Risk stratification of patients with coronary artery disease and left ventricular dysfunction by exercise radionuclide angiography and exercise electrocardiography

Abstract

Background: The results of multicenter trials indicate that patients with left ventricular dysfunction and either three-vessel or left main coronary artery disease have improved prognosis when treated surgically.

Objective: As part of a larger evaluation and follow-up study of coronary artery disease, the objective of this investigation was to determine whether exercise radionuclide angiography can be used, in patients with mild symptoms of coronary artery disease and left ventricular dysfunction at rest, to identify patients with three-vessel or left main coronary artery disease.

Methods and results: Eighty-four consecutive patients were studied with angiographically defined coronary artery disease in whom left ventricular ejection fraction at rest ranged from 20% to 40%. Patients underwent exercise electrocardiography, rest and exercise radionuclide angiography, and 24-hour electrocardiographic monitoring. There were 22 patients with one-vessel, 31 with two-vessel, 27 with three-vessel, and four with left main coronary artery disease. All but four patients had a documented history of myocardial infarction. By univariate analysis, the following parameters were related to the anatomic severity of coronary artery disease: magnitude of ST segment depression with exercise (p < 0.001), magnitude of change in ejection fraction with exercise (p < 0.005), and occurrence of angina during exercise (p < 0.005). However, because of the extensive overlap among anatomic subgroups, no single factor had both a satisfactory sensitivity and a satisfactory specificity in identifying patients with three-vessel and left main coronary artery disease. Multivariate stepwise regression analysis also failed to predict three-vessel or left main coronary artery disease satisfactorily (sensitivity 73% and specificity 73%; positive predictive accuracy 59% and negative predictive accuracy 83%). Nonetheless, this multivariate analysis provided important prognostic information. During medical therapy (mean follow-up 56 months), the patients with a high likelihood of three-vessel or left main coronary artery disease had a greater risk of death or reinfarction than had patients with a low likelihood (p < 0.05). These functional data were better than coronary anatomy alone in providing risk stratification. Four of six patients with two-vessel disease who died were classified incorrectly by the multivariate analysis in the high-likelihood group for three-vessel or left main coronary artery disease, but classified correctly as being at high risk; whereas none of the patients with three-vessel disease who were misclassified in the low-likelihood group died during medical therapy.

Conclusion: Although exercise radionuclide angiography in patients with minimal symptoms of coronary artery disease and left ventricular dysfunction is not precise in predicting three-vessel or left main coronary artery disease, it provides important functional information regarding subsequent prognosis during medical therapy.