PROGNOSTIC VALUE OF HEART-RATE ADJUSTMENT OF EXERCISE-INDUCED ST SEGMENT DEPRESSION IN THE MULTIPLE RISK FACTOR INTERVENTION TRIAL

Objectives. We sought to assess the effect of heart rate adjustment of ST segment depression on risk stratification for the prediction of de ath from coronary artery disease. Background. Standard analysis of the ST segment response to exercise based on a fixed magnitude of horizon tal or downsloping ST segment depression has demonstrated only limited diagnostic sensitivity for the detection of coronary artery disease a nd has variable test performance in predicting coronary artery disease mortality. Heart rate adjustment of the magnitude of ST segment depre ssion has been proposed as an alternative approach to increase the dia gnostic and prognostic accuracy of the exercise electrocardiogram (EGG ). Methods. Exercise ECGs were performed in 5,940 men from the Usual C are Group of the Multiple Risk Factor Intervention Trial at entry into the study. An abnormal ST segment response to exercise was defined ac cording to standard criteria as greater than or equal to 100 mu V of a dditional horizontal or downsloping ST segment depression at peak exer cise. The ST segment/heart rate index was calculated by dividing the c hange in ST segment depression from rest to peak exercise by the exerc ise-induced change in heart rate. An abnormal ST segment/heart rate in dex was defined as >1.60 mu V/beats per min. Results. After a mean fol low-up of 7 years there were 109 coronary artery disease deaths. Using a Cox proportional hazards model, a positive exercise ECG by standard criteria was not predictive of coronary mortality (age-adjusted relat ive risk [RR] 1.5, 95% confidence interval [CI] 0.6 to 3.6, p = 0.39). In contrast, an abnormal ST segment/heart rate index significantly in creased the risk of death from coronary artery disease (age-adjusted R R 4.1, 95% CI 2.7 to 6.0, p < 0.0001). Excess risk of death was confin ed to the highest quintile of ST segment/heart rate index values, and within this quintile, risk was directly related to the magnitude of te st abnormality. After multivariate adjustment for age, diastolic blood pressure, serum cholesterol and cigarettes smoked per day, the ST seg ment/heart rate index remained a significant independent predictor of coronary death (RR 3.6, 95% CI 2.4 to 5.4, p < 0.001). Conclusions, Si mple heart rate adjustment of the magnitude of ST segment depression i mproves the prediction of death from coronary artery disease in relati vely high risk, asymptomatic men. These findings strongly support the use of heart rate-adjusted indexes of ST segment depression to improve the predictive value of the exercise EGG.