Higher processing repeatability of myocardial flow reserve calculated using net retention model compared to one compartment model in SPECT studies

Abstract

AbstractDynamic assessment of myocardial blood flow (MBF) and myocardial flow reserve (MFR) provides additional information that can improve diagnostic accuracy of radionuclide myocardial perfusion imaging in some clinical situations. This study assessed processing repeatability of these parameters calculated using two models—net retention (RET) and one compartment (1CM) in dynamic SPECT studies, using the latest version of Corridor 4DM software (v2024). Data of 107 patients were analyzed retrospectively (57 of whom were assessed in our previous study using 4DM v2015). Dynamic SPECT studies were carried out using a routine two-day rest-dipyridamole protocol. Data was processed in 4DM v2024 twice by one operator and once by another operator. Automatic heart image positioning during post-processing in 4DM v2024 was significantly improved compared to v2015, reducing the number of studies requiring extensive manual corrections from 41 to 12%. This significantly improved interobserver processing repeatability of MFR values in RCA territory compared to our previous study using v2015—from r = 0.67 to 0.85 (p = 0.0034). Interobserver processing repeatability of MBF and MFR in all 107 patients was significantly better in RET model compared to 1CM model. In conclusion, RET model is more reliable for calculating MBF and MFR values based on dynamic SPECT studies.