QUANTIFICATION OF CORONARY-ARTERY CALCIUM BY ELECTRON-BEAM COMPUTED-TOMOGRAPHY FOR DETERMINATION OF SEVERITY OF ANGIOGRAPHIC CORONARY-ARTERY DISEASE IN YOUNGER PATIENTS

Objectives. This study attempted to 1) evaluate five quantitative meas ures of coronary artery calcium and determine which best agreed with c oronary artery disease severity at angiography; and 2) determine optim al quantity cutpoints to distinguish among no, mild and significant di sease. Background. Coronary artery calcium identified noninvasively by electron beam computed tomography is a sensitive marker for atheroscl erosis. Quantitative assessments of calcium could distinguish among pa tients with no, mild and significant disease in clinical, screening an d research settings. Methods. One hundred sixty patients, 23 to 59 yea rs old, underwent coronary angiography and electron beam computed tomo graphy. Coronary artery calcium was defined as dense (>130 Hounsfield units) foci greater than or equal to 2 mm(2) on the tomogram. Regressi on and receiver operating characteristic analyses were used to evaluat e five quantitative measures of calcium as predictors of the largest s tenosis in the coronary arteries and to identify optimal cutpoints for distinguishing among disease categories. No disease was defined as no stenosis, mild disease as 10% to 49% diameter stenosis in one or more major branches and significant disease as greater than or equal to 50 % diameter stenosis in one or more major branches. Results. Ail measur es evaluated performed well. With calcific area as the quantitative me asure, the best cutpoint for discriminating between patients,vith and without disease was the presence of calcium: sensitivity 81%, specific ity 86% and overall accuracy 83%, The best cutpoint for discriminating between patients with and without significant disease was Is mm(2): s ensitivity 86%, specificity 81% and accuracy 83%. Conclusions. Because the ranges of calcium quantity over lapped across disease categories, no cutpoints would distinguish among categories with absolute certain ty. However, selected cutpoints could rule out disease in most healthy subjects and identify most patients with significant disease.