- Email: [email protected]
Congenital Long QT Syndrome in Children Identified by Family Screening
Congenital Long QT Syndrome in Children Identified by Family Screening Colin Petko, MDa, David J. Bradley, MDa, Martin Tristani-Firouzi, MDa, Mitchell I. Cohen, MDb, Shubhayan Sanatani, MDc, Elizabeth V. Saarel, MDa, Cecilia A. Albaro, MDa, and Susan P. Etheridge, MDa,* The diagnosis of congenital long-QT syndrome (LQTS) in the relatives (nonprobands) of index patients (probands) is increasing because of screening. This report documents the clinical courses and outcomes of nonproband pediatric patients with LQTS. All patients aged <18 years with LQTS were identified at 3 pediatric centers. Demographic data, personal and family histories, electrocardiographic data, and genetic diagnoses (if available) were obtained. Probands were defined as the first patients in their families diagnosed with LQTS and nonprobands as those diagnosed by screening. Of 144 patients with LQTS, 84 (58%) were nonprobands aged 6.5 ⴞ 5.4 years with QTc intervals of 479 ⴞ 34 ms. No nonproband presented with resuscitated sudden death, atrioventricular block, or ventricular arrhythmia, but 7 (8.3%) had histories of syncope at presentation. All nonproband patients were treated. During a follow-up period of 4.7 ⴞ 3.9 years, there were no deaths in the nonproband group, but device implantation was performed in 13 (15%), 4 of whom had appropriate shocks. As expected, compared with probands, nonprobands were less symptomatic. Additionally, nonprobands were younger and had shorter QTc intervals. Although device implantation was more common in probands, there was no difference in appropriate implantable cardioverter-defibrillator shocks or mortality between the probands and their affected relatives. In conclusion, children are increasingly identified with LQTS as a result of family screening. Although phenotypic differences exist between probands and nonprobands, survival is excellent in the 2 groups with therapy. Appropriate implantable cardioverter-defibrillator discharges in the nonproband group underscore the importance of follow-up in this relatively asymptomatic population. © 2008 Elsevier Inc. All rights reserved. (Am J Cardiol 2008;101:1756 –1758) Patients with long-QT syndrome (LQTS) identified as a result of family screening practices now represent a sizable portion of the LQTS population managed by pediatric cardiologists. This population, although increasing, has not been systematically evaluated. We sought to describe the clinical courses and outcomes of these nonproband pediatric LQTS patients identified through family screening practices. Methods and Results Institutional review board approval was obtained from the 3 participating institutions. Pediatric cardiology databases were searched for all patients aged ⬍18 years followed with LQTS from January 1989 to January 2007. Demographic data, personal and family histories, electrocardiographic data, and genetic diagnoses (if available) were recorded. Follow-up data collection was closed on October 1, 2007. We reviewed the clinical histories, noting cardiac events related to LQTS, including syncope, sudden death, resuscitated sudden death, seizures, and near drowning. a
University of Utah, Salt Lake City, Utah; bPhoenix Children’s Hospital and Arizona Pediatric Cardiology Consultants/Pediatrix, Phoenix, Arizona; and cUniversity of British Columbia, Vancouver, British Columbia, Canada. Manuscript received December 28, 2007; revised manuscript received and accepted February 2, 2008. *Corresponding author: Tel: 801-662-5400; fax: 801-662-5404. E-mail address: [email protected] (S.P. Etheridge). 0002-9149/08/$ – see front matter © 2008 Elsevier Inc. All rights reserved. doi:10.1016/j.amjcard.2008.02.068
Probands were defined as the index patients or first patients in their families to be diagnosed with LQTS. Nonprobands were identified as patients who presented for evaluation because of known family histories of LQTS; they represented positively diagnosed relatives of probands. Although all nonprobands presented as a result of family screening, some nonprobands acknowledged symptoms at the time of presentation. Each patient underwent baseline 12-lead electrocardiography before therapy and in the absence of medications known to prolong repolarization. The QT interval was corrected using Bazett’s formula. Confirmatory electrocardiography was performed on all patients. Genetic analysis was performed at 1 commercial and various research laboratories. Patients were screened for known LQTS mutations in the KCNH2, KCNE1, KCNE2, KCNQ1, and SCN5A genes. All patients were screened for additional mutations once an initial mutation was found to identify potential compound heterozygotes. In the absence of genetic data, the diagnosis of LQTS was made by a pediatric electrophysiologist and confirmed clinically using exercise testing, additional electrocardiography, electrocardiography from family members, and, rarely, epinephrine challenge. Regardless of symptoms, all 144 patients diagnosed with LQTS in this series were treated. Beta blockers were used in 143 patients, with mexiletine added in 4. There was a single patient with an implantable cardioverter-defibrillator (ICD) who did not receive medicine. All patients were counseled to avoid strenuous activity and medications known to prowww.AJConline.org
Arrhythmias and Conduction Disturbances/LQTS in Nonprobands Table 1 Comparison of proband and nonproband patients with long-QT syndrome Variable Age at diagnosis (yrs) Length of follow-up (yrs) Female QTc interval (ms) Symptoms -blocker therapy Device Positive genetic test results Death
Nonprobands (n ⫽ 84)
Probands (n ⫽ 60)
p Value
6.5 ⫾ 5.4 4.7 ⫾ 3.9 45% 479 ⫾ 34 17% 99% 15% 58% 0%
9.4 ⫾ 5.4 4.4 ⫾ 3.8 60% 502 ⫾ 47 73% 100% 31% 33% 1.6%
0.002 NS NS ⬍0.001 ⬍0.001 NS 0.03 0.005 NS
Data are expressed as mean ⫾ SD or percentages.
long repolarization. ICD or pacemaker implantation was performed at the discretion of the treating electrophysiologist. There was uniformity among contributing centers concerning the following implantation indications: (1) symptoms despite -blocker therapy, (2) a known SCN5A mutation, (3) atrioventricular block, (4) documented ventricular tachycardia or Torsades de Pointes, and (5) presentation with resuscitated sudden death.1 Additionally, devices were placed for profound QTc prolongation and a worrisome family history of sudden death in first-degree relatives, although there was not uniform adherence to these indications. Continuous variables are presented as mean ⫾ SD and categorical variables as percentages. Student’s t test was used to compare continuous data, the chi-square or Fisher’s exact test for categorical data, and Kaplan-Meier analysis to assess survival with differences between groups determined by logrank tests. A p value ⱕ0.05 was considered significant. A total of 144 children from 95 families were identified with LQTS. Table 1 lists the demographic and electrocardiographic data of probands and nonprobands. The total number of families differs from the number of probands because the probands were either adults (and not included in the study) or were ascertained at centers outside of our study group. The diagnosis of LQTS was most commonly made from abnormal electrocardiographic results obtained during screening in patients with family histories of LQTS (84 patients [58%]). Other patients presented with syncope (30 patients), sudden death (3 patients), or QTc prolongation on electrocardiography without symptoms (27 patients). There were 32 patients with either pacemakers (6 patients) or ICDs (26 patients). Genetic testing had been performed in 83 (58%) patients. In 68 of these subjects (82%), missense mutations were identified at a position previously associated with LQTS. In 8 patients, no mutations were identified, and in 7 patients, the genetic results are pending (Figure 1). None of the nonprobands tested has had negative genetic test results. Sudden cardiac death occurred in 1 patient in the proband group. Nonproband patients were identified as a result of family screening, although 7 patients (8.3%) reported ⱖ1 episode of syncope before evaluation and therapy, with 1 event occurring while swimming. No nonproband patient presented with resuscitated sudden death, atrioventricular
1757
block, or ventricular arrhythmia. In the nonproband group, 18 patients (21%) had QTc intervals ⬎500 ms. There was no relation between symptoms at presentation or during follow-up and the QTc interval. All patients in the nonproband group were treated with  blockers and/or device therapy. Beta blockers were used in 83 of 84 nonproband patients (99%), most commonly nadolol, with propranolol and atenolol used less often. Because of described treatment failure, atenolol is used less frequently.2 One nonproband with an SCN5A mutation was treated with an ICD but without a  blocker. Device implantation was undertaken in 13 nonproband patients (1 pacemaker and 12 ICDs). ICDs were implanted in all nonproband patients identified with SCN5A mutations but none with KCNQ1 mutations. ICD indications were syncope despite -blocker therapy in 8 patients, of whom 3 had syncope at presentation and continued episodes despite therapy and 5 developed symptoms while receiving  blockers. ICDs were placed in 3 siblings with family histories of sudden death and 2 asymptomatic patients with SCN5A mutations. In total, genetic diagnoses in these 13 included 2 SCN5A mutations, 3 KCNH2 mutations, 5 with no testing performed, and 3 with results pending. In nonproband patients with ICDs, 4 had appropriate shocks (1 with a KCNH2 mutation and 3 not genotyped); 2 of these also had inappropriate shocks. Device implantation was more common in the proband group compared with nonprobands, but there was no difference in the number of total or appropriate ICD shocks. In the nonproband group, we assessed the relation between symptoms (at presentation or during follow-up) and relationship to the proband. Symptoms were no more prevalent in patients with proband parents or siblings than in those with more distantly related probands. Discussion The identification of the molecular basis of LQTS has facilitated the screening of family members of index patients. Consequently, subjects identified as a result of family screening represent an increasing proportion of patients with LQTS. With enhanced family screening and preparticipation evaluation, this nonproband population is likely to increase further. Indeed, among the 144 pediatric patients with LQTS, the most common reason for diagnosis was abnormal electrocardiographic results obtained during screening because of a family history of LQTS. Nonproband patients were younger at presentation, were less likely to have or to develop symptoms related to LQTS, and had significantly shorter QTc intervals. Mortality was low in the 2 groups, with no deaths among nonprobands. Approximately 30 years ago, LQTS was thought to have a mortality of up to 73% without therapy, with a decrease to 6% with -blocker therapy.3 Children and adolescents were thought to be at the highest risk for LQTS-related sudden death or aborted sudden death.4 The first large series of exclusively pediatric patients with LQTS was published in 1993 and reflects a previous era in LQTS management.5 In that multicenter, retrospective analysis, the 5-year mortality was 8%, with only 82% receiving therapy. Our data and
1758
The American Journal of Cardiology (www.AJConline.org)
Figure 1. The distribution of genetic test results in probands and nonprobands is shown in absolute numbers. The compound heterozygote patient was positive for KCNQ1 and SCN5A mutations.
those of others suggest lower mortality, possibly as a result of early identification and treatment.1,6 Our results reflect recent diagnostic trends and management practices for pediatric patients with LQTS. The approach at the 3 centers was fairly uniform and included family screening and education concerning lifestyle restrictions; -blocker therapy in all patients, regardless of symptoms; and device therapy for patients at high risk. With this approach, there was a single death in a 2-yearold child, a proband, who died despite -blocker therapy and a pacemaker. Although there were no deaths in the group of patients with LQTS identified as a result of screening, some were symptomatic at presentation, and others developed symptoms during follow-up despite therapy. And although they were less likely than probands to require device implantation (15% vs 31%), they received a similar rate of appropriate ICD shocks. Having a proband as a first-degree relative did not confer an increased risk for symptoms compared with a more distant relative. Thus, continued follow-up throughout childhood is mandatory to adequately protect proband and nonproband patients with LQTS. This study was limited by its retrospective design and relatively small sample size. The broad geographic distribution of the patients in this series prevented the evaluation of complete families in many cases. Genetic testing was not
systematically performed in this population but was rather dictated by the availability of research laboratory resources and insurance coverage of commercial genotyping. In the absence of confirmatory genetic testing, diagnoses were made on the basis of clinical assessment and therefore may be subject to error. Comparisons of mortality rates between proband and nonproband groups are limited by their uncommon occurrence. 1. Etheridge SP, Sanatani S, Cohen MI, Albaro CA, Saarel EV, Bradley DJ. Long QT syndrome in children in the era of implantable defibrillators. J Am Coll Cardiol 2007;50:1335–1340. 2. Chatrath R, Bell CM, Ackerman MJ. Beta-blocker therapy failures in symptomatic probands with genotyped long-QT syndrome. Pediatr Cardiol 2004;25:459 – 465. 3. Schwartz PJ, Periti M, Malliani A. The long Q-T syndrome. Am Heart J 1975;89:378 –390. 4. Hobbs JB, Peterson DR, Moss AJ, McNitt S, Zareba W, Goldenberg I, Qi M, Robinson JL, Sauer AJ, Ackerman MJ, et al. Risk of aborted cardiac arrest or sudden cardiac death during adolescence in the long-QT syndrome. JAMA 2006;296:1249 –1254. 5. Garson A, Jr., Dick M II, Fournier A, Gillette PC, Hamilton R, Kugler JD, van Hare GF III, Vetter V, Vick GW III. The long QT syndrome in children. An international study of 287 patients. Circulation 1993;87: 1866 –1872. 6. Priori SG, Napolitano C, Schwartz PJ, Grillo M, Bloise R, Ronchetti E, Moncalvo C, Tulipani C, Veia A, Bottelli G, Nastoli J. Association of long QT syndrome loci and cardiac events among patients treated with beta-blockers. JAMA 2004;292:1341–1344.
University of Utah, Salt Lake City, Utah; bPhoenix Children’s Hospital and Arizona Pediatric Cardiology Consultants/Pediatrix, Phoenix, Arizona; and cUniversity of British Columbia, Vancouver, British Columbia, Canada. Manuscript received December 28, 2007; revised manuscript received and accepted February 2, 2008. *Corresponding author: Tel: 801-662-5400; fax: 801-662-5404. E-mail address: [email protected] (S.P. Etheridge). 0002-9149/08/$ – see front matter © 2008 Elsevier Inc. All rights reserved. doi:10.1016/j.amjcard.2008.02.068
Probands were defined as the index patients or first patients in their families to be diagnosed with LQTS. Nonprobands were identified as patients who presented for evaluation because of known family histories of LQTS; they represented positively diagnosed relatives of probands. Although all nonprobands presented as a result of family screening, some nonprobands acknowledged symptoms at the time of presentation. Each patient underwent baseline 12-lead electrocardiography before therapy and in the absence of medications known to prolong repolarization. The QT interval was corrected using Bazett’s formula. Confirmatory electrocardiography was performed on all patients. Genetic analysis was performed at 1 commercial and various research laboratories. Patients were screened for known LQTS mutations in the KCNH2, KCNE1, KCNE2, KCNQ1, and SCN5A genes. All patients were screened for additional mutations once an initial mutation was found to identify potential compound heterozygotes. In the absence of genetic data, the diagnosis of LQTS was made by a pediatric electrophysiologist and confirmed clinically using exercise testing, additional electrocardiography, electrocardiography from family members, and, rarely, epinephrine challenge. Regardless of symptoms, all 144 patients diagnosed with LQTS in this series were treated. Beta blockers were used in 143 patients, with mexiletine added in 4. There was a single patient with an implantable cardioverter-defibrillator (ICD) who did not receive medicine. All patients were counseled to avoid strenuous activity and medications known to prowww.AJConline.org
Arrhythmias and Conduction Disturbances/LQTS in Nonprobands Table 1 Comparison of proband and nonproband patients with long-QT syndrome Variable Age at diagnosis (yrs) Length of follow-up (yrs) Female QTc interval (ms) Symptoms -blocker therapy Device Positive genetic test results Death
Nonprobands (n ⫽ 84)
Probands (n ⫽ 60)
p Value
6.5 ⫾ 5.4 4.7 ⫾ 3.9 45% 479 ⫾ 34 17% 99% 15% 58% 0%
9.4 ⫾ 5.4 4.4 ⫾ 3.8 60% 502 ⫾ 47 73% 100% 31% 33% 1.6%
0.002 NS NS ⬍0.001 ⬍0.001 NS 0.03 0.005 NS
Data are expressed as mean ⫾ SD or percentages.
long repolarization. ICD or pacemaker implantation was performed at the discretion of the treating electrophysiologist. There was uniformity among contributing centers concerning the following implantation indications: (1) symptoms despite -blocker therapy, (2) a known SCN5A mutation, (3) atrioventricular block, (4) documented ventricular tachycardia or Torsades de Pointes, and (5) presentation with resuscitated sudden death.1 Additionally, devices were placed for profound QTc prolongation and a worrisome family history of sudden death in first-degree relatives, although there was not uniform adherence to these indications. Continuous variables are presented as mean ⫾ SD and categorical variables as percentages. Student’s t test was used to compare continuous data, the chi-square or Fisher’s exact test for categorical data, and Kaplan-Meier analysis to assess survival with differences between groups determined by logrank tests. A p value ⱕ0.05 was considered significant. A total of 144 children from 95 families were identified with LQTS. Table 1 lists the demographic and electrocardiographic data of probands and nonprobands. The total number of families differs from the number of probands because the probands were either adults (and not included in the study) or were ascertained at centers outside of our study group. The diagnosis of LQTS was most commonly made from abnormal electrocardiographic results obtained during screening in patients with family histories of LQTS (84 patients [58%]). Other patients presented with syncope (30 patients), sudden death (3 patients), or QTc prolongation on electrocardiography without symptoms (27 patients). There were 32 patients with either pacemakers (6 patients) or ICDs (26 patients). Genetic testing had been performed in 83 (58%) patients. In 68 of these subjects (82%), missense mutations were identified at a position previously associated with LQTS. In 8 patients, no mutations were identified, and in 7 patients, the genetic results are pending (Figure 1). None of the nonprobands tested has had negative genetic test results. Sudden cardiac death occurred in 1 patient in the proband group. Nonproband patients were identified as a result of family screening, although 7 patients (8.3%) reported ⱖ1 episode of syncope before evaluation and therapy, with 1 event occurring while swimming. No nonproband patient presented with resuscitated sudden death, atrioventricular
1757
block, or ventricular arrhythmia. In the nonproband group, 18 patients (21%) had QTc intervals ⬎500 ms. There was no relation between symptoms at presentation or during follow-up and the QTc interval. All patients in the nonproband group were treated with  blockers and/or device therapy. Beta blockers were used in 83 of 84 nonproband patients (99%), most commonly nadolol, with propranolol and atenolol used less often. Because of described treatment failure, atenolol is used less frequently.2 One nonproband with an SCN5A mutation was treated with an ICD but without a  blocker. Device implantation was undertaken in 13 nonproband patients (1 pacemaker and 12 ICDs). ICDs were implanted in all nonproband patients identified with SCN5A mutations but none with KCNQ1 mutations. ICD indications were syncope despite -blocker therapy in 8 patients, of whom 3 had syncope at presentation and continued episodes despite therapy and 5 developed symptoms while receiving  blockers. ICDs were placed in 3 siblings with family histories of sudden death and 2 asymptomatic patients with SCN5A mutations. In total, genetic diagnoses in these 13 included 2 SCN5A mutations, 3 KCNH2 mutations, 5 with no testing performed, and 3 with results pending. In nonproband patients with ICDs, 4 had appropriate shocks (1 with a KCNH2 mutation and 3 not genotyped); 2 of these also had inappropriate shocks. Device implantation was more common in the proband group compared with nonprobands, but there was no difference in the number of total or appropriate ICD shocks. In the nonproband group, we assessed the relation between symptoms (at presentation or during follow-up) and relationship to the proband. Symptoms were no more prevalent in patients with proband parents or siblings than in those with more distantly related probands. Discussion The identification of the molecular basis of LQTS has facilitated the screening of family members of index patients. Consequently, subjects identified as a result of family screening represent an increasing proportion of patients with LQTS. With enhanced family screening and preparticipation evaluation, this nonproband population is likely to increase further. Indeed, among the 144 pediatric patients with LQTS, the most common reason for diagnosis was abnormal electrocardiographic results obtained during screening because of a family history of LQTS. Nonproband patients were younger at presentation, were less likely to have or to develop symptoms related to LQTS, and had significantly shorter QTc intervals. Mortality was low in the 2 groups, with no deaths among nonprobands. Approximately 30 years ago, LQTS was thought to have a mortality of up to 73% without therapy, with a decrease to 6% with -blocker therapy.3 Children and adolescents were thought to be at the highest risk for LQTS-related sudden death or aborted sudden death.4 The first large series of exclusively pediatric patients with LQTS was published in 1993 and reflects a previous era in LQTS management.5 In that multicenter, retrospective analysis, the 5-year mortality was 8%, with only 82% receiving therapy. Our data and
1758
The American Journal of Cardiology (www.AJConline.org)
Figure 1. The distribution of genetic test results in probands and nonprobands is shown in absolute numbers. The compound heterozygote patient was positive for KCNQ1 and SCN5A mutations.
those of others suggest lower mortality, possibly as a result of early identification and treatment.1,6 Our results reflect recent diagnostic trends and management practices for pediatric patients with LQTS. The approach at the 3 centers was fairly uniform and included family screening and education concerning lifestyle restrictions; -blocker therapy in all patients, regardless of symptoms; and device therapy for patients at high risk. With this approach, there was a single death in a 2-yearold child, a proband, who died despite -blocker therapy and a pacemaker. Although there were no deaths in the group of patients with LQTS identified as a result of screening, some were symptomatic at presentation, and others developed symptoms during follow-up despite therapy. And although they were less likely than probands to require device implantation (15% vs 31%), they received a similar rate of appropriate ICD shocks. Having a proband as a first-degree relative did not confer an increased risk for symptoms compared with a more distant relative. Thus, continued follow-up throughout childhood is mandatory to adequately protect proband and nonproband patients with LQTS. This study was limited by its retrospective design and relatively small sample size. The broad geographic distribution of the patients in this series prevented the evaluation of complete families in many cases. Genetic testing was not
systematically performed in this population but was rather dictated by the availability of research laboratory resources and insurance coverage of commercial genotyping. In the absence of confirmatory genetic testing, diagnoses were made on the basis of clinical assessment and therefore may be subject to error. Comparisons of mortality rates between proband and nonproband groups are limited by their uncommon occurrence. 1. Etheridge SP, Sanatani S, Cohen MI, Albaro CA, Saarel EV, Bradley DJ. Long QT syndrome in children in the era of implantable defibrillators. J Am Coll Cardiol 2007;50:1335–1340. 2. Chatrath R, Bell CM, Ackerman MJ. Beta-blocker therapy failures in symptomatic probands with genotyped long-QT syndrome. Pediatr Cardiol 2004;25:459 – 465. 3. Schwartz PJ, Periti M, Malliani A. The long Q-T syndrome. Am Heart J 1975;89:378 –390. 4. Hobbs JB, Peterson DR, Moss AJ, McNitt S, Zareba W, Goldenberg I, Qi M, Robinson JL, Sauer AJ, Ackerman MJ, et al. Risk of aborted cardiac arrest or sudden cardiac death during adolescence in the long-QT syndrome. JAMA 2006;296:1249 –1254. 5. Garson A, Jr., Dick M II, Fournier A, Gillette PC, Hamilton R, Kugler JD, van Hare GF III, Vetter V, Vick GW III. The long QT syndrome in children. An international study of 287 patients. Circulation 1993;87: 1866 –1872. 6. Priori SG, Napolitano C, Schwartz PJ, Grillo M, Bloise R, Ronchetti E, Moncalvo C, Tulipani C, Veia A, Bottelli G, Nastoli J. Association of long QT syndrome loci and cardiac events among patients treated with beta-blockers. JAMA 2004;292:1341–1344.