Abstract 5821: Impact of Half-Time Image Acquisition with Wide-Beam Reconstruction on the Perfusion Scores and Diagnostic Certainty of Single-Photon Emission Computed Tomography: Comparison with Filtered Back-Projection
Filtered Back Projection (FBP) has been used in SPECT imaging for decades. Half-time acquisition with wide-beam reconstruction (WBR) is a new technology that relies on resolution recovery, and has not been evaluated clinically. To compare stress perfusion scores and diagnostic certainty of WBR and FBP in consecutive pts. referred for gated SPECT. 434 pts. (63+/−13 yrs; 196 F, 238 M) referred for gated SPECT underwent stress Tc-99m sestamibi or tetrofosmin FBP (25 sec/stop) followed by WBR half-time (10 sec/stop) acquisition on a dual-detector 64-stop system (GE Millenium). Images were reconstructed by standard algorithm (QGS, Cedars-Sinai, Los Angeles, CA) for FBP, and by UltraSPECT (Haifa, Israel) for WBR. Summed stress perfusion scores (SSS) were based on a 5-point/17 segment model (0= normal tracer uptake to 4= absent tracer uptake). For all scans, perfusion was considered normal if SSS= 0–1, equivocal if SSS= 2–3, and abnormal if SSS>/= 4. Left ventricular myocardium defect in each of the coronary territory was quantified as %LVLAD,RCA, LCX = SSSLAD, RCA, LCX divided by maximal SSS for all segments in that territory (7 segments for LAD, max SSSLAD= 28; 5 segments for RCA and LXC, max SSSRCA, LCX= 20). Paired t-test and Bland-Altman were used for SSS, %LV, EF. Kappa test was used to compare WBR and FBP for normal, equivocal and abnormal scans and for interobserver agreement. A 2 group t-test was used to compare the magnitude of difference in %LVLAD,RCA, LCX for FBP and WBR discordant and concordant scans. see Table⇓. Interobserver kappa= 0.75–0.78. WBR and FBP kappa= 0.49 SSS and EF and % LVLAD,RCA were less for WBR than for FBP. Interobserver agreement was good, and overall scan agreement moderate. WBR and FBP agreed fairly for SSS, EF, and %LVLAD and poorly for %LV RCA, LCX. There were fewer equivocal scans with WBR, likely due to a decrease in anterior attenuation artifact. Angiographic correlation is currently in progress.