Author + information
- Received April 21, 2008
- Revision received July 7, 2008
- Accepted July 14, 2008
- Published online October 7, 2008.
- Raphael Rosso, MD⁎,
- Evgeni Kogan, MD⁎,
- Bernard Belhassen, MD⁎,
- Uri Rozovski, MD⁎,
- Melvin M. Scheinman, MD§,
- David Zeltser, MD⁎,
- Amir Halkin, MD⁎,
- Arie Steinvil, MD⁎,
- Karin Heller, MD⁎,
- Michael Glikson, MD†,
- Amos Katz, MD‡ and
- Sami Viskin, MD⁎,⁎ ()
- ↵⁎Reprint requests and correspondence:
Dr. Sami Viskin, Tel Aviv Medical Center, Weizman 6, Tel Aviv 64239, Israel
Objectives The purpose of this study was to determine whether J-point elevation is a marker of arrhythmic risk.
Background J-point elevation has been considered an innocent finding among healthy young individuals (the “early repolarization” pattern). However, this electrocardiogram (ECG) finding is increasingly being associated with idiopathic ventricular fibrillation (VF).
Methods In a case-control study, the ECG of 45 patients with idiopathic VF were compared with those of 124 age- and gender-matched control subjects and with those of 121 young athletes. We measured the height of J-point and ST-segment elevation and counted the presence of slurring in the terminal portion of the R-wave.
Results J-point elevation was more common among patients with idiopathic VF than among matched control subjects (42% vs. 13%, p = 0.001). This was true for J-point elevation in the inferior leads (27% vs. 8%, p = 0.006) and for J-point elevation in leads I to aVL (13% vs. 1%, p = 0.009). J-point elevation in V4 to V6 occurred with equal frequency among patients and matched control subjects (6.7% vs. 7.3%, p = 0.86). Male subjects had J-point elevation more often than female subjects and young athletes had J-point elevation more often than healthy adults but less often than patients with idiopathic VF. The presence of ST-segment elevation or QRS slurring did not add diagnostic value to the presence of J-point elevation.
Conclusions J-point elevation is found more frequently among patients with idiopathic VF than among healthy control subjects. The frequency of J-point elevation among young athletes is intermediate (higher than among healthy adults but lower than among patients with idiopathic VF).
Since the description of the Brugada syndrome (1), J-point and ST-segment elevation in the right precordial leads is viewed as a marker of increased arrhythmic risk in patients with no organic heart disease (2,3). In contrast, J-point and ST-segment elevation in the lateral leads is considered an innocent finding because it is often observed in healthy young individuals (the so called “early repolarization” variant) (4).
Recently, several case reports called our attention to the association of idiopathic ventricular fibrillation (VF) with J-point elevation (with or without ST-segment elevation) in the inferior (5–12) or lateral leads (13–15). Therefore, we conducted the present study to determine whether J-point elevation with or without ST-segment elevation—in leads other than V1 to V3—is an innocent finding or a marker associated with a history of VF in patients without heart disease. After completion of this study, a similar study was published by Haissaguerre et al. (16).
The “VF Group” consisted of 45 patients (ages 14 to 69 years, mean age 38 ± 15 years; 71% male). This group consisted of 43 patients with a clinical diagnosis of idiopathic VF on the basis of rigorous criteria (17) and 2 patients eventually diagnosed with Brugada syndrome (see the following text). These patients have been reported in previous publications describing the clinical characteristics of idiopathic VF (18), the mode of onset of their spontaneous arrhythmias (19), their high VF-inducibility rate during electrophysiologic studies (20,21), and their response to quinidine therapy (20,21). Essentially, idiopathic VF was diagnosed by exclusion in cardiac arrest survivors when a complete work-up for alternative etiologies was negative (17) and was diagnosed with confidence in patients with syncope or cardiac arrest when spontaneous polymorphic ventricular tachycardia/VF, with all of the unique characteristics of idiopathic VF (19), was recorded. Familial relations do not exist between the VF patients; none of them has additional family members with malignant ventricular arrhythmias, and none is of South East Asian origin. The main objective of this study was to determine the diagnostic value of J-point elevation in leads other than V1 to V3 in patients with no organic heart disease. Accordingly, we included in the VF group 2 patients ultimately diagnosed with Brugada syndrome. These patients (both male, ages 28 and 42 years) have strictly normal QRS and ST segments in V1 to V3 in the baseline electrocardiogram (ECG) (Fig. 1), and the diagnosis of Brugada syndrome became apparent only after interventions (i.e., they developed type-I ST-segment elevation in V1 to V3  after placement of the V1 to V3 electrodes at a higher intercostal space  or a flecainide challenge test ). In contrast, patients with Brugada syndrome with any pathology in V1 to V3 on their baseline ECG, were excluded.
Two patient cohorts served as control subjects: a “matched control” and a “young-athletes” group. The matched-control group consisted of 3 control subjects matched for age and gender for each VF patient. Control subjects ages 24 to 70 years were recruited from among 3,500 adults participating in the Tel Aviv Medical Screening program in 2007. Screening included history, examination, ECG, and exercise testing of all patients and additional tests as necessary. Control subjects qualified for inclusion if: 1) the screening report mentioned “no history of syncope” and “no history of heart disease”; 2) the ECG was reported as normal (J-point elevation would not be considered abnormal in this screening program); and 3) screening did not reveal heart disease. The first 3 control subjects with these characteristics, who matched each idiopathic VF patient by age (within ±2 years) and gender, were included. For younger control subjects we studied healthy medical personnel or their offspring (again, 3 control subjects/patient). Subsequently, 11 control subjects were excluded for the following reasons: incidental finding of left bundle branch block (1 medical personnel), 1 type-II Brugada tracing (reported as normal ECG in the screening), and ECG with noise in ≥1 lead or wrong age (9 patients). Thus, the matched-control consisted of 124 control subjects matched for age and gender to the VF patient (2 or 3 control subjects for each patient). The “Young-Athletes” control group was studied to test the hypothesis that that J-point elevation is more prevalent among young individuals (23) with increased vagal tone (24). This group consisted of 132 athletes (3 athletes for each VF patient) randomly selected from a pre-participation medical screening program. These were 17- to 19-year-old noncompetitive athletes with different levels of training consisting mainly of endurance running (50% of them were male). Eleven of the ECG tracings were disqualified due to background noise or missing ≥1 lead, so that the “young control group” ultimately consisted of 121 athletes.
To blind ECG interpreters to patient grouping, all tracings were scanned and coded. Segments showing extra-systoles were not used. However, some of the ECG traces of idiopathic VF patients were old, and complete blinding was impossible for all traces. The scanned ECGs were reviewed in random order by 3 investigators. Grading was by consensus. “J-point elevation” or “J waves” were defined as positive “hump-like” deflections immediately after a positive QRS complex at the onset of the ST-segment. Because J-point elevation might be hidden in the QRS complex (25), 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 upright concavity and as “J-point elevation” when a sharp and well-defined hump was noted immediately after the R-wave (23). The height of the J-point and/or ST-segment elevation from the baseline was measured with an electronic caliper after the scanned ECGs were enlarged 4 times. The ST-segment elevation was measured at its most horizontal section.
Continuous variables are displayed as mean ± SD, and categorical variables are presented as number and percent in each group. Comparison of continuous variables between patients and matched control subjects was done with blocked analysis of variance, where the blocks are defined as each case with all its matched control subjects, under the general linear model procedure. Comparison of all dichotomous variables between patients and matched control subjects was done with the conditional logistic regression analysis. Results are displayed as the odds ratio (OR) plus 95% confidence intervals (CIs). All aforementioned analyses were considered significant at p < 0.05 (2-tailed). The SPSS statistical package (SSPS Inc., Chicago, Illinois) was used to perform all statistical evaluations.
Calculating the positive predictive value of J-point elevation for diagnosing idiopathic VF would lead to gross overestimation, because our study population consisted of 1 VF patient for every 2 to 3 control subjects, whereas in the real world idiopathic VF is far more infrequent. To translate the significance of our findings into clinical practice, we used the Bayes' theorem (26) to determine the conditional probability of having idiopathic VF, when J-point elevation is detected, with the formula: where PJ(IVF) is the probability of having idiopathic VF, when J waves are present. The parameters entered into the formula, PIVF(J) and PnotIF(J) (the probability of having a J-point elevation for an individual with idiopathic VF and for an individual with no IVF, respectively), were derived from the results of the present study; P(IVF) (the incidence of idiopathic VF in the population) was entered as 3.4/100,000. This figure was based on epidemiological data showing that the risk of cardiac arrest for the population ages 35 to 45 years is 34 of 100,000 (27), whereas only 10% of cardiac arrest events in this age group are due to idiopathic VF (18). Finally, P(not IVF), which is the incidence of individuals without idiopathic VF was approximated to 1.
J-point elevation was more commonly seen among patients with idiopathic VF than among matched-control subjects (42% vs. 13%, p = 0.001). This was particularly true for J-point elevation in the inferior leads (27% vs. 8%, p = 0.006) and was also true for J-point elevation in leads I to aVL (13% vs. 1%, p = 0.009). In contrast, J-point elevation in V4 to V6 occurred with equal frequency among patients and matched-control subjects (6.7% vs. 7.3%, p = 0.86) (Table 1,Figs. 1 and 2).⇓
J-point elevation >0.05 mV (or >0.1 mV) was also more frequent among idiopathic VF patients (Table 1). When J-point elevation was present, it tended to be taller among patients with idiopathic VF than among control subjects, but the difference did not reach statistical significance (0.14 vs. 0.09 mV, p = 0.099). Moreover, we could not identify a cut-off value of J-point elevation (height) that would reliably distinguish VF patients from control subjects. Excluding the 2 patients ultimately diagnosed with Brugada syndrome from this analysis did not change the results (data not shown).
Patients and control subjects had similar R-R, PR, QT, and QTc intervals (data not shown). The R waves in leads V5 to V6 were slightly taller in the VF group (1.06 mV vs. 0.86 mV, p = 0.006). However, the R waves in those leads where J-point elevation was more prevalent in the VF group were not taller in the VF group (0.85 mV vs. 0.89 mV and 0.62 mV vs. 0.76 mV [p = NS] for the VF and matched-control groups in the inferior and high-lateral leads, respectively). No correlation between R-wave amplitude and J-point elevation was found.
Slurring of the descending limb of the R wave was commonly observed among patients and control subjects (31% vs. 24%, p = 0.4). The presence of “slurring” was not useful for identifying patients with idiopathic VF regardless of the lead where it was observed (Table 2). Moreover, the presence of “QRS slurring and J-point elevation” or the presence of “QRS slurring or J-point elevation” did not add diagnostic value to the presence of J-point elevation alone (Table 2, Fig. 2).
The ST-segment elevation was commonly observed among VF patients and control subjects (33% vs. 24%, p = 0.35). The presence of ST-segment elevation was not useful for identifying patients with idiopathic VF regardless of its magnitude or the leads where it was observed (Table 3). Also, the combination of “ST-segment elevation and J-point elevation” or the presence of “ST-segment elevation or J-point elevation” did not add diagnostic value to the presence of J-point elevation alone (data not shown).
Influence of gender on the incidence of J-point elevation
The influence of gender on the incidence of J waves is best appreciated from Figure 3. Within each of the 3 patient groups, male subjects had J waves more often than female subjects. This difference remained significant even after correcting for gender-related differences in heart rate (not shown) and were particularly strong when only large J waves (>0.1 mV) were counted.
Comparison with young athletes
Figure 4 shows 2 important findings: 1) the frequency of J waves among young athletes was intermediate between that observed among healthy adults and that observed in the VF group; and 2) the distribution of J-point elevation also differed between patient groups: when healthy control subjects had J-point elevation, this occurred in leads V4 to V6 and/or in the inferior leads. In contrast, idiopathic VF patients had J-point elevation mainly in the inferior leads, less commonly in I to aVL and least commonly in leads V4 to V6.
Diagnostic power of J-point elevation
As mentioned previously (see “Statistics”), the estimated risk for idiopathic VF in the general population, ages 35 to 45 years, is roughly 3.4 of 100,000 individuals. On the basis of the results of our study, the probability of having J-point elevation is 0.42 for idiopathic VF patients and 0.13 for control subjects. According to the Bayes' formula of conditional probabilities, finding a J-wave in the ECG of an individual in the 35 to 45 years age range increases the chances of having idiopathic VF from 3.4 of 100,000 individuals to only 11 of 100,000.
J-point elevation is considered an innocent finding among healthy young individuals (4). However, this finding is increasingly being associated with idiopathic VF (5–16). We conducted this study to define whether the presence of J-point elevation is a marker of arrhythmic risk. The ECGs of 45 patients with idiopathic VF were compared with those of 124 healthy control subjects matched for age and gender. This study design was selected to avoid the bias of age and gender, because benign J-point elevation is found predominantly in young male subjects, whereas patients with idiopathic VF are older (mean age = 38 years) and only 2 of 3 are male. We included a second control group of young non-competitive athletes who are expected to have a high incidence of benign J-point elevation (4).
Frequency of J-point elevation
We observed J-point elevation in 42% of VF patients and 13% of matched control subjects. When only J-point elevation ≥0.1 mV was counted, the numbers were 31% vs. 9% (p ≧ 0.002). These results are remarkably similar to those recently reported by Haissaguerre et al. (16), who found J-point and/or ST-segment elevation of that magnitude in 31% of 216 idiopathic VF patients and 5% of 412 control subjects matched for age, gender, race, and level of physical activity. Also, 22% of our healthy “young athletes” had J-point elevation, a figure consistent with the 27% of young male subjects with J-point elevation reported by others (4).
Height of J-point elevation
We speculated that in a manner analogous to the duration of the QT segment, which tells apart patients with long QT syndrome and healthy individuals, the height of J-point elevation, rather than its mere presence, would identify patients with VF. We could not find a cut-off for J-point elevation that would reliably distinguish between idiopathic VF patients and control subjects but recognized a trend toward higher J-point elevation in VF patients. Furthermore, in a recent larger study, Haissaguerre et al. (16) reported that the J waves were indeed taller in the idiopathic VF group. Population-based studies are needed to define the upper limit of the “normal J-point elevation.” The location of the J-wave also varied in the different patient groups (Fig. 4).
Why is J-point elevation more common in idiopathic VF?
Simultaneous recordings of transmural ECG and intracardiac action potentials show that J-point elevation is the electrocardiographic representation of a voltage gradient created during the early phases of repolarization (28). This voltage gradient results from the different shapes of the action potential at different cardiac layers (28). In the epicardium, a higher density of transient outward potassium current (ITo) leads to a prominent notch during phase I of the action potential that is hardly noticeable in the endocardium (28). This normal difference in ITo current density in different myocardial areas plays a role in arrhythmogenesis. The Brugada syndrome represents one extreme of the spectrum. Here, a reduction in sodium inflow current caused by mutant sodium channels causes complete loss of the dome (phase II) of the action potential in the right ventricular epicardium where ITo current density is highest. Consequently, in the Brugada syndrome there is marked J-point and ST-segment elevation in the right precordial leads (V1 to V3) and a large dispersion of repolarization that allows phase-II reentry polymorphic ventricular arrhythmias to occur even in the absence of triggering extra-systoles (29). On the other side of the spectrum, vagal-mediated bradycardia might increase ITo current in the lateral left ventricle, leading to J-point and ST-segment elevation in V4 to V6 (early repolarization pattern) that is essentially not arrhythmogenic even in the presence of ventricular extra-systoles (25,30,31). One could speculate that J-point elevation in the inferior leads represents a moderately arrhythmogenic substrate that facilitates polymorphic ventricular arrhythmias but only in response to critically timed extra-systoles. This reasoning accommodates 2 observations: 1) J-point elevation in the inferior leads is more common in patients with idiopathic VF but is also fairly common in healthy control subjects; and 2) arrhythmias in idiopathic VF are triggered by extra-systoles that typically have a superior axis (6,14,32,33), suggesting an inferior origin. These extra-systoles, which have been mapped to the Purkinje-fibers network (32,33), have a very short coupling interval (19), suggesting that an abnormally short refractory period exists in the surrounding areas.
We might have underestimated the frequency of J waves for 2 reasons: 1) J-point elevation might not be present at all times in a given patient and might occur just before VF onset (8,14,16,34), and 2) Brugada syndrome is especially common in South East Asia (2,3). Similarly, most case reports describing J-point elevation in idiopathic VF are from the Far East (6–8,13–15,34). It is possible that J-point elevation is a genetic characteristic that was under-represented in our series. In contrast, we might have overestimated the frequency of J-point elevation by counting intraventricular conduction delay as “J-point elevation.” Interventions like heart-rate slowing (8,34), drug infusions (6,8,15,34,35) or signal-averaged electrocardiography (16) might be necessary to reliably distinguish intraventricular conduction block from J-point elevation. Of note, systematic evaluation of idiopathic VF patients with signal-averaged electrocardiography (performed in the Haissaguerre et al.  study) showed that only a minority of patients with J-point elevation had “late potentials.” This suggests that J-point elevation represents “early repolarization” rather than “delayed depolarization” abnormalities.
J-point elevation occurred more frequently among our patients with idiopathic VF than among similar control subjects. However, to attach a practical meaning to this finding, one must take into account that idiopathic VF is rare. That is best achieved by using the Bayes' law of conditional probabilities. Accordingly, finding a J-wave in a young adult would increase the probability of idiopathic VF from 3.4:100,000 to 11:100,000, a negligible difference. Thus, the incidental finding of a J-wave during screening should not be interpreted as a marker of “increased risk” because the odds for this fatal disease would still be roughly 1:10,000.
Similar arguments apply to patients with syncope, because the majority do not have idiopathic VF. Idiopathic VF patients generally have inducible VF during electrophysiologic studies. Yet, understanding the prognostic implications of VF induction in a patient without documented spontaneous arrhythmias is problematic (36). Consequently, the decision to perform an electrophysiologic study, in a patient with syncope and apparently normal heart, should be based on a highly malignant clinical history and not on the presence or absence of J-point elevation.
- Abbreviations and Acronyms
- confidence interval
- odds ratio
- ventricular fibrillation
- Received April 21, 2008.
- Revision received July 7, 2008.
- Accepted July 14, 2008.
- American College of Cardiology Foundation
- Brugada P.,
- Brugada J.
- Antzelevitch C.,
- Brugada P.,
- Borggrefe M.,
- et al.
- Nademanee K.,
- Veerakul G.,
- Nimmannit S.,
- et al.
- Viskin S.,
- Belhassen B.
- Belhassen B.,
- Shapira I.,
- Shoshani D.,
- Paredes A.,
- Miller H.,
- Laniado S.
- Belhassen B.,
- Viskin S.,
- Fish R.,
- Glick A.,
- Setbon I.,
- Eldar M.
- Shimizu W.,
- Matsuo K.,
- Takagi M.,
- et al.
- Kambara H.,
- Phillips J.
- Perez-Riera A.P.,
- Hiroshi Uchida A.,
- Schapachnik E.,
- et al.
- Zheng Z.J.,
- Croft J.B.,
- Giles W.H.,
- Mensah G.A.
- Yan G.X.,
- Antzelevitch C.
- Antzelevitch C.,
- Viskin S.
- Haissaguerre M.,
- Shoda M.,
- Jais P.,
- et al.
- Viskin S.,
- Rogowski O.