Header Graphic

                     Back to Patient's Version


Congenital long-QT syndrome is a genetic disorder encompassing a family of mutations that can lead to aberrant ventricular electrical activity.  These genetic mutations are called “channelopathies”; they are the responsible that genes that encode for protein channels that regulate the flow of sodium, potassium, and calcium ions in and out of the cardiac myocyte.  The result of these mutations is an increased risk of ventricular arrhythmia, specifically torsade de pointes, that can lead to syncope, aborted cardiac arrest, and sudden cardiac death.  LQTS manifests on the patient’s electrocardiogram as a prolongation of the corrected QT  (QTc) interval and/or abnormal morphology of the T-wave (Figure 1). 

            When measuring the degree of QT interval prolongation, it is necessary to correct for the patient’s heart rate, termed QTc.   This is done with Bazett’s Formula:


Prior to genetic mapping, LQTS was divided into two syndromes based on inheritance pattern and associated non-cardiac defects.  The Romano-Ward Syndrome (RWS) has an autosomal dominant inheritance and no associated genetic defects.  The second and more rare syndrome, Jervell Lange-Nielsen Syndrome (JLNS), has an autosomal recessive inheritance and the presence of bilateral sensori-neural deafness.  The hearing loss results from disturbed potassium ion handling in the inner ear. JLNS represents a more severe form of cardiac disease with greater risk of fatal arrhythmia.

Currently, LQTS is classified according to10 possible types of ion channel mutations, LQT1-10[1][2], that encompass the prior classification system with some overlap.  LQT1-6, 8-10 correspond to RWS, while LQT 1 & 5 correspond to JLNS when associated with deafness[3].  LQT7, also known as Anderson-Tawil syndrome, is associated with periodic paralysis, dysmorphic features, and cardiac arrhythmias[4].


In the United States, the incidence of congenital LQTS is estimated to be one in 7,000-10,000[5][6].  There is a female preponderance, ranging from 1.6-2.0:1[7][8].  Among patients enrolled in The International Long-QT registry, the average age of presentation is  2115 years.  Presentation most often is with a syncopal episode.   Males often present during pre-adolescence while females present later[9].  Congenital LQTS is also believed to be one of the causes of sudden infant death accounting for 5-10% of cases.[10][11]. 

Of the ten types of channelopathies, the first three, LQT1-3, are the most prevalent and most studied. LQT1 occurs in 30-35%, LQT2 in 25-30%, LQT3 in 5-10%, LQT4 in 1-2%, and LQT5 in 1% of cases.  LQT6-10 are all rare2.


The ventricular myocyte activation cycle, or action potential, is dependent on ion channels (Figure 2).   During rest (phase 4), molecular pumps in the cell membrane of the myocyte push sodium and calcium ions out of the cell and bring potassium ions into the cell.  The cell membrane is impermeable to backflow of the sodium and calcium. However, via the inward rectifying potassium channels (IK1 and IKAch), positively charged potassium ions slowly leak out of the cell and leave the interior negatively charged at -80mV(also known as the membrane potential).  As the membrane potential becomes less negative, it reaches a critical threshold value and depolarizes.  The cell membrane voltage-gated sodium channel, NaV1.5 (INa), then opens allowing positively charged sodium to rapidly flow into the cell.  This is phase 0 or depolarization.  To maintain this depolarization, L and T type calcium channels (ICaL and ICaT) are activated to allow the influx of positive calcium ions.   During the next phase, phase 1, two types of delayed rectifier potassium channels open to allow efflux of potassium (ITo and IKur) and thereby slightly reducing the cells now positive internal charge.  During Phase 2, there is equilibrium between influx of calcium and the efflux of potassium, and the cells internal charge plateaus.  During phase 3, efflux of potassium ions predominates with the activation of the slow and rapid delayed rectifier channels (IKs, and IKr).  At the terminal portion of Phase 3, the inward rectifying potassium channels (IK1 and IKach) are activated and further helps to extrude potassium.  This action results in a return of the membrane potential to its negative resting potential.  Phases 1-4 are termed repolarization[12].

In the various sub-types of LQTS, mutations lead to malfunction of the different ion channels.  In all subtypes, the overall effect is a prolongation of repolarization.  With the prolonged repolarization, L-type calcium channels can be promoted to re-open causing a rise in cellular calcium.  This rise then can lead to a premature depolarization of the myocyte (early after depolarization) resulting in a premature beat .  Because there is heterogeneity of the repolarization in the surrounding cells, this early beat can then lead to depolarization of neighboring myocytes setting off an unstable ventricular tachycardia or torsade de pointes[13].  Syncope occurs when torsade de pointes is short lived and self terminates. However, if the torsade de pointes does not terminate, it can degenerate to ventricular fibrillation and ultimately can result in sudden death.  

Sub-Types (Table 1):




Mutation Effect

ECG finding




K+  Efflux

Broad, late-inset, T wave




K+  Efflux

Widely-split, low-amplitude, T wave




Prolonged Na+ influx

Biphasic or peaked, late-onset, T wave




Build-up of Na+ within cell and Ca2+ outside of cell

Variable Qt interval prolongation




K+  Efflux

Not defined




K+  Efflux

Not defined




K+  Efflux

Modest prolongation of Qt interval




Prolonged Ca2+  influx

Exaggerated Qt interval prolongation




Prolonged Na+ influx

Not defined


Extremely rare, found in 1 family

SCN4 β

Prolonged Na+ influx

Not defined

An expanded table


In this subtype, the slowly activating delayed rectifier potassium channel, IKs, has a loss of function due to various mutations of the gene KVLQT1 (KCNQ1) at locus 11p15.5.  The slow potassium channel is composed of 2 types of proteins: one alpha sub-unit and one beta subunit.  The alpha subunit forms the ion channel and the beta subunit works as a modifier of the channels functioning.  The KVLQT1 gene product is responsible for the formation of the alpha subunit.  Mutations of this gene lead to a prolongation of repolarization by decreasing potassium efflux.  When a homozygous mutation and sensori-neural deafness are present, this combination is labeled JLNS 1 as opposed to LQT1.  The electrocardiogram of patients with LQT1 most often has as a broad-based T-wave or a late-onset normal appearing T-wave7, 3(Figure 2).

Figure 2



Adapted from Moss AJ, Zareba W, Benhorin J, et. al. ECG T wave patterns in genetically distinct forms of the hereditary long QT syndrome. Circulation. 1995;92:2929-34.


Mutations in the Human “Ether-a-go-go” Related Gene (HERG) are responsible for this subtype.  The locus for HERG is 7q35-36 and encodes for the alpha subunit of the rapidly activating delayed rectifier potassium channel, IKr. As in LQT1, the channel protein is composed with an alpha and beta subunit.  Mutations in this gene lead to loss-of-function of this channel and prolonged repolarization.  The electrocardiogram can display widely split or low-amplitude T-waves7,3(Figure 2)


Unlike LQT1 and LQT2, the primary mutation in this subtype leads to continued activation of the voltage-gated sodium channel.  Mutations in gene SCN5A located on chromosome 3p21-24 lead to a mutation of the alpha subunit resulting in a gain-of-function of the channel. This mutation prolongs the influx of sodium, extending repolarization.  The electrocardiogram for these patients will often have late-onset of the T-wave that can be biphasic or peaked7,3(Figure 2)


The genetic mutation in this subtype affects sodium, potassium, and calcium ion flow.  Mapped to chromosome 4q25-27, the ANKB gene encodes for the ankyrin-B adaptor protein.  This protein is responsible in anchoring the Na-K ATPase and Na/Ca exchanger on the cell membrane.  Loss of function of these proteins leads to a build-up of sodium within the cell and calcium outside of the cell.  The electrocardiogram in these patients shows variable QT prolongation and often have a normal QT interval7,3.


Similar to LQT1, the slow delayed rectifier potassium channel (IKs) in this subtype has a loss of function. However, LQT5 has mutations of the beta subunit.   The gene that encodes for the beta subunit is the minK (KCNE1) at locus 21q22.1-2.  As in LQT1, decreased potassium efflux leads to prolonged repolarization.  Also similar to LQT1, when homozygous mutations and sensori-neural deafness are present, this combination is called JLNS2.  There is no known pathonemonic finding on electrocardiogram for LTQ57,3.


Mutations of the MiRP1 gene located on chromosome 21q22.1 lead to a loss of function of the beta subunit of the rapid delayed rectifier potassium channel, (IKr).   Similar to LQT2, potassium efflux is decreased and repolarization is prolonged. Electrocardiogram findings have not yet been defined for this subtype7,3. 


In this subtype, there is a mutation of the KCNJ2 gene located on chromosome 17q23.  This mutation leads to a loss-of-function of the inward rectifying channel (IK1), resulting in decreased potassium efflux.  The electrocardiogram of these patients has a prominent U wave and a normal or modestly prolonged QT interval.

Although LQT7 is currently included among the sub-types of congenital long-QT syndrome, this inclusion is controversial.  It is now believed that the previous studies that led to the inclusion of LQT7 inaccurately measured the QT interval by including the large U wave which is pathonemonic for this mutation3,4,7.


In LQT8, inactivation of the L-type calcium channel causes prolonged calcium inflow and markedly prolonged repolarization.  The gene responsible for this mutation is CACNA1C which is mapped to chromosome 12p13.3 and encodes for the alpha subunit of the L-type calcium channel.  The electrocardiogram of these patients can have an exaggerated prolongation of their QT interval7,3.


As in LQT3, this subtype has prolonged activation of the rapid sodium channel.  The gene CAV3 localized to chromosome 3p25 encodes for the Caveolin-3 protein.  The Caveolin-3 protein forms an invagination in the cell membrane, and the voltage-gated sodium channel co-localizes within the “cave” on the cell membrane.  In a mechanism that has not yet been fully defined, mutations in the CAV3 gene lead to prolonged activation of rapid sodium channels and a prolonged phase 0 of the action potential2.  Electrocardiographic findings for this subtype have not yet been defined.


The gene SCN4β located on chromosome 11q23 encodes for the beta subunit of the voltage-gated sodium channel.  Mutations of this gene lead to a gain-of-function of the sodium channel akin to the mutations that cause LQT32.  This subtype is very rare, and electrocardiographic findings have not been defined.    


Diagnosis of LQTS is challenging.  Patients can be referred for evaluation for many reasons.  Symptomatic patients are those who have had unexplained syncope or aborted sudden cardiac death. Asymptomatic patients have a prolonged QTc on routine electrocardiogram or have a first-degree relative diagnosed with the disease.  In both the asymptomatic and symptomatic groups, the QTc interval can be normal on initial presentation[14].  The asymptomatic patients are also more likely to have a normal QTc duration then the symptomatic patients[15].

A scoring system has been created to aid in the diagnosis of LQTS[16].  A score of ≥ 4 indicates a high probability of LQTS; 2 or 3 indicates intermediate probability; ≤ 1 indicates a low probability:

Clinical Finding


QTc interval

     ≥ 480 ms

     460-470 ms

     450 ms (men)





Torsade de pointes


T-wave alternans


Notched T-wave in 3 leads


Low heart rate for age (children)



     With stress

     Without stress




Congenital deafness


Family member with definite LQTS


Unexplained SCD in immediate family member younger than 30 years old


A score 4 is diagnostic. However, patients with the disease who score lower will be missed by this strict criteria.  If a cut-off of 2 is used, greater than 50% of patients will be diagnosed with only a 10% false positive diagnosis15.  Because use of the electrocardiogram and this scoring system lead to an under-diagnosis, further testing may be indicated for patients with a low to intermediate score.  Patients can be followed over time and have repeat electrocardiograms that may show variability in the QTc interval[17]. 

Among patients with LQT1, there is a higher prevalence of “concealed LQTS” or LQTS with a normal QTc.   Stress testing with either exercise or an epinephrine infusion can unmask the cases of LQT1 by revealing a pathonomonic failure of the QTc interval to shorten with stimulation[18][19][20].  Exercise testing is limited by artifact on the ECG tracing which results in an inaccurate measurement of the QT interval.  For the epinephrine QT stress testing, there are currently two protocols available.  The Mayo Clinic protocol uses a continuous infusion of epinephrine with a doubling of the dose every 5 minutes.  The diagnosis of LQT1 is supported when there is an increase in the QT interval of 30ms at a dose of 0.1 mg/kg/min.  A positive test is 76% predictive of LQT1.  A negative test can almost virtually rule-out LQT1. However, a negative test does not rule-out other subtypes of LQTS19. 

The second epinephrine protocol, the Shimuzu protocol, has the added benefit of being able to not only diagnose LQT1 but also to diagnose some cases of LQT2.  For this protocol, a bolus of epinephrine is given and then followed by a continuous infusion.  If the QTc is prolonged by greater than 35msec above the baseline during the infusion portion, this result is 90% predictive of LQT1.  However, if the QTc does not meet these criteria but is prolonged by greater than 80msec after the bolus, LTQS2 can be 100% accurately diagnosed[21].  It is important to note that with any provocative testing such as exercise or epinephrine infusion, a negative test does not rule-out LQTS. 

Recently, genetic testing has become commercially available for the diagnosis and sub-typing of LQTS and is available along with genetic counseling at our Program.  The current recommendations for genetic testing are to determine the genotype of patients clinically diagnosed with LQTS and to test first-degree relatives of patients with known LQTS[22].  The test can identify 76% of patients with LQT1, LQT2, LQT3, LQT5, and LQT6[23].  In those cases with a high suspicion for LQTS but with a negative test, the false negative result is most likely due to a mutation that has not yet been identified or a mutation that is not tested for with the current test[24].


Treatment of LQTS is guided by the individual’s risk of sudden cardiac death.  Patients who have already had an aborted sudden cardiac arrest are considered to have the highest risk of a recurrent event[25]. In these patients, medical treatment with beta-blockers and placement of an implantable cardioverter-defibrillator is strongly recommended[26][27]. 

For patients without prior cardiac events, therapy is initiated with a beta-blocker medication and lifestyle modifications. This treatment is especially important for those patients with prolonged QTc intervals, as increasing QTc interval is directly related to increased risk of sudden cardiac death[28]. 

Lifestyle modification includes avoiding triggers of cardiac events and medications that prolong the QT interval.   Triggers for LQT1 include stress and exercise, especially swimming.  For LQT2, triggers include auditory stimuli and stress.  For LQT3, the primary triggers are rest and sleep; hence, here there are no specific triggers to avoid[29].   For all other subtypes, information on triggers has not yet been defined. 

If patients continue to suffer from syncope and/or ventricular arrhythmia despite lifestyle modification and beta-blocker therapy, ICD placement is recommended26.  Another controversial option for these patients is left cervicothoracic sympathetic ganglionectomy.  This surgical procedure involves removal of nerve plexi that are believed to modulate sympathetic activity on the heart[30].

Follow-up of a cohort of patients who underwent ICD placement found the average number of ICD shocks to be 1.1 2.2 shocks/patient-year, of which 5.3% were inappropriate.   Among the patients with prior aborted sudden cardiac death, there was a clustering of events with an event rate of 1.3 + 2.5 shocks/patient-year.  For all others, the event rate was lower at 0.2 0.3 shocks/year[31]. With the advent of reliable genetic testing and a better understanding of the pathophysiology of the various mutations, genotype-specific pharmacotherapy is also being developed.  For patients with LQT2, researchers have found that by increasing the extracellular potassium concentration, the QT interval can be shortened.  In one small trial, eight patients were given potassium supplementation and spironolactone with a goal potassium of 1.5meq/l above their baseline.  The average potassium level achieved was 1.2meq/l above the baseline and all but one patient had significant improvement of their QTc interval.  This study was limited by the requirement for frequent blood draws and short follow-up.  It is also not known whether this therapy is clinically significant and leads to a reduction in cardiac events.

For patients with LQT3, use of mexiletine in conjunction with a beta-blocker or ICD has been suggested[32].  In a small clinical trial, patients with LQT2 and LQT3 were placed on mexiletine.  Patients in the LQT3 group had a significant reduction in their QTc interval while LQT2 patients did not.  However, it is not known whether the improvement in the ECG parameters translates to a reduction in cardiac events[33].

For both LQT 7 and 8, there are case reports showing a possible benefit with verapamil.  Use of verapamil was found to reduce the ventricular tachycardia burden in both sub-types. However, the long-term safety is unknown[34][35]. 

There are also special considerations for early ICD implant if the genotype of the patient is known.  Of the three predominant subtypes, LQT3 has the greatest risk of cardiac events at 0.60%/year, followed by LQT2 at 0.56%/year, and LQT1 at 0.30%/year28.  Current guidelines allow for consideration of ICD placement in patients with LQT2 and LQT3. However these guidelines are controversial26.   Priori et. al. recommend using QTc duration and gender to further risk stratify these three subtypes.  When including QTc duration to determine risk of event, as the QTc increases, the risk dramatically increases in LQT2 and LQT3 and in males with LQT1. In these patients when the QTc duration increases to greater than 500 msec the risk of cardiac events before the age of 40 years old is 50%. .  For females with LQT3, the QTc duration does not affect their risk of cardiac events and their overall risk is intermediate (30-49% before the age of 40).  Other genotypes with an intermediate risk include females with LQT2 and males with LQT3 who have a QTc duration  < 500 msec.  All other variants are low risk and have a risk of causing a cardiac event before the age of 40 of < 30%28. 

Figure 3  

Nattel S. and Carlsson L. Innovative approaches to anti-arrhythmic drug therapy.

Nat Rev Drug Discov. 2006 Dec;5(12):1034-49

[1] Robin NH, Taberaux PB, Genetic Testing in Cardiovascular Disease.  J Am Coll Cardiol, 2007; 50:727-   


[2] Lehnart SE, Ackerman MJ, Benson W, et. al.  Inherited arrhythmias: a National Heart, Lung, and Blood Institute and Office of Rare Diseases workshop consensus report about the diagnosis, phenotyping, molecular mechanisms, and therapeutic approaches for primary cardiomyopathies of gene mutations affecting ion channel function.  Circulation. 2007 Nov 13;116(20):2325-45. 

[3] Chiang CE and Roden DM.  The long QT syndromes: genetic basis and clinical implications.  J Am Coll Cardiol.  2000 Jul;36(1):1-12.


[4] Zhang L, Benson W, Tristani-Firouzi M, et al.  Electrocardiographic features in Andersen-Tawil syndrome patients with KCNJ2 mutations.  Characteristic T-U-wave patterns predict the KCNJ2 genotype.  Circulation. 2005l111:2720-2726

[5]Vincent GM. The molecular genetics of the long QT syndrome: genes causing fainting and sudden death.  Annu Rev Med. 1998;49:263-74.

[6] Schwartz, PJ.  The long QT syndrome.  Curr Probl Cardiol. 1997 Jun;22(6):297-351.

[7] Modell SM and Lehmann MH. The long QT syndrome family of cardiac ion channelopathies: a HuGE review.Genet Med. 2006 Mar;8(3):143-55.

[8] Moss AJ, Schwartz PJ, Cramptom RS, et. al. The long QT syndrome. Prospective longitudinal study of 328 families.  Circulation. 1991 Sep;84(3):1136-44 

[9] Locati EH, Zareba W, Moss AJ, et. al.  Age- and sex-related differences in clinical manifestations in patients with congenital long-QT syndrome: findings from the International LQTS Registry.  Circulation. 1998 Jun 9;97(22):2237-44.

[10] Tester DJ, Ackerman MJ.  Sudden infant death syndrome: how significant are the cardiac channelopathies? Cardiovasc Res.  2005;67:388-396.


[11] Arnestad M, Crotti L, Rognum TO, et al.  Prevalence of Long-QT syndrome gene variants in sudden infant death syndrome.  Circulation 2007;115:361-367.

[12] Priori SG, Barhanin J, Hauer RN, et. al. Genetic and molecular basis of cardiac arrhythmias: impact on clinical management part III.  Circulation. 1999 Feb 9;99(5):674-81.

[13] Zipes DP, Libby P, Bonow OR, and Braunwald E.  Braunwald’s heart disease: a textbook of cardiovascular medicine. 2005, 7th edition.

[14] Napolitano C, Prior SG, and Schwartz PJ. Genetic testing in the Long-QT syndrome: development and validation of an efficient approach to genotyping in clinical practice.  JAMA. 2005 Dec 21;294(23):2975-80.

[15] Priori SG, Napolitano C, and Schwartz PJ.  Low penetrance in the Long-QT syndrome: clinical impact.  Circulation.  1999;99(:529-533.


[16] Schwartz PJ, Moss AJ, Vincent GM, and Cramptom RS.  Diagnostic criteria for the Long-QT syndrome: an update.  Circulation 1993;88(2):782-784.

[17] Goldenberg I, Mathew J, Moss AJ, et. al. Corrected QT variability in serial electrocardiograms in Long- QT syndrome: the importance of the maximum corrected QT for risk stratification.  J Am Coll Cardiol. 2006 Sep 5;48(5):1047-52.

[18] Takenaka K, Tomohiko A, Shimizu W, et. al.  Exercise stress test amplifies genotype-phenotype correlation in the LQT1 and LQT2 forms of the Long-QT syndrome.  Circulation. 2003 Feb 18;107(6):838-44.

[19]Vyas H, Hejlik J, and Ackerman MJ.  Epinephrine QT stress testing in the evaluation of congenital long-QT syndrome: diagnostic accuracy of the paradoxical QT response.  Circulation. 2006 Mar 21;113(11):1385-92.

[20] Vyas H and Ackerman MJ.  Epinephrine QT stress testing in congenital long QT syndrome.  Journal of Electrocardiology. 2006;39:s107-s113.


[21] Shimizu W, Noda T, Takaki H, et. al.  Diagnostic value of epinephrine test for genotyping LQT1, LQT2, and LQT3 forms of congenital long QT syndrome.  Heart Rhythm. 2004;3:276-283.

[22] Robin NH, Tabereaux PB, Benza R, and Korf BR.  Genetic testing in cardiovascular disease.  J Am Coll Cardiol. 2007 Aug 21;50(8):727-37 

[23] Taggart NW,  Haglund CM, Tester DJ, and Ackerman MJ. Diagnostic miscues in congenital long-QT syndrome. Circulation. 2007;115:2613-2620.


[24] Roden DM. Long-QT syndrome.  N Eng J Med.  2008;358: 169-176.


[25] Moss AJ, Zareba W, Hall WJ, et. al.  Effectiveness and limitations on beta-blocker therapy in congenital long-QT syndrome.  Circulation.  2000;101:616-623.


[26]Zipes DJ, Camm AJ, Borggreffe M et. al.  ACC/AHA/ESC 2006 Guidelines for Management of Patients With Ventricular Arrhythmias and the Prevention of Sudden Cardiac Death: a report of the American College of Cardiology/American Heart Association Task Force and the European Society of Cardiology Committee for Practice Guidelines (writing committee to develop Guidelines for Management of Patients With Ventricular Arrhythmias and the Prevention of Sudden Cardiac Death): developed in collaboration with the European Heart Rhythm Association and the Heart Rhythm Society.  Circulation. 2006 Sep 5;114(10):e385-484.


[27] Moss AJ, Zareba W, Hall WJ, et al. Effectiveness and limitations of beta-blocker therapy in congenital long-QT syndrome.  Circulation. 2000 Feb 15;101(6):616-23.


[28] Priori SG, Schwartz PJ, Napolitano C, et. al.  Risk stratification in the long-QT syndrome.  N Engl J Med 2003;348:1866-1874


[29] Schwartz PJ, Priori SG, Spazzolini C, et al.  Genotype-phenotype correlation in the long-QT syndrome: gene-specific triggers for life-threatening arrhythmias.  Circulation. 2001 Jan 2;103(1):89-95.


[30] Moss AJ.  Clinical management of patients with the long-QT syndrome: drugs, devices, and gene-specific therapy.  PACE 1997;20:2058-2060.


[31] Monnig G, Kobe J, Loher A, et. al. Implantable cardioverter-defibrillator therapy in patients with congenital long-QT syndrome: a long-term follow-up.  Heart Rhythm.  2005;2:497-504.


[32] Shimizu W, Aiba T, and Antzelevitch C.  Specific therapy based on the genotype and cellular mechanism in  inherited cardiac arrhythmias.  Long Qt syndrome and Brugada syndrome.  Current Pharmaceutical Design.  2005;11:156101572.


[33] Schwartz PJ, Priori SG, Locati EH, et. al  Long Qt syndrome patients with mutations of the SNC5A and HERG genes have differential responses to Na+ channel blockade and to increases in heart rate: implications for gene-specific therapy.  Circulation.  1995;92:3381-3386


[34] Kannankeril PJ, Roden DM, and Fish FA.  Suppression of bidirectional ventricular tachycardia and unmasking of prolonged QT interval with verapamil in Andersen’s syndrome.  J Cardiovasc Electrophysiol.  2004;15:119.


[35] Jacobs A, Knight BP, McDonald KT, Burke MC. Verapamil decreases ventricular tachyarrhythmias in a patient with Timothy syndrome (LQT8). Heart Rhythm. 2006 Aug;3(8):967-70.