Sensitivity of two electron beam tomography protocols for the detection and quantification of coronary artery calcium.

Abstract

The purpose of this study was to compare the sensitivity of two electron beam tomography protocols for detection and quantification of coronary artery calcium. We selected 101 patients (57% men, mean age 53 +/- 10 years) to undergo two consecutive electron beam tomography and acquired imaging with both a 6-mm and a 3-mm slicing protocol. Three pixels (area, 1.03 mm(2)) and a minimal density of 130 H were used for definition of calcified plaque. We found coronary artery calcifications in 46 patients when we used a 6-mm protocol and in 61 patients when we used a 3-mm protocol (p < 0.001). The average total calcium score was 77 (+/-140) with a 6-mm protocol and 251 (+/-395) with a 3-mm protocol (p < 0.005). The average number of calcified lesions per patient was 1.7 for a 6-mm protocol and 3.7 for a 3-mm protocol (p < 0.01). Of 179 individual lesions seen using a 3-mm protocol, 103 (58%) were missed using a 6-mm protocol, and only 27% of the lesions with a calcium score less than or equal to 40 seen with a 3-mm protocol were detected with 6-mm slicing (p < 0.001). The mean lesion attenuation with a 6-mm protocol was 160 (+/-42) H, compared with 218 (+/-44) H with a 3-mm protocol (p < 0.001), indicating a significantly greater partial volume averaging with the former protocol. A 6-mm slicing protocol is significantly less sensitive than a 3-mm protocol for the detection and quantification of coronary artery calcium. Since one third of coronary events occur in patients with low calcium scores, a 6-mm protocol might be unreliable for risk assessment because of substantial loss of information in this calcium score range.