Comparison of misregistration artifacts on cardiac SPECT/CT and PET/CT

Objectives Although CT-based attenuation correction (CTAC) has many advantages it also has disadvantage of sequential acquisition which makes misregistration artifact (MA) on the corrected cardiac single photon emission computed tomography (SPECT) or positron emission tomography (PET) images. The aim of this study was to compare CTAC associated MA in SPECT/CT and PET/CT.

Methods We enrolled 131 patients who underwent myocardial perfusion imaging (MPI) with either Tc-99m MIBI (n=100) or N-13 ammonia (n=31). Low-dose CT during shallow breathing was acquired before rest and stress MPI for CTAC. The automatic registration (AR) between CT and MPI was performed first followed by manual re-registration (RR). We compared the directions (3 axes; x, left-right axis; y, anterior-posterior axis; z, superior-inferior axis), distance (presented by Euclidian distance, √(x2 + y2 + z2)) and perfusion variables (SSS, summed stress score; SRS, summed rest score; SDS, summed difference score) of misregistration between SPECT/CT and PET/CT. Left ventricle was divided into five regions (anterior, septal, inferior, lateral walls and apex) and myocardial radiotracer uptake was measured in each regions.

Results All subjects showed at least one misregistration over 1 mm. Directions of misregistration in SPECT/CT were towards right, anterior and superior, while those of PET/CT towards left, posterior and superior directions. Different direction of misregistration made a significant difference in radiotracer distribution in SPECT/CT and PET/CT: Misregistrated PET/CT images showed significantly decreased uptake in the lateral wall as compared to SPECT/CT (P=0.006 on rest, P=0.007 on stress). Differences of SSS, SRS, SDS and distance of misregistration were not different between SPECT/CT and PET/CT.

Conclusions Manual re-registration was needed in every SPECT/CT and PET/CT studies. Different directions of misregistration between SPECT/CT and PET/CT made a significant difference in radiotracer.

Research Support This study was supported by a grant (A070001) from the Korea National Enterprise for Clinical Trials.