Usefulness of exercise testing shortly after acute myocardial infarction for predicting 10-year mortality

Abstract

Previous studies on the prognostic value of exercise test variables after acute myocardial infarction (AMI) are limited methodologically and have yielded inconsistent results. This study determined whether 6 exercise test variables (systolic blood pressure, ST-segment depression or elevation, exercise capacity, arrhythmias, and angina pectoris) after controlling for age and sex, enhance 6 clinical variables (digoxin, previous AMI, history of systemic hypertension and angina, Killip class, and stress) as predictors of cardiovascular death and act as independent predictors as well. The present study followed 258 patients for 10.6 years, each of whom had AMI between 1977 and 1980 and an exercise test before hospital discharge. By 1988, 71 of the 258 patients had died, 56 of cardiovascular causes. This study is unique because exposure and outcome variables are clearly defined and follow-up was complete and longer than in previous studies. Multivariate survival analysis using an exponential model was tested to evaluate the conditional effects of the exercise test and clinical variables and to control for confounders. The model combined the exercise test and clinical variables. Results are reported with hazard ratios (HR) and 95% confidence intervals (CI). For important clinical risk predictors, the HRs with CIs are: digoxin use, HR 4.0 (CI 1.8, 8.5); history of prior AMI, HR 2.4 (CI 1.2, 4.7); history of systemic hypertension, HR 2.5 (CI 1.3, 4.5); angina, HR 2.4 (CI 1.3, 4.5); and stress, HR 4.2 (CI 2.2, 7.9). Three exercise test variables—hypotensive blood pressure response, HR 5.1 (CI 1.9,13.6); ST-segment depression, HR 1.8 (CI 0.98, 3.6); and ST-segment elevation, HR 2.4 (CI 0.98, 5.9)—were strong independent predictors of cardiovascular death after controlling for age, sex, and clinical variables. This is the only exercise test study with a complete 10.6-year followup in the U.S., a clearly defined outcome, and showing that 3 exercise test variables—hypotensive blood pressure response, and ST-segment depression and elevation—are strong independent predictors of cardiovascular death after controlling for clinical variables. These results support the use of exercise testing after AMI for predicting long-term prognosis.