Inferolateral early repolarization (ER) patterns on standard electrocardiogram (ECG) are associated with increased risk for cardiac and arrhythmic death in general adult population cohorts. We sought to determine the prevalence of inferolateral ER on surface ECG in multiracial pre- and postadolescent populations and to analyze its association with age, race, gender, and ST-segment patterns. A retrospective review was conducted of all ECGs recorded from preadolescent (aged 8–12 years, n = 719) and postadolescent (aged 21–25 years, n = 755) patients seen at a large academic medical center between January 1, 2009, and December 31, 2010. The overall prevalence of inferolateral ER was similar in the preadolescent and postadolescent populations (17% vs 16%, NS). The prevalence of ER increased after puberty in male patients (16% to 25%, p <0.001) and decreased in female patients (18% to 9%, p <0.001). Prevalence of ascending early repolarization (benign variant) also increased in males after puberty (15% to 23%, p <0.004) and decreased in females (11% to 4%, p <0.001). There were no differences in the prevalence of the risk-associated horizontal/descending pattern (3% in both groups). Subgroup analysis was performed on ECGs from the cohort of outpatients without cardiac disease, and the statistical trends remained the same. In conclusion, the overall prevalence of inferolateral ER was higher in pre- and postadolescent populations than in adult populations. However, the prevalence of the risk-associated horizontal/descending ST-segment pattern was only 3%, comparable to prevalence rates in the adult population. The variations in prevalence by gender and age suggest a possible influence of reproductive hormones.
An inferolateral early repolarization pattern (ER) on standard 12 lead electrocardiograms (ECGs) is associated with increased risk for cardiac and arrhythmic death in general population cohorts and is more prevalent in patients with a history of unexplained (idiopathic) ventricular fibrillation than in age- and gender-matched controls. Various studies have shown a prevalence of ER ranging from 5% to 13% in the general population, and it appears even higher in specific populations, such as young athletes and black patients. However, differentiating at-risk subjects from those with a benign variation remains a clinical challenge. Recently, the pattern of a horizontal or descending ST segment after ER was identified as a marker of increased risk for arrhythmic death and for the development of idiopathic ventricular fibrillation, whereas the ascending ST-segment variant, which is more common in young athletes, appears benign. Multiple molecular studies have demonstrated that testosterone augments J-point elevation, and clinical studies suggest that testosterone may affect the incidence and prevalence of ER. We therefore analyzed the ECGs of pre- and postadolescent male and female patients in a multiracial population to determine the gender-specific prevalence and the prevalence of the higher-risk horizontal ST-segment pattern and to identify variations in prevalence associated with puberty.
Methods
All ECGs recorded from both inpatients and outpatients at a large, university-affiliated community hospital (Jackson Memorial Medical Center, University of Miami Miller School of Medicine, Miami, Florida) are maintained in a computerized database. We accessed all ECGs performed on patients in the age groups 8 to 12 years (preadolescent) and 21 to 25 years (postadolescent) from January 1, 2009, through December 31, 2010. Preparticipation screening ECGs among University of Miami athletes were not included in this database. We excluded all ECGs with intraventricular conduction abnormalities and paced rhythms. There were 719 ECGs in the preadolescent population (443 male patients) and 755 in the postadolescent population (351 male patients). Demographic data and ER patterns were recorded. ST-segment morphologies associated with ER were classified as upsloping or horizontal/downsloping. Medical records were reviewed for inpatient and outpatient status, medical history, and indications for ECGs. Indications were divided into 5 general categories: suspected cardiac symptoms; cardiac arrhythmia; structural heart disease; screening, including prechemotherapy, preoperative, and prestimulant therapy for attention-deficit disorder; and noncardiac illness.
To determine if the results could be applied to the general population, subgroup analyses were performed on outpatients who did not have underlying structural or arrhythmic cardiac disease (including hypertension and echocardiographic evidence of left ventricular hypertrophy). Subjects who had ECGs performed for suspected cardiac symptoms and who were subsequently diagnosed with cardiac disease were also excluded. In this cohort, there were 418 subjects in the preadolescent population and 496 in the post-adolescent. This study was approved by the Institutional Review Board of the University of Miami.
All ECG analyses were performed visually by physicians trained in ECG interpretation. ER was defined based on previously reported criteria : a slow deflection (“slurring”) of the downslope of the R wave ≥0.1 mV or a notched J point ≥0.1 mV in 2 contiguous inferior (II, III, aVF) or lateral leads (I, aVL, V 4–6 ), or both. Using criteria introduced by Tikkanen et al, we classified ST-segment morphology after ER as ascending if the ST segment was ≥0.1 mV at 100 milliseconds after the J point and as horizontal/descending if the ST segment was ≤0.1 mV at 100 milliseconds after J-point elevation ( Figure 1 ).
All readers were blinded to ECG indications, initial clinical interpretations, and computerized interpretations of the ECGs. Because of the possibility of inter-reader variation, the first 800 ECGs were analyzed by 2 readers blinded to age, gender, and race as a pilot to test diagnostic accuracy. Kappa score for ER diagnosis was .63, which was considered a statistically acceptable value. Therefore, the remainder of the ECGs were interpreted by a single blinded reader.
Categorical variables are expressed as number and percentages. Continuous variables are presented as mean ± SD. The chi-square test was used to compare dichotomized variables, and a 2-sample t test was used for continuous variables; p value <0.05 was considered statistically significant. The analysis was performed using PASW 17.0 (SPSS, Chicago, Illinois) software.
Results
Demographics, ECG indications, and prevalence of ER in the overall population are presented in Table 1 . There was a higher percentage of ECGs recorded from black and from male patients in the preadolescent population. The increase in the proportion of ECGs recorded from female patients in the postadolescent population may be attributable to the proportion of patients in the postadolescent age group on the obstetrics services. The prevalence of ER in the male population increased after puberty, from 16% in preadolescents (71 of 443) to 25% in postadolescents (88 of 351; p <0.001). In contrast, the overall prevalence in female patients decreased from 18% in preadolescents to 9% after puberty (p <0.001; Figure 2 ). There was no difference in mean heart rate between male pre- and postadolescents (78.0 vs 79.5, p = 0.4) or between female pre- and postadolescents (83.6 vs 83.4, p = 0.9).
| Variable | Preadolescent (Ages 8–12 Yrs, n = 719) | Postadolescent (Ages 21–25 Yrs, n = 755) | p Value |
|---|---|---|---|
| Male | 443 (62%) | 351 (47%) | <0.001 |
| Black | 375 (52%) | 350 (46%) | 0.04 |
| Inpatients | 202 (28%) | 244 (32%) | NS |
| ECG indications | |||
| Suspected cardiac symptoms | 342 (48%) | 312 (41%) | <0.02 |
| Arrhythmia | 43 (6%) | 12 (2%) | <0.001 |
| Structural heart disease | 96 (13%) | 19 (3%) | <0.001 |
| Screening | 162 (23%) | 137 (18%) | <0.04 |
| Noncardiac condition | 76 (11%) | 275 (36%) | <0.001 |
| Prevalence of ER in overall population | |||
| Total ER | 121 (17%) | 123 (16%) | NS |
| Inferior ER | 38 (5%) | 19 (3%) | NS |
| Lateral ER | 51 (7%) | 45 (6%) | NS |
| Inferolateral ER | 32 (4%) | 59 (8%) | <0.01 |
| Ascending | 99 (14%) | 98 (13%) | NS |
| Descending | 22 (3%) | 25 (3%) | NS |
| Lead-specific distribution of ER in ER population | |||
| Inferior ER | 38/121 (31%) | 19/123 (15%) | <0.02 |
| Lateral ER | 51/121 (42%) | 45/123 (37%) | NS |
| Inferolateral ER | 32/121 (26%) | 59/123 (48%) | <0.001 |
| Ascending | 99/121 (82%) | 98/123 (80%) | NS |
| Descending | 22/121 (18%) | 25/123 (20%) | NS |
Similar to the overall prevalence of ER, the overall prevalence of ascending ST-segment pattern increased in male patients after puberty (15% to 23%, p <0.004) and decreased in females (11% to 4%, p <0.001; Figure 2 ). The prevalence of the risk-associated horizontal/descending ST-segment pattern was low in both age groups (3% in each) and there was no difference in the prevalence when stratified by gender (male, 1% to 2%, p = 0.10; females, 7% to 5%, p = 0.19). The prevalence of both ascending and horizontal/descending patterns within the subset of patients with early repolarization was also similar in both groups ( Table 1 ).
Black patients had a higher prevalence of ER than white patients in both age groups. The prevalence was 22% in black preadolescents compared with 11% in white preadolescents, and 22% versus 11% in the respective postadolescent subgroups (p <0.001 for both; Figure 3 ). Black patients had a higher prevalence of ER than white patients in both age groups (23% vs 9% for preadolescent boys and 35% vs 18% for postadolescent men, p <0.0001 for both). In the preadolescent female population, there was no difference in prevalence between black and white girls (20% vs 16%, p <0.39); however, in the postadolescents, black women had a higher prevalence of ER (12% vs 5%, p <0.008), although this remained lower than in black male patients.
