ECG T-Wave Patterns in Genetically Distinct Forms of the Hereditary Long QT Syndrome
Background The long QT syndrome is an inherited disorder with prolonged ventricular repolarization and a propensity to ventricular tachyarrhythmias and sudden arrhythmic death. Recent linkage studies have demonstrated three separate loci for this disorder on chromosomes 3, 7, and 11, and specific mutated genes for long QT syndrome have been identified on two of these chromosomes. We investigated ECG T-wave patterns (phenotypes) in members of families linked to three genetically distinct forms of the long QT syndrome.
Methods and Results Five quantitative ECG repolarization parameters, ie, four Bazett-corrected time intervals (QTonset-c, QTpeak-c, QTc, and Tduration-c, in milliseconds) and the absolute height of the T wave (Tamplitude, in millivolts), were measured in 153 members of six families with long QT syndrome linked to markers on chromosomes 3 (n=47), 7 (n=30), and 11 (n=76). Genotypic data were used to define each family member as being affected or unaffected with long QT syndrome. Affected members of all six families had longer QT intervals (QTonset-c, QTpeak-c, or QTc) than unaffected family members (P<.01). Each of the three long QT syndrome genotypes was associated with somewhat distinctive ECG repolarization features. Among affected individuals, the QTonset-c was unusually prolonged in those individuals with mutations involving the cardiac sodium channel gene SCN5A on chromosome 3 (lead II QTonset-c [mean±SD]: chromosome 3, 341±42 ms; chromosome 7, 290±56 ms; chromosome 11, 243±73 ms; P<.001); Tamplitude was generally quite small in the chromosome 7 genotype (lead II Tamplitude, mV: chromosome 3, 0.36±0.14; chromosome 7, 0.13±0.07; chromosome 11, 0.37±0.17; P<.001); and Tduration was particularly long in the chromosome 11 genotype (lead II Tduration-c: chromosome 3, 187±33 ms; chromosome 7, 191±51 ms; chromosome 11, 262±65 ms; P<.001). Similar ECG findings were observed in leads aVF and V5. A considerable variability exists in the quantitative repolarization parameters associated with each genotype, with overlap in the T-wave patterns among the three genotypes.
Conclusions Three separate genetic loci for the long QT syndrome including mutations in two cardiac ionic channel genes were associated with different phenotypic T-wave patterns on the ECG. This study provides insight into the influence of genetic factors on ECG manifestations of ventricular repolarization.
The familial form of long QT syndrome is predominantly an autosomal-dominant disorder associated with prolonged ventricular repolarization and a propensity to recurrent syncope, polymorphousventricular tachycardia (torsade de pointes), and sudden death.1 2 Linkage studies have demonstrated genetic heterogeneity in the long QT syndrome,3 4 5 6 with at least three separate loci for this inherited disorder, ie, on chromosomes 3p21-24, 7q35-36, and 11p15.5.3 7 Recently, two forms of long QT syndrome have been shown to result from mutations in ionic channel genes on chromosomes 3 and 7.8 9
Several quantitatively abnormal ECG characteristics have been described in patients with long QT syndrome.10 Among individuals with long QT syndrome and QTc prolongation, the morphology of the T wave is frequently unusual, with a spectrum of configurations.11 In two recent studies, the clinical significance of notched or biphasic T waves was reported in patients with long QT syndrome.12 13
In reviewing the ECGs from members of families with long QT syndrome having three different genotypes,3 7 we noted similar morphological repolarization patterns among patients with the same gene linkage and dissimilar repolarization patterns in patients with long QT with linkage to different gene loci. The purpose of this study is to investigate the ECG T-wave patterns (phenotypes) in patients with the long QT syndrome from families showing linkage to three different gene loci.
Five previously reported families3 5 7 and one additional family with autosomal-dominant forms of susceptibility to the long QT syndrome, involving 153 family members with linkage to markers on chromosomes 3 (n=47), 7 (n=30), and 11 (n=76), served as the study population.3 5 7 No patient had congenital neural hearing loss. Genotypic data were used to define individual family members as being affected or unaffected with the long QT syndrome.3 5 7 8 Routine demographic data and 12-lead ECGs were obtained on all subjects. All subjects provided informed consent for the gene linkage studies.
The first 12-lead ECG obtained on each patient was analyzed in terms of specific time intervals and signal amplitudes, with particular attention to the quantitative characteristics of ventricular repolarization. T-Wave alternans was not present on any of the baseline ECGs. Ventricular ectopic beats were rare, and sinus beats following the pause after an ectopic premature beat were not selected for analysis. When marked sinus arrhythmia was present and the recorded lead was long enough to identify this rhythm, the next complex after the shortest RR interval was selected for analysis, as recommended by Martin et al.14 In each subject, measurements describing cycle length, QRS duration, and T-wave morphology were made on one representative complex in leads II, aVF, and V5. The RR (ms) interval duration was measured in the cycle preceding the analyzed repolarization pattern. T-Wave patterns were evaluated using five quantitative parameters, with correction (c) for heart rate using the Bazett formula15 for rate-dependent parameters (parameterc=measured parameter/s). We report the Bazett-corrected parameters in the same units as the original parameter, as recently recommended by Molnar et al.16 The five parameters were as follows: (1) QTonset-c (ms), the time interval from the Q wave to the point at which the T wave departs from the flat portion of the ST segment; (2) QTpeak-c (ms), the time interval from the Q wave to the positive or negative peak of the T wave; (3) QTc (ms), the time interval from the Q wave to the end of the T wave, defined as the point of its merger with the isoelectric line. When a discrete U wave interrupted the T wave before the latter returned to the baseline, the end of the T wave was defined as the nadir of the curve between the T and U waves. (4) Tduration-c (ms), the overall duration of the T wave from the T wave onset to the end of the T wave; and (5) Tamplitude (mV), the absolute amplitude of the T wave from the extrapolated PR segment–baseline to the T-wave peak. In this study population, low-amplitude notches on the downslope of the T wave13 were observed in 17% of the affected and 5% of the unaffected family members; discrete U waves were observed in 21% of the affected and 53% of the unaffected family members. Time and amplitude measurements of T-wave notches and QT-U intervals were not used in the evaluation of T-wave patterns in the current analyses.
Two separate families were associated with linkage to long QT syndrome markers on each of three chromosomes. The affected and unaffected members of each family for each chromosome were identified, with the result that there were 12 subgroups. For continuous variables, a Student’s t test was used to evaluate differences in a given parameter between two subgroups, and a one-way ANOVA test was performed to evaluate differences among multiple subgroups.17 For dichotomous variables, χ2 tests were performed to detect differences in the distribution of the variable among two or more subgroups.
For each of the five repolarization parameters, ANOVA was performed to evaluate family differences and differences between affected and unaffected members from families linked to markers on each chromosome. The parameter Tamplitude was analyzed in logarithmic units due to skewness and increasing variability with increasing mean values; reported probability values for Tamplitude were derived from such analyses. Pairs of subgroups were compared by use of pooled t tests. Because parameters sometimes differed between the two families associated with a single chromosome, data were never pooled across such families; instead, means from two families were averaged and the standard errors then determined (the square root of the sum of squared standard errors, using pooled standard deviations). Such averages were then compared by use of Student’s t tests or F tests. To assess differences in the repolarization parameters between two families, analyses were performed with stratification by affected/unaffected status.
Eight family members were taking β-adrenergic blockers when the ECG was recorded. Exclusion of these eight individuals had a negligible effect on the findings. The reported findings are for the entire study population inclusive of these eight individuals.16
A total of 153 members from six families with long QT syndrome were studied, with 76 members (49.7%) affected and 77 members (50.3%) unaffected by genotypic criteria. Long QT syndrome was associated with chromosome 11 in approximately 53% of the affected family members, with chromosome 3 in 25%, and with chromosome 7 in 22%. The clinical characteristics of the study population, broken down by chromosome, family, and affected status, are presented in Tables 1 through 3⇓⇓⇓. Age distribution was similar across the 12 subgroups (P>.10 by ANOVA). Sex distribution among the affected and unaffected members of the six families was somewhat variable but not different by χ2 analysis (P>.10). At the time of the analyzed ECG, therapy with β-adrenergic blocking agents was relatively infrequent in both unaffected (1.3%) and affected (9.2%) family members. No family member had undergone left cervicothoracic sympathetic ganglionectomy before the baseline ECG recording.
Quantitative ECG repolarization parameters in unaffected and affected members of families with long QT syndrome linked to markers on chromosomes 3, 7, and 11 are presented in Tables 1⇑, 2⇑, and 3⇑, respectively. The ECG data in these tables are for lead II; similar findings were obtained when using quantitative parameters from leads aVF and V5.
In the two families showing linkage to markers on chromosome 3 (Table 1⇑), all affected individuals had identical mutations in the cardiac sodium channel gene SCN5A.8 On average, those affected with this sodium channel gene abnormality had significantly prolonged repolarization parameters (QTonset-c, QTpeak-c, and QTc) compared with unaffected members of the same family (P<.001; Table 1⇑). T-wave duration and amplitude were similar in unaffected and affected members in each of the two families. RR interval was longer in affected than in unaffected members of both families (P<.05). The two families had similar repolarization characteristics except that Tduration-c was longer in family 1 than in family 2 (P<.01).
Long QT syndrome in two families was linked to markers on chromosome 7 (Table 2⇑). On average, the affected members of both of these families had significantly prolonged QTc intervals (P<.01) and reduced T-wave amplitudes (P=.02) compared with unaffected members. Repolarization characteristics of the two families were similar.
In two other families, long QT syndrome was linked to markers on chromosome 11 (Table 3⇑). When compared with unaffected members of the same family, affected family members had significantly prolonged repolarization parameters (QTonset-c, QTpeak-c, and QTc; P<.01), a trend toward a prolonged Tduration-c (P<.01 in one of the two families), and similar T-wave amplitudes. Repolarization characteristics of the two families were similar except that Tduration-c was longer in family 5 than in family 6 (P<.01).
Comparison of Repolarization Characteristics in the Three Genotypes
Comparisons by chromosome of the five repolarization parameters in ECG leads II, aVF, and V5 among affected individuals are presented in Table 4⇓. The T-wave parameters were significantly different among the three genotypes in each of the three leads with the single exception of Tduration-c in lead aVF. Each of the three long QT syndrome genotypes had distinctive repolarization patterns. The QT parameters (QTonset-c, QTpeak-c, and QTc) were most prolonged among affected members with the chromosome 3 genotype. Tamplitude was smallest in the chromosome 7 genotype, and Tduration-c was longest in the chromosome 11 genotype.
Repolarization parameters among the unaffected members of the six families also differed to some degree (Tables 1 through 3⇑⇑⇑). We therefore examined the differences in mean values for repolarization parameters between affected and unaffected family members, taking the average of the two families associated with each chromosome (Table 5⇓). Affected individuals carrying the abnormal gene on chromosome 3 had especially prolonged QT intervals, with T waves of normal duration and amplitude. Affected individuals carrying the abnormal gene on chromosome 7 had moderately prolonged QT intervals and low T-wave amplitude. Affected patients carrying the abnormal gene on chromosome 11 had moderately prolonged QT intervals and prolonged T-wave duration.
Examples of ECGs that reflect the distinguishing repolarization characteristics described in Tables 1 through 5⇑⇑⇑⇑⇑ for affected family members carrying abnormal genes for long QT syndrome on chromosome 3, 7, and 11 are presented in the Figure⇓. It should be emphasized that within a specific genotype, variability exists in the ECG repolarization pattern among affected family members, and there is overlap in the T-wave morphology between affected and unaffected individuals from a given family. Also, overlap exists in T-wave patterns between affected individuals with different genotypes, especially those from families in whom the syndrome is linked to chromosomes 7 and 11. The ECG repolarization pattern was most distinctive in affected individuals with SCN5A mutations on chromosome 3 (the Figure⇓).
In patients with long QT syndrome, the ECG T-wave repolarization pattern is influenced by genotype. Although QT-interval prolongation has been the hallmark of the clinical diagnosis of long QT syndrome, variations in the morphological configuration of prolonged or delayed repolarization have been recognized by several investigators.10 11 12 13 The present study shows that in patients with long QT syndrome, different repolarization patterns on the ECG are associated with different genotypes.
Patients with mutations in the SCN5A sodium channel gene on chromosome 38 have a distinctive, late-appearing T wave that is clearly different from the low-amplitude, moderately delayed T wave observed in affected patients who are carriers of the abnormal gene on chromosome 7. Both of these repolarization patterns are different from the broad-based, prolonged T-wave pattern found in patients who are carriers of the abnormal gene on chromosome 11.
Several ionic currents are operative during ventricular repolarization, and considerable attention has been focused in the past on possible alterations in the delayed rectifier potassium current and the calcium current to explain the electrophysiological phenomena responsible for QT prolongation.18 Recent discoveries by Wang et al8 and Curran et al9 support the hypothesis that at least two forms of long QT syndrome result from mutations in cardiac ion channel genes that are involved in the structure and function of sodium and potassium channels.
Wang et al8 have pointed out that a mutation in the SCN5A gene on chromosome 3 causes deletions in the critical amino acid sequences in the sodium channel region responsible for fast sodium inactivation.19 Subtle abnormalities of sodium channel function, eg, delayed inactivation or altered voltage dependence of channel inactivation, could delay repolarization and lead to the QT prolongation observed in patients with this SCN5A mutation (the Figure⇑).
The mean values for the quantitative repolarization parameters in affected individuals of a given genotype are quite different from the mean values of unaffected family members (Tables 1 through 3⇑⇑⇑) and from the mean values of affected individuals with different genotypes (Table 4⇑). It is quite obvious from Tables 1 through 5⇑⇑⇑⇑⇑ that considerable variability exists in the quantitative repolarization parameters within families, between families, and among the three genotypes for both affected and unaffected individuals. This quantitative variability reflects the spectrum of observed repolarization patterns and the overlap that exists in the ECG phenotypes among the three different genotypes with long QT syndrome. This variability and overlap are not unexpected; the expression of mutated genes is influenced not only by the genetic milieu20 but also by age, sex, heart rate, and various acquired factors. Vincent et al21 previously showed a substantial overlap in the QTc interval among carriers and noncarriers of the abnormal gene on chromosome 11 for long QT syndrome, a finding consistent with variable penetrance. Among the acquired factors, β-adrenergic blocker therapy has been reported to modify the notched T-wave pattern in some patients with long QT syndrome.12 β-Blocker therapy was infrequently used in the affected individuals at the time of the baseline ECG in the present study, and exclusion of these individuals had only a negligible influence on the observed repolarization patterns. Thus, there is no evidence that β-blockers had a meaningful effect on the T-wave patterns in this population.
The present study highlights the relation between genotype and ECG phenotype in the hereditary long QT syndrome. Because multiple genetic and acquired factors can influence ventricular repolarization, it is unlikely that the morphological pattern of the T wave can be used to accurately identify the genotype in patients with suspected long QT syndrome. Rather, this study provides new insight into the genetic determinants of ventricular repolarization as manifested by various phenotypic T-wave patterns on the ECG in three different long QT syndrome genotypes. A larger number of families and affected members with the long QT syndrome is needed to evaluate the clinical implications of the different genotypes and the associated ECG phenotypes regarding the occurrence of arrhythmic cardiac events such as syncope, aborted cardiac arrest, and sudden cardiac death.
This work was supported in part by grants from the National Institutes of Health (RO1-HL-33843 and RO1-HL-51618). We appreciate the technical assistance of Mark Andrews and the secretarial expertise of Mara Leppaluoto. We are indebted to the members of the families with long QT syndrome who participated in this research program.
- Received May 11, 1995.
- Revision received July 12, 1995.
- Accepted August 3, 1995.
- Copyright © 1995 by American Heart Association
Moss AJ, Schwartz PJ, Crampton RS, Tzivoni D, Locati EH, MacCluer J, Hall WJ, Weitkamp L, Vincent M, Garson A Jr, Robinson JL, Benhorin J, Choi S. The long QT syndrome: prospective longitudinal study of 328 families. Circulation. 1991;84:1136-1144.
Keating M, Atkinson D, Dunn C, Timothy K, Vincent GM, Leppert M. Linkage of a cardiac arrhythmia, the long QT syndrome, and the Harvey ras-1 gene. Science. 1991;252:704-706.
Benhorin J, Kalman YM, Medina A, Towbin J, Rave-Harel N, Dyer TD, Blangero J, MacCluer JW, Kerem B. Evidence for genetic heterogeneity in the long QT syndrome. Science. 1993;260:1960-1962.
Towbin JA, Li H, Taggart RT, Lehmann MH, Schwartz PJ, Satler CA, Ayyagari R, Robinson JL, Moss AJ, Hejtmancik JF. Evidence of genetic heterogeneity in Romano-Ward long QT syndrome: analysis of 23 families. Circulation. 1994;90:2635-2644.
Benhorin J, Merri M, Alberti M, Locati E, Moss AJ, Hall WJ, Cui L. Long QT syndrome: new electrocardiographic characteristics. Circulation. 1990;82:521-527.
Moss AJ, Robinson J. Clinical features of the idiopathic long QT syndrome. Circulation. 1992;85(suppl I):I-140-I-144.
Bazett HC. An analysis of the time relations of electrocardiograms. Heart. 1920;7:353-367.
SAS/STAT User’s Guide: Version 6. 4th ed. Cary, NC: SAS Institute; 1990.
Moss AJ, Benhorin J. QT interval prolongation: basic considerations and clinical consequences. In: Braunwald E, ed. Heart Disease Updates. Spring 1993.
West J, Patton D, Scheuer T, Wang Y, Goldin A, Catterall W. A cluster of hydrophobic amino acid residues required for fast Na+-channel inactivation. Proc Natl Acad Sci U S A. 1992;89:10910-10914.