The prevalence of early repolarization (ER) ranges from 2.3% to 29.3%, varying with age, race, and sex.¹ ER is also commonly found in high-performance athletes, with an incidence is around 1/3. However, the yearly incidence of sudden cardiac death in this population is very low.² For many years, ER was considered a benign finding on the electrocardiogram (ECG), but recent studies suggest that this pattern in leads other than V1-V3 (especially inferior leads) is associated with a higher risk of ventricular arrhythmias and sudden death.^1,3-5

Haïssaguerre et al.⁴ found a high prevalence of ER in patients who were resuscitated after cardiac arrest due to ventricular fibrillation (VF), and patients with ER presented twice the risk of VF recurrence.⁴ In the CASPER study, patients who had aborted sudden death with structurally normal hearts, ER was found as the cause in 7.9% of the sample.⁶ The degree of elevation of point J in the inferior leads can correlate with the magnitude of risk. The relative risk of death was 1.3 in those with 0.1 mV and three times in those with 0.2 mV;¹ the risk of ventricular arrhythmia was bigger in those with 0.2 mV, with the inclination of the ST segment resulting in a meaningful prognosis, with a horizontal or downward orientation associated with a higher arrhythmic risk.⁷

Screening and management of patients with ER and the risk of sudden death are a challenge.⁸ The American Heart Association (AHA) recommends that, when detected, standard ER should be investigated in case of unexplained syncope or family history of sudden death and arrhythmogenic ECG features.⁸ The low incidence of arrhythmic events also presents a major challenge for effective interventions that can reduce arrhythmic risk.²

Although quinidine and isoproterenol can be effective in managing patients with VF, the data supporting their use remains limited to small samples.^9,10 Due to the limited data in the literature, especially concerning the Brazilian population, this study aims to evaluate whether patients with different types of ER present a higher risk of ventricular arrhythmias and death.

Temporal changes in the ER pattern were not assessed in this study; only a single ECG was evaluated. ER was classified based on the morphology observed in leads: lateral (leads I, aVL, V5-V6), inferior (leads II, III, aVF), and infero-lateral (combination of these leads).

This retrospective cohort study included 478 patients from a university hospital with an electrocardiographic diagnosis of early ventricular repolarization between 2011 and 2022. The study was conducted in accordance with regulatory guidelines involving human subjects, and the research project was approved by the Institutional Research Ethics Committee (number 2021/0585).

Inclusion criteria encompassed all patients diagnosed (by cardiologists) with early ventricular repolarization (J point elevation ≥ 1 mm in at least two contiguous leads, with a slurring or notching pattern, excluding leads V1 to V3 – Figure 1) who underwent ECG evaluation at the university hospital and were older than 18 years between 2011 and 2022. Exclusion criteria comprised the impossibility of performing an electrocardiogram for measures of ventricular repolarization parameters, secondary changes from ventricular repolarization, and patients with pacemakers and other cardiac devices.

Figure 1

The search for the term “early repolarization” through the hospital database retrieved 2,172 patients. Of those, 923 were selected for review based on the inclusion criteria. After thorough evaluation, 478 patients were diagnosed with ER.

During follow-up, ventricular arrhythmias included sustained ventricular tachycardia and ventricular fibrillation as defined by ICD-10. Causes of death were primarily cardiac conditions, including sudden cardiac arrest.

A total of 2172 patients with a previous diagnosis of ER on the ECG were initially identified from the hospital database. Of these, 941 patients were selected for review based on specific inclusion criteria. After thorough evaluation by three cardiologists, 478 patients were confirmed to have a genuine diagnosis of ER and were included in the study. The remaining patients who did not fulfill the criteria for ER were used as controls (Figure 2 and Central Illustration).

Figure 2

Central Illustration

The study population was predominantly male (74%) with an average age of 45.6 years. Among them, 21.5% were active smokers, and 43% had some high school education. The prevalence of comorbidities such as diabetes mellitus, systemic arterial hypertension, dyslipidemia, coronary artery disease, and heart failure with reduced ejection fraction varied within the sample, with the lateral ER group showing a higher prevalence of systemic arterial hypertension (Table 1).

Table 1

	Lateral	Inferior	Inferolateral	Total
Variables	(n = 170)	(n = 98)	(n = 210)	n = 478
Age	48.5±4.95	45.5±7.78	43.2±5.46	45.57±6.56
Education
Some primary education	79 (46)	40 (41)	87 (41)	206
Primary education	36 (20.4)	19 (19)	45 (21)	100
Secondary education	29 (17)	19 (19)	38 (18)	86
Tertiary education	11 (6.4)	9 (9.1)	12 (5.7)	32
Unknown	7 (4.1)	9 (9.1)	17 (8.9)	33
Sex
Male	130 (76)	70 (71)	154 (73)	354
Female	40 (24)	28 (29)	56 (27)	124
Diabetes Mellitus	35 (20)	16 (16)	42 (20)	93
SAH	86 (50)	32 (33)	75 (36)	193
Dyslipidemia	14 (8.2)	5 (5.1)	12 (5.7)	31
Active smoker	35 (20)	19 (19)	49 (23)	103
CAD	17 (10)	12 (12)	14 (6.7)	43
HFrEF	5 (3)	2 (2)	2 (1)	9

SAH: Systemic arterial hypertension; CAD: coronary artery disease; HFrEF: heart failure with reduced ejection fraction.

During the follow-up period, 13 patients (2.7%) experienced ventricular arrhythmias, with varying incidence between ER groups. Significant differences were observed in age, sex, and heart failure with reduced ejection fraction between the groups Regarding mortality, 11 patients (2.3%) died, with differences in age between the ER groups (Table 2).

Table 2

	All-cause mortality		Ventricular arrhythmia
	(n=11)	p-valuec	(n=13) p-valued
Age	59±10.24	0.0012b	35.3±17.50	0.0519b
Education		0.3140a		0.2813a
Some primary education	3 (27)		8 (62)
Primary education	2 (18)		2 (15)
Secondary education	6 (55)		3 (23)
Tertiary education	0		1 (8)
Unknown	0		0
Type of ER		0.6201a		0.4758a
Lateral	5 (45)		5 (38)
Inferior	1 (9)		4 (31)
Inferolateral	5 (45)		4 (31)
Sex		1a		0.0255a
Male	8 (73)		13 (100)
Female	3 (27)		0
Diabetes mellitus	2 (18)	1a	1 (8)	0.4786a
SAH*	5 (45)	0.9709a	6 (46)	0.7765a
Dyslipidemia	0	1a	1 (8)	0.5865a
Active smoker	3 (27)	0.8334a	2 (15)	0.1588a
CAD	0	0.6099a	3 (23)	0.1031a
HFrEF	1 (9)	0.4504a	2 (15)	0.0221a

*SAH: Systemic arterial hypertension; CAD: coronary artery disease; HFrEF: heart failure with reduced ejection fraction; ER: early repolarization; a: chi-square test; b: Student’s t-test; c: compared with the non-death group; d: compared with the group without arrhythmia ventricular.

The Kaplan-Meier survival analysis demonstrated no statistically significant differences in survival rates between the groups of patients with different types of ER and the control group (p-value = 0.7 and p-value = 0.5 for mortality and ventricular arrhythmia, respectively). However, the analysis revealed varying cumulative risk for death and ventricular arrhythmia between different ER groups (Figure 3).

Figure 3

Cox regression analysis showed no statistically significant hazard ratios for different types of ER regarding mortality or ventricular arrhythmias. Only age demonstrated a significant hazard ratio for both outcomes (Table 3).

Table 3

	Death		Arrhythmia
	HR Gross	HR Adjusted	HR Gross	HR Adjusted
Early repolarization
Lateral	HR= 1.000	HR=1.000	HR=1.000	HR= 1.000
Inferior	HR=0.419 (IC= 0.049 - 3.589)	HR= 0.528 (IC= 0.061 - 4.580)	HR= 1.537 (IC= 0.412 - 5.731)	HR= 1.278 (IC= 0.336 - 4.856)
Inferolateral	HR=0.928 (IC= 0.268 - 3.211)	HR= 1.242 (IC= 0.354 - 4.365)	HR= 0.670 (IC= 0.180 - 2.497)	HR= 0.543 (IC= 0.143 - 2.054)
Age		HR= 1.080 (IC= 1.027- 1.135)	HR= 0.959 (IC= 0.925 - 0.995)
Sex
Male		HR=1.000
Female		HR= 0.988 (IC= 0.2605 - 3.750)

Recent findings from Liu et al.¹¹ suggest that ER is a transient phenomenon, appearing in baseline ECGs but potentially resolving in subsequent measurements. This transient nature could indicate a benign course, as discussed in their cohort study.

The elevation of the J point in the ECG in 12 leads, compatible with ER, has traditionally been viewed as an innocent finding in healthy individuals.^12,13

The study sample, as expected, consisted of middle-aged individuals with fewer comorbidities than typically seen in cardiology studies. Although not a randomized study, the characteristics of the groups were very similar, with only the lateral ER group showing a higher incidence of hypertensive patients, reducing the effect of selection bias, although it still exists. Previous studies have reported average ages of ER populations equal to or very similar to the sample in this study.¹

When analyzing the characteristics of the sample that died, age resulted in a statistically significant difference, confirming the expected association between older age and higher mortality risk. Regarding the incidence of ventricular arrhythmias, sex and heart failure with reduced ejection fraction (HFrEF) resulted in statistically significant differences. Patients with ventricular arrhythmia consisted of young males, with a notable percentage presenting HFrEF, indicating a predisposition to ventricular arrhythmogenesis through various cardiomyopathy mechanisms. Age, after adjustment in the Cox model, emerged as a protective factor against ventricular arrhythmia.³

Survival curves analyzed using the Kaplan-Meier method revealed high survival rates across all groups, with no statistically significant differences (p=0.7). However, the group with inferolateral ER exhibited the highest cumulative risk. Consistent with the findings from Tikkanen et al.¹ in 2009, patients with inferior ER showed a higher rate of cardiovascular mortality, especially when the J point was higher than 2mm. However, when analyzing mortality from all causes, no significant difference was found between the groups with lateral ER and inferior ER. The percentage of the sample free of ventricular arrhythmia was high across all groups, with no statistically significant differences (p=0.5). Unlike mortality, patients with inferior ER exhibited a greater cumulative risk of ventricular arrhythmia, while the infero-lateral group showed the lowest risk. Similar findings were reported by Tikkanen et al.¹ in 2009 regarding ventricular arrhythmia, with a greater incidence observed in the inferior ER group when the J point was higher than 2mm.¹

Although mortality did not significantly differ between groups, it was interesting to note the difference observed in the HR adjusted for age and sex in the group with inferolateral ER. This finding suggests a potential risk factor with 1.2 times more chance of death. However, the wide confidence interval, including zero, may be attributed to the small number of deaths. More recently, Haïssaguerre et al.⁴ observed a higher frequency of inferolateral ER in patients with sudden death aborted by ventricular tachycardia, corroborating the findings of this study that ER category is associated with a greater risk of ventricular arrhythmia and death.⁴

Despite making important contributions to this field of knowledge, this study has several limitations. Data were obtained retrospectively through the analysis of medical records, and many details depended on third-party descriptions, which sometimes resulted in incomplete or missing information. Data extraction from medical records was also limited by technical issues and restricted access to these documents. Additionally, the use of secondary data, such as the SIM, due to scarce information regarding deaths made it challenging to categorize the cause of death. Another limitation was the low number of outcomes (death and ventricular arrhythmia), which could compromise the sample’s power to detect statistically significant differences. It is essential for clinicians, particularly cardiologists, to recognize the significance of this new evidence and pay close attention to patients with these electrocardiographic findings.

Also, the design of a retrospective cohort study has several limitations that warrant consideration. Firstly, its retrospective design relying on pre-existing data from medical records introduces potential biases and limitations in data completeness and accuracy. The relatively small sample size and limited number of outcomes, particularly deaths and ventricular arrhythmias, may have impacted the study’s statistical power and ability to detect significant differences between ER groups. Additionally, the study’s single-center nature and exclusion criteria may limit the generalizability of findings to broader populations. Data quality and availability, and potential confounding factors not fully controlled, further add complexity to the interpretation of results. The observational nature of the study precludes the establishment of causal relationships between ER and outcomes. Longer follow-up periods and prospective, multicenter studies are needed to validate findings and address these limitations comprehensively.

In this study, we observed that patients with early ventricular repolarization, particularly in the lateral and inferior leads, did not exhibit a significantly higher incidence of ventricular arrhythmias or mortality. These findings suggest that the presence of ER alone may not be a strong predictor of adverse outcomes in our study population. However, further prospective and randomized studies are warranted to validate these results and inform clinical practice regarding the management of patients with early ventricular repolarization.