[Evaluation of coronary calcifications with 64-slice CT - variability of the scores and the influence of the reconstruction interval].

PURPOSE

To evaluate the variability of coronary calcium scores depending on the image reconstruction interval using a 64-slice CT scanner.

MATERIALS AND METHODS

30 patients (18 male, 12 female; mean age 57 +/- 9 yrs; mean heart rate 66 +/- 10 bpm) underwent coronary calcium scoring using a 64-slice CT scanner (Somatom Sensation 64, Siemens Medical Solutions, Erlangen) and a standardized scanning protocol. Oral beta-blockers were administered to 12 patients with a baseline heart rate > 70 bpm. Images were reconstructed in 10 % increments from 10 - 100 % of the RR interval. Two blinded experienced observers independently calculated Agatston (AS), calcium mass (MS) and volume scores (VS) for every reconstructed image series. The results were compared to similar studies for 16-slice CT scanners.

RESULTS

The mean values and mean coefficients of variation among all patients were as follows: AS, 397 +/- 829, 109 % MS, 88 +/- 225, 154 % VS, 335 +/- 669, 100 %. Regarding the reconstruction intervals, the mean coefficients of variation were as follows: 107 % (AS), 97 % (VS), 116 % (MS). No specific image reconstruction interval with statistically significant lower variability for each score could be identified. High inter-observer agreement was achieved (K = 0.98). With statistical significance (p < 0.05) 10/30 patients (pts) were able to be allocated to more than one risk group (RG): 6 pts = 2 RG; 3 pts = 3 RG; 1 pts = 4 RG. The scores for 5/30 patients were zero for at least one reconstruction interval, but further reconstructions revealed calcifications. The number of patients assignable to different risk groups was significantly lower compared to published data using a 16-slice scanner (p < 0.05).

CONCLUSION

Coronary calcium scores determined using a 64-slice scanner display a wide range of variability depending on the image reconstruction interval as already described for 16-slice CT scanners. However, compared to previous studies, our data indicate that this vendor's generation of scanners reduces the influence of score variations on the risk stratification.