A Prospective Study of Persons with the Early Repolarization Normal Variant Electrocardiogram
Sometimes confused with abnormal electrocardiogram (ECG) patterns, especially acute myocardial infarction, the poorly understood “early repolarization normal variant” (VAR) ECG has been considered benign, but, to our knowledge, has not been studied in relationship to later cardiovascular (CV) problems. We collected 2081 possible VAR and normal (NORM) ECG’s from among those done at health examinations of 72,830 members of a pre-paid health care program in 1983-85. All ECG’s were reinterpreted independently as VAR, Borderline, or NORM by 2 cardiologists blinded to all identifying and clinical data; agreement by both was required for categorization as VAR. Comparisons of subsequent hospitalizations and deaths through 1998 among all 72,830 examinees used Cox proportional hazards models with 7 covariates. Outpatient diagnoses were available from 1995 through 1999 among a subset of 45,528 examinees still in the program in 1995; these were compared in sex & age-stratified groups by chi-square tests. The blinded readings yielded 670 VAR ECG s, 330 borderline ECG s, and 1081 NORM ECG s. Comparison of VAR to NORM showed increased VAR proportions of men, persons < 40 years old, and blacks or Asians (all with p<0.001). Among 21,826 persons hospitalized in the entire cohort there were 136 with VAR; among 7974 persons who died there were 37 with VAR; among 15,422 persons with outpatient diagnoses there were 95 with VAR. No significant differences in risk for VAR vs all persons were found for hospitalization or death for all causes, CV causes, or non-CV causes. For outpatient diagnoses, persons with VAR had lower rates/1000 persons than the entire population for any CV diagnosis (339/1000 for all; 256/1000 for VAR; p < 0.001). Diagnosis rates among VAR were significantly lower for hypertension (p < 0.02) arrhythmias (p < 0.01) and non-specific CV symptoms (p < 0.05) and also were nonsignificantly lower among VAR for coronary disease, cardiomyopathy, heart failure, stroke, and valvular disease. These data confirm the benign nature of VAR, suggesting that VAR not only carries no increased risk for development of CV conditions but may be associated with decreased CV risk.