BACKGROUND: Coronary artery disease remains the leading cause of death in the United States. Currently clinicians rely on traditional models of risk-factor analysis to predict coronary outcomes. EBCT has recently been identified as a tool for measuring calcium within the coronary arteries and promoted as a means of predicting coronary risk. The use of EBCT as a prognostic or screening tool is based on the premise of a causal and incremental association between coronary artery calcium and atherosclerosis.1,2 More coronary calcium means more atherosclerotic heart disease, which in turn means a higher risk for coronary events. The objective of this study was to review the literature on EBCT as a noninvasive method for predicting subsequent coronary events.
POPULATION STUDIED: The average age of a study participant in the 5 identified studies was 57 years, and 74% were men. The study setting (ie, primary care or referral) and the subject ethnicity were not reported. The baseline cardiovascular risk of participants was inconsistently reported.
STUDY DESIGN AND VALIDITY: This study was a meta-analysis. The authors searched the literature for articles pertaining to EBCT, heart disease, and prognosis. Studies were included if they were performed on an asymptomatic adult population with adequate follow-up (minimum of 36 months) and assessment of coronary outcomes (myocardial infarction [MI] or death) was reported. The authors included data from ongoing and unpublished studies. Incomplete data were imputed in a conservative fashion to limit bias toward the alternative hypothesis. Two authors independently reviewed the data, and differences were reconciled by group consensus. A random effects model was used to calculate summary estimates of the risk ratios.
OUTCOMES MEASURED: The primary outcome measures were risk ratios for hard coronary events (MI and cardiac death) and combined events (MI, cardiac death, and revascularization procedures).
RESULTS: Nine studies (4 published articles and 5 abstracts) were identified. Three were duplicate publications that reported the same data as another study, and one had only 33% follow-up; these were appropriately excluded. The remaining 5 studies with 4348 patients were included. There was significant heterogeneity between studies, with the best designed study having among the lowest risk ratios (2.3). The summary estimates calculated using a random effects model showed that patients with higher calcium scores by EBCT were at an increased risk of hard events (summary risk ratio=4.2; 95% confidence interval [CI], 1.6-11.3) and combined events (summary risk ratio=8.7; 95% CI, 2.7-28.1). However, these calculations should not have been reported in the first place because of the broad methodologic differences between studies and their significant heterogeneity. These major flaws greatly weaken the conclusions that can be drawn from this meta-analysis.
EBCT is a relatively costly test ($300-$400) in search of a clinical niche. It is no better at predicting coronary outcomes than traditional risk-factor modeling or the use of Framingham data. There is no evidence to support the routine use of EBCT as a screening tool for coronary disease in an asymptomatic population.