OUP user menu

★ Clinical update ★

Clinical significance of variants of J-points and J-waves: early repolarization patterns and risk

M. Juhani Junttila, Solomon J. Sager, Jani T. Tikkanen, Olli Anttonen, Heikki V. Huikuri, Robert J. Myerburg
DOI: http://dx.doi.org/10.1093/eurheartj/ehs110 2639-2643 First published online: 29 May 2012


The variations in the electrocardiographic patterns of J-point elevations, and the complex of J-points and J-waves in early repolarization (ER), in conjunction with disparities in associated sudden cardiac death (SCD) risk, have lead to a recognition of the need to carefully classify the spectrum of these observations. Many questions about the pathogenesis of J-wave patterns, and the associated magnitudes of risk, remain unanswered, especially in regard to the risk implications in certain high-prevalence subpopulations such as athletes, children, and adolescents. Interest in these electrocardiography (ECG) patterns has grown dramatically in recent years, in large part because of the frequency with which these patterns are observed on routine ECGs. In this review, we discuss the current knowledge on the prevalence of different J-point/J-wave patterns and estimates of the magnitude of mortality and SCD risk associated with J-point elevations and J-waves, in what has become known as ER patterns.

  • Early repolarization
  • Brugada syndrome
  • Sudden cardiac death


The J-point on the electrocardiographic waveform is historically defined as the junction between the end of the QRS complex and the beginning of the ST-segment.1,2 In 1953, Osborn3 described the presence of broad positive deflections originating from an elevated J-point, induced by experimental hypothermia and associated with ventricular fibrillation. He considered them currents of injury. They came to be known as J-waves bearing his name (Osborn waves) and have become a generally accepted marker for clinical hypothermia. However, during the same period of time, others noted less dramatic J-point elevations with concomitant J-wave deflections, primarily in the anterior leads, on electrocardiograms recorded from normal young individuals. This pattern was considered benign, even though the Osborn J-wave variant had been considered clinically significant at the same time.4,5 The normal variants in young subjects became defined by the term ‘early repolarization’ (ER). The ER pattern is more prevalent among males, African-Americans, and adolescents, and it is accentuated by vagal tone and hypothermia.610 In more recent years, the notion that the J-wave of ER was a universally benign normal variant, except when induced by exogenous factors came into question, as data emerged on variants of J-point elevation and J-waves that were associated with conditions that carried a risk of sudden cardiac death (SCD). These included the electrocardiographic patterns observed in the right precordial leads in the Brugada syndrome, and inferolateral ER or J-waves recently associated with an increased risk of mortality and SCD in case–control and general population studies.1118

The variations in the electrocardiographic patterns of J-point elevations, and the complex of J-points and J-waves in ER, in conjunction with disparities in associated SCD risk, have lead to a recognition of the need to carefully classify the spectrum of these observations.19 Many questions about the pathogenesis of J-wave patterns, and associated magnitudes of risk, remain unanswered, especially in regard to the risk implications in certain high-prevalence subpopulations such as athletes, children, and adolescents. Interest in these electrocardiography (ECG) patterns has grown dramatically in recent years, in large part because of the frequency with which these patterns are observed on routine ECGs. In this review, we discuss the current knowledge on the prevalence of different J-point/J-wave patterns and estimates of the magnitude of mortality and SCD risk associated with J-point elevations and J-waves, in what has become known as ER patterns.

Anterior J-point elevations and J-waves in the Brugada syndrome

The Brugada syndrome is characterized by a right bundle branch block pattern with ST-segment elevation and inversion of the terminal part of T-wave in the right precordial leads.18,20 The J-wave pattern may vary from time to time within the series of ECGs in individual patients, tends to be accentuated by increased vagal tone and fever, and can be unmasked or enhanced by the Class I membrane-active antiarrhythmic drugs, ajmaline, flecainide, and procainamide.2024 Exercise, catecholamine stimulation, and quinidine have been shown to normalize the Brugada ECG pattern, although some conflicting results have been reported between the Asian and European Brugada syndrome populations.20,21,25 There is also a considerable male predominance in the incidence of the syndrome.19

The ECG patterns within the spectrum of the Brugada syndrome are classified into three categories (Figure 1). J-point elevations are characteristic of all three patterns of Brugada-associated ECGs, the distinctions between the patterns reflected primarily in the J-waveforms following the J-points, and extending into the ST-segments and T-waves. The Type I pattern, J-point elevation with a coved J-wave–ST-segment configuration, is the most specific and considered the highest risk pattern, especially in symptomatic patients. The Type I ECG pattern is the only diagnostic ECG for the Brugada syndrome.20 Asymptomatic patients with Type I patterns are also at increased risk compared with the general population without Type I patterns or compared with those with Type II patterns, but the magnitude of risk is considerably lower. Moreover, some ECG patterns are associated with higher risk of symptoms or life-threatening events in the Brugada syndrome: specifically, higher J-point elevations, QRS durations >100 ms, and a prominent r′' in lead aVR.26 To date, no data suggest that athletes have a higher prevalence of Brugada ECG patterns, even though many athletes have ER patterns in the right precordial leads.27 Interestingly, inferolateral ER patterns have also been reported to be present in as high as 11–15% of Brugada patients and have been shown to have a strong adverse effect on the prognosis.2830

Figure 1

Brugada electrocardiography patterns. Brugada syndrome subtypes are shown. (Left) Typical Type I (coved type) Brugada syndrome electrocardiography pattern, (middle) Type II (saddleback) Brugada electrocardiography pattern, and (right) Type III Brugada electrocardiography pattern. Types II and III have an ascending ST-segment after the J-wave but diagnostic Type I electrocardiography has a descending ST-segment with T-wave inversions after the J-wave. Paper speed 50 mm/s, gain 10 mm/mV. Reprinted from Junttila et al. Prevalence and prognosis of subjects with Brugada-type electrocardiography pattern in a young and middle-aged Finnish population. Eur Heart J 2004;25:874–8. ©Oxford University Press.

The prevalence of the diagnostic Brugada ECG is ethnicity-dependent. Among the Asian population, the prevalence of the Type I Brugada pattern has been estimated to be ∼0.4% and in the European population 0–0.01%.3135 The annual incidence of life-threatening arrhythmias has been estimated to be 7.7% among patients with aborted SCD, 1.9% among patients with syncope, and 0.5% in asymptomatic subjects.36 Although the incidence of life-threatening arrhythmias in asymptomatic Brugada patients remains a debated issue,37,38 the majority of life-threatening or fatal arrhythmias in patients with the Brugada syndrome are nocturnal, likely due to the association of vagal tone with the amplitude of the J-wave.20,39

Although most investigators consider the pathophysiology of Brugada syndrome-associated J-waves to be regional ER, recent data suggest the possibility of delayed depolarization in the right ventricular outflow tract as a contributing mechanism to J-waves, arrhythmia expression, or perhaps both.4042

Inferolateral J-point elevations and J-waves–risk-associated early repolarization patterns

Inferolateral ER is characterized by a deflection in the R-wave descent (slurred pattern) or a positive deflection with a secondary r′' wave (notching pattern) in the terminal part of the QRS complex in at least two inferior (II, III, aVF) leads, in two lateral (I, aVL, V4–6) leads, or both. The association between SCD and inferolateral ER was first observed by Haissaguerre et al.11 in a case–control study of 206 patients who survived ventricular fibrillation in the absence of a defined cause (idiopathic ventricular fibrillation, IVF) and 412 matched control subjects. Inferolateral ER was observed in 31% of the IVF group, compared with 5% of the controls. The IVF patients had significantly greater amplitude of J-point elevation than controls, and subjects with extreme J-point elevation (>5 mm) had the highest occurrences of VF episodes.11 Multiple case–control studies have confirmed this association and revealed that inferolateral ER amplitude is strongly associated with vagal tone and hypothermia.1214,43,44 Additionally, exercise, quinidine, and catecholamines abolish the J-wave.8 Two independent family studies have recently suggested the inheritance of propensity to ER patterns. In a study from the Framingham Heart Study population, the siblings of ER subjects were twice as likely to have ER than non-ER subjects (OR 2.22, P< 0.05), and in a study on over 500 British families drawn from a general population cohort, ER was over two times more likely to occur in children of the family (OR 2.54, P= 0.005) if one of the parents had an ER ECG pattern.45,46 A recent paper also demonstrated the familial occurrence of the ER pattern in subjects with IVF and ER.47

The prevalence and prognosis of inferolateral ER has been studied extensively in three general population studies.1517 The first was conducted in Finland and included 10 864 middle-aged subjects representative of the general population who were enrolled into a population-based study of coronary heart disease between 1966 and 1972, with a mean follow-up of 30 ± 11 years. The prevalence of inferolateral ER recorded on ECGs at entry was 5.8% (inferior ER 3.5%, lateral ER 2.4%, or both 0.1%). Inferior ER was associated with an increased risk of cardiac mortality [risk ratio (RR) 1.28, P= 0.03], and inferior ER patterns with J-point elevations >0.2 mV was associated with cardiac mortality (RR 2.98, P< 0.001) and sudden arrhythmic death (RR 2.92, P= 0.01). In the same population, QTc durations >440 ms in males and 460 ms in females were associated with a smaller magnitude of increased risk for cardiac mortality (RR 1.20, P= 0.03), as was the Sokolow–Lyon voltage criteria for left ventricular hypertrophy (RR 1.60, P= 0.004). Recently, an additional study was published from the same population, where the inferior ER patterns were subgrouped into notched or slurred J-wave patterns and into ascending or horizontal/descending ST-segments following the J-wave.48 The risk for arrhythmic death did not differ between notched and slurred J-wave ER patterns, but the ST segment morphology distinguished high-risk patterns from benign patterns. Horizontal/descending ST-segment in the inferior leads was associated with a significant risk of arrhythmic death (RR 1.62, 95% CI 1.19–2.21), and this pattern combined with a 2-mm J-point elevation further increased the risk (RR 3.37, 95% CI 1.75–6.51). Ascending ST-segments after J-waves did not carry an increased risk (RR 1.01, P= NS). Coincidentally, ascending ST-segments after J-waves was the most prevalent pattern in athletes. Similar results were described recently from IVF populations by the Tel Aviv group where horizontal/descending ST-segment after J-point improved the ability to distinguish IVF patients from matched controls.49

The second study on ER prevalence and prognosis was conducted in a German population subset of 1945 subjects (age range 35–74 yrs) from the KORA/MONICA cohort.16 The prevalence of inferolateral ER (13.1%), with an inferior ER prevalence of 7.6%, was higher than observed in the Finnish study. In the German study, the risk of cardiac death was increased among inferior ER carriers and was strongly associated with male gender and younger age at the time of the ECG recording. Similar to the Finnish data, males with inferior ER in the age group of 35–54 years had over four-fold risk of cardiac death.

The third general population study of prevalence and incidence of ER was conducted in the Nagasaki area in Japan in a population of atomic bomb survivors.17 Subjects in this study had biennial physical examination including ECG during the total follow-up period of 46 years. Interestingly, incident ER findings during the follow-up were found in 779 subjects and ‘stable’ ER in 650 subjects resulting in a total prevalence of 29.3%. The mortality rates in this study were also surprising. Subjects with ER did not have increased risk of all-cause mortality or cardiac death, but the risk of sudden unexpected death was significantly elevated among ER subjects, as in studies conducted in Western populations.

Another recent study explored the prognostic significance of ER among chronic coronary disease patients with ICD.50 In this case–control study, the prevalence of inferior ER was significantly higher among patients who had appropriate ICD therapy for ventricular arrhythmias than in patients who were arrhythmia free (28 vs. 8%, P= 0.011) even after adjustment for LV ejection fraction.

One noteworthy phenomenon is the considerable age dependency of ER prevalence among males. Early repolarization is overrepresented among young males compared with females, but the higher prevalence in males declines rapidly during middle age. This suggests a potential influence of testosterone as a modifier of J-wave/ER expression, an association also observed in the Brugada syndrome.45,51 This male preponderance as a function of age is primarily attributable to the frequency of the benign ascending ST-segment pattern of ER.

Benign patterns of J-point elevation and J-waves/early repolarization

The prognostic significance of ER was first comprehensively studied in a general population, in which subjects with J-point elevation in any lead, including anterior leads, was the only inclusion criterion.27 The presence of J-waves was not required. In this study, ER was not found to be associated with increased mortality risk. Thus, it appears that J-point elevation or ST-segment elevation itself is not a mortality risk factor in the absence of notching and/or slurring of the terminal portion of QRS or the formation of apparent J-waves. Among the inferolateral ER pattern carriers, ascending ST-segment was not associated with increased mortality risk, as mentioned before.48 Similarly in Brugada syndrome ECGs, the pattern of the ST-segment morphology plays a role in risk assessment. Several general population-based studies have shown that Type II or III Brugada ECG finding in a routine ECG screening in otherwise healthy individual without personal or family history of SCD or life-threatening arrhythmias is a benign finding.26

J-point elevations and J-waves/early repolarization in athletes

These patterns have been observed in routine ECG recordings from asymptomatic athletes for many years and have been considered to be normal variants. It is still generally accepted that the most are indeed benign. However, the association of inferior ER with SCD has been recently described in an athlete sudden cardiac arrest (SCA) population from Italy.52 Inferior ER prevalence was significantly higher in athlete SCA population compared with control athlete population (14.3 vs. 2.1%, P= 0.017). In the same study, ST-elevation absence after the J-point elevation was overrepresented in the SCA victims. Similar findings regarding ST-segment morphology were found in the Finnish study and in the study among IVF patients by the Tel Aviv group et al.48,49 In a recent study of ER among athletes, many distinctive characteristics between inferior and lateral ER carriers were observed. The overall prevalence of ER (30%) was strikingly high, as was as the prevalence of inferior ER (20%) in the mixed ethnicity athlete population from South Florida.6 The left ventricular hypertrophy (LVH) pattern determined by Sokolow–Lyon voltage criteria was present in 25% of inferior ER carriers, but was even more frequent among lateral ER carriers (40%). Interestingly, African-American athletes did not have higher prevalence of inferior ER, but did have significantly more lateral ER. The proposed ‘benign’ ascending ST-segment morphology with the ER pattern was overrepresented among the athletes with only 4% overall prevalence of horizontal/descending ST-segment. The majority (88%) of the ER patterns with horizontal/descending were detected in inferior leads.36 In this study during the 10 years of preparticipation ECG screening, no SCDs or symptomatic ventricular arrhythmias occurred. Another recent study demonstrated similar findings of ER prevalence in another athlete population and also showed that the prevalence of ER increases during peak training season.53 The increased ER occurrence was independent of echocardiographic findings related to ‘athletes heart’, i.e. increased LV remodelling.


Although there are many common features among the various patterns of J-point elevations and J-waves on electrocardiograms that demonstrate ER, the associated clinical risks vary from common benign incidental findings on routine ECGs to patterns suggesting an increased risk of SCD, as in Type I Brugada ECG patterns and hypothermia-induced J-wave changes. The precise incidence of, or propensity to, J-point elevation and J-wave generation among a normal general population, might be even more common than thought, expressing variably, associated with both specific circumstances or occurring in an apparently random fashion. For example, J-waves are very common during hypothermia and during very high vagal tone in some circumstances (e.g. trained athletes), and less common during routine ECG screening. The Brugada syndrome and inferolateral ER share the same pathophysiological characteristics of temporal variation in the ECG pattern, normalization during adrenergic states and quinidine administration, and male predominance (Table 1).

View this table:
Table 1

Common features of early repolarization and Brugada syndrome

In addition to hypothermia, there are at least three different clinically relevant J-wave/ER patterns reported to date. In the Brugada syndrome, the influence of the ECG pattern is relatively straightforward: high penetrance, familial inheritance pattern, and high risk associated with the Type I pattern. It is noteworthy that J-point elevations are observed in all three types of Brugada ECG patterns, the differences between the three being in the J-waves and ST–T wave patterns. The second association, inferolateral patterns associated with IVF, seems to be an entity that resembles the Brugada syndrome, with similar risk magnitude and clinical features. Finally, the inferolateral ER pattern associated with increased mortality risk in the general population may be a risk modifier interacting with other mechanisms of transient peaks of risk, such as ischaemia or other proarrhythmic events. As the Kaplan–Meier curves of inferior ER show in the Finnish study,12 the temporal incidence of sudden arrhythmic death increased around the same age (50–70 years) as the mean age of SCD among coronary disease patients, thus suggesting that this more prevalent modality of inferolateral ER may indeed serve a modifying role during acute coronary events. All of the population-based studies suggesting the adverse prognostic value of ER focused on the presence of specific J-waves, not only the presence of consistent, ascending ST-segment elevation.1517 Also in the case–control studies only the presence of J-waves have distinguished patients with idiopathic VF from controls, whereas ST-segment elevation have not provided any prognostic value. Therefore, it seems legitimate to draw conclusions that it is' mainly the presence of J-waves in ER patients that result in an increased risk of arrhythmic events. Additionally, the recent categorization of ascending ST-segment after J-waves in general population as a benign trait supports this hypothesis.48

Although clinicians must be aware of the arrhythmogenic potential of ER, there are emerging data on the ST-segment morphology that relieve some of the pressure on clinical interpretation of the relevance of incidental ER patterns created by the publications from Haissaguerre et al.11 and Tikkanen et al.15 There is also increasing data to suggest that lateral ER may be a benign finding.1517 In athletes, lateral ER is associated with LVH voltage and African-American ethnicity, suggesting a normal physiologic variant to the pattern. Inferior ER amongst athletes lacked such correlation.6

In a recent commentary, the problem of confusing terminology was explored.19 Although the terminology surrounding the terms of J-point elevation and J-waves, vs. ER, can be misleading in many ways, part of our intent in this review and discussion is to clarify the language and its implications, blending historical concepts with contemporary observations. The specific patterns that have been explored in recent years have been shown to convey risk for arrhythmic death in two separate general population-based studies. Therefore, regardless of the terminology, this variant has the potential for arrhythmia prediction that can be seen in an ECG tracing several years or even decades before the actual intermittent risk for sudden death. The patterns of J-point elevations and J-waves or ER patterns appear to reflect a continuum of risk for arrhythmias (Figure 2). Regardless of whether the patterns have their origins in variants of depolarization or repolarization processes,54 or contributions by either or both, the patterns appear to have usefulness for risk prediction in a number of clinical circumstances.

Figure 2

Inferolateral early repolarization patterns and magnitude of sudden cardiac death risk. Figure illustrates the estimated cardiac mortality risk associated with the corresponding electrocardiography pattern (highest risk on the top of the pyramid and lowest on the bottom) and also the estimated prevalence of the pattern in general population (width of the pyramid).

The inferolateral ER patterns evoke the inverse problem of electrophysiology in a new light. The subtle yet important differences in ECG patterns seem to reflect variable and intermittent pathophysiological molecular mechanisms. Understanding these mechanisms, their ECG manifestation and the corresponding risk profile are an ongoing challenge.


This study was funded by Fondation Leducq, Paris, France, and Finnish Foundation for Cardiovascular Research, Helsinki, Finland.

Conflict of interest: none declared.


View Abstract