Impact of follow-up time and re-measurement of the electrocardiogram and conventional cardiovascular risk factors on their predictive value for myocardial infarction

Objective. To explore if the predictive power of mid-life electrocardiogram (ECG) abnormalities and conventional cardiovascular risk factors for future myocardial infarction, change over a 30-year follow-up period and if repeated examination improves their predictive power. Design and setting. Longitudinal population-based study. Participants. A total of 2322 men aged 50 years, with a total follow-up period of 30 years and 1221 subjects were re-examined at age of 70 years. Main outcome measure. Risk for fatal and nonfatal myocardial infarction (n = 470) analysed at age 50 and 70 years, and separately for 1-10, 11-20 and 21-30 years after first examination. Investigations included a 12-lead ECG, blood pressure, body mass index, smoking habits, and glucose and lipid variables. Results. Whilst high blood pressure and dyslipidaemia variables, such as apolipoprotein B/apolipoprotein A1 ratio, measured at age 50 showed an almost unchanged predictive power during the three decades of follow up, ECG abnormalities, insulin and BMI were only significant predictors during the two first decades of follow up. Despite increased prevalences of ECG abnormalities and obesity with age, they predicted for future myocardial infarction in midlife only whilst fasting insulin and smoking regained their predictive power when re-measured at age 70. Conclusion. Whilst hypertension and dyslipidaemia were consistent as risk factors, length of follow-up period and age at baseline investigations affected the predictive power of ECG abnormalities, fasting insulin, BMI and smoking. In the elderly, ECG abnormalities did not contribute to the prediction of myocardial infarction but smoking and fasting insulin may be important in the pathophysiology leading to MI especially in this age group.