drawing conclusions from this group because of its relatively small size. Vincent (2003) noted that even with the important work of Priori et al. (2003), risk prediction remained difficult, which he illustrated with several cases. A 13-year-old boy with the LQT1 genotype died suddenly while running, with no prior symptoms. His electrocardiogram, obtained 2 weeks earlier as part of family screening, was normal, with a QTc of 450 ms. A 20-year-old woman with the LQT2 genotype died in her sleep; her electrocardiogram has been found to be normal, with a QTc of 460 ms.

Westenskow et al. (2004) analyzed the KCNQ1, KCNH2, SCN5A, KCNE1, and KCNE2 genes in 252 probands with long QT syndrome and identified 19 with biallelic mutations in LQTS genes, of whom 18 were either compound heterozygous (monogenic) or double heterozygous (digenic) and 1 was homozygous. They also identified 1 patient who had triallelic digenic mutations (see 152427.0021). Compared with probands who had 1 or no identified mutation, probands with 2 mutations had longer QTc intervals (p less than 0.001) and were 3.5-fold more likely to undergo cardiac arrest (p less than 0.01). All 20 probands with 2 mutations had experienced cardiac events. Westenskow et al. (2004) concluded that biallelic mono- or digenic mutations (which the authors termed 'compound mutations') cause a severe phenotype and are relatively common in long QT syndrome. The authors noted that these findings support the concept of arrhythmia risk as a multi-hit process and suggested that genotype can be used to predict risk.

Brink et al. (2005) studied an LQTS founder population (SA-A341V) consisting of 22 apparently unrelated South African kindreds of Afrikaner origin, including the LQTS pedigrees designated 161, 162, 163, 164, and 166 that were previously reported by Wang et al. (1996) and de Jager et al. (1996), all traceable to a single founding couple of mixed Dutch and French Huguenot origin who married in approximately 1730. Among the 166 carriers of the A341V mutation in the KCNQ1 gene (607542.0010), 131 (79%) were symptomatic, with a median age of 6 years at first cardiac event, and 23 (14%) had sudden cardiac death before 20 years of age. The mutation carriers exhibited a wide range of QTc values, from 406 to 676 ms, with 12% having a normal QTc (440 ms or less); QTc was longer in symptomatic than asymptomatic individuals. Both QTc of 500 ms or greater and heart rate of 73 bpm or greater were significant risk factors for experiencing cardiac events after controlling for other covariates. Brink et al. (2005) compared the Afrikaner patients to the general LQT1 population (Priori et al., 2003) and found that the SA-A341V group exhibited a significantly more severe form of the disease, with an earlier age of onset, longer QTc intervals, and an increased incidence of first cardiac event by age 20 years. Functional analysis in CHO cells demonstrated that coexpression of the A341V mutant reduced the magnitude of the wildtype channel repolarizing current I(Ks) by approximately 50%, indicating that the mutation exerts a dominant-negative effect. Brink et al. (2005) noted that this effect on I(Ks), which activates during increased heart rate and is essential for QT interval adaptation during tachycardia, might explain why 79% of lethal arrhythmic episodes in LQT1 patients with mutations impairing I(Ks) occur during exercise. In contrast, most lethal episodes in LQT2 and LQT3 patients occur during startle reaction and at rest or during sleep, respectively.

Nomenclature

It has been pointed out by many, e.g., Martini (1998), that this syndrome should be called Romano-Ward syndrome since Romano et al. (1963) described it 1 year before Ward (1964).

History

Itoh et al. (1982) reported a family in which 10 persons with the Ward-Romano syndrome had the same HLA haplotype and suggested that a chromosome 6 gene may cause the disorder. Weitkamp and Moss (1985) did a lod score analysis of the Itoh family and of a second family studied by them and arrived at a maximum lod score of 3.68 at theta 0.04 and 0.05. The single recombinant showed no evidence of recombination within the region demarcated by the loci HLA-A, -B, -C, and -DR and the GLO locus. Therefore, the LQT locus was thought to be outside the HLA-A:GLO segment on 6p. Melki et al. (1987) provided further data corroborating the linkage to HLA. However, analysis of HLA haplotypes in kindreds with the long QT syndrome forced Weitkamp et al. (1989) to conclude that the LQT locus is in fact not linked to HLA. Giuffre et al. (1990) and Keating et al. (1991) also excluded linkage to HLA.

Weitkamp et al. (1994) presented an analysis of HLA haplotype sharing among affected pedigree members, showing an excess of haplotype sharing in a previously published Japanese pedigree and possibly also in 15 families of European descent. The haplotypes shared by affected persons derived from both affected and unaffected parents. They also found a nonrandom distribution of the HLA-DR gene in patients with LQT compared with controls, suggesting an association between LQT phenotype and specific HLA-DR genes. Their data indicated that DR2 has a protective effect and, particularly in males, that DR7 may increase susceptibility to the LQT syndrome. Thus, LQT syndrome may be influenced by genes on chromosomes 11 and 6, possibly with a sex-specific effect.