An algorithm for noninvasive identification of angiographic three-vessel and/or left main coronary artery disease in symptomatic patients on the basis of cardiac risk and electron-beam computed tomographic calcium scores.

OBJECTIVES

We sought to model an algorithm for noninvasive identification of angiographically obstructive three-vessel and/or left main disease based on conventional cardiac risk assessment and site and extent of coronary calcium determined by electron-beam computed tomography (EBCT).

BACKGROUND

Such an algorithm would greatly facilitate clinical triage in symptomatic patients with no previous diagnosis of coronary artery disease (CAD).

METHODS

We examined 291 patients with suspected, but not previously diagnosed, CAD who underwent coronary angiography for clinical indications. Cardiac risk factors were determined as defined by the National Cholesterol Education Program. An EBCT scan was performed in all patients, and a coronary calcium score (Agatston method) was computed. Total per-patient calcium scores and separate scores for the major coronary arteries were generated. These scores were also analyzed for localization of coronary calcium in the more distal versus proximal tomographic sections. These parameters and the risk factors were considered for the model described in the following section.

RESULTS

Sixty-eight patients (23%) had angiographic three-vessel and/or left main CAD. Multiple logistic regression analysis determined male sex, presence of diabetes and left anterior descending (LAD) and circumflex (LCx) coronary calcium scores, independent from more distal calcium localization, as independent predictors for identification of three-vessel and/or left main CAD. Based on this four variable model, a simple noninvasive index (NI) was constructed as the following: loge(LAD score) + log(e)(LCx score) + 2[if diabetic] + 3[if male]. Receiver operating characteristic curve analysis for this NI yielded an area under the curve of 0.88+/-0.03 (p < 0.0001) for separating patients with, versus without, angiographic three-vessel and/or left main CAD. Various NI cutpoints demonstrated sensitivities from 87-97% and specificities from 46-74%. The NI values >14 increased the probability of angiographic three-vessel and/or left main CAD from 23% (pretest) to 65-100% (posttest), and NI values <10 increased the probability of no three-vessel and/or left main CAD from 77% (pretest) to 95-100% (posttest).

CONCLUSIONS

On the basis of a simple algorithm ("noninvasive index"), EBCT calcium scanning in conjunction with risk factor analysis can rule in or rule out angiographically severe disease, i.e., three-vessel and/or left main CAD, in symptomatic patients.