- Email: [email protected]
Distinguishing “benign” from “malignant early repolarization”: The value of the ST-segment morphology
Distinguishing “benign” from “malignant early repolarization”: The value of the ST-segment morphology Raphael Rosso, MD,*‡ Eran Glikson,*‡ Bernard Belhassen, MD,* Amos Katz, MD,† Amir Halkin, MD,* Arie Steinvil, MD,* Sami Viskin, MD* From the *Department of Cardiology, Tel Aviv Sourasky Medical Center and, Sackler School of Medicine, Tel Aviv University, Tel Aviv, †Department of Cardiology, Barzilai Hospital, Ashkelon, and Ben Gurion University of the Negev, Beer-Sheva, Israel. BACKGROUND Means for distinguishing the very common “benign early repolarization” from the very rare but malignant form are needed. Recently, the presence of early repolarization with “horizontal ST segment” was found to predict arrhythmic death during long-term follow-up in a large population study. We therefore speculated that the combination of “J waves with horizontal ST segment” would correlate with a history of idiopathic ventricular fibrillation (VF) better than the mere presence of J waves. OBJECTIVES To determine whether the morphology of the ST segment adds diagnostic value to the mere presence of J waves in a case– control series of idiopathic VF. METHODS We reanalyzed our case– control study showing that the presence of J waves strongly correlates with a history of idiopathic VF among 45 patients with this disorder, 124 controls matched for age and gender (“matched-control” group), and 121 young athletes. This time we focused only on those patients with J waves and graded their ST-segment morphology
The presence of J waves and ST-segment elevation on the electrocardiogram (ECG), jointly termed “the early repolarization pattern,” was harmoniously considered a benign finding devoid of clinical significance.1–5 This perception changed, however, as case reports,6 –11 case– control studies,12–15 and large population studies16 –19 established that the presence of “J waves” is strongly associated with an increased risk for arrhythmic death in general16,17,19 and with idiopathic ventricular fibrillation (VF) in particular.12–15 These findings have raised concerns among physicians who are now asked to estimate the arrhythmic risk of a patient who has early repolarization that was incidentally discovered on a routine ECG recording.20 Thus, reliable means for distinguishing the very common “benign early repolarization” from the very rare malignant form are needed. Unfortunately, no clinical or ECG features evalu-
† Raphael Rosso and Eran Glikson contributed equally to this manuscript. Address for reprints and correspondence: Dr Sami Viskin, MD, Department of Cardiology, Tel-Aviv Medical Center, Tel Aviv University, Weizman 6 St., Tel-Aviv 64239, Israel. E-mail address: [email protected].
as either “horizontal” or “ascending” according to predefined criteria. RESULTS The presence of J waves was associated with a history of idiopathic VF with an odds ratio of 4.0 (95% confidence intervals ⫽ 2.0 –7.9), but having both J waves and horizontal ST segment yielded an odds ratio of 13.8 (95% confidence intervals ⫽ 5.1–37.2) for having idiopathic VF. CONCLUSIONS We report, for the first time, that the combination of J waves with horizontal/descending ST segment improved our ability to distinguish patients with idiopathic VF from controls matched by gender and age. KEYWORDS Idiopathic ventricular fibrillation; Early repolarization; J waves ABBREVIATIONS CI ⫽ confidence interval; ECG ⫽ eectrocardiogram; OR ⫽ odds ratio; VF ⫽ ventricular fibrillation (Heart Rhythm 2012;9:225–229) © 2012 Heart Rhythm Society. All rights reserved.
ated thus far, such as male gender, a history of familial sudden death, or the amplitude and distribution of J waves in the ECG, have sufficient prognostic value.21 Recently, Tikkanen et al17 reanalyzed their original population-based study that established that early repolarization is a statistically significant predictor of arrhythmic death. In the later publication, these investigators analyzed the contour of the ST segment and found that an “upsloping” (or “rapidly ascending”) ST segment is quasi-universal among young athletes with early repolarization and has no adverse clinical implications. In contrast, the combination of J waves or slurred R waves with a flat (ie, horizontal/descending) ST segment was strongly associated with an increased arrhythmic risk at long-term follow-up.22 We therefore reanalyzed our own case– control series13 to determine whether the combination of “J waves with plain ST segment” would correlate with a history of idiopathic VF better than the mere presence of J waves.
Methods Our study population was recently described in detail.13 It consisted of 45 patients with idiopathic VF (aged 14 – 69
1547-5271/$ -see front matter © 2012 Heart Rhythm Society. All rights reserved.
doi:10.1016/j.hrthm.2011.09.012
226
Heart Rhythm, Vol 9, No 2, February 2012
years, mean age 38 ⫾ 15 years; 71% males), a “matchedcontrol group” consisting of 124 healthy individuals matched by age and gender to the idiopathic VF patients, and an “athletes group” of 121 noncompetitive athletes (aged 17–19 years, 50% males) from a preparticipation medical screening program. “J-point elevation” or “J waves” were defined as positive “humplike” deflections immediately following a positive QRS complex at the onset of the ST segment. Since J-point elevation may be hidden in the QRS complex,23 we also noted the presence of “slurring” at the terminal part of the QRS complex. The transition from the QRS complex to the ST segment was defined as “slurred” when the R wave gradually became the ST segment with an upright concavity and as “J-point elevation” when a sharp and well-defined hump was noted immediately following the R wave.24 On the basis of these definitions, we found that J-point elevation was more common among patients with idiopathic VF than among matched controls and that young athletes had J-point elevation more often than healthy adults but less often than patients with idiopathic VF (see below).13 For the present study, we focused on those idiopathic VF patients and controls who had J-point elevation in the form of “J waves” or “slurred R waves” and classified the morphology of their ST segment by using the definitions of Tikkanen et al.22 Accordingly, ST-segment patterns after the J point were coded as “concave/rapidly ascending” (when there was ⬎0.1-mV elevation of the ST segment within 100 ms after the J point and the ST segment merged gradually with the T wave) (Figure 1A) or as “horizontal/descending” (when the ST-segment elevation was ⱕ0.1 mV within 100 ms after the J point and continued as a flat ST segment until the onset of the T wave) (Figure 1B and Figure 2). Based on the data by Tikkanen et al,22 showing that only horizontal/descending ST segment is of prognostic significance in patients with early repolarization, whenever the ST segment was “ascending” in some leads and “horizontal/descending” in others (looking only at those leads with J-point eleva-
Figure 1
tion or slurred J waves), we counted the patient as having “horizontal/descending” early repolarization.
Statistics Continuous variables are displayed as mean ⫾ standard deviation, and categorical variables are presented as number and percentage in each group. Comparison of continuous variables between patients and a young-athletes control group was done by using independent students’ t test. Comparison of all dichotomous variables between patients and young-athletes control group was done by using 2 statistics. Comparison of continuous variables between patients and matched controls was done by using blocked analysis of variance, where the blocks are defined as each case with all its matched controls under the general linear model procedure. Comparison of all dichotomous variables between patients and matched controls was done by using the conditional logistic regression analysis. The predictive value of J waves and the morphology of the ST segment were assessed by using binary logistic regressions. Results are displayed as the odds ratio (OR) plus 95% confidence intervals (CIs). All the above analyses were considered significant at P ⬍.05 (2 tailed). The SPSS statistical package was used to perform all statistical evaluations (SSPS, Inc, Chicago, IL).
Results As previously reported,13 J waves were more prevalent among patients with idiopathic VF than among age- and gender-matched controls (42% vs 13%; P ⬍.001) or among young athletes (42% vs 22%; P ⫽ .013). J waves were more prevalent among males, and this was true for idiopathic VF patients as well as for controls (Table 1). However, once J waves were present, gender had no statistically significant effect on the contour of the ST segment (Table 1) and the only characteristic associated with J waves with the ST-segment pattern was a history of idiopathic VF (Table 1). Figure 3 shows the percentage of patients with horizontal/descending vs rapidly ascending ST segment among
Rapidly ascending (A) and horizontal (B) ST segment in the leads deploying J waves (J waves marked with arrowhead).
Rosso et al
Distinguishing “Benign” from “Malignant Early Repolarization”
227
Figure 2 Electrocardiogram of a male patient with idiopathic ventricular fibrillation with J waves in the inferior leads. Leads II, III, and aVF are also enlarged to show the horizontal/descending ST segment.
patients with J waves in the 3 patient groups: 13 (68.4%) of the 19 idiopathic VF patients with J waves had horizontal/ descending ST segments; in contrast, only 4 (25%) of the 16 age-matched controls and 4 (14.3%) of the 28 young athletes with J waves had a horizontal ST segment. Thus, the vast majority of healthy controls with J-point elevation had ST segment of the rapidly ascending type and this was particularly true for the young-athletes control group (Figure 3). The reverse was true for idiopathic VF group (Figure 3). The difference between the partition of patients with J waves into horizontal/descending vs rapidly ascending ST segment was highly significant (P ⬍.02 and P ⬍.001 for the idiopathic VF vs the matched-control and the young-athletes control groups, respectively). The differences persisted to a lesser degree when the percentage of patients with
horizontal vs ascending ST segments was compared only for patients with J-point elevation (ie, not counting those with slurred R waves) (Figure 3B). Here, only the difference between idiopathic VF and the young athletes remained statistically significant (P ⬍.001). Slurred R waves did not correlate with a history of idiopathic VF (already reported).13 In our sample of individuals, the presence of J waves was associated with a history of idiopathic VF with an OR of 4.0 (95% CI ⫽ 2.0–7.9) and having both J waves and horizontal/descending ST segment yielded an OR of 13.8 (95% CI ⫽ 5.1–37.2) for having idiopathic VF. One should emphasize, however, that patients with idiopathic VF are greatly overrepresented in our series (in comparison to real-life scenario). As explained in detail elsewhere,13 better appreciation of the risk associated with J waves and horizontal ST segment can be obtained with the
Table 1 Clinical and electrocardiographic characteristics of patients with idiopathic ventricular fibrillation and controls matched for age and gender
Age (y) RR (ms) QT (ms) Corrected QT interval (ms) Male gender Idiopathic ventricular fibrillation
No J waves (n ⫽ 133)
J wave (n ⫽ 35)
P value
Ascending ST (n ⫽ 18)
Horizontal ST (n ⫽ 17)
P value
39 ⫾ 13 866 ⫾ 140 355 ⫾ 28 384 ⫾ 25 90 (68%) 26 (20%)
33 ⫾ 15 888 ⫾ 149 355 ⫾ 26 379 ⫾ 26 31 (89%) 19 (54%)
.024 .423 .904 .297 .019 <.001
31 ⫾ 17 890 ⫾ 132 355 ⫾ 24 378 ⫾ 31 17 (94) 6 (33)
36 ⫾ 14 885 ⫾ 170 355 ⫾ 29 380 ⫾ 20 14 (82) 13 (77)
.359 .928 .993 .887 .338 .018
Bold numbers indicate values that are statistically significant.
228
Heart Rhythm, Vol 9, No 2, February 2012
What is wrong with early repolarization?
Figure 3 Distribution of patients with early repolarization according to the morphology of their ST segment. Patients with either J-point elevation or slurred R wave are shown in panel A, whereas patients only with J-point elevation are shown in panel B. VF ⫽ ventricular fibrillation.
Bayes’ formula of conditional probabilities. Accordingly, if the estimated odds for developing idiopathic VF for an individual in the age range of 35 to 45 years are 3.4 in 100,000, the risk increases to 11 in 100,000 once J waves are found in the ECG and to 30.4 in 100,000 if the J waves are followed by a horizontal ST segment.
Discussion Early repolarization is a common ECG finding, with a prevalence ranging from 6% to 24% in population-based studies.16 –19 Moreover, series including 1000 athletes or more report that as many as 40% of healthy athletes have early repolarization.21 On the other hand, this finding has been strongly associated with idiopathic VF,12–15,25 a rare but highly lethal disease.26 Therefore, diagnostic tools capable of distinguishing the very common “benign early repolarization” from the very rare malignant form are needed.20,21 On the basis of data recently published by Tikkanen et al,22 we speculated that early repolarization with horizontal/descending ST segment would tell apart idiopathic VF patients from matched controls more accurately than the presence of early repolarization per se.
Main findings The combination of J waves with horizontal/descending ST segment improved our ability to distinguish patients with idiopathic VF from controls matched by gender and age: Although the OR for having idiopathic VF for patients with J waves was 4.0 (95% CI ⫽ 2.0–7.9), having J waves and horizontal ST segment yielded an OR of 13.8 (95% CI ⫽ 5.1–37.2). Thus, our findings support the concept presented by Tikkanen et al22 regarding the malignant nature of early repolarization with horizontal ST segment.
As explained in detail elsewhere,21 we reason that the increased risk of arrhythmic death identified by Tikkanen et al17,22 for those with early repolarization within the general population does not necessarily reflect an association with idiopathic VF. Given the fact that coronary disease is the most common cause of death in cohorts such as those studied by Tikkanen et al, the noted association is more likely to represent an increased risk for “ischemic VF” among adults eventually developing coronary disease.21 Experimental studies suggest that early repolarization is a marker of increased dispersion of repolarization that increases the risk for VF but only in the presence of certain triggers.27 In experimental settings, such triggers are timely delivered external electrical stimuli,27 whereas in real life, these are premature spontaneous extrasystoles falling on the descending limb of the T wave: the R-on-T phenomenon. This phenomenon typically occurs during “ischemic VF,”28,29 Brugada syndrome,15 or idiopathic VF.30 This might explain why early repolarization is a marker for increased arrhythmic mortality across a heterogeneous clinical spectrum, including the general adult-population cohort studied by Tikkanen et al,17,22 Brugada syndrome,31 and idiopathic VF.12–15,21,25 Having said that, it is important to recognize that many aspects of the “early repolarization theory” are still a matter of unresolved debate.32–37
Not all forms of early repolarization are created equal J waves followed by rapidly ascending ST elevation is the quasi-universal pattern among healthy athletes with early repolarization, is the most common form in healthy adults, and does not appear to have any adverse prognostic significance. Only when the J waves are followed by horizontal/ descending ST segment does a very strong association with arrhythmic risk exist. This observation, first noticed by Tikkanen et al for an unselected adult-population cohort, is now reported, for the first time, in this case– control series of idiopathic VF. It is tempting to speculate that J waves followed by horizontal/descending ST segment are the ECG representation of very steep localized repolarization gradients, as recently suggested in case studies using noninvasive electrocardiographic imaging,38 whereas J waves with ascending ST segment represent a more gradual transition.
Study limitations 1. It is possible that we were biased in our interpretation of ECGs because some traces of idiopathic VF patients are old and look different from those of the controls even after scanning. Also, since we have a limited number of ECG traces for our patients prior to the onset of quinidine therapy, we did not test for day-to-day variability of ST-segment elevation type. 2. When rapidly ascending and horizontal/descending ST segments were present in different leads, we arbitrarily counted the ECG as showing “horizontal/descending” pattern. Rather than accepting our study as conclusive, we interpret our findings as further support of the important observation made by Tikkanen et al in a more rigorously conducted study including
Rosso et al
Distinguishing “Benign” from “Malignant Early Repolarization”
all the ECG traces of a prospectively followed large cohort. Importantly, when attempting to diagnose short or long QT syndrome, we accept that the transition from “abnormally short” to “normal” and then to “long” QT intervals is gradual with no single corrected QT interval value distinguishing the healthy from the genetically ill.39 There is no reason to believe that distinguishing normal from abnormal early repolarization will be any easier.
Clinical implications
Idiopathic VF is a rare disease.26 Therefore, to truly understand the diagnostic significance of “early repolarization with horizontal/ descending ST segment,” one must use formulas of conditional probabilities.13 Accordingly, the estimated odds for developing idiopathic VF for an individual in the age range of 35 to 45 years, even if he or she has J waves and even if these are followed by horizontal ST segment, are only 34 in 100,000 or 0.03%. To put these numbers in perspective, it is worth revisiting the controversy surrounding the best clinical approach to asymptomatic Brugada syndrome.40 Here, those in favor and those against the use of electrophysiologic studies for risk stratification quote annual risks of death for patients with asymptomatic Brugada syndrome who have positive tests as being 5% and ⬍1%, respectively.40–45 These figures are higher by orders of magnitude than the 0.03% risk estimated here for asymptomatic individuals with J waves and horizontal/descending ST segment. Thus, despite the advancements in risk stratification of patients with early repolarization based on ST-segment morphology, first proposed by Tikkanen et al22 and now confirmed here, we ought to recognize that we still cannot reliably distinguish the very common “benign early repolarization” from the truly rare malignant form.
References 1. Fenichel NN. A long-term study of concave RS-T elevation⫺a normal variant of the electrocardiogram. Angiology 1962;13:360 –366. 2. Goldman MJ. RS-T segment elevation in mid- and left precordial leads as a normal variant. Am Heart J 1953;46:817– 820. 3. Grusin H. Peculiarities of the African’s electrocardiogram and the changes observed in serial studies. Circulation 1954;9:860 – 867. 4. Klatsky AL, Oehm R, Cooper RA, Udaltsova N, Armstrong MA. The early repolarization normal variant electrocardiogram: correlates and consequences. Am J Med 2003;115:171–177. 5. Wasserburger RH, Alt WJ. The normal RS-T segment elevation variant. Am J Cardiol 1961;8:184 –192. 6. Aizawa Y, Tamura M, Chinushi M, et al. Idiopathic ventricular fibrillation and bradycardia-dependent intraventricular block. Am Heart J 1993;126:1473–1474. 7. Kalla H, Yan GX, Marinchak R. Ventricular fibrillation in a patient with prominent J (Osborn) waves and ST segment elevation in the inferior electrocardiographic leads: a Brugada syndrome variant? J Cardiovasc Electrophysiol 2000;11:95–98. 8. Riera AR, Ferreira C, Schapachnik E, Sanches PC, Moffa PJ. Brugada syndrome with atypical ECG: downsloping ST-segment elevation in inferior leads. J Electrocardiol 2004;37:101–104. 9. Sahara M, Sagara K, Yamashita T, et al. J wave and ST segment elevation in the inferior leads: a latent type of variant Brugada syndrome? Jpn Heart J 2002;43:55–60. 10. Shinohara T, Takahashi N, Saikawa T, Yoshimatsu H. Characterization of J wave in a patient with idiopathic ventricular fibrillation. Heart Rhythm 2006;3:1082–1084. 11. Sugao M, Fujiki A, Nishida K, et al. Repolarization dynamics in patients with idiopathic ventricular fibrillation: pharmacological therapy with bepridil and disopyramide. J Cardiovasc Pharmacol 2005;45:545–549. 12. Haissaguerre M, Derval N, Sacher F, et al. Sudden cardiac arrest associated with early repolarization. New Engl J Med 2008;358:2016 –2023. 13. Rosso R, Kogan E, Belhassen B, et al. J-point elevation in survivors of primary ventricular fibrillation and matched control subjects: incidence and clinical significance. J Am Coll Cardiol 2008;52:1231–1238. 14. Abe A, Ikeda T, Tsukada T, et al. Circadian variation of late potentials in idiopathic ventricular fibrillation associated with J waves: insights into alternative pathophysiology and risk stratification. Heart Rhythm 2010;7:675– 682.
229
15. Nam GB, Ko KH, Kim J, et al. Mode of onset of ventricular fibrillation in patients with early repolarization pattern vs. Brugada syndrome. Eur Heart J 2009;31:330 –339. 16. Sinner MF, Reinhard W, Muller M, et al. Association of early repolarization pattern on ECG with risk of cardiac and all-cause mortality: a population-based prospective cohort study (MONICA/KORA). PLoS Med 2010;7:e1000314. 17. Tikkanen JT, Anttonen O, Junttila MJ, et al. Long-term outcome associated with early repolarization on electrocardiography. New Engl J Med 2009;361:2529–2537. 18. Olson KA, Viera AJ, Soliman EZ, Crow RS, Rosamond WD. Long-term prognosis associated with J-point elevation in a large middle-aged biracial cohort: the ARIC study. Eur Heart J. In press. 19. Haruta D, Matsuo K, Tsuneto A, et al. Incidence and prognostic value of early repolarization pattern in the 12-lead electrocardiogram. Circulation 2011;123:2931–2937. 20. Viskin S. Idiopathic ventricular fibrillation “Le Syndrome d’Haissaguerre” and the fear of J waves. J Am Coll Cardiol 2009;53:620 – 622. 21. Rosso R, Adler A, Halkin A, Viskin S. Risk of sudden death among young individuals with J waves and early repolarization: putting the evidence into perspective. Heart Rhythm 2011;8:923–929. 22. Tikkanen JT, Junttila MJ, Anttonen O, et al. Early repolarization: electrocardiographic phenotypes associated with favorable long-term outcome. Circulation 2011;123:2666 –2673. 23. Hlaing T, DiMino T, Kowey PR, Yan GX. ECG repolarization waves: their genesis and clinical implications. Ann Noninvasive Electrocardiol 2005;10:211–223. 24. Kambara H, Phillips J. Long-term evaluation of early repolarization syndrome (normal variant RS-T segment elevation). Am J Cardiol 1976;38:157–156. 25. Merchant FM, Noseworthy PA, Weiner RB, Singh SM, Ruskin JN, Reddy VY. Ability of terminal QRS notching to distinguish benign from malignant electrocardiographic forms of early repolarization. Am J Cardiol 2009;104:1402–1406. 26. Viskin S, Belhassen B. Idiopathic ventricular fibrillation. Am Heart J 1990;120: 661– 671. 27. Antzelevitch C, Yan GX. J wave syndromes. Heart Rhythm 2010;7:549 –558. 28. Di Diego JM, Antzelevitch C. Ischemic ventricular arrhythmias: experimental models and their clinical relevance. Heart Rhythm 2011;8:1963–1968. 29. Wolfe CL, Nibbley C, Bhandari A, Chatterjee K, Scheinman MM. Polymorphous ventricular tachycardia associated with acute myocardial infarction. Circulation 1991;84:1543–1551. 30. Viskin S, Lesh M, Eldar M, et al. Mode of onset of malignant ventricular arrhythmias in idiopathic ventricular fibrillation. J Cardiovasc Electrophysiol 1997;8:1115–1120. 31. Kamakura S, Ohe T, Nakazawa K, et al. Long-term prognosis of probands with Brugada-pattern ST-elevation in leads V1-V3. Circ Arrhythm Electrophysiol 2009;2:495–503. 32. Wilde AA, Postema PG, Di Diego JM, et al. The pathophysiological mechanism underlying Brugada syndrome: depolarization versus repolarization. J Mol Cell Cardiol 2010;49:543–553. 33. Janse MJ, Coronel R, Opthof T. Rebuttal to M cells are present in the ventricular myocardium: counterpoint. Heart Rhythm 2011;8:1100. 34. Janse MJ, Coronel R, Opthof T. Counterpoint: M cells do not have a functional role in the ventricular myocardium of the intact heart. Heart Rhythm 2011;8:934–937. 35. Wilson LD, Jennings MM, Rosenbaum DS. Rebuttal to M cells are not present in the ventricular myocardium. Point. Heart Rhythm 2011;8:1099. 36. Wilson LD, Jennings MM, Rosenbaum DS. Point: M cells are present in the ventricular myocardium. Heart Rhythm 2011;8:930 –933. 37. Nattel S, Antzelevitch C, Noble D. Resolving the M-cell debate: why and how. Heart Rhythm 2011;8:1293–1295. 38. Ghosh S, Cooper DH, Vijayakumar R, et al. Early repolarization associated with sudden death: insights from noninvasive electrocardiographic imaging. Heart Rhythm 7:534 –537. 39. Viskin S. The QT interval: too long, too short or just right. Heart Rhythm 2009;6:711–715. 40. Viskin S, Rogowski O. Asymptomatic Brugada syndrome: a cardiac ticking time-bomb? Europace 2007;9:707–710. 41. Viskin S, Rosso R. Risk of sudden death in asymptomatic Brugada syndrome: not as high as we thought and not as low as we wished. . .but the contrary. J Am Coll Cardiol 2010;56:1585–1588. 42. Brugada J, Brugada R, Brugada P. Electrophysiologic testing predicts events in Brugada syndrome patients. Heart Rhythm. 2011;8:1595–1597. 43. Brugada J, Brugada R, Brugada P. Rebuttal to EP testing does not predict cardiac events in patients with Brugada syndrome. Heart Rhythm 2011;8:1796. 44. Wilde AA, Viskin S. EP testing does not predict cardiac events in Brugada syndrome. Heart Rhythm. 2011;8:1598 –1600. 45. Wilde AA, Viskin S. Rebuttal to EP testing predicts cardiac events in patients with Brugada syndrome. Heart Rhythm. 8:1797.
The presence of J waves and ST-segment elevation on the electrocardiogram (ECG), jointly termed “the early repolarization pattern,” was harmoniously considered a benign finding devoid of clinical significance.1–5 This perception changed, however, as case reports,6 –11 case– control studies,12–15 and large population studies16 –19 established that the presence of “J waves” is strongly associated with an increased risk for arrhythmic death in general16,17,19 and with idiopathic ventricular fibrillation (VF) in particular.12–15 These findings have raised concerns among physicians who are now asked to estimate the arrhythmic risk of a patient who has early repolarization that was incidentally discovered on a routine ECG recording.20 Thus, reliable means for distinguishing the very common “benign early repolarization” from the very rare malignant form are needed. Unfortunately, no clinical or ECG features evalu-
† Raphael Rosso and Eran Glikson contributed equally to this manuscript. Address for reprints and correspondence: Dr Sami Viskin, MD, Department of Cardiology, Tel-Aviv Medical Center, Tel Aviv University, Weizman 6 St., Tel-Aviv 64239, Israel. E-mail address: [email protected].
as either “horizontal” or “ascending” according to predefined criteria. RESULTS The presence of J waves was associated with a history of idiopathic VF with an odds ratio of 4.0 (95% confidence intervals ⫽ 2.0 –7.9), but having both J waves and horizontal ST segment yielded an odds ratio of 13.8 (95% confidence intervals ⫽ 5.1–37.2) for having idiopathic VF. CONCLUSIONS We report, for the first time, that the combination of J waves with horizontal/descending ST segment improved our ability to distinguish patients with idiopathic VF from controls matched by gender and age. KEYWORDS Idiopathic ventricular fibrillation; Early repolarization; J waves ABBREVIATIONS CI ⫽ confidence interval; ECG ⫽ eectrocardiogram; OR ⫽ odds ratio; VF ⫽ ventricular fibrillation (Heart Rhythm 2012;9:225–229) © 2012 Heart Rhythm Society. All rights reserved.
ated thus far, such as male gender, a history of familial sudden death, or the amplitude and distribution of J waves in the ECG, have sufficient prognostic value.21 Recently, Tikkanen et al17 reanalyzed their original population-based study that established that early repolarization is a statistically significant predictor of arrhythmic death. In the later publication, these investigators analyzed the contour of the ST segment and found that an “upsloping” (or “rapidly ascending”) ST segment is quasi-universal among young athletes with early repolarization and has no adverse clinical implications. In contrast, the combination of J waves or slurred R waves with a flat (ie, horizontal/descending) ST segment was strongly associated with an increased arrhythmic risk at long-term follow-up.22 We therefore reanalyzed our own case– control series13 to determine whether the combination of “J waves with plain ST segment” would correlate with a history of idiopathic VF better than the mere presence of J waves.
Methods Our study population was recently described in detail.13 It consisted of 45 patients with idiopathic VF (aged 14 – 69
1547-5271/$ -see front matter © 2012 Heart Rhythm Society. All rights reserved.
doi:10.1016/j.hrthm.2011.09.012
226
Heart Rhythm, Vol 9, No 2, February 2012
years, mean age 38 ⫾ 15 years; 71% males), a “matchedcontrol group” consisting of 124 healthy individuals matched by age and gender to the idiopathic VF patients, and an “athletes group” of 121 noncompetitive athletes (aged 17–19 years, 50% males) from a preparticipation medical screening program. “J-point elevation” or “J waves” were defined as positive “humplike” deflections immediately following a positive QRS complex at the onset of the ST segment. Since J-point elevation may be hidden in the QRS complex,23 we also noted the presence of “slurring” at the terminal part of the QRS complex. The transition from the QRS complex to the ST segment was defined as “slurred” when the R wave gradually became the ST segment with an upright concavity and as “J-point elevation” when a sharp and well-defined hump was noted immediately following the R wave.24 On the basis of these definitions, we found that J-point elevation was more common among patients with idiopathic VF than among matched controls and that young athletes had J-point elevation more often than healthy adults but less often than patients with idiopathic VF (see below).13 For the present study, we focused on those idiopathic VF patients and controls who had J-point elevation in the form of “J waves” or “slurred R waves” and classified the morphology of their ST segment by using the definitions of Tikkanen et al.22 Accordingly, ST-segment patterns after the J point were coded as “concave/rapidly ascending” (when there was ⬎0.1-mV elevation of the ST segment within 100 ms after the J point and the ST segment merged gradually with the T wave) (Figure 1A) or as “horizontal/descending” (when the ST-segment elevation was ⱕ0.1 mV within 100 ms after the J point and continued as a flat ST segment until the onset of the T wave) (Figure 1B and Figure 2). Based on the data by Tikkanen et al,22 showing that only horizontal/descending ST segment is of prognostic significance in patients with early repolarization, whenever the ST segment was “ascending” in some leads and “horizontal/descending” in others (looking only at those leads with J-point eleva-
Figure 1
tion or slurred J waves), we counted the patient as having “horizontal/descending” early repolarization.
Statistics Continuous variables are displayed as mean ⫾ standard deviation, and categorical variables are presented as number and percentage in each group. Comparison of continuous variables between patients and a young-athletes control group was done by using independent students’ t test. Comparison of all dichotomous variables between patients and young-athletes control group was done by using 2 statistics. Comparison of continuous variables between patients and matched controls was done by using blocked analysis of variance, where the blocks are defined as each case with all its matched controls under the general linear model procedure. Comparison of all dichotomous variables between patients and matched controls was done by using the conditional logistic regression analysis. The predictive value of J waves and the morphology of the ST segment were assessed by using binary logistic regressions. Results are displayed as the odds ratio (OR) plus 95% confidence intervals (CIs). All the above analyses were considered significant at P ⬍.05 (2 tailed). The SPSS statistical package was used to perform all statistical evaluations (SSPS, Inc, Chicago, IL).
Results As previously reported,13 J waves were more prevalent among patients with idiopathic VF than among age- and gender-matched controls (42% vs 13%; P ⬍.001) or among young athletes (42% vs 22%; P ⫽ .013). J waves were more prevalent among males, and this was true for idiopathic VF patients as well as for controls (Table 1). However, once J waves were present, gender had no statistically significant effect on the contour of the ST segment (Table 1) and the only characteristic associated with J waves with the ST-segment pattern was a history of idiopathic VF (Table 1). Figure 3 shows the percentage of patients with horizontal/descending vs rapidly ascending ST segment among
Rapidly ascending (A) and horizontal (B) ST segment in the leads deploying J waves (J waves marked with arrowhead).
Rosso et al
Distinguishing “Benign” from “Malignant Early Repolarization”
227
Figure 2 Electrocardiogram of a male patient with idiopathic ventricular fibrillation with J waves in the inferior leads. Leads II, III, and aVF are also enlarged to show the horizontal/descending ST segment.
patients with J waves in the 3 patient groups: 13 (68.4%) of the 19 idiopathic VF patients with J waves had horizontal/ descending ST segments; in contrast, only 4 (25%) of the 16 age-matched controls and 4 (14.3%) of the 28 young athletes with J waves had a horizontal ST segment. Thus, the vast majority of healthy controls with J-point elevation had ST segment of the rapidly ascending type and this was particularly true for the young-athletes control group (Figure 3). The reverse was true for idiopathic VF group (Figure 3). The difference between the partition of patients with J waves into horizontal/descending vs rapidly ascending ST segment was highly significant (P ⬍.02 and P ⬍.001 for the idiopathic VF vs the matched-control and the young-athletes control groups, respectively). The differences persisted to a lesser degree when the percentage of patients with
horizontal vs ascending ST segments was compared only for patients with J-point elevation (ie, not counting those with slurred R waves) (Figure 3B). Here, only the difference between idiopathic VF and the young athletes remained statistically significant (P ⬍.001). Slurred R waves did not correlate with a history of idiopathic VF (already reported).13 In our sample of individuals, the presence of J waves was associated with a history of idiopathic VF with an OR of 4.0 (95% CI ⫽ 2.0–7.9) and having both J waves and horizontal/descending ST segment yielded an OR of 13.8 (95% CI ⫽ 5.1–37.2) for having idiopathic VF. One should emphasize, however, that patients with idiopathic VF are greatly overrepresented in our series (in comparison to real-life scenario). As explained in detail elsewhere,13 better appreciation of the risk associated with J waves and horizontal ST segment can be obtained with the
Table 1 Clinical and electrocardiographic characteristics of patients with idiopathic ventricular fibrillation and controls matched for age and gender
Age (y) RR (ms) QT (ms) Corrected QT interval (ms) Male gender Idiopathic ventricular fibrillation
No J waves (n ⫽ 133)
J wave (n ⫽ 35)
P value
Ascending ST (n ⫽ 18)
Horizontal ST (n ⫽ 17)
P value
39 ⫾ 13 866 ⫾ 140 355 ⫾ 28 384 ⫾ 25 90 (68%) 26 (20%)
33 ⫾ 15 888 ⫾ 149 355 ⫾ 26 379 ⫾ 26 31 (89%) 19 (54%)
.024 .423 .904 .297 .019 <.001
31 ⫾ 17 890 ⫾ 132 355 ⫾ 24 378 ⫾ 31 17 (94) 6 (33)
36 ⫾ 14 885 ⫾ 170 355 ⫾ 29 380 ⫾ 20 14 (82) 13 (77)
.359 .928 .993 .887 .338 .018
Bold numbers indicate values that are statistically significant.
228
Heart Rhythm, Vol 9, No 2, February 2012
What is wrong with early repolarization?
Figure 3 Distribution of patients with early repolarization according to the morphology of their ST segment. Patients with either J-point elevation or slurred R wave are shown in panel A, whereas patients only with J-point elevation are shown in panel B. VF ⫽ ventricular fibrillation.
Bayes’ formula of conditional probabilities. Accordingly, if the estimated odds for developing idiopathic VF for an individual in the age range of 35 to 45 years are 3.4 in 100,000, the risk increases to 11 in 100,000 once J waves are found in the ECG and to 30.4 in 100,000 if the J waves are followed by a horizontal ST segment.
Discussion Early repolarization is a common ECG finding, with a prevalence ranging from 6% to 24% in population-based studies.16 –19 Moreover, series including 1000 athletes or more report that as many as 40% of healthy athletes have early repolarization.21 On the other hand, this finding has been strongly associated with idiopathic VF,12–15,25 a rare but highly lethal disease.26 Therefore, diagnostic tools capable of distinguishing the very common “benign early repolarization” from the very rare malignant form are needed.20,21 On the basis of data recently published by Tikkanen et al,22 we speculated that early repolarization with horizontal/descending ST segment would tell apart idiopathic VF patients from matched controls more accurately than the presence of early repolarization per se.
Main findings The combination of J waves with horizontal/descending ST segment improved our ability to distinguish patients with idiopathic VF from controls matched by gender and age: Although the OR for having idiopathic VF for patients with J waves was 4.0 (95% CI ⫽ 2.0–7.9), having J waves and horizontal ST segment yielded an OR of 13.8 (95% CI ⫽ 5.1–37.2). Thus, our findings support the concept presented by Tikkanen et al22 regarding the malignant nature of early repolarization with horizontal ST segment.
As explained in detail elsewhere,21 we reason that the increased risk of arrhythmic death identified by Tikkanen et al17,22 for those with early repolarization within the general population does not necessarily reflect an association with idiopathic VF. Given the fact that coronary disease is the most common cause of death in cohorts such as those studied by Tikkanen et al, the noted association is more likely to represent an increased risk for “ischemic VF” among adults eventually developing coronary disease.21 Experimental studies suggest that early repolarization is a marker of increased dispersion of repolarization that increases the risk for VF but only in the presence of certain triggers.27 In experimental settings, such triggers are timely delivered external electrical stimuli,27 whereas in real life, these are premature spontaneous extrasystoles falling on the descending limb of the T wave: the R-on-T phenomenon. This phenomenon typically occurs during “ischemic VF,”28,29 Brugada syndrome,15 or idiopathic VF.30 This might explain why early repolarization is a marker for increased arrhythmic mortality across a heterogeneous clinical spectrum, including the general adult-population cohort studied by Tikkanen et al,17,22 Brugada syndrome,31 and idiopathic VF.12–15,21,25 Having said that, it is important to recognize that many aspects of the “early repolarization theory” are still a matter of unresolved debate.32–37
Not all forms of early repolarization are created equal J waves followed by rapidly ascending ST elevation is the quasi-universal pattern among healthy athletes with early repolarization, is the most common form in healthy adults, and does not appear to have any adverse prognostic significance. Only when the J waves are followed by horizontal/ descending ST segment does a very strong association with arrhythmic risk exist. This observation, first noticed by Tikkanen et al for an unselected adult-population cohort, is now reported, for the first time, in this case– control series of idiopathic VF. It is tempting to speculate that J waves followed by horizontal/descending ST segment are the ECG representation of very steep localized repolarization gradients, as recently suggested in case studies using noninvasive electrocardiographic imaging,38 whereas J waves with ascending ST segment represent a more gradual transition.
Study limitations 1. It is possible that we were biased in our interpretation of ECGs because some traces of idiopathic VF patients are old and look different from those of the controls even after scanning. Also, since we have a limited number of ECG traces for our patients prior to the onset of quinidine therapy, we did not test for day-to-day variability of ST-segment elevation type. 2. When rapidly ascending and horizontal/descending ST segments were present in different leads, we arbitrarily counted the ECG as showing “horizontal/descending” pattern. Rather than accepting our study as conclusive, we interpret our findings as further support of the important observation made by Tikkanen et al in a more rigorously conducted study including
Rosso et al
Distinguishing “Benign” from “Malignant Early Repolarization”
all the ECG traces of a prospectively followed large cohort. Importantly, when attempting to diagnose short or long QT syndrome, we accept that the transition from “abnormally short” to “normal” and then to “long” QT intervals is gradual with no single corrected QT interval value distinguishing the healthy from the genetically ill.39 There is no reason to believe that distinguishing normal from abnormal early repolarization will be any easier.
Clinical implications
Idiopathic VF is a rare disease.26 Therefore, to truly understand the diagnostic significance of “early repolarization with horizontal/ descending ST segment,” one must use formulas of conditional probabilities.13 Accordingly, the estimated odds for developing idiopathic VF for an individual in the age range of 35 to 45 years, even if he or she has J waves and even if these are followed by horizontal ST segment, are only 34 in 100,000 or 0.03%. To put these numbers in perspective, it is worth revisiting the controversy surrounding the best clinical approach to asymptomatic Brugada syndrome.40 Here, those in favor and those against the use of electrophysiologic studies for risk stratification quote annual risks of death for patients with asymptomatic Brugada syndrome who have positive tests as being 5% and ⬍1%, respectively.40–45 These figures are higher by orders of magnitude than the 0.03% risk estimated here for asymptomatic individuals with J waves and horizontal/descending ST segment. Thus, despite the advancements in risk stratification of patients with early repolarization based on ST-segment morphology, first proposed by Tikkanen et al22 and now confirmed here, we ought to recognize that we still cannot reliably distinguish the very common “benign early repolarization” from the truly rare malignant form.
References 1. Fenichel NN. A long-term study of concave RS-T elevation⫺a normal variant of the electrocardiogram. Angiology 1962;13:360 –366. 2. Goldman MJ. RS-T segment elevation in mid- and left precordial leads as a normal variant. Am Heart J 1953;46:817– 820. 3. Grusin H. Peculiarities of the African’s electrocardiogram and the changes observed in serial studies. Circulation 1954;9:860 – 867. 4. Klatsky AL, Oehm R, Cooper RA, Udaltsova N, Armstrong MA. The early repolarization normal variant electrocardiogram: correlates and consequences. Am J Med 2003;115:171–177. 5. Wasserburger RH, Alt WJ. The normal RS-T segment elevation variant. Am J Cardiol 1961;8:184 –192. 6. Aizawa Y, Tamura M, Chinushi M, et al. Idiopathic ventricular fibrillation and bradycardia-dependent intraventricular block. Am Heart J 1993;126:1473–1474. 7. Kalla H, Yan GX, Marinchak R. Ventricular fibrillation in a patient with prominent J (Osborn) waves and ST segment elevation in the inferior electrocardiographic leads: a Brugada syndrome variant? J Cardiovasc Electrophysiol 2000;11:95–98. 8. Riera AR, Ferreira C, Schapachnik E, Sanches PC, Moffa PJ. Brugada syndrome with atypical ECG: downsloping ST-segment elevation in inferior leads. J Electrocardiol 2004;37:101–104. 9. Sahara M, Sagara K, Yamashita T, et al. J wave and ST segment elevation in the inferior leads: a latent type of variant Brugada syndrome? Jpn Heart J 2002;43:55–60. 10. Shinohara T, Takahashi N, Saikawa T, Yoshimatsu H. Characterization of J wave in a patient with idiopathic ventricular fibrillation. Heart Rhythm 2006;3:1082–1084. 11. Sugao M, Fujiki A, Nishida K, et al. Repolarization dynamics in patients with idiopathic ventricular fibrillation: pharmacological therapy with bepridil and disopyramide. J Cardiovasc Pharmacol 2005;45:545–549. 12. Haissaguerre M, Derval N, Sacher F, et al. Sudden cardiac arrest associated with early repolarization. New Engl J Med 2008;358:2016 –2023. 13. Rosso R, Kogan E, Belhassen B, et al. J-point elevation in survivors of primary ventricular fibrillation and matched control subjects: incidence and clinical significance. J Am Coll Cardiol 2008;52:1231–1238. 14. Abe A, Ikeda T, Tsukada T, et al. Circadian variation of late potentials in idiopathic ventricular fibrillation associated with J waves: insights into alternative pathophysiology and risk stratification. Heart Rhythm 2010;7:675– 682.
229
15. Nam GB, Ko KH, Kim J, et al. Mode of onset of ventricular fibrillation in patients with early repolarization pattern vs. Brugada syndrome. Eur Heart J 2009;31:330 –339. 16. Sinner MF, Reinhard W, Muller M, et al. Association of early repolarization pattern on ECG with risk of cardiac and all-cause mortality: a population-based prospective cohort study (MONICA/KORA). PLoS Med 2010;7:e1000314. 17. Tikkanen JT, Anttonen O, Junttila MJ, et al. Long-term outcome associated with early repolarization on electrocardiography. New Engl J Med 2009;361:2529–2537. 18. Olson KA, Viera AJ, Soliman EZ, Crow RS, Rosamond WD. Long-term prognosis associated with J-point elevation in a large middle-aged biracial cohort: the ARIC study. Eur Heart J. In press. 19. Haruta D, Matsuo K, Tsuneto A, et al. Incidence and prognostic value of early repolarization pattern in the 12-lead electrocardiogram. Circulation 2011;123:2931–2937. 20. Viskin S. Idiopathic ventricular fibrillation “Le Syndrome d’Haissaguerre” and the fear of J waves. J Am Coll Cardiol 2009;53:620 – 622. 21. Rosso R, Adler A, Halkin A, Viskin S. Risk of sudden death among young individuals with J waves and early repolarization: putting the evidence into perspective. Heart Rhythm 2011;8:923–929. 22. Tikkanen JT, Junttila MJ, Anttonen O, et al. Early repolarization: electrocardiographic phenotypes associated with favorable long-term outcome. Circulation 2011;123:2666 –2673. 23. Hlaing T, DiMino T, Kowey PR, Yan GX. ECG repolarization waves: their genesis and clinical implications. Ann Noninvasive Electrocardiol 2005;10:211–223. 24. Kambara H, Phillips J. Long-term evaluation of early repolarization syndrome (normal variant RS-T segment elevation). Am J Cardiol 1976;38:157–156. 25. Merchant FM, Noseworthy PA, Weiner RB, Singh SM, Ruskin JN, Reddy VY. Ability of terminal QRS notching to distinguish benign from malignant electrocardiographic forms of early repolarization. Am J Cardiol 2009;104:1402–1406. 26. Viskin S, Belhassen B. Idiopathic ventricular fibrillation. Am Heart J 1990;120: 661– 671. 27. Antzelevitch C, Yan GX. J wave syndromes. Heart Rhythm 2010;7:549 –558. 28. Di Diego JM, Antzelevitch C. Ischemic ventricular arrhythmias: experimental models and their clinical relevance. Heart Rhythm 2011;8:1963–1968. 29. Wolfe CL, Nibbley C, Bhandari A, Chatterjee K, Scheinman MM. Polymorphous ventricular tachycardia associated with acute myocardial infarction. Circulation 1991;84:1543–1551. 30. Viskin S, Lesh M, Eldar M, et al. Mode of onset of malignant ventricular arrhythmias in idiopathic ventricular fibrillation. J Cardiovasc Electrophysiol 1997;8:1115–1120. 31. Kamakura S, Ohe T, Nakazawa K, et al. Long-term prognosis of probands with Brugada-pattern ST-elevation in leads V1-V3. Circ Arrhythm Electrophysiol 2009;2:495–503. 32. Wilde AA, Postema PG, Di Diego JM, et al. The pathophysiological mechanism underlying Brugada syndrome: depolarization versus repolarization. J Mol Cell Cardiol 2010;49:543–553. 33. Janse MJ, Coronel R, Opthof T. Rebuttal to M cells are present in the ventricular myocardium: counterpoint. Heart Rhythm 2011;8:1100. 34. Janse MJ, Coronel R, Opthof T. Counterpoint: M cells do not have a functional role in the ventricular myocardium of the intact heart. Heart Rhythm 2011;8:934–937. 35. Wilson LD, Jennings MM, Rosenbaum DS. Rebuttal to M cells are not present in the ventricular myocardium. Point. Heart Rhythm 2011;8:1099. 36. Wilson LD, Jennings MM, Rosenbaum DS. Point: M cells are present in the ventricular myocardium. Heart Rhythm 2011;8:930 –933. 37. Nattel S, Antzelevitch C, Noble D. Resolving the M-cell debate: why and how. Heart Rhythm 2011;8:1293–1295. 38. Ghosh S, Cooper DH, Vijayakumar R, et al. Early repolarization associated with sudden death: insights from noninvasive electrocardiographic imaging. Heart Rhythm 7:534 –537. 39. Viskin S. The QT interval: too long, too short or just right. Heart Rhythm 2009;6:711–715. 40. Viskin S, Rogowski O. Asymptomatic Brugada syndrome: a cardiac ticking time-bomb? Europace 2007;9:707–710. 41. Viskin S, Rosso R. Risk of sudden death in asymptomatic Brugada syndrome: not as high as we thought and not as low as we wished. . .but the contrary. J Am Coll Cardiol 2010;56:1585–1588. 42. Brugada J, Brugada R, Brugada P. Electrophysiologic testing predicts events in Brugada syndrome patients. Heart Rhythm. 2011;8:1595–1597. 43. Brugada J, Brugada R, Brugada P. Rebuttal to EP testing does not predict cardiac events in patients with Brugada syndrome. Heart Rhythm 2011;8:1796. 44. Wilde AA, Viskin S. EP testing does not predict cardiac events in Brugada syndrome. Heart Rhythm. 2011;8:1598 –1600. 45. Wilde AA, Viskin S. Rebuttal to EP testing predicts cardiac events in patients with Brugada syndrome. Heart Rhythm. 8:1797.