Abstract 4355631: Comparison of SPECT/CT Quantification to Planar Heart-to-Contralateral Ratio of 99mTc-Pyrophosphate CT Scans for Prediction of Heart Failure Hospitalization in Patients with Transthyretin Cardiac Amyloidosis (ATTR-CA)

Abstract

Background: ATTR-CA is associated with heart failure (HF) and premature death. It is diagnosed non-invasively with radionuclide imaging, most commonly using 99mTc-pyrophosphate (PYP). Recent advances in SPECT/CT offer potential for more precise quantification than planar imaging utilizing heart-to-contralateral (H/CL) ratio. We compare SPECT/CT-based metrics to H/CL in terms of prediction of HF hospitalization and correlation with echocardiographic (TTE) and laboratory abnormalities. Methods: All patients from 10/2023-10/2024 referred for PYP scans at Columbia University underwent SPECT/CT and planar imaging using an NM/CT 870 DR camera 3h after PYP injection. Positive scans were quantified by 2 board-certified nuclear cardiologists using Xeleris software (GE HealthCare, Milwaukee, WI). Readers selected the voxel with greatest PYP uptake. Regions of interest (ROIs) encompassing tracer uptake in the heart were generated. The reader selected a threshold to maximize heart activity and minimize bone activity. Standardized uptake values (SUV; LV radiotracer concentration normalized to body weight and injected dose), percent injected dose (%ID; proportion of injected radiotracer retained in the heart) and cardiac amyloid activity (CAA; total myocardial radiotracer burden appearing in the ROI) were determined. Mean and max SUV in the ROI (SUVmean, SUVmax) were determined. Clinical and TTE variables, from the most proximate TTE to PYP scanning, were acquired through chart review. Imaging variables were analyzed with respect to the clinical variables, and Cox hazards analysis of each quantitative variable, controlling for age and sex, predicted risk of hospitalization. Results: Of 333 patients, 43 with positive scans were included. When adjusted for age and sex, SUVmean, SUVmax, %ID, and CAA were significantly associated with hospitalization risk while H/CL was not. All quantitated variables were strongly correlated with TTE variables LVPW, IVS, and diastolic dysfunction, as well as with troponin T. Only %ID was significantly correlated with LVEF on TTE. Conclusion: Emerging quantitative SPECT/CT PYP variables were associated with significantly increased hospitalization risk in ATTR-CA patients, while H/CL was not. SPECT/CT variables were also associated with clinical variables such as LVEF, LV wall thickness and troponin T. These findings suggest that SPECT/CT quantification of PYP scans may better identify individuals at risk for hospitalization than planar imaging.