Prediction of mortality during the first year after acute myocardial infarction from clinical variables and stress test at hospital discharge.

The predictive value of a predischarge symptom-limited stress test was studied in 405 consecutive survivors of acute myocardial infarction (AMI). Three hundred patients performed bicycle ergometry; 105 could not perform it. Among these latter 105 patients, the stress test was contraindicated in 43 because of angina or heart failure and in 62 because of noncardiac limitations. One-year survival was 44% in the "cardiac-limited" group (19 of 43) and 92% in the "non-cardiac-limited" group (57 of 62). One-year survival among the patients who performed an exercise test at discharge was 93% (280 out of 300). The best stress test predictor of mortality by univariate analysis was the extent of blood pressure (BP) increase: 42 +/- 24 mm Hg in 280 survivors vs 21 +/- 14 mm Hg in 20 nonsurvivors (p less than 0.001). Among the 212 patients in whom BP increased 30 mm Hg or more, mortality was 3% (n = 6), while it was 16% (n = 14) among the 88 patients in whom BP increased less than 30 mm Hg. Angina, ST changes and arrhythmias were not as predictive. Stepwise discriminant function analysis showed inadequate BP increase to be an independent predictor of mortality. A high-risk group can be identified at discharge on clinical grounds in patients unable to perform a stress test, whereas intermediate- and low-risk groups can be identified by the extent of BP increase during exercise.