Assessment of severity of coronary artery stenosis in a canine model using the PET agent 18F-fluorobenzyl triphenyl phosphonium: comparison with 99mTc-tetrofosmin.

UNLABELLED Myocardial perfusion imaging plays an important role in clinical management of coronary artery disease, but the most commonly used radionuclides significantly underestimate the severity of coronary artery stenosis. The objective of this study was to evaluate the potential clinical utility of the PET compound (18)F-fluorobenzyl triphenyl phosphonium ((18)F-FBnTP) and characterize its capacity to assess the severity of coronary artery stenosis in a canine model in vivo and ex vivo. METHODS (18)F-FBnTP myocardial uptake was measured in 17 dogs with various degrees of stenosis of the left anterior descending (LAD) or circumflex (LCx) coronary arteries during adenosine vasodilation, using dynamic PET and gamma-well counting. True myocardial blood flow in ischemic (IS) and nonischemic (NIS) beds of the left ventricle was determined with radioactive microspheres. (18)F-FBnTP and (99m)Tc-tetrofosmin activities were compared in 8 dogs ex vivo. RESULTS The quantitative assessment of the perfusion defect was significantly (P < 0.03) more accurate with (18)F-FBnTP than with (99m)Tc-tetrofosmin, in mild (IS/NIS; 0.72 +/- 0.08, 0.93 +/- 0.07, respectively, mean +/- SE) and severe stenosis (0.42 +/- 0.05, 0.64 +/- 0.08, respectively), compared with microsphere flow (mild, 0.43 +/- 0.06; severe, 0.22 +/- 0.04). The IS/NIS ratio of both radionuclides correlated linearly with microsphere flow disparity with a similar slope. Flow defect contrast was 2.7 times greater for (18)F-FBnTP than for (99m)Tc-tetrofosmin, as inferred from the regression line intercept (0.14 vs. 0.38, respectively). The (18)F-FBnTP PET IS/NIS ratio (mild, 0.70 +/- 0.04; severe, 0.46 +/- 0.02), did not differ statistically (P >or= 0.330) from that measured ex vivo. A nearly identical qualitative and quantitative estimate of stenosis severity was obtained by early, short (5-15-min) and delayed, prolonged (30-60-min) (18)F-FBnTP PET scans. The stenotic area measured by PET was 16% smaller than that defined by tissue staining. CONCLUSION (18)F-FBnTP PET is a promising new technology for rapid noninvasive detection and assessment of perfusion defect severity in the myocardium.