sudden cardiac death, torsades de pointes, T wave alternans, syncope, genetic mutation

Transcription

1 Reprinted with permission from JOURNAL OF CARDIOVASCULAR ELECTROPHYSIOLOGY Volume 12 No. 4 April 2001 Copyright 2001 by Futura Publishing Company Inc. Armonk NY Electrocardiographic Prediction of Abnormal Genotype in Congenital Long QT Syndrome: Experience in 101 Related Family Members ELIZABETH S. KAUFMAN M.D. SILVIA G. PRIORI M.D. PH.D.* CARLO NAPOLITANO M.D.* PETER J. SCHWARTZ M.D. SUDHA IYENGAR PH.D. ROBERT C. ELSTON PH.D. AUDREY H. SCHNELL PH.D. EIRAN Z. GORODESKI B.A. GUHAN RAMMOHAN B.S. NAEL O. BAHHUR B.A. DAVID CONNUCK M.D. LINDA VERRILLI R.N. B.S.N. M.ED. DAVID S. ROSENBAUM M.D. and ARTHUR M. BROWN M.D. PH.D. From the The Heart and Vascular Research Center MetroHealth Campus Case Western Reserve University Cleveland Ohio; *Fondazione S. Maugeri Pavia Italy; the Department of Cardiology Policlinico S. Matteo IRCCS and University of Pavia Pavia Italy; and the Department of Epidemiology and Biostatistics Rammelkamp Center for Education and Research MetroHealth Campus Case Western Reserve University Cleveland Ohio Prediction of Congenital Long QT Syndrome. Introduction: Previous studies showed that diagnosing congenital long QT syndrome (LQTS) is dif cult due to variable penetrance and genetic heterogeneity especially when subjects from multiple families with diverse mutations are combined. We hypothesized that a combination of clinical and ECG techniques could identify gene carriers within a single family with congenital LQTS. Methods and Results: One hundred one genotyped members of a family with LQTS including 26 carriers of a HERG mutation underwent history and ECG analysis. Forty-eight family members also underwent exercise testing with QT and T wave alternans (TWA) analysis and 24-hour Holter monitoring with QT and heart rate variability analysis. A logistic regression model which included age gender QTc and QTc by age provided the best prediction of gene carrier status although there was substantial overlap (78%) of QTc among subjects with and without the mutation. QTc was not helpful as a discriminator in children 13 years. TWA (observed infrequently) did not add signi cantly to the model s ability to predict abnormal genotype. Conclusion: Even in this homogeneous LQTS population the phenotype was so variable that clinical and detailed ECG analyses did not permit an accurate diagnosis of gene carrier status especially in children. Sustained microvolt TWA was a speci c (100%) but insensitive (18%) marker for LQTS. Its ability to predict risk of arrhythmia in this population remains to be determined. Genetic testing serves an essential role in screening for carriers of LQTS. (J Cardiovasc Electrophysiol Vol. 12 pp April 2001) sudden cardiac death torsades de pointes T wave alternans syncope genetic mutation 455 Introduction Congenital long QT syndrome (LQTS) is a group of genetically transmitted diseases in which patients have abnormal myocardial repolarization (often re ected in a prolonged QT interval on the ECG) and present with syncope or sudden death due to torsades de pointes. 1-3 The accurate This work was supported in part by U.S. Public Health Service Research Grant GM28356 from the National Institute of General Medical Sciences and Resource Grant RR03655 from the National Center for Research Resources; a GCRC Grant from NIH: Grant MO1RR00080 awarded to MetroHealth Medical Center Cleveland Ohio; Telethon Grant 1058; and CEE Grant BMH4-CT This study also was supported in part by National Institutes of Health Grants HL36930 HL55404 HL61642 and DK Address for correspondence: Elizabeth S. Kaufman M.D. Heart and Vascular Research Center Hamann Third Floor MetroHealth Campus Case Western Reserve University 2500 MetroHealth Drive Cleveland OH Fax: ; ekaufman@metrohealth.org Manuscript received 21 August 2000; Accepted for publication 30 January diagnosis of LQTS is critically important because appropriate treatment substantially reduces the risk of sudden death. Accurate diagnosis is made dif cult by limitations of the ECG as a diagnostic tool because the QT intervals of affected and unaffected individuals overlap substantially 4 and by the variable penetrance of LQTS even within a single family. 5 These dif culties have led to a search for better diagnostic techniques. Multiple gene abnormalities that cause LQTS have been identi ed 6 but genotyping has several limitations. Genotyping LQTS is a complex time-consuming expensive process with restricted availability. Even when genotyping is attempted a mutation may not be found. Presumably some LQTS families have mutations of genes not previously identi ed with LQTS. Finally although the genotype is useful in identifying unaffected individuals the mere presence of an abnormal genotype may be only weakly predictive of an individual s clinical risk (and need for treatment). Numerous clinical techniques have been used in an attempt to diagnose LQTS and predict risk more accurately. 7 These methods have included QT measurement 8 T wave

2 456 Journal of Cardiovascular Electrophysiology Vol. 12 No. 4 April 2001 morphology analysis ventricular wall-motion analysis 1 1 QT dispersion 1 2 Holter monitoring 1 3 and exercise testing. 1 4 Potentially useful but not yet widely tested is T wave alternans (TWA) analysis. Grossly visible TWA has been observed in LQTS and is known to presage torsades de pointes in LQTS patients. 1 5 Microvolt-level TWA de ned as a subtle and visibly undetectable beat-to-beat alteration in the amplitude of the T wave which is known to predict arrhythmic events in patients with structural heart disease is an evolving method that has potential value as a risk predictor in LQTS This study was designed to test the following hypothesis: a combination of clinical and noninvasive ECG techniques including microvolt-level TWA can identify genetically affected patients with LQTS when applied to a single large family. Whereas previous studies grouped together subjects with different genetic mutations (known to cause different clinical manifestations) this study examined the clinical phenotype of a genetically homogeneous group a large family with a single LQTS mutation. Methods A large family (n 5 101) with LQTS was identi ed after a 31-year-old woman with a history of syncope died suddenly and her paternal aunt was diagnosed with LQTS. After obtaining informed consent (approved by and in accordance with the guidelines of the Institutional Review Board at MetroHealth Medical Center) related family members underwent a medical history (focused on syncopal episodes) 12-lead ECG recording and blood testing for genetic analysis. A subset of these individuals (those able to travel to the medical center and ride a stationary bicycle) also underwent 24-hour (Holter) ambulatory ECG monitoring and TWA testing during exercise on a stationary bicycle (n 5 48). Electrocardiography Standard supine 12-lead ECGs were examined in a blinded fashion for RR QT and QTc intervals; abnormal T wave morphology; and QT dispersion. QT interval was calculated by measuring and averaging the QT in lead II on the 12-lead ECG. The QT interval was corrected for heart rate to yield QTc by dividing by the square root of the RR interval (Bazett s formula). Manual and digitized measurements by separate readers were compared to assess accuracy. T wave morphology was judged normal or abnormal (by a blinded observer) based on T wave notching diffuse low amplitude broad base prolonged at ST segment or multiple U waves. 2 QT dispersion was calculated by subtracting the shortest QT from the longest QT interval on the 12-lead ECG. Ambulatory ECG Analysis Holter monitoring was performed with subjects not taking beta blockers. Marquette 8500 (Marquette Electronics Milwaukee WI USA) ve-channel Holter ECG recordings scanned and edited on a Marquette SXP Holter scanner were examined for inappropriate sinus bradycardia or pauses ventricular tachycardia and QTc intervals at fastest and slowest heart rates over the 24-hour period. After labeling of ectopic beats and noise the edited RR interval data were transferred to a Sun workstation where heart rate variability analysis was performed. Spectral data were produced by fast Fourier transformation computed on 5-minute segments. Parameters measured included mean RR standard deviation of RR (SDRR) and log transformed highfrequency (0.15 to 0.40 Hz) and low-frequency (0.04 to 0.15 Hz) power (Ln HFP and Ln LFP). Exercise Testing and TWA Analysis High-resolution ECG recordings were made during seated rest and during exercise on a stationary bicycle (Cambridge Heart 2000 Bedford MA USA). Beta blockers were withheld for 24 hours before testing. Exercise tests were performed according to a heart rate-guided protocol with an effort to record at least 5 minutes at rest 2 to 3 minutes at heart rates of and. 110 beats/min and 7 minutes of gradual recovery. QTc was calculated during peak exercise (QTc Stress) and throughout recovery with the maximum recovery QTc chosen as QTc Recovery. TWA studies were classi ed as normal abnormal or nondiagnostic according to previously established criteria. 1 9 TWA analysis is classi ed as abnormal if there is sustained alternans of at least 1.9 mv lasting for at least 1 minute and consistently present while heart rate exceeds some patient-speci c heart rate threshold value that does not exceed 110 beats/min with noise #1.8 mv alternans-to-noise ratio $3 and ectopics #10%. Genetic Typing Mononuclear cells were isolated from EDTA anticoagulated whole blood by hypotonic lysis of red cells and total DNA was isolated from each subject as described by Miller et al. 2 0 Polymorphisms at four candidate genes [KvLQT1 (LQT1) HERG (LQT2) SCN5A (LQT3) and MinK (LQT5)] were examined for linkage with the LQTS in this family and mutation detection was performed using singlestrand conformational polymorphism analysis and direct sequencing to determine and con rm the speci c mutation. Genetic typing was performed concurrently with clinical testing. The clinical investigators were blinded to the results of genetic typing until all the clinical data had been collected and classi ed. To analyze the functional consequences of the identi ed mutation voltage-clamp techniques were used to compare the mutant and wild-type genes expressed in Xenopus oocytes and in mammalian cells con rming that the identi ed mutation was able to cause QT prolongation in an experimental model. The details of these ndings are described elsewhere. 2 1 Statistical Analysis Logistic regression was used to determine what clinical variables (assessed by interpreters blinded to the genotype results) predicted an abnormal genotype. QT measurements and age were analyzed as continuous variables. Unpaired t-tests were used to compare heart rate variability measures in gene carriers and noncarriers. Fisher s exact test was used to test for signi cance of the association between abnormal TWA and syncope. To compare the diagnostic value of QTc in children (#13 years) versus adults (. 13 years) means were calculated separately by genotype status for children and adults. For children and for adults standardized mean differences in QTc between genetically normal and abnormal individ-

3 Kaufman et al. Prediction of Congenital Long QT Syndrome 457 TABLE 1 History of Syncope by Genotype Status Syncope uals were calculated separately using log transforms to stabilize the standard deviations. To test for this age by genotype interaction a pooled estimate of the standard error(s) was calculated with 97 degrees of freedom and the following t-statistic calculated as given in Equation 1: (QTc AA 2 QTc NA ) 2 (QTc AC 2 QTc NC ) s(1/n AA 1 1/n NA 1 1/n AC 1 1/n NC ) 1/ 2 (1) where AA 5 abnormal adults NA 5 normal adults AC 5 abnormal children NC 5 normal children and abnormal and normal apply to gene status. Log scale was used. All tests were two sided. Results No Syncope Abnormal genotype 4 22 Normal genotype Among 101 related family members (45 male and 56 female) who completed ECG and DNA analysis 48 individuals (27 male; age years range 11 to 71 years) also wore Holter monitors and completed bicycle testing with TWA analysis. The family members studied spanned three generations. The proband was a 31-year-old woman who had a history of fainting spells over a 15-year period. She was 3.3 months postpartum at the time of her death which occurred just after she arose from bed in the night. Twenty-six individuals (13 male and 13 female) were found to have a genetic abnormality (R752W) in the C-terminus of HERG on chromosome 7. No other sudden deaths had occurred in the three generations studied. Syncope occurred in a minority of genetically affected and unaffected family members. Table 1 shows the occurrence of syncope in subjects with normal and abnormal genotype. Table 2 lists a detailed description of the subjects with syncope. Clinical and ECG Predictors of Abnormal Genotype When logistic regression was used to determine what clinical and ECG variables predicted presence of the genetic mutation a model including age gender QTc and the product of QTc 3 age provided the best prediction. (This standard method used in regression analysis for evaluating an interaction between two variables indicated that the strength of QTc as a predictor varied with age). History of syncope mean RR interval uncorrected QT abnormal T Subject TABLE 2 Clinical Characteristics of Subjects with Syncope Age (years) Sex QTc (sec) Genotype Description of Syncope 1 33 F 0.41 Normal Multiple sudden episodes 2 66 F 0.50 Abnormal Once when dehydrated 3 21 F 0.42 Normal Once no detail available 4 61 M 0.49 Abnormal Several emotional stress-induced 5 22 F 0.44 Normal Multiple emotional stress-induced 6 38 M 0.42 Normal Multiple abnormal tilt table test 7 32 M 0.40 Normal Multiple sudden episodes 8 23 F 0.41 Normal Multiple sudden episodes 9 35 F 0.42 Normal Once while pregnant M 0.41 Normal Once with exertion F 0.41 Normal Once with gastroenteritis M 0.41 Normal Multiple breath-holding spells F 0.41 Normal Multiple breath-holding spells F 0.48 Abnormal Once with cholecystitis F 0.42 Normal Multiple with phlebotomy and exercise M 0.44 Abnormal Multiple: in church giving blood in the heat F 0.43 Normal Multiple emotional stress-induced M 0.41 Normal Multiple emotional stress-induced F 0.40 Normal Once emotional stress-induced M 0.42 Normal Once during exertion TABLE 3 Individual Variables as Predictors of Abnormal Genotype Variable N Sensitivity Speci city PPV NPV RR P Value QTc ECG QTc stress QTc recovery QT Morphology QT dispersion NS Syncope NS TWA NS NPV 5 negative predictive value; PPV 5 positive predictive value; TWA 5 T wave alternans.

4 458 Journal of Cardiovascular Electrophysiology Vol. 12 No. 4 April 2001 TABLE 4 Sensitivity and Speci city of QTc for Predicting Abnormal Genotype QTc (sec) Sensitivity (%) Speci city (%).. Male subjects 13 years Female subjects 13 years wave morphology QT dispersion and presence of TWA did not add signi cantly to the model. Whereas QT measurements and age were used as continuous variables in the logistic regression model Table 3 shows the performance of individual variables as predictors of abnormal genotype using prede ned xed cutoff values based on established norms. In Table 3 QT seconds for females and seconds for males was de ned as abnormal. QT dispersion was de ned as abnormal if the maximum minus the minimum QT measured on the baseline ECG exceeded 100 msec. 2 2 Sensitivity speci city positive predictive value (PPV) negative predictive value (NPV) relative risk (RR) number of subjects in whom the variable was measured (N) and P value are shown for each variable. When children #13 years were excluded age became insigni cant and only QTc and gender predicted presence of the genetic mutation (Table 4). Note that there was a signi cant degree of overlap of QTc carriers and noncarriers (Fig. 1) even among adults. In 30% of individuals (9 adult carriers and 10 adult noncarriers) QTc was in the ambiguous overlap zone of 0.42 to 0.46 seconds. When both adults and children were included in the analysis 78% of individuals had QTc in the overlap zone of 0.40 to QTc in Children versus Adults Among 101 family members means and standard deviations were calculated separately by genotype status for 38 children (#13 years) and 63 adults (. 13 years). Twenty-six subjects were found to have the genetic mutation (5 children and 21 adults). Among affected adults QTc seconds (mean 6 SD; range 0.42 to 0.56); for unaffected Figure 1. QTc for children and adults by gene status. QTc is shown for genetically abnormal adults (age. 13 years) normal adults abnormal children and normal children. Circles represent females and squares represent males. Figure 2. ECGs from an adult and a child with the HERG mutation. The top panel shows the ECG of an adult with the HERG mutation. QTc seconds. The bottom panel shows the ECG of a 3-year-old child with the HERG mutation. QTc seconds. adults QTc seconds (range 0.38 to 0.46). Among affected children QTc seconds (range 0.40 to 0.45); for unaffected children QTc seconds (range 0.39 to 0.44) (Figs. 1 and 2). On the log scale the mean QTc for affected adults differed from that of unaffected adults by 2.48 SD. In contrast the mean QTc of affected children differed from that of unaffected children by only 1.15 SD. This interaction was signi cant (t with 97 degrees of freedom; P 0.01). Ambulatory ECG Analysis Forty-eight family members including 20 gene carriers underwent Holter monitoring. There were no abnormal bradycardic rhythms and only one three-beat run of ventricular tachycardia among all the recordings. No signi cant differences were found between gene carriers and noncarriers for any heart rate variability parameter measured (Table 5). Exercise Tests and TWA Analysis Forty-eight family members (26 male and 22 female; age years range 11 to 71) underwent exercise testing and TWA analysis. No signi cant arrhythmias occurred during exercise testing. Seventeen of 20 mutation positive and 24 of 28 mutation negative family members had interpretable TWA tests and were included in data analysis. Among genetically affected family members 18% had abnormal TWA (Fig. 3). None of the genetically unaffected members had abnormal TWA (Table 6). Among the genetically abnormal individuals QTc was not signi cantly different in those with and those without abnormal TWA.

5 Kaufman et al. Prediction of Congenital Long QT Syndrome 459 TABLE 5 Heart Rate Variability Results by Mutation Status Age (years) Gender QTc (sec) Mean RR (msec) SDRR Ln HFP Ln LFP HERG M 11 F HERG M 12 F P 0.05 P NS NS NS NS Abbreviations as de ned in text. Discussion We studied a relatively homogeneous population (a single large family with congenital LQTS due to a mutation in HERG) and systematically analyzed their clinical phenotype. In contrast previous studies of LQTS phenotype combined families with diverse mutations (and many studies combined families with mutations of different genes). Despite the genetic homogeneity of our study population 26 of whom had the identical HERG mutation the clinical expression of this mutation varied substantially among individuals ranging from normal to asymptomatic QT prolongation to syncope and sudden death (in the proband). Despite the wide variation of QTc (0.40 to 0.56 sec) among affected family members (Fig. 1) and the substantial (78%) overlap between QTc of carriers and noncarriers we found QTc to be the best clinical test for predicting presence of the abnormal genotype. Other ECG measurements (including combinations of uncorrected QT RR interval and QT dispersion) and even history of syncope did not add signi cantly to the ability to predict genotype. This latter nding was surprising but this family was mildly affected (only 4/26 genetically affected individuals had syncope) and the relatively frequent occurrence of syncope (a common symptom in the general population) in the genetically unaffected family members confounded the usefulness of this symptom in our study. It should be noted that syncope is a highly signi cant clinical event in the overwhelming majority of LQTS families; syncope in LQTS relatives must be considered arrhythmic and treated accordingly. We demonstrated a signi cant interaction between age and genotype on QTc. QTc did not predict the mutation in children #13 years. Use of uncorrected QT or QT in combination with RR was no better so it is unlikely that limitations of Bazett s correction formula can account for this nding. This has not been described previously and may be unique to our family or to a subset of families with HERG mutations. Whereas the ECG remains the initial screening tool for LQTS in children (as well in adults) and Figure 3. Examples of abnormal and negative T wave alternans (TWA) tests. On the left is an example of abnormal TWA and on the right is a normal TWA test. The top panel shows heart rate (HR). Note that the vertical lines overlying the heart rate graph represent sudden large uctuations in HR and indicate the presence of ectopic beats. The next four panels show the magnitude of TWA (shaded regions) in precordial leads V 2 to V 6. The mean noise in each lead is shown by the dotted line. The bottom panel shows ECG lead II during each test. Note the absence of visible TWA on the ECG on the left despite simultaneous recording of abnormal microvolt-level TWA.

6 460 Journal of Cardiovascular Electrophysiology Vol. 12 No. 4 April 2001 TABLE 6 TWA Results by Mutation Status HERG Positive HERG Negative TWA1 3 0 TWA TWA 5 T wave alternans. a prolonged QTc should be interpreted as evidence of LQTS a normal QTc cannot be equated with absence of the genetic mutation. It is not known whether the increase in QTc we observed at adolescence is hormonally determined or due to another developmental process. Our ndings support the practice of obtaining serial ECGs in children and suggest that young children in LQTS families might bene t from presumptive treatment regardless of QTc. The family we studied had an unusually low rate of arrhythmic events. This may be due to the cellular mechanism of the mutation (abnormal intracellular processing of functional I Kr channels). 2 1 The mild phenotype of this family was re ected in the low rate of sudden death and relatively few markedly prolonged QT intervals yet the mutation was lethal in one young adult. Therefore other gene carriers may be at risk. The presence or absence of microvolt-level TWA did not add to our ability to predict presence of the abnormal gene in this study. This was probably due to the small number of people who completed TWA testing and to the low incidence of abnormal TWA even within genetically affected family members. Zareba et al. 1 7 found that the presence of grossly visible TWA identi ed a subset of LQTS patients at high risk for clinical events; however grossly visible TWA in their study was strongly associated with the degree of QTc prolongation and thus did not make a signi cant independent contribution to the risk of cardiac events. In contrast we found that among the genetically abnormal individuals QTc was not signi cantly different in those with and those without abnormal TWA. We observed sustained microvolt-level TWA only among a minority of LQTS subjects. However the optimal technique for measuring and interpreting microvolt-level TWA in the LQTS population has not been established. Shimizu and Antzelevitch 2 3 showed in a LQTS model (using an I Kr blocking agent) that isoproterenol at constant heart rate initially prolongs then shortens repolarization (with a transient increase in dispersion). This suggests that microvolt TWA in patients with a HERG mutation might be transient and thus not ful ll the conventional criteria for abnormal sustained TWA. Therefore limitations of our study include the lack of established criteria for inducing and interpreting TWA in LQTS patients and the lack of data correlating microvolt TWA with adverse arrhythmic events in this population. Furthermore we were unable to measure TWA in young children due to (1) the size of the stationary bicycle and (2) the marked heart rate variability seen in children. It is clear from previous studies that congenital LQTS is not one disease: mutations in different LQTS genes lead to marked phenotypic variation. 2 4 Therefore we cannot extrapolate our results to LQTS families with mutations in other genes. It is likely that different mutations in the HERG gene itself may lead to signi cantly different phenotypes 2 5 an issue obscured by (1) the phenotypic variation within families and (2) the small size of most families studied. Conclusion In this large family with a HERG mutation no clinical parameter or combination of measurements accurately predicted carrier status. Gender and QTc provided the best prediction of presence of the abnormal genotype in adults and adolescents although substantial ambiguity remained. In contrast no parameter including QTc was able to predict genotype among young children. Thus even within a single genetically homogeneous family there was pronounced phenotypic variability and even QTc the best clinical parameter was unreliable. In the absence of an adequate clinical surrogate genetic testing serves an essential role in screening for carriers of LQTS. The medical community should consider how to make this complex and expensive testing available in the future. Microvolt TWA was a speci c but insensitive marker for LQTS but it may hold promise as a predictor of individual susceptibility to arrhythmias. Future studies and more extended follow-up are required to determine if microvoltlevel TWA is a reliable prognostic tool that can predict high risk in this population. A test capable of predicting individual risk among asymptomatic family members is needed because then effective treatment options including the implantable cardioverter de brillator can be targeted to highrisk individuals. Congenital LQTS is a group of disorders caused by diverse mutations. Speci c clinical features caused by one mutation cannot be extrapolated to LQTS families with other mutations. The large family described in this study had an unusually low incidence of major arrhythmic events and cannot be taken as a paradigm for families with LQTS or even with HERG mutations. The relatively benign clinical course of the gene carriers in this family highlights the diversity of LQTS and points to the need for studies that contrast the phenotypes produced by different LQTS mutations. References 1. Schwartz PJ: Idiopathic long QT syndrome: Progress and questions. Am Heart J 1985;109: Moss AJ Robinson J: Clinical features of the idiopathic long QT syndrome. Circulation 1992;85(Suppl I):I-140 I Schwartz PJ Priori SG Napolitano C: The long QT syndrome. In Zipes DP Jalife J eds: Cardiac Electrophysiology: From Cell to Bedside. Third Edition. WB Saunders Philadelphia 2000 pp Vincent GM Timothy KW Leppert M Keating M: The spectrum of symptoms and QT intervals in carriers of the gene for the long-qt syndrome. N Engl J Med 1992;327: Priori SG Napolitano C Schwartz PJ: Low penetrance in the long-qt syndrome: Clinical impact. Circulation 1999;99: Priori SG Barhanin J Hauer RNW Haverkamp W Jongsma HJ Kleber AG McKenna WJ Roden DM Rudy Y Schwartz K Schwarz PJ Towbin JA Wilde AM: Genetic and molecular basis of cardiac arrhythmias: Impact on clinical management. Part I and II. Circulation 1999;99: ; Part III. Circulation 1999;99: Schwartz PJ Moss AJ Vincent GM Crampton RS: Diagnostic criteria for the long QT syndrome: An update. Circulation 1993;88: Moss AJ: Measurement of the QT interval and the risk associated with QTc interval prolongation: A review. Am J Cardiol 1993;72:23B-25B. 9. Malfatto G Beria G Sala S Bonazzi O Schwartz PJ: Quantitative analysis of T wave abnormalities and their prognostic implications in the idiopathic long QT syndrome. J Am Coll Cardiol 1994;23:

7 Kaufman et al. Prediction of Congenital Long QT Syndrome Priori SG Mortara DW Napolitano C Diehl L Paganini V Cantù F Cantù G Schwartz PJ: Evaluation of the spatial aspects of T-wave complexity in the long QT syndrome. Circulation 1997;96: Nador F Beria G De Ferrari GM Stramba-Badiale M Locati EH Lotto A Schwartz PJ: Unsuspected echocardiographic abnormality in the long QT syndrome: Diagnostic prognostic and pathogenetic implications. Circulation 1991;84: Swan H Saarinen K Kontula K Tiovonen L Viitasalo M: Evaluation of QT interval duration and dispersion and proposed clinical criteria in diagnosis of long QT syndrome in patients with a genetically uniform type of LQT1. J Am Coll Cardiol 1998;32: Eggeling T Osterhues HH Hoeher M Gabrielsen FG Weismueller P Hombach V: Value of Holter monitoring in patients with the long QT syndrome. Cardiology 1992;81: Swan H Viitasalo M Piippo K Laitinen P Kontula K Tiovonen L: Sinus node function and ventricular repolarization during exercise stress test in long QT syndrome patients with KvLQT1 and HERG potassium channel defects. J Am Coll Cardiol 1999;34: Schwartz PJ Malliani A: Electrical alternation of the T-wave: Clinical and experimental evidence of its relationship with the sympathetic nervous system and with the long Q-T syndrome. Am Heart J 1975; 89: Rosenbaum DS Jackson LE Smith JM Garan H Ruskin JN Cohen RJ: Electrical alternans and vulnerability to ventricular arrhythmias. N Engl J Med 1994;330: Zareba W Moss AJ Le Cessie S Hall WJ: T wave alternans in idiopathic long QT syndrome. J Am Coll Cardiol 1994;23: Platt SB Vijgen JM Albrecht P Van Hare GF Carlson MD Rosenbaum DS: Occult T wave alternans in long QT syndrome. J Cardiovasc Electrophysiol 1996;7: Hohnloser SH Klingenheben T Li Y Zabel M Peetermans J Cohen RJ: T wave alternans as a predictor of recurrent ventricular tachyarrhythmias in ICD recipients: Prospective comparison with conventional risk markers. J Cardiovasc Electrophysiol 1998;9: Miller SA Dykes DD Polesky HF: A simple salting out procedure for extracting DNA from human nucleated cells. Nucl Acids Res 1988; 16: Ficker E Thomas D Viswanathan PC Dennis AT Priori SG Napolitano C Memmi M Wible BA Kaufman ES Iyengar S Schwartz PJ Rudy Y Brown AM: Novel characteristics of a misprocessed mutant HERG channel linked to hereditary long QT syndrome. Am J Physiol 2000;279 (Heart Circ Physiol):H1748-H Priori SG Napolitano C Diehl L Schwartz PJ: Dispersion of the QT interval: A marker of therapeutic ef cacy in the idiopathic long QT syndrome. Circulation 1994;89: Shimizu W Antzelevitch C: Differential effects of b-adrenergic agonists and antagonists on transmural dispersion of repolarization and torsade de pointes in LQT1 LQT2 and LQT3 models of the long QT syndrome. Circulation 1998;98:I Zareba W Moss AJ Schwarz PJ Vincent GM Robinson JL Priori SG Benhorin J Locati E Towbin JA Keating MT Lehmann MH Hall WJ for the International Long-QT Syndrome Registry Research Group: In uence of the genotype on the clinical course of the long-qt syndrome. N Engl J Med 1998;339: Sanguinetti MC Curran ME Spector PS Keating MT: Spectrum of HERG K 1 -channel dysfunction in an inherited cardiac arrhythmia. Proc Natl Acad Sci U S A 1996;93: