Predictability of sudden death from resting electrocardiogram. Effect of previous manifestations of coronary heart disease.

The sudden death rate from coronary heart disease over a mean period of 4 years was related to the electrocardiographic findings in 3 groups of subjects, survivors of myocardial infarction, employed men, and employed men with no symptoms or history of coronary disease. Within each group the sudden death rate correlated with the number of electrocardiographic findings, particularly Q, ST, and T wave items. However, between groups there were large differences in sudden death rates in subjects with the same findings. These were greatest in the case of ventricular conduction disturbances and disturbances of rhythm and rate which appeared to be benign in those free of symptoms but ominous after infarction. Findings predictive of sudden death were also predictive of non-sudden coronary deaths. It is concluded that the electrocardiogram is only one of several aids to the diagnosis and assessment of severity of disease and not a substitute. Prognoses derived from clinical case series are inappropriate to symptomless individuals in whom isolated electrocardiographic findings denote little increase in risk.