Long QT Syndrome Ilan Goldenberg, MD, Wojciech Zareba, MD, PhD, and Arthur J. Moss, MD Abstract: The hereditary Long QT syndrome (LQTS) is a genetic channelopathy with variable penetrance that is associated with increased propensity for poly- morphic ventricular tachyarrhythmias and sudden cardiacdeathinyoungindividualswithnormalcardiac morphology. The diagnosis of this genetic disorder relies on a constellation of electrocardiographic, clini- cal, and genetic factors. Accumulating data from re- cent studies indicate that the clinical course of affected LQTS patients is time-dependent and age-specific, demonstrating important gender differences among age groups. Risk assessment should consider age– gender interactions, prior syncopal history, QT-inter- val duration, and genetic factors. Beta-blockers consti- tute the mainstay therapy for LQTS, while left cardiac sympathetic denervation and implantation of a cardio- verter defibrillator should be considered in patients whoremainsymptomaticdespitebeta-blockertherapy. Current and ongoing studies are also evaluating geno- type-specific therapies that may reduce the risk for life-threatening cardiac events in high-risk LQTS patients. (Curr Probl Cardiol 2008;33:629-694.) Melvin M. Scheinman: The comment that beta-blockers remain the main- stay of therapy for patients with the Long QT syndrome (LQTS) is true as a generalization. As will be made abundantly clear by the authors, beta- blockersappeartobelesseffectiveinsomeformsoftheLongQTsyndrome (LQT2) and are without value for other forms (LQT3). Theauthorshavenoconflictsofinteresttodisclose. CurrProblCardiol2008;33:629-694. 0146-2806/$–seefrontmatter doi:10.1016/j.cpcardiol.2008.07.002 CurrProblCardiol,November2008 629 T he Long QT syndrome (LQTS) is a hereditary disorder in which most affected family members have delayed ventricular repolar- ization manifest on the electrocardiogram (ECG) as QT prolon- gation.1,2 The disorder is associated with an increased propensity to arrhythmogenic syncope, polymorphous ventricular tachycardia (torsade de pointes), and sudden arrhythmic death. LQTS is due to mutations involving principally the myocyte ion-channels, and this monogenetic disorder has an autosomal-dominant inheritance pattern. About 85% of the reported cases are inherited from one of the parents, with the remaining 15% of the affected patients having de novo mutations. The disease is relatively infrequent with an overt prevalence estimated at about1:3,000-1:5,000inthegeneralpopulation.Thedisorderhasvariable penetrance. In recent studies, two LQTS mutations were identified in approximately 10% of genotyped LQTS patients,3 and this finding suggests that this genetic disorder may be considerably more frequent than is generally appreciated. LQTS patients may be especially suscep- tible to drug-induced cardiac arrhythmias. Clinical and genetic studies of patients with LQTS have provided unique insight into the electrophysi- ology of the heart and basic arrhythmogenic mechanisms.4 In 1957, two Norwegian physicians, Drs. A. Jervell and F. Lange- Nielsen, reported a family of six siblings, four of whom were deaf with recurrentfaintingattacks.5Oneofthedeafchildrendiedsuddenlybefore the family was evaluated. The other three deaf children had QT prolon- gation on the ECG, and two of them died suddenly. A copy of the ECG recorded a few months before one of the children died suddenly during a syncopal episode is presented in Fig 1. The parents and the two non-deaf childrenwerehealthywithnormalECGs.Theverynextyear,Levineand Woodworth published a similar case in a deaf patient with syncopal attacks, QT prolongation, and sudden death.6 The disorder was initially described as the surdo-cardiac syndrome7 but subsequently was referred toastheJervellandLange-Nielsensyndrome(JLS).Itisnowappreciated that this syndrome is due to homogozygous mutations involving the KCNQ1 gene in which the extreme severity of the cardiac condition is due to inheritance of two KCNQ1 mutations (one from each parent, ie, a double-dominant disorder). The deafness reflects a recessive disorder with the two KCNQ1 mutations resulting in sensory hearing loss due to involvement of the auditory nerves.8 AfewyearsaftertheJervellandLange-Nielsenpublication,Romanoet al in 19639 and Ward in 196410 reported independent families in which affected members had QT prolongation, recurrent syncope, and sudden 630 CurrProblCardiol,November2008 FIG1.ECGrecordedonJuly20,1953inthefirstreportedpatientwithdeafness,recurrentsyncope, andQTprolongation.Thepaperspeedis50mm/s,QT(cid:1)0.48s,RR(cid:1)0.90s,andQTc(Bazett)(cid:1) 0.51 s. (Reproduced with permission from Jervell A, Lange-Nielsen F. Congenital deal-mutism, functional heart disease with prolongation of the Q-T interval and sudden death. Am Heart J 1957;54:59-68.)5 CurrProblCardiol,November2008 631 death without deafness, with an autosomal-dominant pattern of inheri- tance. This LQTS disorder without deafness was considerably more frequent than the Jervell and Lange-Nielsen syndrome and has been referred to as the Romano–Ward syndrome. During the 1960s several individuals and families with LQTS, mostly with the Romano–Ward variant, were reported in the literature. In 1969, one of the authors of this publication (A.J.M.) saw in consultation a 39-year-old women with normal hearing who experienced recurrent syncope and had marked QT prolongation (QTc (cid:1) corrected QT (cid:1) 0.69s).WithknowledgeofthereportbyYanowitzetalin196611thatleft stellate stimulation in canine studies prolonged the QT interval, a left cervico-thoracic sympathetic ganglionectomy (left 7th cervical through left 2nd thoracic, including the stellate ganglion) was performed through a left supraclavicular approach.12 The ECG-QT findings before any interventions, with local left and right stellate block, and 6 months after left cervicothoracic sympathetic ganglionectomy are presented in Fig 2. The patient remains alive and well with modest QT prolongation and without recurrent syncope 37 years after the ganglionectomy surgery. In a recent report relating the clinical experience with left cervicothoracic sympatheticganglionectomysurgeryforrefractoryLQTSin147patients, the overall long-term experience has been very favorable.13 MelvinM.Scheinman:TheoriginalarticlebyYanowitzrelativetotheeffects of left stellate stimulation was transient and, although left stellate ganglion resection has been used therapeutically, the genesis of the LQTS is clearly not due to sympathetic imbalance, as was postulated in the 1970s. The effectiveness of beta-blockers in the treatment of LQTS was appreciated in the mid-1970s, and it has now become the treatment of choice for this disorder.14 Following the report by the Rochester, NY group of the successful therapyofLQTSwithleftcervicothoracicsympatheticganglionectomyin 1971,12alargenumberofpatientswithLQTSwerereferredtotheauthors for clinical evaluation and therapy. About this same time, Dr. Peter Schwartz in Milan, Italy and Dr. Richard Crampton in Charlottesville, Virginia reported their experience with LQTS and the link between the left stellate ganglion and this disorder.15,16 Melvin M. Scheinman: A positive experience with left stellate ganglion resectionhasnotbeenuniversal.Inajointexperiencebetweenourselvesand 632 CurrProblCardiol,November2008 FIG2.LeadIIofECGs(paperspeed,25mm/s)takenatrest1dayaftersyncopalepisode(A:QT(cid:1) 0.64s);afterlocalleftstellate-ganglionblock(B:QT(cid:1)0.46s);afterlocalrightstellate-ganglionblock(C: QT(cid:1)0.72s);and6monthsafterleftcervicothoracicsympatheticganglionectomy(D:QT(cid:1)0.44s). (Reproduced with permission from Moss AJ, McDonald J. Unilateral cervicothoracic sympathetic ganglionectomyforthetreatmentoflongQTintervalsyndrome.NEnglJMed1971;285:903-4).12 CurrProblCardiol,November2008 633 colleaguesatStanford,wefoundrecurrenceofsymptoms,includingaborted sudden death, was common after prolonged follow-up. We also found that theHorner’ssyndromeresolvedafteryearsoffollow-up.Returnofautonomic functionhasbeenwelldocumentedafterorthotopiccardiactransplantation. In1979,thethreeofus(Moss,Schwartz,andCrampton)establishedthe International Long QT Syndrome Registry.17 During the past 28 years, the Registry has enrolled 1276 proband-identified LQTS families involv- ingover3600clinicallyorborderline-affectedfamilymembers,withover 2000 of these family members having genetically confirmed LQTS mutations. Publications from the International LQTS Registry have providedinsightintoriskmechanisms,genotype–phenotypeassociations, risk stratification by age, gender, and genotype, and the importance of syncopeasacardiaceventthatfrequentlyprecedesabortedcardiacarrest (ACA) or sudden cardiac death (SCD). Molecular-Genetic Aspects of LQTS Ion-Channel Genes and Currents QT prolongation is the hallmark of LQTS and may arise from either a reduction in the outward potassium current during phase 3 of the action potential (referred to as a “loss of function”) or an augmented late entry of sodium or calcium ions into the myocytes due to malfunctioning sodium or calcium channels (referred to as a “gain of function”).18 The most common genetic causes of LQTS involve mutations in genes regulating the (cid:1)-subunits of the following: (1) the slowly activating potassium repolarization channel (KCNQ1; LQT1) resulting in a reduc- tion in IKs current; (2) the rapidly activating potassium repolarization channel (KCNH2; LQT2) resulting in a reduction in IKr current; or (3) the sodium channel (SCN5A; LQT3) resulting in an increase in late INa current.LQTShasbeenidentifiedinfrequentlyinpatientswithmutations involving the auxiliary (cid:2)-subunits of KCNQ1 (minK; LQT5) and of KCNH2 (MiRP1; LQT6), although there is not full agreement regarding the function of MiRP1. A summary of LQT1, LQT2, LQT3, LQT5, and LQT6 genotypes, their altered ion-channel currents, and their relative frequency in mutation-identified LQTS patients is presented in Table 1. At the present time, several hundred different mutations have been identified in these five LQTS genes, and it is this group of ion-channel genes that has characterized LQTS as a “channelopathy.” LQT1, LQT2, and LQT3 genotypes account for 97% of the mutations identified in LQTS patients. A schematic diagram of the prolongation of the myocyte 634 CurrProblCardiol,November2008 TABLE1.Mutationsinion-channelgenescausingLongQTsyndrome Ion- Frequencyin Affected channel mutation-identified Genotype Chromosome gene Channelprotein current LQTSpatients(%) LQT1 11 KCNQ1 4(cid:1)-subunitseach 2I 45 Ks (Kv7.1) with6 membrane- spanning segments LQT2 7 KCNH2 4(cid:1)-subunitseach 2I 45 Kr (hERG) with6 (K11.1) membrane- v spanning segments LQT3 3 SCN5A 1(cid:1)-subunitwith 1lateI 7 Na (Na1.5) 24membrane- v spanning segments LQT5 21 KCNE1 (cid:2)-subunitof 2I (cid:3)1 Ks (MinK) KCNQ1with1 membrane- spanning segment LQT6 21 KCNE2 (cid:2)-subunitof 2I (cid:3)1 Kr (MiRP1) KCNH2with1 membrane- spanning segment LQT7 17 KCNJ2 2membrane- 2I (cid:3)1 K1 (Kir2.1) spanning segments LQT8 6 CACNA1C 1(cid:1)-subunitwith 1I (cid:3)1 1 Ca (Ca1.2) 24membrane- v spanning segments LQT10 11 SCN4B (cid:2)-subunitof 1lateI (cid:3)1 Na (NaB4) SCN5Awith1 v membrane- spanning segment action potential that results from a reduction in IKs (LQT1) or IKr (LQT2) currents or from an increase in late INa (LQT3) current is presented in Fig 3. During the past few years, mutations in three other ion-channel genes havebeenidentifiedinasmallnumberofLQTSfamilies(LQT7,(cid:2)8,and (cid:2)10; Table 1): (1) mutation in the KCNJ2 gene results in a reduction in Kir2.1 current, QT prolongation, and a phenotype dominated by skeletal CurrProblCardiol,November2008 635 FIG3.Schematicdiagramoftheinfluenceofalteredion-channelcurrentsontheactionpotential duration in LQTS. Inward currents are indicated below the line and outward currents are indicated above the line. The hatched rectangles denote the timing location of the effect of mutations in LQT1, (cid:2)2, and (cid:2)3 genes on sodium and potassium ion-channel currents. The actionpotentialisprolonged(horizontalarrow)whenthereisinappropriategainoffunction (GOF)inlatesodiumcurrentINaorlossoffunction(LOF)inslowly(IKs)orrapidly(IKr)acting repolarizingpotassiumcurrents. abnormalities(Andersen–Tawilsyndrome)(LQT7)19;(2)mutationinthe CACNA1C gene results in an increase in Ca 1.2 current, QT prolonga- v tion, and a phenotype characterized by syndactyly in both hands and feet and multiorgan dysfunction (Timothy’s syndrome) (LQT8)20; and (3) mutation in the SCN4B gene (LQT10), one of the (cid:2)-subunits of SCN5A, results in an increased late entry of sodium current into the cell with associated QT prolongation.21 Itisnowappreciatedthatmutationsinnon-ionchannelgenescanaffect ion-channel currents with resultant prolongation in ventricular repolar- ization and the QT interval. LQT4 is caused by mutations in the ankyrin B gene22 that produces a protein that functions as a cytoskeletal membrane adapter and is involved in the cellular organization of the 636 CurrProblCardiol,November2008 TABLE2.Mutationsinchannelopathic-relatedgenescausingLongQTsyndrome Frequencyin Electrophysiologic mutation-identified Label Chromosome Affectedgene effects LQTSpatients(%) LQT4 4 Ankyin-B(ANK2) Sodiumpumpand (cid:3)1 Na/Caexchanger LQT9 3 CAV3(Caveolin) Alteredgatingkinetics (cid:3)1 ofNa1.5 v sodium pump, the sodium/calcium exchanger, and the inositol-1,4,5- triphosphate receptors.23 Recently, mutations in caveolin-3 were identi- fied in patients with LQTS, and this new LQTS-related gene has been labeled LQT9.24 Electrophysiologic studies have revealed altered gating kineticsinthecardiacsodiumchannelwithresultantincreaseinsustained late sodium current (Na 1.5) probably due to direct protein–protein v interactions.25 Thus, LQT4 and LQT9 may be considered “channelo- pathic-relateddisorders”(Table2),anditislikelythatmutationsinother channelopathic-relatedgeneswillbeidentifiedinthefuturetoaccountfor some of the 25% or so LQTS patients who are negative for mutations in the eight classical channelopathy genes. Melvin M. Scheinman: In addition to the 10 recognized genetic mutations causing LQTS, two additional genetic mutations were described at the last American Heart Association meeting. One involved the Yotiao protein com- plex (Chen M, Marquardt ML, Tester DJ, et al. Circulation 2007;116:11-653; abstract), which produces abnormalities in the IKs current and mutations in alpha-1-syntrophin (Vatta M, Ai T, Wu G, et al. Circulation 2007;116:11- 653; abstract), which produced abnormality in the Na(cid:4) channel. More genetic abnormalities will likely be discovered as 25% of patients with the LQTS currently do not have a specific mutation. Ion-Channel Structure and Function Ion channels are formed by transcription of exonic DNA from the gene into mRNA with translation into a sequence of amino acids that make up channel-protein subunits (Fig 4). The subunits are co-assembled into a structural channel protein that is transported to the cell surface through a process called trafficking. The protein is then anchored into the myocyte membrane so it can function as an ion channel. The protein channel undergoes controlled degradation and appropriate replacement. Thus, a mutation in the genetic DNA can produce an abnormal or deficient ion CurrProblCardiol,November2008 637 FIG 4. Schematic diagram of ion-channel formation. Exons (black) are transcripted into messengerRNA(mRNA)thattranslatestheinformationintoachannelproteininvolvingachain ofaminoacidsconsistingofanN-terminusregion,sixmembrane-spanningsegments,anda C-terminus region. The pore area is located between the 5th and 6th membrane-spanning segments. The X in the LQTS gene denotes a DNA mutation that appears as an altered amino-acidsequenceinthechannelprotein. channel through a variety of processes that may include altered subunit assembly, trafficking deficiencies, impaired anchoring of the channel in the myocyte membrane, abnormal amino acid sequence in the protein channel, or an imbalance in the degradation-formation rates of the channelprotein.TraffickingdeficienciesareespeciallycommoninLQTS patients with mutations in the KCNH2 (LQT2) gene.26 The putative topology of each voltage-gated cardiac ion channel consists of a pore-forming (cid:1)-subunit and one or more attached (cid:2)-sub- units. The protein structure of the (cid:1)-subunits of the LQT1 and LQT2 potassiumchannels(KCNQ1,KCNH2)aresimilarandconsistofaseries of amino acids with an N-terminus region, six membrane-spanning segments (S1-S6) with connecting intracellular cytoplamic loops (S2-S3 and S4-S5) and a pore area (S5-loop-S6), and a C-terminus region. Four (cid:1)-subunits join together to form a tetrameric potassium channel around a 638 CurrProblCardiol,November2008