Objective: To examine the relation between resting electrocardiographic (ECG) abnormalities and risk of coronary heart disease (CHD).
Design and setting: This was a prospective study of 7735 middle-aged men aged 40-59 years at entry (British Regional Heart Study). At baseline assessment each man completed a modified World Health Organization (WHO) (Rose) chest-pain questionnaire, gave details of his medical history and had a three-lead orthogonal electrocardiogram recorded. "Symptomatic CHD' refers to a history of anginal chest pain and/or a prolonged episode of central chest pain on WHO questionnaire and/or recall of a doctor diagnosis of CHD (angina or myocardial infarction).
Main outcome measures: These were the first major CHD events, i.e. fatal CHD and non-fatal myocardial infarction, occurring during 9.5 years of follow-up.
Results: Of 611 first major CHD events during follow-up, 243 (40%) were fatal. After adjustment for age, other ECG abnormalities and symptomatic CHD, the ECG abnormalities most strongly associated with risk of a major CHD event were definite myocardial infarction (relative risk 2.5; 95% confidence interval 1.8-7.5) and definite myocardial ischaemia (1.9; 1.1-2.9). Other ECG abnormalities independently associated with a statistically significant increase in risk were left ventricular hypertrophy (2.2; 1.5-3.3), left axis deviation (1.3; 1.1-1.6) and ectopic beats, particularly if these were ventricular (1.6; 1.1-2.4). Three ECG abnormalities associated with a marked increase in CHD case-fatality rate were pre-existing myocardial infarction (67%), major conduction defect (71%) and arrhythmia (67%); the rate in men with none of these abnormalities was 32%. The relative risks associated with each ECG abnormality were similar in men with and without symptomatic CHD. The increase in risk in the presence of symptomatic CHD (2.4-fold) and ECG evidence of definite myocardial infarction (2.5-fold) was similar; the presence of both factors increased risk more than six-fold. The most serious ECG abnormalities-definite myocardial infarction and ischaemia-were useful predictors of future major CHD events only in men with symptomatic CHD.
Conclusion: The prognostic importance of major ECG abnormalities is strongly influenced by the presence of symptomatic CHD. In men with symptomatic CHD the resting electrocardiogram may help to define a group at high risk who may benefit from intervention. However, it has little or no value as a screening tool in middle-aged men without symptomatic CHD.