CLINICAL STUDIES
The Early Repolarization Normal Variant
Electrocardiogram: Correlates and Consequences
Arthur L. Klatsky, MD, Rudolph Oehm, MD, Robert A. Cooper, MD, Natalia Udaltsova, PhD,
Mary Anne Armstrong, MA
PURPOSE: We compared the characteristics and outcomes of
patients with early repolarization electrocardiograms (ECGs)
with those who had normal ECGs.
METHODS: In 1983 to 1985, we collected photocopies of 2234
selected ECGs from 73,088 patients undergoing health exami-
nations. Excluding 153 ECGs with missing data or that were
judged to be abnormal, the remaining ECGs were reinterpreted
in 2000 by cardiologists as showing early repolarization (n
670), or being borderline (n 330) or normal (n 1081). Char-
acteristics and outcomes of persons with early repolarization ECGs
were compared with those who had normal ECGs using analysis of
variance, logistic regression, or proportional hazards models. In-
formation on exercise was available in 325 patients.
RESULTS: Patients with early repolarization were more likely
to be male (81% [n 583] vs. 33% [n 360]), 40 years old
F
irst described only 34 years after the Dutch physi-
ologist Willem Einthoven invented the electrocar-
diogram (ECG) in 1902 (1), the early repolariza-
tion has also been called unusual RT segment
deviation and normal RST elevation variant and may
be related to the persistent juvenile pattern character-
ized by right precordial T inversions (2 4). Its prevalence
has been estimated between 1% and 5% of healthy adults
(1,3,57), and is thought to be more common in young
men, perhaps especially those who are athletic (8,9).
Some (5,10 12) have reported a higher prevalence in
blacks, whereas others (6,13,14) question this associa-
tion.
Although not considered a marker of cardiovascular
disease, early repolarization has practical importance be-
cause it may mimic the ECG of acute myocardial infarc-
tion, pericarditis, ventricular aneurysm, hyperkalemia, or
From the Division of Cardiology, Department of Medicine, and the
Division of Research, Kaiser Permanente Medical Care Program, Oak-
land, California.
This work was supported by a grant from the Kaiser Foundation
Research Institute, Oakland, California. Data collection in 1983 to 1985
was supported by a grant from the Alcoholic Beverage Medical Research
Foundation, Baltimore, Maryland.
Requests for reprints should be addressed to Arthur L. Klatsky, MD,
Kaiser Permanente Medical Center, 280 West MacArthur Boulevard,
Oakland, California 94611, or hartmavn@pacbell.net.
Manuscript submitted September 4, 2002, and accepted in revised
form April 11, 2003.
2003 by Excerpta Medica Inc.
All rights reserved.
(60% [n 441] vs. 37% [n 403]), black (48% [n 384] vs.
26% [n 280]), and more athletically active (mean [ SD],
10.4 1.3 hours per week of activity vs. 6.4 1.2 hours per
week of activity) than those with normal ECGs. Patients with
early repolarization were not more likely to be hospitalized
(hazard ratio [HR] 1.0; 95% confidence interval [CI]: 0.9 to
1.2) or to die (HR 0.8; 95% CI: 0.6 to 1.2) during follow-up
than those with normal ECGs. Outpatient diagnoses were not
more common in those with early repolarization; arrhythmias
were actually less common (P 0.01).
CONCLUSION: Although especially prevalent in young, ath-
letic, black men, early repolarization is not rare in other pa-
tients. The long-term prognosis of early repolarization is
benign. Am J Med. 2003;115:171177. 2003 by Excerpta
Medica Inc.
hypothermia. A dilemma may occur when a patient with
ST variant presents with chest pain and no prior available
electrocardiogram (1517). Unnecessary or incorrect di-
agnostic tests, therapeutic decisions, including adminis-
tration of thrombolytic drugs (18), and hospital admis-
sions might result from misinterpretation.
Recognition is relatively easy in a healthy young person
without evidence of heart disease. These patients should
be reassured as harm may result from concern about po-
tentially serious heart disease (5). After observing several
patients with early repolarization-related hospital admis-
sions, we sought to clarify the characteristics and out-
comes of patients with early repolarization, using a col-
lection of ECGs that had been obtained at screening
health examinations. We were particularly interested in
confirming our observation that these patients were at
increased risk of hospitalization for chest pain.
METHODS
Subjects and Questionnaires
The Institutional Review Board of the Kaiser Permanente
Medical Care Program approved the study protocols. The
study ECGs were derived from 73,088 adults who volun-
tarily took health examinations in Oakland, California, in
1983 to 1985. Of these examinees, 32,047 (44%) were
men, 40,286 (55%) were white, 22,979 (31%) were black,
0002-9343/03/$see front matter 171
doi:10.1016/S0002-9343(03)00355-3
 Early Repolarization Normal Variant Electrocardiogram/Klatsky et al

Table 1. Characteristics of Patients with Normal, Borderline, and Early Repolarization Electrocardiograms
Electrocardiogram Group
Early
Normal Borderline Repolarization Early Repolarization versus Normal
(n 1081) (n 330) (n 670)
Difference or Odds Ratio
Characteristic Number (%) or Mean ( SD) (95% Confidence Interval)* P Value
Male sex 360 (33) 222 (67) 583 (87) 15 (12 to 20) 0.001
Race
White 668 (62) 143 (43) 223 (33) 0.36 (0.29 to 0.49) 0.001
Black 280 (26) 131 (40) 348 (52) 2.9 (2.3 to 3.6) 0.001
Hispanic 43 (4) 12 (4) 14 (2) 0.43 (0.18 to 0.81) 0.005
Asian 86 (8) 45 (14) 73 (11) 1.3 (0.8 to 2.1) 0.2
Age (years) 46 16 42 14 37 13 10 (8 to 11) 0.001
Age 40 years 403 (37) 163 (50) 441 (66) 3.2 (2.6 to 4.0) 0.001
Smoke cigarettes 250 (23) 89 (27) 31 (209) 1.2 (0.9 to 1.5) 0.1
Drink alcohol daily 275 (24) 89 (27) 193 (29) 1.4 (1.1 to 1.9) 0.01
Body mass index (kg/m2) 26 5 25 4 24 3 1 (0.4 to 1.4) 0.03
Systolic blood pressure (mm Hg) 125 18 123 16 121 16 5 (6 to 2) 0.001
Diastolic blood pressure (mm Hg) 75 10 74 10 73 9 2 (3 to 1) 0.02
Total cholesterol (mg/dL) 218 50 214 46 200 45 18 (23 to 13) 0.001
Glucose (mg/dL) 97 28 96 21 93 24 5 (8 to 2) 0.01
* Estimated by age-adjusted logistic regression or general linear models (except for age itself).

and 7486 (10%) were Asian. The examination included
questions about demographic characteristics (including
self-classified race), habits and medical history, and
health measurements, including ECGs (19). Certain
medical history items were compressed into composites
(20) of cardiovascular history (yes to any of 12 ques-
tions) and gastrointestinal history (yes to any of nine
questions). Recreational exercise information was sup-
plied on a supplementary questionnaire in 1984 to 1985
in 12,513 (17%) of examinees. A composite variable of
vigorous exercise was constructed, consisting of eight
types: jogging, cycling (outdoor and indoor), tennis,
swimming, aerobic dance, exercise class, and calisthenics.
Follow-up medical events were ascertained by computer
search of databases through 1998, including hospitaliza-
tions within the Kaiser system and death certificate diag-
noses from the California Automated Linkage System
(21). Computerized outpatient diagnostic data were
available from 1995 to 1999 in 45,528 examinees (26,245
women and 19,283 men).
Electrocardiographic Data
All ECGs were from the outpatient clinic and consisted of
nine leads: I, II, III, aVR, aVL, aVF, V1, V3, and V5. These
were interpreted in 1983 to 1985 by five cardiologists us-
ing a mark-sense coding form that included a normal
variant but no early repolarization option. Coding
rules allowed several possibilities, including several

Table 2. Selected Electrocardiographic Correlates of Normal, Borderline, and Early Repolarization Patterns
Electrocardiogram Group
Early
Normal Borderline Repolarization Early Repolarization versus Normal
(n 1081) (n 330) (n 670)
Difference or Odds Ratio
Correlate Number (%) or Mean SD (95% Confidence Interval)* P Value
Heart rate (beats per minute) 68 11 63 10 61 10 7 (6 to 9) 0.001
50 beats per minute 19 (2) 18 (6) 45 (7) 4.1 (2.6 to 6.5) 0.001
60 beats per minute 143 (13) 82 (25) 191 (29) 2.5 (2.0 to 3.2) 0.001
Frontal plane axis (degrees) 43 25 50 24 56 23 13 (11 to 16) 0.001
Slurred R wave (J wave) 6 (0.6) 13 (4) 193 (29) 10 (5 to 23) 0.001
Any precordial T wave 10 mm* 25 (4) 22 (10) 145 (37) 3.6 (2.6 to 5.1) 0.001
SV1 plus RV5 (mm)* 22 7 27 9 31 10 11 (8 to 14) 0.001
SV1 plus RV5 35 mm* 36 (6) 28 (13) 160 (40) 10 (7 to 15) 0.001
* Estimated by age-adjusted logistic regression or general linear models.

172 August 15, 2003 THE AMERICAN JOURNAL OF MEDICINE Volume 115
 Early Repolarization Normal Variant Electrocardiogram/Klatsky et al

Table 3. Recreational Exercise Habits among 325 Patients Who Responded to a Supplemental
Questionnaire in 1984 to 1985
Early
Normal Repolarization
(n 209) (n 116) Odds Ratio
Exercise Type Number (%) or Mean SD (95% Confidence Interval)* P Value
Jogging 16 (8) 26 (22) 2.7 (1.45.6) 0.004
Walking 77 (37) 46 (40) 1.4 (0.92.3) 0.16
Calisthenics 27 (13) 24 (21) 1.6 (0.82.9) 0.16
Exercise class 19 (9) 7 (6) 0.5 (0.21.2) 0.09
Gym workout 14 (7) 15 (13) 1.5 (0.73.3) 0.33
Tennis 3 (1) 13 (11) 6.3 (1.725) 0.005
Bike outdoors 11 (5) 10 (9) 1.2 (0.53.1) 0.66
Stationary bike 18 (9) 4 (3) 0.4 (0.11.3) 0.14
Swimming 14 (7) 7 (6) 1.0 (0.42.6) 0.96
Any vigorous exercise 75 (36) 56 (48) 1.5 (1.02.4) 0.11
Hours per week of any 6.4 1.2 10.4 1.3 5.8 (1.89.9) 0.02
vigorous exercise
* Estimated with age-adjusted logistic regression or general linear models.

Eight types: jogging, cycling (outdoor and indoor), tennis, swimming, aerobic dance, exercise class, calisthen-
ics.

non-ST variations of normal. Most ECGs (57,311/
73,088 [78.5%]) were called normal. Thus, early repo-
larization ECGs could not be identified reliably from the
original ECG readings. Because of the desire to study
early repolarization, from 1982 to 1984 the cardiologists
set aside copies of ECGs that had possible early repolar-
ization, plus the next two consecutive ECGs as controls;
this yielded 2234 ECGs.
Blinded Reinterpretation of Electrocardiograms
Before rereading in 2000, all ECGs were shuffled and
identifying information was removed. Written com-
ments were covered with opaque tape. Two cardiologists
interpreted each ECG, classifying them as abnormal, nor-
mal, borderline early repolarization (defined as 0.5- to
0.9-mm ST elevation), or early repolarization (defined as
1.0-mm ST elevation). Disparities were resolved by
consensus, with no ECG classified as early repolariza-
tion without agreement. Tracings with missing data
(usually race, n 122) or that were judged as abnormal
(n 31) were excluded, leaving 2081 ECGs for analysis.
Readings included heart rate, mean frontal plane elec-
trical axis, subjective (yes/no) ascertainment of slurred
R-wave downslope (J wave) and, for early repolariza-
tion ECGs, the lead with the maximum ST elevation. A
sample of 1245 persons had additional measurements for
each lead, including magnitude of R, S, and T waves and
ST elevation to the nearest 0.5 mm. Duration of the QRS
complex was not measured. Cutpoints chosen for analy-
sis included precordial lead T waves 10 mm, ST eleva-
tions 2 mm (marked) or 3 mm (very marked),
and SV1 plus RV5 35 mm.
Statistical Analysis
Characteristics of patients with early repolarization ECG
patterns were compared with those of patients who had
normal ECGs using age-adjusted general linear models

Table 4. Hospitalizations Through 1998 among 670 Patients with Early Repolarization versus Remaining 72,418 Patients Who
Underwent Health Examinations in 1983 to 1985
All Early
Examinees Repolarization
Cause (9th International Hazard Ratio
Classification of Diseases Codes) Number (%) (95% Confidence Interval)* P Value
Any (except pregnancy-related) 21,826 (30) 136 (20) 1.0 (0.71.1) 0.9
Cardiovascular (390459) 7335 (10) 46 (6.8) 1.3 (1.01.7) 0.1
Noncardiovascular 19,543 (27) 117 (17) 0.9 (0.81.1) 0.5
Coronary disease (410414) 3262 (4.5) 19 (2.8) 1.1 (0.71.7) 0.8
Symptoms and signs (780799) 2728 (3.7) 16 (2.4) 1.0 (0.61.6) 1.0
* Adjusted for age, sex, race, education, marital status, and smoking using proportional hazards models.

August 15, 2003 THE AMERICAN JOURNAL OF MEDICINE Volume 115 173
 Early Repolarization Normal Variant Electrocardiogram/Klatsky et al

for mean values and using age-adjusted logistic regres- Table 5. Selected Outpatient Diagnoses from 1995 to 1999 in
sion analyses (reported as odds ratios with 95% confi- Men with Early Repolarization versus All Examinees by Age at
dence intervals) for categorical values. Stratified analyses Examination*
were performed (if possible) by age, sex, and race. Some All Early
descriptive data about patients with borderline ECGs are Persons Repolarization
presented, but the statistical comparisons involve only Diagnosis Rate Per 1000 Men P Value
those with definite early repolarization or normal ECGs,
unless otherwise noted. Hypertension
40 years old 99 70 0.2
Because only hospitalizations in Kaiser facilities could
40 years old 201 183 0.5
be determined, patients were followed until leaving the
Coronary disease
Kaiser program or December 1998, whichever came first. 40 years old 11 5 0.4
Cox proportional hazards models were used to compare 40 years old 100 84 0.5
the 670 patients with early repolarization with the other Arrhythmia
72,418 examinees. These models had seven covariates 40 years old 6 0 0.3
(age, sex, race, education [no college, some college, col- 40 years old 44 8 0.03
lege graduate], marital status [married, never, formerly], Cardiovascular symptoms
smoking [never, ex-smokers, less than one pack per day, 40 years old 9 0 0.2
one pack per day], and alcohol intake [never, ex- 40 years old 30 15 0.3
drinker, one drink per day, one drink per day]), and Heart failure
40 years old 2 0 0.5
are reported as hazard ratios. Mortality was ascertained
40 years old 26 23 0.4
using records from California. Comparisons of outpa-
tient diagnoses were made with chi-squared tests in age- * 19,283 male examinees remaining in program (7411 40 years old);
stratified groups. P values 0.05 (two-sided) were con- 318 of whom had early repolarization (187 40 years old).
sidered statistically significant.
those with normal ECGs (OR 1.10; 95% CI: 0.89 to
1.35).
RESULTS
Electrocardiographic Correlates
The blinded readings yielded 1081 normal, 330 border- Those with early repolarization had lower mean heart
line, and 670 early repolarization ECGs. The prevalence rates and more rightward mean frontal plane axes than
of early repolarization was 0.9% overall (670/73,088), those with normal ECGs (Table 2). The pattern of early
1.8% (583/32,047) in men, 0.2% (87/41,041) in women, repolarization was more evident in the limb than in the
0.6% (223/40,286) in whites, 1.5% (348/22,979) in precordial leads in 99 (15%) of the 670 patients; 494
blacks, and 1.6% (73/4486) in Asians; in men 40 years (74%) had marked ST elevations (2 mm), and 199
old, the prevalence was 1.5% (137/9339) in whites, 4.4% (30%) had very marked elevations (3 mm). Although
(211/4840) in blacks, and 3.1% (35/1112) in Asians. Men present in only 29% (193/670) of patients, the J wave
comprised 87% (n 583) of those with early repolariza- was fairly specific for early repolarization (Table 2).
tion (Table 1); this male preponderance was similar in Giant T waves (10 mm) were much more prevalent in
whites (187/214 [87%]), blacks (299/349 [86%]), and those with early repolarization (Table 2); high precordial
Asians (49/55 [89%]). Patients with early repolarization QRS voltage was somewhat more common.
were more likely than those with normal ECGs to be black
Exercise Habits
(age-adjusted odds ratio [OR] in men 3.3; 95% confi-
Most types of recreational exercise were reported more
dence interval [CI]: 2.6 to 4.1; age-adjusted OR in women
often by those with early repolarization (Table 3). The
3.6; 95% CI: 2.2 to 6.1) than white. Asians were more
proportions reporting no recreational exercise were 33%
likely than whites to have borderline but not definite early
(68/209) of those with normal ECGs, 29% (20/68) of
repolarization ECGs. Smoking and alcohol habits did not
those with borderline ECGs, and 23% (27/116) of those
differ significantly in the ECG groups. Those with early
with early repolarization (P 0.01).
repolarization had slightly lower body mass index, sys-
tolic and diastolic blood pressures, and total blood cho- Subsequent Illness and Mortality
lesterol and glucose levels. Patients with early repolariza- Patients with early repolarization were not at significantly
tion were more likely to report no illnesses (OR 1.31; greater risk of hospitalization for any cause, or for se-
95% CI: 1.04 to 1.65) and less likely to have gastrointes- lected diagnoses, including cardiovascular conditions
tinal illnesses (OR 0.74; 95% CI: 0.59 to 0.93). There (Table 4). Of all 2728 patients hospitalized for symptoms,
was no difference in cardiovascular illnesses at baseline 53% (n 1455) were hospitalized for chest pain, includ-
when comparing patients with early repolarization and ing 10 (1.5%) of those with early repolarization, and 1445

174 August 15, 2003 THE AMERICAN JOURNAL OF MEDICINE Volume 115
 Early Repolarization Normal Variant Electrocardiogram/Klatsky et al

Figure. Electrocardiogram of a sedentary 43-year-old black man. Features of the early repolarization pattern include counterclock-
wise electrical rotation of the precordial leads (no S in V3), concave upward ST-segment elevations (most marked in V3), and distinct
notching of the R wave downstroke (J wave), most marked in V5 and V3.

(2.0%) of the remaining patients. There were 37 deaths
through 1998 in the 660 patients with early repolariza-
tion, 11 (27%) of which were attributed to cardiovascular
conditions, and 7956 deaths in the remaining patients
(3071 [39%] from cardiovascular conditions), yielding
an adjusted mortality ratio of 0.8 (95% CI: 0.6 to 1.1; P
0.2) comparing the two groups.
Computerized data about outpatient diagnoses from
1995 to 1999 included 26,245 women (53 with early re-
polarization) and 19,283 men (318 with early repolariza-
tion); because of the small number of women, we present
age-stratified data only in men (Table 5). For most car-
diovascular conditions, there were slightly lower rates for
those with early repolarization, although the difference
was statistically significant only for arrhythmias in men
aged 40 years.
DISCUSSION
As described by early observers (35,22,23), early repo-
larization consists of concave upward ST-segment eleva-
tions (1.0 mm) like a cord suspended from the de-
scending R limb at one end and the apex of the T wave at
the other (24). These are usually in the precordial leads
V2 to V5 (16,18) but are sometimes in leads II, III, and
aVF. Other features include bradycardia, tall R or T waves
(or both), counterclockwise electrical rotation of precor-
dial leads with rapid QRS transition (i.e., abrupt change
from an rS to R or Rs), and a notched or slurred R wave
downstroke (J wave) (Figure). Occasionally, T waves
may be inverted or biphasic in one or more leads. Except
in aVR, there are no reciprocal ST depressions suggestive
of localized epicardial injury or PR depressions suggestive
of pericarditis (25). The pattern may normalize with ex-
ercise or isoproterenol administration (5,6,26 28).
The electrophysiologic basis for early repolarization is
not known (2,29). Explanations include hypervagoto-
nia (7), asthenic habitus (7,24), and early epicardial
repolarization before depolarization of the entire myo-
cardium (4,5). Normalization by isoproterenol and exag-
geration by beta-blockade have led to the suggestion that
it represents enhanced activity of the right sympathetic
nerve (28,30). In the absence of data supporting the con-
cept that early repolarization represents selective early
epicardial repolarization (28,30), a recent editorial was
entitled Early repolarization: an underinvestigated mis-
nomer (2).
Not much is known about the epidemiology of early
repolarization (2). Most studies have reported that the
condition is associated with a good prognosis, that there

August 15, 2003 THE AMERICAN JOURNAL OF MEDICINE Volume 115 175
 Early Repolarization Normal Variant Electrocardiogram/Klatsky et al
is variability from one ECG to another in the same pa-
tient, and that there is a tendency for the abnormality to
resolve with time (5,10,24,31). Recently, there has been
renewed interest because of similarities to the arrhyth-
mogenic Brugada syndrome (29,32).
Our data confirm the high prevalence of early repolar-
ization in young black men (5,10 12). Some selection
bias is possible because, at the time of original ECG selec-
tion, age, sex, and name were on the ECGs, and bias about
likelihood of early repolarization might have increased
the selection of young men, Asians, and blacks. However,
we think it unlikely that the demographic associations
that we observed are spurious.
Our data bolster evidence (8,9) that people who engage
in vigorous physical exercise are more likely to have early
repolarization. Possible explanations include predispos-
ing genetic traits common to both athleticism and early
repolarization, or that early repolarization is a conse-
quence of exercise. Our data suggest that those with early
repolarization were more fit than those with normal
ECGs; they had lower body mass index, blood pressure,
blood cholesterol levels, and heart rates.
These data show no evidence that early repolarization
is associated with a greater likelihood of fatal or nonfatal
cardiovascular conditions. Our results for mortality and
hospitalization data were compared with a large control
group, and were adjusted for several confounders. Our
outpatient diagnosis data, which were limited to 5 years
near the end of follow-up, are less robust; we tried to
minimize confounding by presenting only age-stratified
data for men. However, we cannot completely rule out
confounding by lifestyle. Healthy lifestyle traits and
higher socioeconomic status cluster together and are as-
sociated with favorable health experience (3335).
Refuting our prestudy hypothesis, patients with early
repolarization were neither more likely to be admitted for
chest pain nor to be diagnosed with chest pain as outpa-
tients. This suggests that there is not an increase in the
likelihood of chest pain in these patients, or that emer-
gency departments recognized the ECG changes (1517).
Our data confirm that the SV1 RV5 35 mm voltage
criterion is not specific for left ventricular hypertrophy
(36).
Our study has several limitations. Residual confound-
ing is possible; for example, we did not adjust for smoking
intensity. Another limitation is that we had only nine of
the usual 12 ECG leads, which might have contributed to
underascertainment of early repolarization.
We conclude that early repolarization is more preva-
lent in young, athletic, black men, perhaps suggesting
both genetic and environmental factors, but that it is not
rare in women, older persons, whites, and inactive per-
sons. The condition has a benign prognosis, without an
increased risk of cardiovascular disease.
176 August 15, 2003 THE AMERICAN JOURNAL OF MEDICINE
ACKNOWLEDGMENT
We are grateful to Stephen Wagner, MD, Ellen Killebrew, MD,
Edward Strisower, MD (deceased), Kay Yamamoto, Deloris Iv-
ery, and Brenda Lewis for help in collecting the electrocardio-
grams; to Harald Kipp for help with programming; and to Sally
McBride Allen and Teresa Klask for technical assistance.
REFERENCES
1. Shipley RA, Hallaran WR. The four-lead electrocardiogram in two
hundred men and women. Am Heart J. 1936;11:325345.
2. Spodick DH. Early repolarization: an underinvestigated misnomer.
Clin Cardiol. 1997;20:913914.
3. Wasserburger RH, Alt WJ, Lloyd CJ. The normal RS-T segment
elevation variant. Am J Cardiol. 1961;8:84 92.
4. Grant RP, Estes EH, Doyle JT. Spatial vector electrocardiography:
the clinical characteristics of S-T and T vectors. Circulation. 1951;
3:182197.
5. Goldman MJ. Normal variants in the electrocardiogram leading to
cardiac invalidism. Am Heart J. 1950;59:7177.
6. Gottschalk CW, Craige E. A comparison of the precordial S-T and T
waves in the electrocardiograms of 600 healthy young Negro and
white adults. South Med J. 1956;49:453457.
7. Vacanti LJ. Thoracic pain and early repolarization syndrome
at the cardiologic emergency unit. Arq Bras Cardiol. 1996;67:335
338.
8. Hanne-Paparo N, Drory Y, Schoenfeld Y, et al. Common ECG
changes in athletes. Cardiology. 1976;61:267278.
9. Balady GJ, Cadigan JB, Ryan TJ. Electrocardiogram of the athlete:
an analysis of 289 professional football players. Am J Cardiol. 1984;
53:1339 1343.
10. Grusin H. Peculiarities of the Africans electrocardiogram and
the changes observed in serial studies. Circulation. 1954;9:860
867.
11. Seriki O, Smith AJ. The electrocardiogram of young Nigerians. Am
Heart J. 1966;72:153157.
12. Walker RP, Walker BF. The bearing of race, sex, age, and nu-
tritional state on the precordial electrocardiograms of young
South African Bantu and Caucasian subjects. Am Heart J. 1969;77:
441459.
13. Akhmedov NA. Early ventricular repolarization syndrome and
heart function in the inhabitants of Asia, Africa, and Latin America.
Kardiologiia. 1986;26:6365.
14. Mehta MC, Jain AC. Early repolarization on scalar electrocardio-
gram. Am J Med Sci. 1995;309:305311.
15. Eastaugh JA. The early repolarization syndrome. J Emerg Med.
1989:257-262.
16. Brady WJ, Chan TC. Electrocardiographic manifestations: benign
early repolarization. J Emerg Med. 1999;17:473478.
17. Brady WJ, Perron AD, Chan T. Electrocardiographic ST-segment
elevation: correct identification of acute myocardial infarction
(AMI) and non-AMI syndromes by emergency physicians. Acad
Emerg Med. 2001;8:382385.
18. Hollander JE, Lozano M, Fairweather P, et al. Abnormal
electrocardiograms in patients with cocaine-associated chest
pain are due to normal variants. J Emerg Med. 1994;12:199
205.
19. Collen MF, Davis LF. The multitest laboratory in health care. J
Occup Med. 1969;11:355360.
20. Klatsky AL, Armstrong MA, Friedman GD. Relations of alcoholic
beverage use to subsequent coronary artery disease hospitaliza-
tions. Am J Cardiol. 1986;58:710 714.
Volume 115
 Early Repolarization Normal Variant Electrocardiogram/Klatsky et al
21. Arellano MG, Peterson GR, Petitti DB, Smith RE. The California
Automated Mortality Linkage System (CAMLIS). Am J Public
Health. 1984;74:1324 1330.
22. Myers GB, Klein HA, Stofer BE, Hiratzka T. Normal variations in
multiple precordial leads. Am Heart J. 1947;34:785808.
23. Chelton LG, Burchell HB. Unusual RT segment deviations in
electrocardiograms of normal persons. Am J Med Sci. 1955;230:54
60.
24. Fenichell NN. A long term study of concave RS-T elevationa
normal variant of the electrocardiogram. Angiology. 1962;13:360
366.
25. Spodick DH. Differential characteristics of the electrocardiogram
in early repolarization and acute pericarditis. N Engl J Med. 1976;
295:523526.
26. Alimurung BN, Gilbert CA, Felner JM, Schlant RC. The influence
of early repolarization variant on the exercise electrocardiogram: a
correlation with coronary arteriograms. Am Heart J. 1980;99:739
745.
27. Przybojewski JZ, Becker PH. Pseudonormalization of the
normal variant pattern on the ECG of black subjects after
intermittent acute myocardial ischaemia. S Afr Med J. 1987;71:110
113.
28. Morace G, Padeletti L, Porciani MC, Fantini F. Effect of isoproter-
enol on the early repolarization syndrome. Am Heart J. 1979;97:
343347.
August 15, 2003
29. Gussak I, Antzelevitch C. Early repolarization syndrome: clinical
characteristics and possible cellular and ionic mechanisms. J Elec-
trocardiol. 2000;33:299 309.
30. Tseng YZ. Additional data on body surface potential maps of ven-
tricular repolarization in normal adults. Jpn Heart J. 1997;38:473
485.
31. Kambara H, Phillips J. Long-term evaluation of early repolarization
syndrome (normal variant RS-T segment elevation). Am J Cardiol.
1976;38:157166.
32. Bianco M, Bria S, Gianfelici A, et al. Does early repolarization of the
athlete have analogies with the Brugada Syndrome? Eur Heart J.
2001;22:448 449.
33. Kaplan GA, Keil JE. Socioeconomic factors and cardiovascular
disease: a review of the literature. Circulation. 1993;88:19731998.
34. Marmot M, Wilkerson RG, eds. Social Determinants of Health. New
York: Oxford University Press; 1999.
35. Kilander L, Berglund L, Boberg M, et al. Education, lifestyle factors
and mortality from cardiovascular disease and cancer. A 25-year
follow-up of Swedish 50-year old men. Int J Epidemiol. 2001;30:
1119 1126.
36. Levy D, Labib SB, Anderson KM, Christiansen JC, Kannel WB,
Castelli WB. Determinants of sensitivity and specificity of electro-
cardiographic criteria for left ventricular hypertrophy. Circulation.
1990;81:815820.
THE AMERICAN JOURNAL OF MEDICINE Volume 115 177