Electrocardiographic Features in Andersen-Tawil Syndrome Patients With KCNJ2 Mutations
http://www.100md.com
《循环学杂志》
the LDS Hospital, Salt Lake City, Utah (L.Z., G.S., G.M.V.)
Cincinnati Children’s Hospital Medical Center, Cincinnati, Ohio (D.W.B., G.A.)
University of Utah, Salt Lake City (L.Z., M.T.F., G.M.V.)
Howard Hughes Medical Institute, University of California, San Francisco (L.J.P., Y.H.F.)
University of Rochester School of Medicine and Dentistry, Rochester, NY (R.T.)
University of Pavia and Policlinico S. Matteo, IRCCS, Pavia, Italy (P.J.S.)
Vanderbilt University School of Medicine, Nashville, Tenn (A.L.G.)
Shiga University of Medical Science, Seta-Tsukinowa, Japan (M.H.)
Mayo Clinic College of Medicine (M.J.A.), Rochester, Minn.
Abstract
Background— The ECG features of Andersen-Tawil syndrome (ATS) patients with KCNJ2 mutations (ATS1) have not been systematically assessed. This study aimed to define ECG features of KCNJ2 mutation carriers, to determine whether characteristic T-U–wave patterns exist, and to establish whether T-U patterns predict the ATS1 genotype.
Methods and Results— In phase I, evaluation of T-U morphology in ECGs of 39 KCNJ2 mutation carriers identified characteristic T-U patterns: prolonged terminal T downslope, wide T-U junction, and biphasic and enlarged U waves. In phase II, ATS1 genotype prediction by T-U pattern was evaluated in the next 147 ECGs (57 other KCNJ2 mutation carriers, 61 unaffected family members, and 29 ATS patients without KCNJ2 mutations), with a sensitivity of 84% and specificity of 97%. Characteristic T-U patterns were present in 91% (87/96), in whom an enlarged U wave was predominant (73%). In phase III, QTc, QUc, and T- and U-wave duration/amplitude were compared in the 96 ATS1, 29 non-KCNJ2 ATS, and 75 normal subjects. In ATS1 patients, QUc, U-wave duration and amplitude, and QTc were all increased (P<0.001), but median QTc and interquartile range (IQR) were just 440 ms (IQR, 28 ms) compared with 420 ms (IQR, 20 ms) in normal subjects and 425 ms (IQR, 48 ms) in ATS non-KCNJ2 patients.
Conclusions— In ATS1 patients, gene-specific T-U–wave patterns resulting from decreased IK1 owing to KCNJ2 mutations can aid diagnosis and direct genotyping. The normal QTc, distinct ECG, and other clinical features distinguish ATS1 from long-QT syndrome, and it is best designated as ATS1 rather than LQT7.
Key Words: arrhythmia ; electrocardiography ; genetics
Introduction
Andersen-Tawil syndrome (ATS) is a heterogeneous, autosomal dominant genetic or sporadic disorder resulting in periodic paralysis, cardiac arrhythmias, and dysmorphic features.1–4 One gene, KCNJ2, has been identified so far. This genotype has been labeled ATS1, designating type 1 ATS, and >20 mutations have been reported.5–11 ATS1 has also been proposed as LQT7,6 but there has been some uncertainty about this designation because some previously published tracings appeared to include the U wave in the QT measurement.
KCNJ2 encodes the inward rectifier Kir2.1, a member of the Kir2.x subfamily, which is the critical subunit of cardiac IK1, the inward potassium rectifier current.12 Most KCNJ2 mutations cause loss of function and dominant-negative suppression of the Kir2.1 channel function.5–11 Cardiac IK1 exists as a homotetramer of KCNJ2-encoded subunits, but heterotetrameric assembly of a number of different Kir2 subfamily members is possible.12–14 Consistent with this idea, some KCNJ2 mutants cause dominant-negative suppression of other Kir2.x channels such as Kir2.2 and Kir2.3.15
In the heart, IK1 plays an important role in stabilizing resting potential and determining the shape of the terminal portion of the cardiac action potential.12 During the terminal phase of repolarization and during diastole, IK1 dominates membrane conductance and modulates cell excitability.12–14 We hypothesized, therefore, that reduced IK1 resulting from KCNJ2 mutations might uniquely alter late cardiac repolarization and result in both distinctive T-U–wave morphology on ECG and an increased propensity to ventricular arrhythmias. Therefore, this multicenter, multinational study aimed to evaluate the spectrum of the ECG phenotype of ATS1 patients and to determine whether characteristic ECG patterns exist and, if so, whether they predict the ATS1 genotype.
Methods
Study Sample
In accordance with guidelines of the Institutional Review boards, all participants consented, and personal identifiers were removed to protect their identities throughout the study.
Ninety-six ATS1 patients from 33 unrelated kindreds carrying 24 distinct KCNJ2 mutations throughout the Kir2.1 channel were represented in this study (Table 1). A de novo mutation was found in 27% (9 of 33) of kindreds. ATS1 patients were from 12 countries and represented diverse racial/ethnic groups (80 whites, 11 Hispanics, 4 Asians, and 1 Pacific Islander). The clinical manifestations and functional effects of KCNJ2 mutations were previously described in detail5-8,10,11 and are briefly summarized in Tables 1 and 2. The study consisted of the evaluation of ECGs on these 96 ATS1 patients and those of several referent groups, provided by the authors.
ECGs were collected in 3 phases. In phase I, the first ECGs collected were used as a training set for recognition and characterization of ECG patterns. They were from 39 ATS1 subjects of 26 kindreds, hosting 19 distinct KCNJ2 mutations that covered the entire Kir2.1 channel. In phase II, a test set, which included 147 ECGs, was used to examine whether the characteristic ECG T-U patterns found in the phase I study were genotype specific and could be used to predict the ATS1 genotype. The test set included the next 57 ATS1 subjects (7 KCNJ2 mutations), 61 unaffected members (negative for KCNJ2 mutations) from the ATS1 families, and 29 phenotypically positive ATS subjects from 13 kindreds who were negative for KCNJ2 mutations (ATS non-KCNJ2). The diagnosis of ATS in these 29 patients was based on a history of periodic paralysis (19 of 29), presence of dysmorphic features (19 of 29), and presence of ventricular arrhythmias (8 of 29). In phase III, a third set included ECGs from 75 normal subjects (66 whites, 6 Hispanics, and 3 and Asians) selected from the LDS Hospital LQTS database. Comparisons of quantitative ECG variables were made between the 96 ATS1 patients, 29 ATS non-KCNJ2 patients, and normal subjects. Selection of normal subjects required a normal physical examination and a normal ECG, a genotype negative for known LQTS genes, the presence of a visible U wave, and race, age, and sex similar to those of ATS1 subjects.
Phase I: T-U–Wave Pattern Recognition
The morphology of QRS and T and U waves of 39 ATS1 ECGs was carefully examined in the 39 ATS1 patient ECGs. In a manner similar to pattern identification for bundle-branch blocks, myocardial infarction, etc, characteristic T-U–wave patterns not seen in the 3 most common forms of inherited long-QT syndrome (LQT1 through LQT3)16 were identified.
Phase II: Genotype Prediction by T-U–Wave Patterns
To determine whether the characteristic T-U–wave patterns found in phase I were genotype specific and could be used to predict the ATS1 genotype, the test set of 147 (57+61+29) ECGs were read by 1 author (L.Z.) who was blinded to genotype, kindred, age, sex, and other clinical information. By visual inspection of ECG patterns, ATS1 genotype was predicted if 1 of the characteristic T-U–wave pattern(s) identified in phase I were present in the ECG.
Phase III: Quantitative ECG Analysis
In phase III, the ECG variables R-R, PR, QRS, QT, QU, and Tp-Up (T peak to the U peak) intervals and T- and U-wave duration and amplitude were compared in 96 ATS1 patients, 29 ATS non-KCNJ2 patients, and 75 normal referents. Measurements of T and U waves were made in the lead with the highest amplitudes, usually in V2 or V3.17 QT interval was defined from the onset of QRS to the end of the T wave, at the point at which the steepest T downslope crossed the isoelectric line.17,18 The U wave was defined as an early diastolic deflection after the end of the T wave.19 QU interval (onset of QRS to the end of the U wave), U-wave duration (end of the T wave to the end of the U wave),17 and U-wave amplitude (magnitude of positive deflection) were also measured. A U wave was considered enlarged if its amplitude was 0.15 mV and its duration was 210 ms. For comparison among subjects with different heart rates, QT and QU intervals were corrected (QTc and QUc) by use of Bazett’s formula and were averaged from 2 to 3 consecutive sinus beats.
The prevalence of other ECG abnormalities, including ventricular arrhythmia, was also assessed.
Statistical Analysis
The accuracy of genotype prediction (phase II) was measured by sensitivity, specificity, and positive and negative predictive values. Sensitivity was defined as the percent of subjects who carried a KCNJ2 mutation and were scored "yes" or positive for the presence of typical ECG patterns. Specificity was defined as the percent of subjects who did not carry KCNJ2 mutations and were scored "no" or negative by ECG patterns. Positive predictive value was defined as the percent of those assigned positive by ECG patterns who in fact did have the disease. Negative predictive value was defined as the percent of those assigned negative by ECG patterns who in fact did not have the disease.
For the quantitative measures, values are shown as median and interquartile range (IQR). ECG variables were compared pairwise between 3 groups (96 ATS1, 29 ATS non-KCNJ2, and 75 normal subjects) using permutation tests that randomly assigned kindreds between the 2 groups (200 000 replicates using S-PLUS 6.2, Insightful Corps) to test whether the observed difference could be due to chance. A 2-sided value of P<0.05 was considered significant.
Results
Clinical Characteristics of 96 KCNJ2 Mutation Carriers
The clinical characteristics of the ATS1 patients are summarized in Table 1 by each of the KCNJ2 mutations and in Table 2 by the mutation locations in the Kir2.1 channel. Overall, 74% (71 of 96) of KCNJ2 mutation carriers showed dysmorphic features and other developmental defects. Fifty percent (48 of 96) had a history of periodic paralysis; 40% (38 of 96) had both dysmorphic features and periodic paralysis. On baseline ECGs, 41% (39 of 96) showed cardiac arrhythmias, but only 5% (2 of 39) complained of palpitations. Eleven percent had syncope (n=10) and/or cardiac arrest (n=3); the cardiac arrest patients had been treated with implantable cardioverter-defibrillators. A family history of sudden death was present in 4 of 33 kindreds. Although not statistically significant, ATS1 patients with C-terminal mutations (n=31) seemed to have a higher prevalence of periodic paralysis and dysmorphic features (Table 2). The apparent trend may be due to the fact that in this group families tended to be small and 61% (19 of 31) of patients were probands, which have more severe clinical features than other affected family members, which is why they are the first identified.
Characteristic T-U–Wave Patterns and Their Usefulness in Genotype Prediction
One or more characteristic T-U–wave patterns (Figure 2A and 2B) were identified in 91% (87 of 96) of KCNJ2 mutation carriers but were absent in the 29 ATS non-KCNJ2 patients. Abnormal T-U–wave patterns were manifested as a prolonged terminal portion of T-wave downslope in 70% (67 of 96), a wide T-U junction17 in 43% (41 of 96), biphasic U waves in 16% (15 of 96), and enlarged U waves in 73% (70 of 96). A high predictive accuracy of characteristic T-U–wave patterns for ATS1 genotype prediction was demonstrated by the sensitivity, specificity, and positive and negative predictive values of 84%, 97%, 94%, and 91%, respectively (Table 3). Nonspecific T-wave abnormalities such as low amplitude and inverted T waves were present in 12% and 7% of patients, respectively.
Quantitative Measures in ATS1, ATS Non-KCNJ2 Patients, and Normal Subjects
Quantitative ECG analyses are shown in Tables 2 and 4. In ATS1 patients, prominent findings included prolonged QUc and Tp-Up intervals, U-wave duration, and increased U-wave amplitude (all P<0.001 versus ATS non-KCNJ2 patients or ATS1 versus normal subjects). Although statistically significant (Table 4), the median QTc in ATS1 was longer by 20 ms, at 440 ms (IQR, 28 ms) versus 420 ms (IQR, 20 ms) in normal subjects, and the QTc distribution largely (87%) overlapped with normal subjects (Figure 3; 64% had a QTc <440 ms whereas only 17% had a QTc >460 ms). QTc was not significantly different between ATS1 and ATS non-KCNJ2 patients (Table 4). T-wave duration in ATS1 was longer, 220 ms (IQR, 40 ms) versus 190 ms (IQR, 50 ms) in ATS non-KCNJ2 and 200 ms (IQR, 20 ms) in normal subjects (all P<0.01). These quantitative changes are consistent with the morphology features of enlarged U waves, prolonged terminal T-wave downslope, and a wide T-U junction in ATS1 patients. No significant difference was found in any measurements between ATS non-KCNJ2 and normal groups.
Ventricular Arrhythmias in ATS1 Patients
Frequent premature ventricular contractions (PVCs) were present in 41% (39 of 96) of ATS1 patients, with 59% (23 of 39) in bigeminy. Nonsustained polymorphic ventricular tachycardia (VT) was found in 23% (22 of 96) of patients; 68% (15 of 22) were bidirectional VT (Figure 2C). Torsade de pointes was documented in 3% (3 of 96) of patients.
Conduction Abnormalities
Twenty-three percent (22 of 96) of ATS1 patients (age, 34±21 years; 13 female) in 12 kindreds with 10 distinct mutations showed conduction abnormalities. First-degree AV block was seen in 7, right bundle-branch block in 9 (incomplete in 4), left bundle-branch block in 2, bifascicular block in 2, and nonspecific intraventricular conduction delay in 2 patients. In contrast, only 2% of normal subjects showed conduction abnormality, expressed as incomplete right bundle-branch block.
Discussion
This study tested the hypothesis that in ATS1 reduced IK1 as a result of mutations in KCNJ2 would yield unique T-U–wave patterns that distinguish KCNJ2 mutation carriers from noncarriers. Indeed, we observed genotype-specific T-U–wave patterns in the KCNJ2 mutation carriers that included prolonged terminal T-wave downslope, a wide T-U junction, and biphasic and enlarged U waves. These T-U–wave patterns were not seen in the ATS non-KCNJ2 patients or normal subjects. Quantitative ECG analysis supported the qualitative T-U–wave observations. KCNJ2 mutations caused lengthening of the T-wave duration by prolongation of the final downslope of the T wave; a longer Tp-Up interval yielding a separation of the T and U waves, thereby creating a wide T-U junction (Figure 2); and markedly greater U-wave duration and amplitude compared with normal control subjects and ATS non-KCNJ2 patients. For the ATS non-KCNJ2 patients, prospective ECG characterization will await the identification of their disease-causing genes.
Previously, we demonstrated characteristic ST-T–wave patterns in LQT1, LQT2, and LQT3 that could be used for genotype prediction in LQTS.16 These patterns allow a cost-effective strategy for genetic testing in that the gene predicted by ECG patterns can be the first one sequenced, with a high expectation that the selected gene will show the mutation. Similarly, characteristic T-U–wave patterns identified in this study predicted the ATS1 genotype with excellent sensitivity and specificity. Thus, mutations in cardiac ion channel genes are often associated with gene-specific ECG patterns, which may facilitate improved clinical recognition and cost-effective genetic testing for diagnosis of the appropriate arrhythmia syndrome.
Prior reports have proposed ATS1 to be LQT76 on the basis of apparently long QTc intervals and associated arrhythmias. We found the degree of QTc prolongation in ATS1 to be very modest (20 ms) and the median QTc value of 440 ms to be within the normal range. In addition, the distribution of QTc intervals in ATS1 patients largely (87%) overlapped with normal subjects, and only 17% of ATS1 had a QTc >460 ms (Figure 3). Furthermore, the U wave is markedly abnormal in ATS1 yet normal in LQTS. Consequently, the ATS1 ECG phenotype distinctly differs from LQT1 through LQT3 in which the QT prolongation is the primary ECG manifestation. We suggest that the reason for reports of more pronounced QTc prolongation in previous studies is that the U wave was often included in the QT measurement. Furthermore, our large cohort of ATS1 subjects demonstrated other fundamental ECG and clinical differences between ATS1 and LQTS. In ATS1, patients generally seek medical attention because of periodic paralysis1–4,7,8,11 rather than because of the ventricular arrhythmia, consisting of frequent PVCs, mainly in bigeminy, and nonsustained polymorphic, often bidirectional, VT (Tables 1 and 2 and Figure 2), which are usually asymptomatic. Cardiac conduction abnormalities, infrequently seen in LQTS patients, were observed in 23% of ATS1 patients. Torsade de pointes, cardiac arrest, and sudden death are infrequently reported in ATS patients.6,9,11 In LQTS, in contrast, QT prolongation, syncope, and sudden cardiac death resulting from torsade de pointes and ventricular fibrillation, often initiated by gene-specific triggers,20 are the primary manifestations, and PVCs and asymptomatic VT are very uncommon. On the basis of these fundamental differences, even though ATS1 and LQT1 through LQT3 are "channelopathies," it is apparent that they are quite different diseases with different clinical manifestations, ECG characteristics, and outcomes. Hence, we conclude that ATS1 is not a subtype of long-QT syndrome and recommend the annotation of KCNJ2-positive ATS individuals as ATS1 rather than LQT7.
The primary effect of decreased IK1 resulting from KCNJ2 mutations in ATS1 is on the U-wave morphology (Tables 1, 2 and 4 and Figure 2). The enlarged U wave often appeared to be associated with frequent PVCs and nonsustained VT, implying a relationship between the U wave and increased vulnerability to ventricular arrhythmia. A role for U waves in arrhythmogenesis has previously been proposed for right ventricular outflow tract VT and organic heart disease.21–23 However, arrhythmia risk in these settings was related to postextrasystolic augmentation of the U wave. In contrast, the enlarged U waves in ATS1 subjects were observed during sinus rhythm and were distinct from the preceding T wave.
The origin of the U wave is still unknown. KCNJ2 mutation–induced reduction in IK1 could augment the U wave by several possible mechanisms. First, reduced IK1 could increase spatial heterogeneity by influencing action potential duration to a greater degree in 1 region over another. In this scenario, the U wave would represent delayed repolarization of a specific region(s) of the myocardium. Indeed, IK1 blockade produces marked lengthening of action potential duration in rabbit and feline Purkinje fibers, with less prolongation in ventricular myocytes.24,25 Such spatial dispersion of repolarization has been proposed to contribute to the genesis of the U wave.26,27 There are problems with the Purkinje hypothesis, however. Purkinje tissue is a small mass of conduction system that fails to register on the ECG in most animal model experiments. In normal human subjects, the median U-wave duration is 170 ms (Table 2), which is greater than the difference of functional refractory periods of the His-Purkinje system and ventricular myocytes.19 Moreover, if the U wave resulted from delayed Purkinje fiber or ventricular myocyte repolarization, it should appear as a prolonged or notched T wave28,29 rather than as a separate deflection arising from the isoelectric baseline.19
U-wave timing depends heavily on mechanical events. The mechanical interval of ventricular relaxation (isovolumic and early diastolic periods) is estimated to be 200 ms, which temporally approximates the duration of the U wave.19 The duration of ventricular action potentials and the duration of contraction in the heart are on the same order of magnitude.30 In general, moderately large passive length changes in ventricular muscle can induce changes in membrane potential, so-called mechanoelectric feedback.30 Experimental evidence indicates that a variety of channels are activated by changes in cell volume and/or cell stretch, namely the mechanosensitive channels (Ims). They include chloride (Cl–) channels, the delayed rectifier K+ channel (IK), the ATP-sensitive K+ channel (IK,atp), and the Na-KATPase and Na-Ca (Inaca) exchanger. The outward component of IK1 plays an important role in determining the mechanosensitive conductance threshold in the various models. IK1 also dominates membrane conductance during diastolic resting potential. In ventricular myocytes, Ims currents are closely mirrored and counteracted by IK1.31 This ion channel balance31 may contribute to the genesis of the U wave, and one might speculate that decreased IK1 could result in an augmentation of Ims and thereby an enlarged U wave, as seen in ATS1 patients. Detailed analysis of ionic currents underlying the U wave under normal circumstances and in the setting of reduced IK1 may require development of novel pharmacological tools and/or transgenic manipulation of cardiac ion channel genes.
Conclusions
Mutations of KCNJ2 cause the ATS1 genotype and result in gene-specific T-U–wave abnormalities that can be used to distinguish carriers of KCNJ2 mutations from normal subjects, ATS patients without KCNJ2 mutations, and LQTS1 through LQTS3 patients. They can also be used to direct cost-effective strategies for KCNJ2 genetic testing.
The median QTc interval is within the normal range in ATS1. The QTc prolongation reported in earlier studies on ATS1 was due to inclusion of the U wave in the QT measurement. The normal median QTc and other cardiac and noncardiac manifestations differentiate ATS1 from LQTS. Thus, ATS1 is unlike LQTS and should be designated as ATS1 rather than LQT7.
The observations reported in this study also highlight the importance of cardiac IK1 in the terminal repolarization and diastolic phases of cardiac potentials.
Acknowledgments
This work was supported by grant HD39946 from the Muscular Dystrophy Association and a Sandler Neurogenetics grant. We thank Professor Kenneth W. Spitzer for his insight into IK1 physiology, Katherine W. Timothy and Judy L. Jensen for their data contributions, and Diana L. Handrahan for statistical assistance.
References
Andersen E, Krasilnikoff PA, Overvad H. Intermittent muscular weakness, extrasystoles, and multiple developmental anomalies: a new syndrome; Acta Paediatr Scand. 1971; 60: 559–564.
Tawil R, Ptacek LJ, Pavlakis SG, DeVivo DC, Penn AS, Ozdemir C, Griggs RC. Andersen’s syndrome: potassium-sensitive periodic paralysis, ventricular ectopy, and dysmorphic features. Ann Neurol. 1994; 35: 326–330.
Sansone V, Griggs RC, Meola G, Ptacek LJ, Barohn R, Iannaccone S, Bryan W, Baker N, Janas SJ, Scott W, Ririe D, Tawil R. Andersen’s syndrome: a distinct periodic paralysis. Ann Neurol. 1997; 42: 305–312.
Canun S, Perez N, Beirana LG. Andersen syndrome autosomal dominant in three generations. Am J Med Genet. 1999; 85: 147–156.
Plaster NM, Tawil R, Tristani-Firouzi M, Canun S, Bendahhou S, Tsunoda A, Donaldson MR, Iannaccone ST, Brunt E, Barohn R, Clark J, Deymeer F, George AL Jr, Fish FA, Hahn A, Nitu A, Ozdemir C, Serdaroglu P, Subramony SH, Wolfe G, Fu YH, Ptacek LJ. Mutations in Kir2.1 cause the developmental and episodic electrical phenotypes of Andersen’s syndrome. Cell. 2001; 105: 511–519.
Tristani-Firouzi M, Jensen JL, Donaldson MR, Sansone V, Meola G, Hahn A, Bendahhou S, Kwiecinski H, Fidzianska A, Plaster N, Fu YH, Ptacek LJ, Tawil R. Functional and clinical characterization of KCNJ2 mutations associated with LQT7 (Andersen syndrome). J Clin Invest. 2002; 110: 381–388.
Andelfinger G, Tapper AR, Welch RC, Vanoye CG, George AL Jr, Benson DW. KCNJ2 mutation results in Andersen syndrome with sex-specific cardiac and skeletal muscle phenotypes. Am J Hum Genet. 2002; 71: 663–668.
Ai T, Fujiwara Y, Tsuji K, Otani H, Nakano S, Kubo Y, Horie M. Novel KCNJ2 mutation in familial periodic paralysis with ventricular dysrhythmia. Circulation. 2002; 105: 2592–2594.
Donaldson MR, Jensen JL, Tristani-Firouzi M, Tawil R, Bendahhou S, Suarez WA, Cobo AM, Poza JJ, Behr E, Wagstaff J, Szepetowski P, Pereira S, Mozaffar T, Escolar DM, Fu YH, Ptacek LJ. PIP2 binding residues of Kir2.1 are common targets of mutations causing Andersen syndrome. Neurology. 2003; 60: 1811–1816.
Bendahhou S, Donaldson MR, Plaster NM, Tristani-Firouzi M, Fu YH, Ptacek LJ. Defective potassium channel Kir2.1 trafficking underlies Andersen-Tawil syndrome. J Biol Chem. 2003; 278: 51779–51785.
Chun TU, Epstein MR, Dick M II, Andelfinger G, Ballester L, Vanoye CG, George AL Jr, Benson DW. Polymorphic ventricular tachycardia and KCNJ2 mutations. Heart Rhythm. 2004; 1: 235–241.
Lopatin AN, Nichols CG. Inward rectifiers in the heart: an update on I(K1). J Mol Cell Cardiol. 2001; 33: 625–638.
Wang Z, Yue L, White M, Pelletier G, Nattel S. Differential distribution of inward rectifier potassium channel transcripts in human atrium versus ventricle. Circulation. 1998; 98: 2422–2428.
Liu GX, Derst C, Schlichthorl G, Heinen S, Seebohm G, Bruggemann A, Kummer W, Veh RW, Daut J, Preisig-Muller R. Comparison of cloned Kir2 channels with native inward rectifier K+ channels from guinea-pig cardiomyocytes. J Physiol. 2001; 532: 115–126.
Preisig-Muller R, Schlichthorl G, Goerge T, Heinen S, Bruggemann A, Rajan S, Derst C, Veh RW, Daut J. Heteromerization of Kir2.x potassium channels contributes to the phenotype of Andersen’s syndrome. Proc Natl Acad Sci U S A. 2002; 99: 7774–7779.
Zhang L, Timothy KW, Vincent GM, Lehmann MH, Fox J, Giuli LC, Shen J, Splawski I, Priori SG, Compton SJ, Yanowitz F, Benhorin J, Moss AJ, Schwartz PJ, Robinson JL, Wang Q, Zareba W, Keating MT, Towbin JA, Napolitano C, Medina A. The spectrum of ST-T wave patterns and repolarization parameters in congenital long QT syndrome: ECG findings identify genotype. Circulation. 2000; 102: 2849–2855.
Lepeschkin E, Surawicz B. The measurement of the QT interval of the electrocardiogram. Circulation. 1952; 6: 378–388.
Vincent GM, Timothy KW, Leppert M, Keating M. The spectrum of symptoms and QT intervals in the carriers of the gene for the long-QT syndrome. N Engl J Med. 1992; 327: 846–852.
Surawicz B. U wave: facts, hypotheses, misconceptions, and misnomers. J Cardiovasc Electrophysiol. 1998; 9: 1117–1128.
Schwartz PJ, Priori SG, Spazzolini C, Moss AJ, Vincent GM, Napolitano C, Denjoy I, Guicheney P, Breithardt G, Keating MT, Towbin JA, Beggs AH, Brink P, Wilde AA, Toivonen L, Zareba W, Robinson JL, Timothy KW, Corfield V, Wattanasirichaigoon D, Corbett C, Haverkamp W, Schulze-Bahr E, Lehmann MH, Schwartz K, Coumel P, Bloise R. Genotype-phenotype correlation in the long-QT syndrome: gene-specific triggers for life-threatening arrhythmias. Circulation. 2001; 103: 89–95.
Nakagawa M, Ooie T, Hara M, Ichinose M, Nobe S, Yonemochi H, Saikawa T. Dynamics of T-U wave in patients with idiopathic ventricular tachycardia originating from the right ventricular outflow tract. Pacing Clin Electrophysiol. 2004; 27: 148–155.
Viskin S, Zeltser D, Antzelevitch C. When U say "U waves," what do U mean; Pacing Clin Electrophysiol. 2004; 27: 145–147.
Viskin S, Heller K, Barron HV, Kitzis I, Hamdan M, Olgin JE, Wong MJ, Grant SE, Lesh MD. Postextrasystolic U wave augmentation, a new marker of increased arrhythmic risk in patients without the long QT syndrome. J Am Coll Cardiol. 1996; 28: 1746–1752.
Cordeiro JM, Spitzer KW, Giles WR. Repolarizing K+ currents in rabbit heart Purkinje cells. J Physiol. 1998; 508 (pt 3): 811–823.
Sanchez-Chapula JA, Salinas-Stefanon E, Torres-Jacome J, Benavides-Haro DE, Navarro-Polanco RA. Blockade of currents by the antimalarial drug chloroquine in feline ventricular myocytes. J Pharmacol Exp Ther. 2001; 297: 437–445.
Hoffman BF, Cranefield PF. Electrophysiology of the Heart. New York, NY: McGraw-Hill; 1960: 201–202.
Watanabe Y. Purkinje repolarization as a possible cause of the U wave in the electrocardiogram. Circulation. 1975; 51: 1030–1037.
Shimizu W, Antzelevitch C. Sodium channel block with mexiletine is effective in reducing dispersion of repolarization and preventing torsade de pointes in LQT2 and LQT3 models of the long-QT syndrome. Circulation. 1997; 96: 2038–2047.
Yan GX, Antzelevitch C. Cellular basis for the normal T wave and the electrocardiographic manifestations of the long-QT syndrome. Circulation. 1999; 100: 1660–1666.
Lab MJ. Contraction-excitation feedback in myocardium physiological basis and clinical relevance. Circ Res. 1982; 50: 757–766.
Rice JJ, Winslow RL, Dekanski J, McVeigh E. Model studies of the role of mechano-sensitive currents in the generation of cardiac arrhythmias. J Theor Biol. 1998; 190: 295–312.(Li Zhang, MD; D. Woodrow )
Cincinnati Children’s Hospital Medical Center, Cincinnati, Ohio (D.W.B., G.A.)
University of Utah, Salt Lake City (L.Z., M.T.F., G.M.V.)
Howard Hughes Medical Institute, University of California, San Francisco (L.J.P., Y.H.F.)
University of Rochester School of Medicine and Dentistry, Rochester, NY (R.T.)
University of Pavia and Policlinico S. Matteo, IRCCS, Pavia, Italy (P.J.S.)
Vanderbilt University School of Medicine, Nashville, Tenn (A.L.G.)
Shiga University of Medical Science, Seta-Tsukinowa, Japan (M.H.)
Mayo Clinic College of Medicine (M.J.A.), Rochester, Minn.
Abstract
Background— The ECG features of Andersen-Tawil syndrome (ATS) patients with KCNJ2 mutations (ATS1) have not been systematically assessed. This study aimed to define ECG features of KCNJ2 mutation carriers, to determine whether characteristic T-U–wave patterns exist, and to establish whether T-U patterns predict the ATS1 genotype.
Methods and Results— In phase I, evaluation of T-U morphology in ECGs of 39 KCNJ2 mutation carriers identified characteristic T-U patterns: prolonged terminal T downslope, wide T-U junction, and biphasic and enlarged U waves. In phase II, ATS1 genotype prediction by T-U pattern was evaluated in the next 147 ECGs (57 other KCNJ2 mutation carriers, 61 unaffected family members, and 29 ATS patients without KCNJ2 mutations), with a sensitivity of 84% and specificity of 97%. Characteristic T-U patterns were present in 91% (87/96), in whom an enlarged U wave was predominant (73%). In phase III, QTc, QUc, and T- and U-wave duration/amplitude were compared in the 96 ATS1, 29 non-KCNJ2 ATS, and 75 normal subjects. In ATS1 patients, QUc, U-wave duration and amplitude, and QTc were all increased (P<0.001), but median QTc and interquartile range (IQR) were just 440 ms (IQR, 28 ms) compared with 420 ms (IQR, 20 ms) in normal subjects and 425 ms (IQR, 48 ms) in ATS non-KCNJ2 patients.
Conclusions— In ATS1 patients, gene-specific T-U–wave patterns resulting from decreased IK1 owing to KCNJ2 mutations can aid diagnosis and direct genotyping. The normal QTc, distinct ECG, and other clinical features distinguish ATS1 from long-QT syndrome, and it is best designated as ATS1 rather than LQT7.
Key Words: arrhythmia ; electrocardiography ; genetics
Introduction
Andersen-Tawil syndrome (ATS) is a heterogeneous, autosomal dominant genetic or sporadic disorder resulting in periodic paralysis, cardiac arrhythmias, and dysmorphic features.1–4 One gene, KCNJ2, has been identified so far. This genotype has been labeled ATS1, designating type 1 ATS, and >20 mutations have been reported.5–11 ATS1 has also been proposed as LQT7,6 but there has been some uncertainty about this designation because some previously published tracings appeared to include the U wave in the QT measurement.
KCNJ2 encodes the inward rectifier Kir2.1, a member of the Kir2.x subfamily, which is the critical subunit of cardiac IK1, the inward potassium rectifier current.12 Most KCNJ2 mutations cause loss of function and dominant-negative suppression of the Kir2.1 channel function.5–11 Cardiac IK1 exists as a homotetramer of KCNJ2-encoded subunits, but heterotetrameric assembly of a number of different Kir2 subfamily members is possible.12–14 Consistent with this idea, some KCNJ2 mutants cause dominant-negative suppression of other Kir2.x channels such as Kir2.2 and Kir2.3.15
In the heart, IK1 plays an important role in stabilizing resting potential and determining the shape of the terminal portion of the cardiac action potential.12 During the terminal phase of repolarization and during diastole, IK1 dominates membrane conductance and modulates cell excitability.12–14 We hypothesized, therefore, that reduced IK1 resulting from KCNJ2 mutations might uniquely alter late cardiac repolarization and result in both distinctive T-U–wave morphology on ECG and an increased propensity to ventricular arrhythmias. Therefore, this multicenter, multinational study aimed to evaluate the spectrum of the ECG phenotype of ATS1 patients and to determine whether characteristic ECG patterns exist and, if so, whether they predict the ATS1 genotype.
Methods
Study Sample
In accordance with guidelines of the Institutional Review boards, all participants consented, and personal identifiers were removed to protect their identities throughout the study.
Ninety-six ATS1 patients from 33 unrelated kindreds carrying 24 distinct KCNJ2 mutations throughout the Kir2.1 channel were represented in this study (Table 1). A de novo mutation was found in 27% (9 of 33) of kindreds. ATS1 patients were from 12 countries and represented diverse racial/ethnic groups (80 whites, 11 Hispanics, 4 Asians, and 1 Pacific Islander). The clinical manifestations and functional effects of KCNJ2 mutations were previously described in detail5-8,10,11 and are briefly summarized in Tables 1 and 2. The study consisted of the evaluation of ECGs on these 96 ATS1 patients and those of several referent groups, provided by the authors.
ECGs were collected in 3 phases. In phase I, the first ECGs collected were used as a training set for recognition and characterization of ECG patterns. They were from 39 ATS1 subjects of 26 kindreds, hosting 19 distinct KCNJ2 mutations that covered the entire Kir2.1 channel. In phase II, a test set, which included 147 ECGs, was used to examine whether the characteristic ECG T-U patterns found in the phase I study were genotype specific and could be used to predict the ATS1 genotype. The test set included the next 57 ATS1 subjects (7 KCNJ2 mutations), 61 unaffected members (negative for KCNJ2 mutations) from the ATS1 families, and 29 phenotypically positive ATS subjects from 13 kindreds who were negative for KCNJ2 mutations (ATS non-KCNJ2). The diagnosis of ATS in these 29 patients was based on a history of periodic paralysis (19 of 29), presence of dysmorphic features (19 of 29), and presence of ventricular arrhythmias (8 of 29). In phase III, a third set included ECGs from 75 normal subjects (66 whites, 6 Hispanics, and 3 and Asians) selected from the LDS Hospital LQTS database. Comparisons of quantitative ECG variables were made between the 96 ATS1 patients, 29 ATS non-KCNJ2 patients, and normal subjects. Selection of normal subjects required a normal physical examination and a normal ECG, a genotype negative for known LQTS genes, the presence of a visible U wave, and race, age, and sex similar to those of ATS1 subjects.
Phase I: T-U–Wave Pattern Recognition
The morphology of QRS and T and U waves of 39 ATS1 ECGs was carefully examined in the 39 ATS1 patient ECGs. In a manner similar to pattern identification for bundle-branch blocks, myocardial infarction, etc, characteristic T-U–wave patterns not seen in the 3 most common forms of inherited long-QT syndrome (LQT1 through LQT3)16 were identified.
Phase II: Genotype Prediction by T-U–Wave Patterns
To determine whether the characteristic T-U–wave patterns found in phase I were genotype specific and could be used to predict the ATS1 genotype, the test set of 147 (57+61+29) ECGs were read by 1 author (L.Z.) who was blinded to genotype, kindred, age, sex, and other clinical information. By visual inspection of ECG patterns, ATS1 genotype was predicted if 1 of the characteristic T-U–wave pattern(s) identified in phase I were present in the ECG.
Phase III: Quantitative ECG Analysis
In phase III, the ECG variables R-R, PR, QRS, QT, QU, and Tp-Up (T peak to the U peak) intervals and T- and U-wave duration and amplitude were compared in 96 ATS1 patients, 29 ATS non-KCNJ2 patients, and 75 normal referents. Measurements of T and U waves were made in the lead with the highest amplitudes, usually in V2 or V3.17 QT interval was defined from the onset of QRS to the end of the T wave, at the point at which the steepest T downslope crossed the isoelectric line.17,18 The U wave was defined as an early diastolic deflection after the end of the T wave.19 QU interval (onset of QRS to the end of the U wave), U-wave duration (end of the T wave to the end of the U wave),17 and U-wave amplitude (magnitude of positive deflection) were also measured. A U wave was considered enlarged if its amplitude was 0.15 mV and its duration was 210 ms. For comparison among subjects with different heart rates, QT and QU intervals were corrected (QTc and QUc) by use of Bazett’s formula and were averaged from 2 to 3 consecutive sinus beats.
The prevalence of other ECG abnormalities, including ventricular arrhythmia, was also assessed.
Statistical Analysis
The accuracy of genotype prediction (phase II) was measured by sensitivity, specificity, and positive and negative predictive values. Sensitivity was defined as the percent of subjects who carried a KCNJ2 mutation and were scored "yes" or positive for the presence of typical ECG patterns. Specificity was defined as the percent of subjects who did not carry KCNJ2 mutations and were scored "no" or negative by ECG patterns. Positive predictive value was defined as the percent of those assigned positive by ECG patterns who in fact did have the disease. Negative predictive value was defined as the percent of those assigned negative by ECG patterns who in fact did not have the disease.
For the quantitative measures, values are shown as median and interquartile range (IQR). ECG variables were compared pairwise between 3 groups (96 ATS1, 29 ATS non-KCNJ2, and 75 normal subjects) using permutation tests that randomly assigned kindreds between the 2 groups (200 000 replicates using S-PLUS 6.2, Insightful Corps) to test whether the observed difference could be due to chance. A 2-sided value of P<0.05 was considered significant.
Results
Clinical Characteristics of 96 KCNJ2 Mutation Carriers
The clinical characteristics of the ATS1 patients are summarized in Table 1 by each of the KCNJ2 mutations and in Table 2 by the mutation locations in the Kir2.1 channel. Overall, 74% (71 of 96) of KCNJ2 mutation carriers showed dysmorphic features and other developmental defects. Fifty percent (48 of 96) had a history of periodic paralysis; 40% (38 of 96) had both dysmorphic features and periodic paralysis. On baseline ECGs, 41% (39 of 96) showed cardiac arrhythmias, but only 5% (2 of 39) complained of palpitations. Eleven percent had syncope (n=10) and/or cardiac arrest (n=3); the cardiac arrest patients had been treated with implantable cardioverter-defibrillators. A family history of sudden death was present in 4 of 33 kindreds. Although not statistically significant, ATS1 patients with C-terminal mutations (n=31) seemed to have a higher prevalence of periodic paralysis and dysmorphic features (Table 2). The apparent trend may be due to the fact that in this group families tended to be small and 61% (19 of 31) of patients were probands, which have more severe clinical features than other affected family members, which is why they are the first identified.
Characteristic T-U–Wave Patterns and Their Usefulness in Genotype Prediction
One or more characteristic T-U–wave patterns (Figure 2A and 2B) were identified in 91% (87 of 96) of KCNJ2 mutation carriers but were absent in the 29 ATS non-KCNJ2 patients. Abnormal T-U–wave patterns were manifested as a prolonged terminal portion of T-wave downslope in 70% (67 of 96), a wide T-U junction17 in 43% (41 of 96), biphasic U waves in 16% (15 of 96), and enlarged U waves in 73% (70 of 96). A high predictive accuracy of characteristic T-U–wave patterns for ATS1 genotype prediction was demonstrated by the sensitivity, specificity, and positive and negative predictive values of 84%, 97%, 94%, and 91%, respectively (Table 3). Nonspecific T-wave abnormalities such as low amplitude and inverted T waves were present in 12% and 7% of patients, respectively.
Quantitative Measures in ATS1, ATS Non-KCNJ2 Patients, and Normal Subjects
Quantitative ECG analyses are shown in Tables 2 and 4. In ATS1 patients, prominent findings included prolonged QUc and Tp-Up intervals, U-wave duration, and increased U-wave amplitude (all P<0.001 versus ATS non-KCNJ2 patients or ATS1 versus normal subjects). Although statistically significant (Table 4), the median QTc in ATS1 was longer by 20 ms, at 440 ms (IQR, 28 ms) versus 420 ms (IQR, 20 ms) in normal subjects, and the QTc distribution largely (87%) overlapped with normal subjects (Figure 3; 64% had a QTc <440 ms whereas only 17% had a QTc >460 ms). QTc was not significantly different between ATS1 and ATS non-KCNJ2 patients (Table 4). T-wave duration in ATS1 was longer, 220 ms (IQR, 40 ms) versus 190 ms (IQR, 50 ms) in ATS non-KCNJ2 and 200 ms (IQR, 20 ms) in normal subjects (all P<0.01). These quantitative changes are consistent with the morphology features of enlarged U waves, prolonged terminal T-wave downslope, and a wide T-U junction in ATS1 patients. No significant difference was found in any measurements between ATS non-KCNJ2 and normal groups.
Ventricular Arrhythmias in ATS1 Patients
Frequent premature ventricular contractions (PVCs) were present in 41% (39 of 96) of ATS1 patients, with 59% (23 of 39) in bigeminy. Nonsustained polymorphic ventricular tachycardia (VT) was found in 23% (22 of 96) of patients; 68% (15 of 22) were bidirectional VT (Figure 2C). Torsade de pointes was documented in 3% (3 of 96) of patients.
Conduction Abnormalities
Twenty-three percent (22 of 96) of ATS1 patients (age, 34±21 years; 13 female) in 12 kindreds with 10 distinct mutations showed conduction abnormalities. First-degree AV block was seen in 7, right bundle-branch block in 9 (incomplete in 4), left bundle-branch block in 2, bifascicular block in 2, and nonspecific intraventricular conduction delay in 2 patients. In contrast, only 2% of normal subjects showed conduction abnormality, expressed as incomplete right bundle-branch block.
Discussion
This study tested the hypothesis that in ATS1 reduced IK1 as a result of mutations in KCNJ2 would yield unique T-U–wave patterns that distinguish KCNJ2 mutation carriers from noncarriers. Indeed, we observed genotype-specific T-U–wave patterns in the KCNJ2 mutation carriers that included prolonged terminal T-wave downslope, a wide T-U junction, and biphasic and enlarged U waves. These T-U–wave patterns were not seen in the ATS non-KCNJ2 patients or normal subjects. Quantitative ECG analysis supported the qualitative T-U–wave observations. KCNJ2 mutations caused lengthening of the T-wave duration by prolongation of the final downslope of the T wave; a longer Tp-Up interval yielding a separation of the T and U waves, thereby creating a wide T-U junction (Figure 2); and markedly greater U-wave duration and amplitude compared with normal control subjects and ATS non-KCNJ2 patients. For the ATS non-KCNJ2 patients, prospective ECG characterization will await the identification of their disease-causing genes.
Previously, we demonstrated characteristic ST-T–wave patterns in LQT1, LQT2, and LQT3 that could be used for genotype prediction in LQTS.16 These patterns allow a cost-effective strategy for genetic testing in that the gene predicted by ECG patterns can be the first one sequenced, with a high expectation that the selected gene will show the mutation. Similarly, characteristic T-U–wave patterns identified in this study predicted the ATS1 genotype with excellent sensitivity and specificity. Thus, mutations in cardiac ion channel genes are often associated with gene-specific ECG patterns, which may facilitate improved clinical recognition and cost-effective genetic testing for diagnosis of the appropriate arrhythmia syndrome.
Prior reports have proposed ATS1 to be LQT76 on the basis of apparently long QTc intervals and associated arrhythmias. We found the degree of QTc prolongation in ATS1 to be very modest (20 ms) and the median QTc value of 440 ms to be within the normal range. In addition, the distribution of QTc intervals in ATS1 patients largely (87%) overlapped with normal subjects, and only 17% of ATS1 had a QTc >460 ms (Figure 3). Furthermore, the U wave is markedly abnormal in ATS1 yet normal in LQTS. Consequently, the ATS1 ECG phenotype distinctly differs from LQT1 through LQT3 in which the QT prolongation is the primary ECG manifestation. We suggest that the reason for reports of more pronounced QTc prolongation in previous studies is that the U wave was often included in the QT measurement. Furthermore, our large cohort of ATS1 subjects demonstrated other fundamental ECG and clinical differences between ATS1 and LQTS. In ATS1, patients generally seek medical attention because of periodic paralysis1–4,7,8,11 rather than because of the ventricular arrhythmia, consisting of frequent PVCs, mainly in bigeminy, and nonsustained polymorphic, often bidirectional, VT (Tables 1 and 2 and Figure 2), which are usually asymptomatic. Cardiac conduction abnormalities, infrequently seen in LQTS patients, were observed in 23% of ATS1 patients. Torsade de pointes, cardiac arrest, and sudden death are infrequently reported in ATS patients.6,9,11 In LQTS, in contrast, QT prolongation, syncope, and sudden cardiac death resulting from torsade de pointes and ventricular fibrillation, often initiated by gene-specific triggers,20 are the primary manifestations, and PVCs and asymptomatic VT are very uncommon. On the basis of these fundamental differences, even though ATS1 and LQT1 through LQT3 are "channelopathies," it is apparent that they are quite different diseases with different clinical manifestations, ECG characteristics, and outcomes. Hence, we conclude that ATS1 is not a subtype of long-QT syndrome and recommend the annotation of KCNJ2-positive ATS individuals as ATS1 rather than LQT7.
The primary effect of decreased IK1 resulting from KCNJ2 mutations in ATS1 is on the U-wave morphology (Tables 1, 2 and 4 and Figure 2). The enlarged U wave often appeared to be associated with frequent PVCs and nonsustained VT, implying a relationship between the U wave and increased vulnerability to ventricular arrhythmia. A role for U waves in arrhythmogenesis has previously been proposed for right ventricular outflow tract VT and organic heart disease.21–23 However, arrhythmia risk in these settings was related to postextrasystolic augmentation of the U wave. In contrast, the enlarged U waves in ATS1 subjects were observed during sinus rhythm and were distinct from the preceding T wave.
The origin of the U wave is still unknown. KCNJ2 mutation–induced reduction in IK1 could augment the U wave by several possible mechanisms. First, reduced IK1 could increase spatial heterogeneity by influencing action potential duration to a greater degree in 1 region over another. In this scenario, the U wave would represent delayed repolarization of a specific region(s) of the myocardium. Indeed, IK1 blockade produces marked lengthening of action potential duration in rabbit and feline Purkinje fibers, with less prolongation in ventricular myocytes.24,25 Such spatial dispersion of repolarization has been proposed to contribute to the genesis of the U wave.26,27 There are problems with the Purkinje hypothesis, however. Purkinje tissue is a small mass of conduction system that fails to register on the ECG in most animal model experiments. In normal human subjects, the median U-wave duration is 170 ms (Table 2), which is greater than the difference of functional refractory periods of the His-Purkinje system and ventricular myocytes.19 Moreover, if the U wave resulted from delayed Purkinje fiber or ventricular myocyte repolarization, it should appear as a prolonged or notched T wave28,29 rather than as a separate deflection arising from the isoelectric baseline.19
U-wave timing depends heavily on mechanical events. The mechanical interval of ventricular relaxation (isovolumic and early diastolic periods) is estimated to be 200 ms, which temporally approximates the duration of the U wave.19 The duration of ventricular action potentials and the duration of contraction in the heart are on the same order of magnitude.30 In general, moderately large passive length changes in ventricular muscle can induce changes in membrane potential, so-called mechanoelectric feedback.30 Experimental evidence indicates that a variety of channels are activated by changes in cell volume and/or cell stretch, namely the mechanosensitive channels (Ims). They include chloride (Cl–) channels, the delayed rectifier K+ channel (IK), the ATP-sensitive K+ channel (IK,atp), and the Na-KATPase and Na-Ca (Inaca) exchanger. The outward component of IK1 plays an important role in determining the mechanosensitive conductance threshold in the various models. IK1 also dominates membrane conductance during diastolic resting potential. In ventricular myocytes, Ims currents are closely mirrored and counteracted by IK1.31 This ion channel balance31 may contribute to the genesis of the U wave, and one might speculate that decreased IK1 could result in an augmentation of Ims and thereby an enlarged U wave, as seen in ATS1 patients. Detailed analysis of ionic currents underlying the U wave under normal circumstances and in the setting of reduced IK1 may require development of novel pharmacological tools and/or transgenic manipulation of cardiac ion channel genes.
Conclusions
Mutations of KCNJ2 cause the ATS1 genotype and result in gene-specific T-U–wave abnormalities that can be used to distinguish carriers of KCNJ2 mutations from normal subjects, ATS patients without KCNJ2 mutations, and LQTS1 through LQTS3 patients. They can also be used to direct cost-effective strategies for KCNJ2 genetic testing.
The median QTc interval is within the normal range in ATS1. The QTc prolongation reported in earlier studies on ATS1 was due to inclusion of the U wave in the QT measurement. The normal median QTc and other cardiac and noncardiac manifestations differentiate ATS1 from LQTS. Thus, ATS1 is unlike LQTS and should be designated as ATS1 rather than LQT7.
The observations reported in this study also highlight the importance of cardiac IK1 in the terminal repolarization and diastolic phases of cardiac potentials.
Acknowledgments
This work was supported by grant HD39946 from the Muscular Dystrophy Association and a Sandler Neurogenetics grant. We thank Professor Kenneth W. Spitzer for his insight into IK1 physiology, Katherine W. Timothy and Judy L. Jensen for their data contributions, and Diana L. Handrahan for statistical assistance.
References
Andersen E, Krasilnikoff PA, Overvad H. Intermittent muscular weakness, extrasystoles, and multiple developmental anomalies: a new syndrome; Acta Paediatr Scand. 1971; 60: 559–564.
Tawil R, Ptacek LJ, Pavlakis SG, DeVivo DC, Penn AS, Ozdemir C, Griggs RC. Andersen’s syndrome: potassium-sensitive periodic paralysis, ventricular ectopy, and dysmorphic features. Ann Neurol. 1994; 35: 326–330.
Sansone V, Griggs RC, Meola G, Ptacek LJ, Barohn R, Iannaccone S, Bryan W, Baker N, Janas SJ, Scott W, Ririe D, Tawil R. Andersen’s syndrome: a distinct periodic paralysis. Ann Neurol. 1997; 42: 305–312.
Canun S, Perez N, Beirana LG. Andersen syndrome autosomal dominant in three generations. Am J Med Genet. 1999; 85: 147–156.
Plaster NM, Tawil R, Tristani-Firouzi M, Canun S, Bendahhou S, Tsunoda A, Donaldson MR, Iannaccone ST, Brunt E, Barohn R, Clark J, Deymeer F, George AL Jr, Fish FA, Hahn A, Nitu A, Ozdemir C, Serdaroglu P, Subramony SH, Wolfe G, Fu YH, Ptacek LJ. Mutations in Kir2.1 cause the developmental and episodic electrical phenotypes of Andersen’s syndrome. Cell. 2001; 105: 511–519.
Tristani-Firouzi M, Jensen JL, Donaldson MR, Sansone V, Meola G, Hahn A, Bendahhou S, Kwiecinski H, Fidzianska A, Plaster N, Fu YH, Ptacek LJ, Tawil R. Functional and clinical characterization of KCNJ2 mutations associated with LQT7 (Andersen syndrome). J Clin Invest. 2002; 110: 381–388.
Andelfinger G, Tapper AR, Welch RC, Vanoye CG, George AL Jr, Benson DW. KCNJ2 mutation results in Andersen syndrome with sex-specific cardiac and skeletal muscle phenotypes. Am J Hum Genet. 2002; 71: 663–668.
Ai T, Fujiwara Y, Tsuji K, Otani H, Nakano S, Kubo Y, Horie M. Novel KCNJ2 mutation in familial periodic paralysis with ventricular dysrhythmia. Circulation. 2002; 105: 2592–2594.
Donaldson MR, Jensen JL, Tristani-Firouzi M, Tawil R, Bendahhou S, Suarez WA, Cobo AM, Poza JJ, Behr E, Wagstaff J, Szepetowski P, Pereira S, Mozaffar T, Escolar DM, Fu YH, Ptacek LJ. PIP2 binding residues of Kir2.1 are common targets of mutations causing Andersen syndrome. Neurology. 2003; 60: 1811–1816.
Bendahhou S, Donaldson MR, Plaster NM, Tristani-Firouzi M, Fu YH, Ptacek LJ. Defective potassium channel Kir2.1 trafficking underlies Andersen-Tawil syndrome. J Biol Chem. 2003; 278: 51779–51785.
Chun TU, Epstein MR, Dick M II, Andelfinger G, Ballester L, Vanoye CG, George AL Jr, Benson DW. Polymorphic ventricular tachycardia and KCNJ2 mutations. Heart Rhythm. 2004; 1: 235–241.
Lopatin AN, Nichols CG. Inward rectifiers in the heart: an update on I(K1). J Mol Cell Cardiol. 2001; 33: 625–638.
Wang Z, Yue L, White M, Pelletier G, Nattel S. Differential distribution of inward rectifier potassium channel transcripts in human atrium versus ventricle. Circulation. 1998; 98: 2422–2428.
Liu GX, Derst C, Schlichthorl G, Heinen S, Seebohm G, Bruggemann A, Kummer W, Veh RW, Daut J, Preisig-Muller R. Comparison of cloned Kir2 channels with native inward rectifier K+ channels from guinea-pig cardiomyocytes. J Physiol. 2001; 532: 115–126.
Preisig-Muller R, Schlichthorl G, Goerge T, Heinen S, Bruggemann A, Rajan S, Derst C, Veh RW, Daut J. Heteromerization of Kir2.x potassium channels contributes to the phenotype of Andersen’s syndrome. Proc Natl Acad Sci U S A. 2002; 99: 7774–7779.
Zhang L, Timothy KW, Vincent GM, Lehmann MH, Fox J, Giuli LC, Shen J, Splawski I, Priori SG, Compton SJ, Yanowitz F, Benhorin J, Moss AJ, Schwartz PJ, Robinson JL, Wang Q, Zareba W, Keating MT, Towbin JA, Napolitano C, Medina A. The spectrum of ST-T wave patterns and repolarization parameters in congenital long QT syndrome: ECG findings identify genotype. Circulation. 2000; 102: 2849–2855.
Lepeschkin E, Surawicz B. The measurement of the QT interval of the electrocardiogram. Circulation. 1952; 6: 378–388.
Vincent GM, Timothy KW, Leppert M, Keating M. The spectrum of symptoms and QT intervals in the carriers of the gene for the long-QT syndrome. N Engl J Med. 1992; 327: 846–852.
Surawicz B. U wave: facts, hypotheses, misconceptions, and misnomers. J Cardiovasc Electrophysiol. 1998; 9: 1117–1128.
Schwartz PJ, Priori SG, Spazzolini C, Moss AJ, Vincent GM, Napolitano C, Denjoy I, Guicheney P, Breithardt G, Keating MT, Towbin JA, Beggs AH, Brink P, Wilde AA, Toivonen L, Zareba W, Robinson JL, Timothy KW, Corfield V, Wattanasirichaigoon D, Corbett C, Haverkamp W, Schulze-Bahr E, Lehmann MH, Schwartz K, Coumel P, Bloise R. Genotype-phenotype correlation in the long-QT syndrome: gene-specific triggers for life-threatening arrhythmias. Circulation. 2001; 103: 89–95.
Nakagawa M, Ooie T, Hara M, Ichinose M, Nobe S, Yonemochi H, Saikawa T. Dynamics of T-U wave in patients with idiopathic ventricular tachycardia originating from the right ventricular outflow tract. Pacing Clin Electrophysiol. 2004; 27: 148–155.
Viskin S, Zeltser D, Antzelevitch C. When U say "U waves," what do U mean; Pacing Clin Electrophysiol. 2004; 27: 145–147.
Viskin S, Heller K, Barron HV, Kitzis I, Hamdan M, Olgin JE, Wong MJ, Grant SE, Lesh MD. Postextrasystolic U wave augmentation, a new marker of increased arrhythmic risk in patients without the long QT syndrome. J Am Coll Cardiol. 1996; 28: 1746–1752.
Cordeiro JM, Spitzer KW, Giles WR. Repolarizing K+ currents in rabbit heart Purkinje cells. J Physiol. 1998; 508 (pt 3): 811–823.
Sanchez-Chapula JA, Salinas-Stefanon E, Torres-Jacome J, Benavides-Haro DE, Navarro-Polanco RA. Blockade of currents by the antimalarial drug chloroquine in feline ventricular myocytes. J Pharmacol Exp Ther. 2001; 297: 437–445.
Hoffman BF, Cranefield PF. Electrophysiology of the Heart. New York, NY: McGraw-Hill; 1960: 201–202.
Watanabe Y. Purkinje repolarization as a possible cause of the U wave in the electrocardiogram. Circulation. 1975; 51: 1030–1037.
Shimizu W, Antzelevitch C. Sodium channel block with mexiletine is effective in reducing dispersion of repolarization and preventing torsade de pointes in LQT2 and LQT3 models of the long-QT syndrome. Circulation. 1997; 96: 2038–2047.
Yan GX, Antzelevitch C. Cellular basis for the normal T wave and the electrocardiographic manifestations of the long-QT syndrome. Circulation. 1999; 100: 1660–1666.
Lab MJ. Contraction-excitation feedback in myocardium physiological basis and clinical relevance. Circ Res. 1982; 50: 757–766.
Rice JJ, Winslow RL, Dekanski J, McVeigh E. Model studies of the role of mechano-sensitive currents in the generation of cardiac arrhythmias. J Theor Biol. 1998; 190: 295–312.(Li Zhang, MD; D. Woodrow )