PET imaging standards for the Ontario Provincial Registry of Cardiac Viability Imaging with FDG PET (CADRE)

Objectives The Ontario Ministry of Health is conducting a multi-center trial to evaluate the role of FDG PET viability imaging in the management of patients with severe left ventricular (LV) dysfunction. There is a need to ensure consistent PET image quality and interpretation across the 4 sites.

Methods Qualifying scans of ‘matched’ perfusion and FDG defects were performed at each site using a Data Spectrum torso phantom to 1) verify accurate definition of the perfusion defect, 2) verify the expected FDG–perfusion mismatch (hibernation) and match (scar) scores, and 3) qualify the imaging methods used at each site. FDG and perfusion images were reconstructed with equivalent resolution and processed using FlowQuant© software to determine LV mismatch and match scores.

Results The mean perfusion defect score of 15±1% was higher than the known defect size of 9% LV (P<0.05) due to residual image heterogeneity. The mean match score was 12±2% (P<0.05 vs 9%), and mean mismatch score was 3±2% (P<0.05 vs 0%). Inter-operator match and mismatch scores were highly repeatable (Δ<1%). There was 3% over-estimation of FDG mismatch using SPECT vs PET perfusion (P<0.05), but less than 3% difference in mismatch and match scores using SPECT perfusion with or without AC to interpret FDG viability.

Conclusions The FDG viability imaging standards produced match and mismatch viability scores with <3% inaccuracy and variability across the CADRE imaging sites.