The congenital long-qt syndrome (LQTS) is an inherited

Transcription

1 Clinical Implications for Affected arents and Siblings of robands With Long-QT Syndrome John Kimbrough, MD, hd; Arthur J. Moss, MD; Wojciech Zareba, MD, hd; Jennifer L. Robinson, MS; W. Jackson Hall, hd; Jesaia Benhorin, MD; Emanuela H. Locati, MD, hd; Aharon Medina, MD; Carlo Napolitano, MD; Silvia riori, MD, hd; eter J. Schwartz, MD; Katherine Timothy, BA; Jeffrey A. Towbin, MD; G. Michael Vincent, MD; Li Zhang, MD Background Whenever a proband is identified with long-qt syndrome (LQTS), his or her parents and siblings should be evaluated regarding the possibility of carrying the disorder. In the majority of cases, one of the proband s parents and one or more siblings are affected. The aim of this study was (1) to determine whether the clinical severity of LQTS in the proband is useful in identifying first-degree family members at high risk for cardiac events, and (2) to evaluate the clinical course of affected parents and siblings of LQTS probands. Methods and Results The clinical and ECG characteristics of 211 LQTS probands and 791 first-degree relatives (422 parents and 369 siblings) were studied to determine if the clinical profile of the proband is useful in determining the clinical severity of LQTS in affected parents and siblings. Affected female parents of an LQTS proband had a greater cumulative risk for a first cardiac event than affected male parents. The probability of a parent or sibling having a first cardiac event was not significantly influenced by the severity of the proband s clinical symptoms. sex and QT c duration were risk factors for cardiac events among affected parents, and QT c was the only risk factor for cardiac events in affected siblings. Conclusions The severity profile of LQTS in a proband was not found to be useful in identifying the clinical severity of LQTS in affected first-degree relatives of the proband. (Circulation. 2001;104: ) Key Words: electrocardiography long-qt syndrome risk factors sex genetics The congenital long-qt syndrome (LQTS) is an inherited disorder with prolonged ventricular repolarization predisposing to ventricular arrhythmias causing syncope, aborted cardiac arrest, and death. The clinical presentation of LQTS is influenced by an individual s age, sex, and genotype. After the initial study linking one form of LQTS to chromosome 11 in 1991, 1 many different mutations have been identified in the cardiac sodium and potassium channel genes responsible for different forms of LQTS. To date, these genotypes include LQT1:KVLQT1 on chromosome 11, LQT2:HERG on chromosome 7, LQT3:SCN5A on chromosome 3, LQT4 associated with a locus on chromosome 4, LQT5:KCNE1, the -subunit of KVLQT1 on chromosome 21, and LQT5:KCNE2, a component of HERG on chromosome The effects of LQTS genotypes on ECG T-wave morphology and clinical outcome have been reported Given the genetic nature of the disease, the identification of a proband with LQTS should involve an assessment of other family members for LQTS. This study focuses on probandidentified LQTS families from parents born before 1959 who lived to procreate an LQTS proband. This subset of the International LQTS Registry was selected to enhance our understanding of the clinical course of this disorder in first-degree relatives of identified probands with LQTS. The purpose of this study was to determine if the clinical profile of the proband is useful in determining the severity of LQTS in parents and siblings. Methods Study opulation The study population was drawn from the International LQTS Registry and consisted of proband-identified families in which both Received March 13, 2001; revision received May 18, 2001; accepted May 21, From the Department of Medicine, Vanderbilt University Medical Center, Nashville, Tenn (J.K.); the Cardiology Unit, Department of Medicine (A.J.M., W.Z., J.L.R.), and the Department of Biostatistics (W.J.H.), University of Rochester Medical Center, Rochester, NY; the Cardiology Department, Bikur Cholim Hospital, Jerusalem, Israel (J.B., A.M.); the Section of Cardiology, Universita Degli Studi Di erugia, erugia, Italy (E.H.L.); the Department of Cardiology, University of avia and oliclinical San Matteo, IRCCS, avia, Italy (.J.S.); the Molecular Cardiology Laboratory, IRCCS Fondazione Maugeri, avia, Italy (C.N., S..); the Department of Medicine, LDS Hospital, Salt Lake City, Utah (K.T., G.M.V., L.S); and the Department of ediatrics, Baylor College of Medicine, Texas Children s Hospital, Houston (J.A.T.). Guest Editor was Hein J.J. Wellens, MD, Maastricht, the Netherlands. Correspondence to Arthur J. Moss, MD, Heart Research Follow-up rogram, Box 653, University of Rochester Medical Center, Rochester, NY heartajm@heart.rochester.edu 2001 American Heart Association, Inc. Circulation is available at 557

2 558 Circulation July 31, 2001 TABLE 1. Clinical Characteristics of LQTS robands by Sex Clinical Variables robands (n 64) robands (n 147) Demographics Age at first ECG, y Cardiac event by age 40 y, n (%) 53 (83) 126 (86) Cardiac arrest or death by age 40 y, n (%) 9 (14) 46 (31) Age at first cardiac event, y 9 6(n 53) 14 8(n 126) Therapy (anytime), n (%) ICD 6 (9) 17 (12) acemaker 11 (17) 41 (28) LCTSG 11 (17) 25 (17) -Blockers 52 (81) 135 (92) 0.03 ECG characteristics Heart rate (beats/min) QT c,ms QT c 500 ms, n (%) 27 (42) 64 (44) Values are mean SD or n (%). ICD indicates implanted cardioverter-defibrillator; LCTSG, left cervicothoracic stellate ganlionectomy. Only values 0.05 are shown. Note: Age at first ECG is the age when the proband was enrolled into the LQTS Registry. robands were followed for varying lengths of time, with a few but not all to age 40 years. Cardiac events and cardiac arrest or death were censored at age 40 years to avoid the confounding influence of non-lqts comorbidity on events after age 40 years. proband s parents were born before The parental birth requirement was implemented to obtain a full 40-year follow-up of parents, thus allowing for a sufficiently long exposure of parents to draw clinically meaningful conclusions regarding their risk of cardiac events. In this study, the proband was defined as the first living family member identified with LQTS by ECG QT c criteria. 11 robands were usually brought to the attention of the registry because of symptoms and were usually identified at a relatively young age (children, adolescents, young adults); their family pedigrees consisted of parents and siblings but rarely their own children. This analysis focuses on parents and siblings of identified LQTS probands but excludes probands with congenital deafness (Jervell and Lange-Nielsen syndrome) because these subjects have a double TABLE 2. Clinical Variables Clinical Characteristics of arents of LQTS robands (n 112) Affected arents (n 113) mutation with inheritance of mutant LQTS alleles from each parent. There were 422 parents and 369 siblings of 211 LQTS probands. Subjects were categorized as affected or unaffected with LQTS, based on previously published QT c criteria for age and sex (affected children 10 years [QT c 0.44 seconds]; affected male subjects 10 years [QT c 0.43 seconds]; affected female subjects 10 years [QT c 0.45 seconds]). 11 Clinical and ECG Variables Routine clinical and ECG information was acquired at each subject s enrollment into the International LQTS Registry. The history of the occurrence of any cardiac events (defined as syncope, aborted cardiac arrest, or cardiac death) and the subject s age at first event (n 99) Unaffected arents (n 98) Demographics and therapy Age at first ECG, y Cardiac event by age 40 y, n (%) 9 (8) 31 (27) (5) 6 (6) Cardiac arrest or death by age 40 y 0 4 (4%) Age at first cardiac event, y Blockers anytime, n (%) 5 (4) 8 (7) 1 (1) 3 (3) ECG characteristics Heart rate, beats/min QT c,ms QT c 500 ms, n (%) 17 (15) 22 (19) 0 0 Only values 0.05 are shown. Note: Age at first ECG is the age when the parent was enrolled into the LQTS Registry. All parents of probands were born before 1959 and were 40 years of age at the time of this analysis. Cardiac events and cardiac arrest or death were censored at age 40 years to avoid the confounding influence of non-lqts comorbidity on events after age 40 years.

3 Kimbrough et al arents and Siblings of robands With LQTS 559 Figure 1. Cumulative probability of LQTS-related first cardiac event (syncope, aborted cardiac arrest, or death) from birth to age 40 years for affected female and male parents of LQTS probands. were obtained. ECG parameters included baseline QT c and heart rate. Follow-up data about cardiac events were acquired after enrollment from various family members and recorded on prespecified forms. The clinical course of all enrolled subjects was based on the occurrence of cardiac events from birth until the date of last follow-up. Information about LQTS therapy at enrollment and at periodic follow-up contact was also obtained. Statistical Analysis Associations between demographic, ECG, and therapeutic characteristics among the probands, parents, and siblings were evaluated with the use of standard statistical techniques. Time-dependent cardiac events were assessed by the Kaplan-Meier life-table method. The Cox proportional hazard regression model 12 was used to evaluate the independent contribution of specified clinical characteristics to time-dependent cardiac events. Resulting hazard ratios are reported with 95% confidence intervals for both nominal and continuous variables. Results robands The clinical characteristics of 211 LQTS probands subdivided by sex are presented in Table 1. More than 80% of the TABLE 3. Clinical Variables Clinical Characteristics of Siblings of LQTS robands (n 86) Figure 2. A, Cumulative probability of LQTS-related first cardiac event (syncope, aborted cardiac arrest [ACA], or death) from birth to age 40 years for affected parents by presence or absence of aborted cardiac arrest or LQTS-related death in probands. B, Cumulative probability of LQTS-related first cardiac event (syncope, aborted cardiac arrest, or death) from birth to age 40 years for affected parents by QT c value of probands. probands were identified after a syncopal or cardiac arrest event. probands had their first cardiac event at a younger mean age than did female probands. Despite the younger age presentation by male probands, female probands displayed a higher frequency of cardiac arrest or LQTSrelated death by age 40 years. ECG characteristics were Affected Siblings (n 108) (n 86) Unaffected Siblings (n 89) Demographics and therapy Age at first ECG, y Cardiac event by age 40 y, n (%) 31 (36) 36 (33) 1 (1) 10 (11) 0.01 Cardiac arrest or death by age 40 y, n (%) 8 (9) 11 (10) 0 0 Age at first cardiac event, y (n 1) Blockers anytime, n (%) 28 (33) 41 (38) 2 (2) 9 (10) 0.03 ECG characteristics Heart rate, beats/min QT c,ms QT c 500 ms, n (%) 18 (21) 26 (24) 0 0 Only values 0.05 are shown. Note: Age at first ECG is the age when the sibling was enrolled into the LQTS Registry. Siblings were followed for varying lengths of time, with a few but not all to age 40 years and beyond. Cardiac events and cardiac arrest or death were censored at age 40 years to avoid the confounding influence of non-lqts comorbidity on events after age 40 years.

4 560 Circulation July 31, 2001 similar in male and female LQTS probands, and -blocker use was more frequent in female probands. arents The clinical characteristics of parents of LQTS probands are presented in Table 2. The QT c interval was significantly longer in affected female than male parents. This sex difference in QT c duration was also present in unaffected parents. Mothers more often than fathers had a cardiac event during the first 40 years of their life, but the first cardiac event occurred earlier in life in LQTS fathers than mothers. LQTSrelated aborted cardiac arrest or death occurred in 4% of the affected female parents but not in any of the affected male parents. Affected female parents exhibited an increasing cumulative probability of a first cardiac event with age, whereas affected male parents showed a leveling off of cumulative risk in their early twenties, with a near elimination of future risk of a first event after that age (Figure 1). The probability of a parent having a first cardiac event was not significantly associated with the occurrence of lifethreatening events in the proband (Figure 2A) or to the QT c duration of the proband (Figure 2B). Siblings The clinical characteristics of the siblings of LQTS probands separated by LQTS status are presented in Table 3. Approximately one third of the affected siblings had a cardiac event, with 10% having an aborted cardiac arrest or death by age 40 years. Unaffected siblings had a low frequency of cardiac events, exclusively syncope, but no aborted cardiac arrest or death. The clinical features of the affected siblings were similar in males and females. The probability of a sibling having a first cardiac event was not related to the occurrence of life-threatening events in the proband (Figure 3A) or to the QT c duration of the proband (Figure 3B). Risk Analysis A Cox proportional hazards model was used to identify clinical risk factors for cardiac events among first-degree relatives of the proband (Table 4). For affected parents, female sex and length of the QT c interval of the parents were found to be significant risk factors for cardiac events; for affected siblings, the QT c duration of the sibling was the only significant risk factor associated with cardiac events. No proband characteristics entered the risk models. Discussion We examined the clinical course of family members of LQTS probands whose parents were born before The severity of the LQTS disorder in the proband was not found to be useful in identifying the clinical severity of LQTS in affected first-degree relatives (siblings or parents) of the proband. Affected mothers of LQTS probands displayed an ongoing risk for cardiac events well after the birth of the proband but affected fathers did not. The length of the QT c interval in Figure 3. A, Cumulative probability of LQTS-related first cardiac event (syncope, aborted cardiac arrest [ACA], or death) for affected siblings by presence or absence of aborted cardiac arrest or LQTS-related death in probands. B, Cumulative probability of LQTS-related first cardiac event for affected siblings by QT c value of probands. affected parents and siblings was associated with a significant risk of LQTS-related cardiac events. During our 20-year experience with the International LQTS Registry, the majority of LQTS probands have been identified after a symptomatic cardiac event, either syncope or aborted cardiac arrest. It had been our conventional thinking that a life-threatening presentation of LQTS in a proband would be useful for identifying affected family members, assuming they were accurately classified as affected gene carriers by QT c criteria, with an increased risk of having similar life-threatening cardiac events. The findings from this study do not support this clinical reasoning. Rather, these findings are consistent with variable intrafamily penetrance of this genetic disorder. 13 Most likely, other genetic and environmental factors played a role in modulating and modifying the clinical manifestations of LQTS in different members of the same family. As previously shown, genotype, age, and sex influence the course of LQTS in affected family members. 9,14,15 It is interesting that a sex difference exists in event rates for parents but not for siblings, with higher event rates in mothers than fathers of the proband but similar event rates in sisters

5 Kimbrough et al arents and Siblings of robands With LQTS 561 TABLE 4. Risk Factors for LQTS-Related Cardiac Events in Affected arents and Siblings of LQTS robands Affected arents* Affected Siblings Risk Factor Hazard Ratio 95% CI Hazard Ratio 95% CI Sex (female/male) 3.59 (1.64, 7.88) QT c, per 10-ms increase 1.13 (1.04, 1.23) (1.04, 1.17) *Thirty-six events in 194 affected parents. Sixty events in 164 affected siblings. Note: Variables that did not enter either risk model at 0.10 include the following 4 proband characteristics: QT c, sex, age, and cardiac arrest or LQTS-related death. and brothers of the proband. The discrepant findings in event rates in parents and siblings by sex may relate to the fact that the parents are a generation older than the siblings. A more complete follow-up time exposure is present in parents than siblings of the proband during ages 15 to 40 years, a time period when the cardiac event rate progressively increased in mothers of the proband but not in fathers (Figure 1). Also, a higher percentage of siblings were using -blockers than were parents, and this therapy is known to reduce the subsequent rate of cardiac event in those with higher pre blocker event rates. 16 Study Limitations Family members were categorized as affected or unaffected with LQTS by age- and sex-adjusted QT c criteria. 11 This classification of subjects as affected and unaffected on the basis of QT c alone is somewhat imprecise, and in subjects with borderline-prolonged QT c values there may be some misclassification in those categorized as affected or unaffected. In this regard, from the 211 probands there were 225 parents who were categorized as being affected. This finding suggests that there may have been 14 probands with a double mutation or that 14 parents were incorrectly classified as being affected, or some combination of these two possibilities. When we eliminated the 14 proband families with two affected parents by ECG criteria, the study findings were essentially unchanged. Although proband characteristics did not predict LQTSrelated cardiac events among the affected parents, a trend to increasing risk in parents of probands having more serious symptoms is evident in Figure 2A. The absence of a significant difference in this trend may be due to the fact that the study was underpowered to detect such differences. Unaffected parents and unaffected siblings had a low frequency of syncope, averaging 6%, and none of the unaffected family members had aborted cardiac arrest or premature sudden death. This syncopal event rate of 1% to 11% in unaffected family members may reflect imprecision in the classification of unaffected by the QT c criteria. On the other hand, the syncopal events in these unaffected subjects may have been due to causes unrelated to LQTS, so-called background noise. In an ongoing analysis from the LQTS Registry, we have found that a history of syncope occurs in 6% of genotype-negative members of genotyped LQTS families. Genotype data were available in only 70 subjects in this data set. Appropriate family analyses of disease severity by proband genotype were not possible because of the small number of currently available genotyped subjects. The use of -blocker therapy in affected siblings was in the range of 35%. -Blockers are associated with some reduction in the frequency of syncopal cardiac events, 16 and this therapy may be a confounding factor in the lack of association observed in disease severity between probands and affected siblings. Conclusions This study reveals that an ongoing risk for a first cardiac event persists throughout adulthood for affected mothers of LQTS probands but not for fathers. Although proband presentation varies with respect to sex, age, and disease severity, multivariate analysis did not demonstrate that these proband characteristics predict LQTS-related cardiac events among affected parents or siblings. Because the clinical severity of LQTS in the proband does not predict disease severity in family members, treatment decisions for a first-degree relative of a proband should not be based on severity of LQTS in the proband. Rather, parental sex and parental and sibling QT c duration are important risk characteristics that should be used in treatment decisions for family members of probands, regardless of LQTS severity in the proband. Acknowledgments This study was supported in part by research grant HL and HL from the National Institutes of Health, Bethesda, Md. We thank Mark Andrews for his expert assistance in data management and analysis. References 1. Keating M, Dunn C, Atkinson D, et al. Consistent linkage of the long-qt syndrome to the Harvey ras-1 locus on chromosome 11. Am J Hum Genet. 1991;49: Wang Q, Curran ME, Splawski I, et al. ositional cloning of a novel potassium channel gene: KVLQT1 mutations cause cardiac arrhythmias. Nat Genet. 1996;12: Curran ME, Splawski I, Timothy KW, et al. A molecular basis for cardiac arrhythmia: HERG mutations cause long QT syndrome. Cell. 1995;80: Wang Q, Shen J, Splawski I, et al. SCN5A mutations associated with an inherited cardiac arrhythmia, long QT syndrome. Cell. 1995;80:

6 562 Circulation July 31, Schott JJ, Charpentier F, eltier S, et al. Mapping of a gene for long QT syndrome to chromosome 4q Am J Hum Genet. 1995;57: Splawski I, Tristani-Firouzi M, Lehmann MH, et al. Mutations in the hmink gene cause long QT syndrome and suppress IKs function. Nat Genet. 1997;17: Abbott GW, Sesti F, Splawski I, et al. MiR1 forms IKr potassium channels with HERG and is associated with cardiac arrhythmia. Cell. 1999;97: 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: Zareba W, Moss AJ, Schwartz J, et al. Influence of genotype on the clinical course of the long-qt syndrome. N Engl J Med. 1998;339: Schwartz J, riori SG, Spazzolini C, et al. Genotype-phenotype correlation in the long-qt syndrome: gene-specific triggers for life-threatening arrhythmias. Circulation. 2001;103: Moss AJ, Robinson JL. The long QT syndrome: genetic considerations. Trends Cardiovasc Med. 1992;2: Cox D. Regression models and life-tables. J Stat Soc. 1972;34: riori SG, Napolitano C, Schwartz J. Low penetrance in the long QT syndrome: clinical impact. Circulation. 1999;99: Zareba W, Moss AJ, LeCessie S, et al. Risk of cardiac events in family members of patients with long QT syndrome. J Am Coll Cardiol. 1995; 26: Locati EH, Zareba W, Moss AJ, et al. Age and gender-related differences in cardiac events in patients with congenital long QT syndrome: findings from the International LQTS Registry. Circulation. 1998;97: Moss AJ, Zareba W, Hall WJ, et al. Effectiveness and limitations of beta-blocker therapy in congenital long-qt syndrome. Circulation. 2000; 101: