Electron beam computed tomography (EBCT) is a method for measuring coronary calcification and has been promoted as a possible non-invasive screening/diagnostic tool for coronary artery disease (CAD). Our objective was to carry out a systematic review and meta-analysis of EBCT for the screening of asymptomatic patients and the diagnosis of symptomatic patients for CAD.
Studies were identified from the PUBMED, MEDLINE, EMBASE, Current Contents, INAHTA and Cochrane Collaboration databases. We identified studies published in English evaluating EBCT using: (1) a prospective design among asymptomatic patients where CAD was measured in terms of clinical outcomes (e.g. myocardial infarction, death, revascularization); and (2)a cross-sectional design among symptomatic patients where CAD was measured by coronary angiography. We compared the risk of CAD in EBCT score categories defined as low (0–10), moderate (11–400) and high (>400). A hierarchical meta-analysis was used to pool risk ratios comparing categories across studies.
We identified 9 studies of asymptomatic patients and 10 studies of symptomatic patients. In both types of studies, we found variability in EBCT category distribution and risk of CAD within categories. For studies of asymptomatic patients we estimated the following risk ratios (95% credible intervals): moderate versus low 3.5 (2.4, 5.1) and high versus low 9.9 (5.3, 17.6). Similar results were obtained for studies of symptomatic patients. Ratios comparing the risk of no CAD among symptomatic patients were as follows: moderate versus low 0.5 (0.3, 0.8) and high versus low 0.12 (0.05, 0.2).
Increasing EBCT scores indicate higher risk for CAD in both asymptomatic and symptomatic patients. In general, asymptomatic patients with EBCT scores in the high category can perhaps be considered for preventive medical therapy and risk factor modification. Symptomatic patients with EBCT scores in the low category can perhaps, at least temporarily, avoid invasive coronary angiography. However, the non-uniform quality of studies and the lack of availability of individual-level data preclude the extension of our results to individual patients.