Providing high sensitivity and resolution, the cadmium-zinc-telluride (CZT) -Camera represents a new standard for myocardial perfusion imaging (MPI). However, modern Anger cameras using resolution recovery reconstruction algorithms also provide an improvement in image quality. The aim of our study was an intraindividual comparison of both techniques.
A two-day stress-rest protocol for MPI was performed, using Regadenoson as stress agent. 250-350 MBq Tc99m-Sestamibi were intravenously injected, depending on body weight. SPECT scans were sequentially performed on 30 consecutive patients using a CZT-Camera (Discovery NM 530c; 3 and 7 min acquisition time) and a dual-headed Anger camera (Discovery NM/CT 670; 20 min acquisition time). Images acquired on CZT camera were reconstructed with and without CT-based attenuation correction (AC). Those acquired on Anger Camera were reconstructed 1)with FBP and 2) with OSEM algorithm, with and without AC, applying scatter correction and resolution recovery (RR). Three experiences readers scored images on a 17-segments polar map, according to ASNM guidelines (0=normal, 4=absent uptake). Myocardial counts (measured by ROI-analysis), summed stress-(SSS), rest-(SRS) and difference-scores (SDS) and LVEF were compared.
All acquired parameters correlated significantly between camera systems. The coefficients of correlation were consistently higher for the non-corrected image data (NC) than for the data including AC and RR for the Anger camera (e.g.: SSS for NC – R2=0.74 vs AC-RR – R2=0.54). The coefficients of correlation between the Anger camera and the 3 min CZT-acquisitions were consistently lower than those with the 7 min acquisitions.
Semiquantitative MPI SPECT results of conventional Anger cameras and CZT-cameras are significantly correlated. Nevertheless, new resolution recovery reconstruction algorithms may amplify artifacts. Individual properties of the imaging procedures have to be taken into account for the interpretation of the results.