Introduction ^99mTc-3,3-diphosphono-1,2-propanodicarboxylic acid (DPD) scintigraphy offers a sensitive, non-invasive test for cardiac transthyretin amyloid. The Perugini grading system relies on a visual score of the planar image – from grade 0 (negative) to 3 (strongly positive). SPECT allows the 3D visualisation of radioactivity within the body. SPECT/CT quantification may improve diagnostic accuracy and offer a means of measuring amyloid burden.

Methods The software algorithm (Hermes Medical) converts recorded counts per voxel into activity per unit volume, enabling quantitative reconstructions and display of standardised uptake values (SUV). Volumes of interest were placed over the heart, adjacent vertebra, paraspinal muscle and liver and peak SUV (SUV_peak) recorded. P-values were calculated using a one-way ANOVA.

Results We identified 32 patients (aged 82±9 years, 59% male), with a range of DPD results (10 grade 0, 6 grade 1, 12 grade 2, 4 grade 3). Cardiac SUV_peak increased from Perugini grade 0 to 2, but plateaued at 3 (1.39±0.47 vs 3.82±1.66 vs 12.89±4.39 vs 10.79±1.12; p<0.001).

Vertebral SUV_peak decreased from Perugini grade 0 to 3 (11.75±3.79 vs 9.60±2.28 vs 8.34±2.25 vs 5.85±1.53; p<0.006). Paraspinal muscle SUV_peak increased from Perugini grade 0 to 3 (0.50±0.13 vs 0.57±0.13 vs 0.76±0.27 vs 1.29±0.24; p<0.001). Liver SUV_peak did not vary between grades (p=0.952). On ROC analysis, a cardiac SUV_peak of 2.3 diagnosed amyloidosis with 95% sensitivity and 90% specificity.

Conclusion SPECT/CT quantification in DPD scintigraphy has diagnostic and discriminatory power. Competition for tracer between skeletal muscle and the heart in grade 3 cases may confound quantification of cardiac amyloid burden.