Prognostic value of coronary calcification and angiographic stenoses in patients undergoing coronary angiography

Objectives. This investigation sought to determine the relative prognostic value of coronary calcific deposits and coronary angiographic findings for predicting coronary heart disease-related events in patients referred for angiography. Background. The relation among coronary calcification, coronary stenoses and coronary heart disease-related events is of interest on a clinical as well as a pathophysiologic basis. Methods. Four hundred ninety-one symptomatic patients underwent coronary angiography and electron beam computed tomography at five different centers between April 1989 and December 1993. The electron beam computed tomograms were interpreted by a cardiologist with no knowledge of the coronary angiographic and clinical data. Receiver operating characteristic (ROC) curves were constructed to determine the relation between electron beam computed tomographic and coronary angiographic findings. A follow-up telephone survey was completed in 86% of patients. The records for all patients who died or were admitted to the hospital for chest pain or suspected myocardial infarction were reviewed by three other cardiologists with no knowledge of the coronary angiographic and electron beam computed tomographic study results. Results. The mean (+/-SE) area under the ROC curve was 0.75 +/- 0.02 for the coronary calcium score, indicating moderate discriminatory power for this score for predicting angiographic findings. Thirteen coronary heart disease-related deaths and eight nonfatal acute infarctions occurred over 30 +/- 13 months. Scores were sorted in ascending order and divided into quartiles of equal size. One patient in the first quartile had a fatal myocardial infarction (coronary calcium score range 0 to 2.1); 2 in the second quartile (range 2.1 to 75.3), 8 in the third quartile (range 75.3 to 397.1) and 10 in the fourth quartile (>397.1) had a coronary heart disease related event. Application of bivariate logistic regression showed that log score but not number of angiographically dis eased vessels significantly predicted the probability of a coronary heart disease-related event occurring during follow up. Conclusions. Electron beam computed tomographic calcium scores correlate moderately well with angiographic findings. These scores predict coronary heart disease-related events in patients undergoing angiography as well as do the number of angiographically affected arteries.