Author + information
- Yuka Otaki,
- Julian Betancur,
- Guido Germano,
- Damini Dey,
- Daniel Berman and
- Piotr Slomka
Background: We aimed to assess the effect of patient motion and expert frame-by frame motion correction on quantitative myocardial blood flow assessment from dynamic PET.
Methods: 84 patients (67 ± 12 years) that underwent dynamic cardiac stress/rest 82Rb PET were assessed for patient motion during the dynamic PET scan. The frequency, magnitude (in mm) of motion, its direction and its effect of myocardial flow reserve (MFR) quantification were assessed. Patient motion was stratified into moderate (≥ 5mm) or severe (≥ 10mm). Motion in early (<90 seconds) and late frames (≥90 seconds) was compared. The effect of frame-by-frame visual motion correction (MoCo) was assessed by the comparison of global and per-vessel blood-to-myocardium spill-over fraction and by evaluating change in diagnosis (MFR >2.0 normal, MFR < 2.0 abnormal) as compared to the original images.
Results: Severe motion was observed in 38% of stress and 48% rest images. Magnitude of motion was higher in early frames than in late frames for stress (early: 8 ± 5 mm vs. late: 3 ± 3 mm) and rest (9 ± 4 mm vs. 2 ± 3 mm) (p<0.0001 for both). Motion was more frequent in early frames than in late frames (stress: 80% vs. 15%, rest: 88 % vs. 11%),(stress: 38% vs. 1%, rest: 46% vs. 2% for severe) (P<0.001 for both). The most common direction of motion was superior/inferior for stress and left/right for rest. Spillover fraction (before 0.37 for global stress and before 0.32 for global rest, after 0.34 for global stress and 0.31 for global rest p < 0.002 for all) was reduced after MoCo for stress and rest, both globally and at each vessel territory. The most common reduction for spillover was seen in RCA territory (before 0.37 vs. after 0.33, P<0.0001). After MoCo, diagnosis was changed in 13% of left anterior descending, 12% left circumflex, and 5% right coronary territory assessments. Per patient diagnosis by MFR was changed in 13 % of cases.
Conclusions: Severe patient motion is common during dynamic PET and affects quantitative results. Frame-by-frame dynamic PET motion correction is not commonly used but should be required for a reliable analysis of myocardial flow during dynamic PET.
Poster Hall, Hall C
Sunday, March 19, 2017, 9:45 a.m.-10:30 a.m.
Session Title: Nuclear Cardiology: Coronary Flow
Abstract Category: 30. Non Invasive Imaging: Nuclear
Presentation Number: 1288-201
- 2017 American College of Cardiology Foundation