Comparison between short-acquisition myocardial perfusion SPECT reconstructed with a new algorithm and conventional acquisition with filtered backprojection processing

Authors:
  • LFLea Miriam B. Fonseca
    Lea Miriam B. Fonseca
  • MLMaria Carolina P.P. Landesmann
    Maria Carolina P.P. Landesmann
  • RLRonaldo S.L. Lima
    Ronaldo S.L. Lima

Abstract
To compare a novel reconstruction package (Evolution for Cardiac), used to improve image quality in myocardial perfusion single photon emission computed tomography (MPS) and enable shorter scan acquisitions, to filtered backprojection used for conventional time acquisitions. Forty-four patients who underwent 2-day Tc-99m sestamibi MPS had both standard [SD: 14-min scan and filtered backprojection processing] and short (SH: 6-min scan and iterative reconstruction with Evolution for Cardiac) studies. In a 17-segment model of the left ventricle, each segment was automatically scored 0-4 (normal to absent radiotracer activity); summed stress, rest and difference scores (SSS, SRS, SDS) were obtained. Spearman's test assessed the correlation between scores from SD and SH studies, and Bland-Altman plots evaluated absolute and mean differences of scores between these studies. The SSS, SRS, and SDS were not statistically different between SD and SH studies (SSS: 5.9 + or - 6.6 vs. 5.8 + or - 6.5; SRS: 2.6 + or - 3.5 vs. 2.6 + or - 2.9; SDS: 3.3 + or - 5.8 vs. 3.2 + or - 5.7). Significant correlations were observed between SD and SH studies regarding SSS (R(2)=0.93), SRS (R(2)=0.85), and SDS (R(2)=0.9) (all P<0.001). Bland-Altman plots showed that differences in SSS, SRS, and SDS fell within 2 standard deviations of the average differences between SD and SH studies. Overall agreement between clinical readings of SD and SH scans was good (up to 100% in abnormal scans). A new reconstruction package (Evolution for Cardiac) enabled short MPS acquisition without changes in quantitative perfusion results when compared with conventional acquisition and processing.