Long-term prognosis of painless exercise-induced ischemia in stable patients with previous myocardial infarction.

OBJECTIVES: This study was designed to better understand the functional correlates and the prognostic relevance of exercise-induced painless ischemia relative to painful ischemia in patients with stable coronary artery disease and previous myocardial infarction (MI). BACKGROUND: The usefulness of exercise testing (ET) for predicting cardiac events, years after MI, although suggested and widely applied, is questionable. In particular, previous studies have reached conflicting conclusions as to whether exercise-induced painless ischemia is related to a less severe myocardial ischemia or to a different prognosis than painful ischemia. METHODS AND RESULTS: Seven hundred sixty-six consecutive stable patients (mean age 57+/-8.6 years, 89% men) with previous MI (mean time from MI 2.8+/-0.75 years) who underwent a Bruce treadmill test and whose data were prospectively entered into our institutional database were enrolled. Patients were followed up for an average of 7+/-0.6 years. End points were (1) cardiac death, (2) cardiac death or nonfatal reinfarction (primary), (3) cardiac death, nonfatal reinfarction, or unstable angina (secondary), and (4) cardiac death, nonfatal reinfarction, unstable angina, or revascularization procedures (secondary, restricted). These patients were retrospectively classified into 4 groups according to exercise test results: (1) painless ischemia, 156 patients; (2) painful ischemia, 75 patients; (3) negative ET, 99 patients; and (4) nondiagnostic ET, the remaining 436 patients. Patients with painless ischemia had less functional impairment and less exercise ischemia than the symptomatic patients (longer exercise duration [P < .001], higher double product [P < .001], higher ischemic threshold [P < .001], and shorter time to ST normalization [P < .001]). Patients with painful ischemia had significantly (P < .0005) increased 6-year risk rates of secondary and restricted end points (49% and 64%, respectively) versus those with painless ischemia (28% and 35%), no inducible ischemia (25% and 27%), or nondiagnostic ET (32% and 37%). Adverse outcomes were mainly the result of higher incidence of unstable angina or revascularization procedures. At multivariate analysis, neither painless nor painful exercise-induced ischemia were independent predictors of end points. CONCLUSIONS: Stable patients with previous MI represent a very low-risk population. In this subset, painless exercise-induced ischemia signifies less severe ischemia than the symptomatic one and has a limited prognostic power. Thus painless exercise-induced ischemia in stable patients with previous MI does not identify patients at increased risk.

Can exercise testing in patients with a history of myocardial infarction predict fatal and non-fatal reinfarction?

BACKGROUND AND OBJECTIVES: Exercise testing (ET) is the preferred initial strategy for risk stratification in patients who are able to exercise and have an interpretable electrocardiogram. However, although it is often suggested and widely applied, its usefulness years after myocardial infarction (MI) is questionable. Therefore, this study was designed to assess the value of exercise testing in predicting the risk of fatal or non-fatal reinfarction in patients with chronic stable coronary artery disease (CAD) due to old myocardial infarction. METHODS: Our study involved 766 consecutive stable subjects [mean (SD) age 57 (8.6) years; male: 89%] with stable CAD due to old MI [mean time from MI: 2.8 (0.75) years], who underwent a Bruce treadmill test and whose data were prospectively entered into our institutional database. Patients were followed up for an average of 7 (0.6) years. RESULTS: Reinfarction was observed in 62 patients; 54 non-fatal and 8 (13%) fatal. Relative risk (RR) of cardiac death for subjects with reinfarction was 4.02 [95% confidence interval (CI): 2.46 to 6.55]. Univariate predictors of fatal or non-fatal reinfarction were: multivessel disease (RR 7.99, CI 1.12 to 56.82), EF < 40% (RR 2.91, CI 1.64 to 7.17), ST depression on rest ECG (RR 2.4, CI 1.30 to 4.45), BP increase with exercise < 10 mmHg (RR 2.36, CI 1.41 to 3.93), BP/HR interaction < 10 mmHg + < 85% max (RR 2.16, CI 1.24 to 3.76). Markers of reduced risk of recurrence included low-risk Duke Treadmill Score (RR 0.55, CI 0.33 to 0.91) and EF > or = 40% (RR 0.34, CI 0.19 to 0.60). A Cox regression model with clinical and exercise parameters detected ST depression on rest ECG (RR 1.47, CI 1.07 to 2.02), BP increase with exercise < 10 mmHg (RR 1.41, CI 1.07 to 1.87), low-risk Duke Treadmill Score (RR 0.79, CI 0.60 to 1.02). A model with coronary anatomy and ejection fraction was also able to identify multivessel disease (RR 2.95, CI 1.43 to 6.09), EF < 40% (RR 1.62, CI 1.17 to 2.25) and BP increase with exercise < 10 mmHg (RR 2.53, CI 1.35 to 4.71). CONCLUSIONS: Stable patients with a history of MI represent a very low-risk population in whom reinfarction continues to have a severe prognosis. ET is unable to identify subjects in whom there is a risk of recurrence, especially if only ischemic parameters are evaluated (in this setting, a clinical or anatomic risk stratification may be better). The application of the Duke Treadmill Score could help to identify a very low-risk group in which no additional testing is required. Therefore, routine ET in stable patients with a history of MI is better at identifying a very low-risk group than in predicting recurrence.