Background: The aim of this study was to assess the long-term predictive values of supine bicycle exercise stress echocardiography (ESE), and the ESE additional role compared to other traditional clinical and rest echocardiographic variables, in 607 patients with low, intermediate and high pretest risk of cardiac events.
Methods: Clinical status and long-term outcome were assessed for a mean period of 46 months (range 12-60 months). ESE was performed for the diagnosis of suspected coronary artery disease (CAD) in 267 patients (43.9%), and for risk stratification of known CAD in 340 patients (56.1%). At baseline, the mean value of wall motion score index (WMSI) was 1.22 +/- 0.36, and the mean left ventricular ejection fraction was 58.5 +/- 10.9%.
Results: ESE was positive for ischemia in 210 patients (34.9%), while ECG was suggestive for ischemia in 157 patients (25.8%). During the test only 97 patients (15.9%) experienced angina. At peak effort, the mean WMSI was 1.38 +/- 0.46. A low workload was achieved by 158 patients (26.1%). During the follow-up period there were 222 events, including 82 hard events (36.9%), 48 deaths (21.6%) and 34 acute non-fatal myocardial infarction (15.3%). At stepwise multivariate model, cigarette smoking (p < 0.01), peak WMSI (p < 0.001), ESE positive for ischemia (p < 0.001) and low workload (p < 0.01) were the only independent predictors of cardiac death, while positive ESE, peak WMSI, angina during the test and hypercholesterolemia were the only independent determinants of hard cardiac events. The cumulative 5-year mean survival rate according to ESE response was 95.9% in patients with negative ESE, and 83.7% in patients with positive ESE (log rank 13.6; p < 0.00001).
Conclusions: ESE yields prognostic information in known or suspected CAD, especially in patients with intermediate pretest risk level. The combined evaluation of clinical variables and other ESE variables, such as peak WMSI and exercise capacity, may further select patients at greatest risk of cardiac death in the overall population.