The noninvasive prediction of cardiac mortality in men and women with known or suspected coronary artery disease

PURPOSE: The association between myocardial perfusion imaging defects and c ardiac mortality in women is undefined. We examined whether myocardial perf usion imaging predicted cardiac mortality in men and women and compared thi s with other variables influencing prognosis. SUBJECTS AND METHODS: Six academic institutions with high-volume nuclear ca rdiology laboratories consecutively studied 5,009 men aged 62 +/- 12 years (mean ISD) and 3,402 women aged 66 +/- 11 years with symptomatic known or s uspected coronary artery disease undergoing exercise (n = 7,486) or pharmac ologic stress (n = 925) myocardial perfusion imaging. A pretest clinical ri sk index was calculated from age, history of myocardial infarction, diabete s, hypertension, and hypercholesterolemia. Myocardial perfusion images were analyzed for stress-induced defects or any defect in the territories of th e three major coronary arteries. RESULTS: Stress-induced perfusion defects were seen in 39% of men and 25% o f women (P = 0.0001). Extensive stress-induced or fixed defects (>2 vascula r territories) were less common in women than men (10% vs 19%, and 4% vs 18 %, both P = 0.0001). During a mean of 2.4 +/- 1.5 years of follow-up, 143 p atients died of cardiac causes. The clinical risk index and number of terri tories with perfusion defects were associated with cardiac mortality in wom en and men. In women undergoing exercise myocardial perfusion imaging, the number of abnormal territories remained the strongest correlate of mortalit y after adjustment for exercise variables. CONCLUSIONS: The results of myocardial perfusion imaging are important, ind ependent predictors of survival in both women and men. Am J Med. 1999;106:1 72-178. (C) 1999 by Excerpta Medica, Inc.