Hybrid CT angiography and quantitative (15)O-water PET for assessment of coronary artery disease: comparison with quantitative coronary angiography

Purpose: CT angiography (CTA) can rule out significant stenoses with a very high reliability, whereas its ability to confirm significant stenoses is suboptimal. In contrast, measurements of myocardial blood flow (MBF) provide information on the haemodynamic consequences of stenoses. Therefore, a combination of the two might improve diagnostic accuracy. We conducted a head-to-head comparison of CTA, measurement of MBF by ¹⁵O-water PET, and hybrid PET/CTA for the detection of significant coronary artery stenoses. Methods: The study group comprised 44 outpatients scheduled for invasive coronary angiography (ICA) with an intermediate pretest likelihood of coronary artery disease. The patients underwent 64-slice CTA and baseline and hyperaemic PET before ICA with quantitative coronary angiography analysis. Results: On a per-patient basis, the negative predictive values (NPV; 95 % confidence intervals in parentheses) were 88 % (64 - 97 %) for CTA, 90 % (71 - 97%) for PET and 92 % (74 - 98%) for PET/CTA, and the positive predictive values (PPV) were 71 % (53 - 85%) for CTA, 87 % (68 - 95%) for PET and 100 % (84 - 100%) for PET/CTA. Similarly, on a per-vessel basis the NPVs (which were generally high) were 97 % (94 - 100%) for CTA, 95 % (90 - 99%) for PET and 97 % (95 - 100%) for PET/CTA, and the PPVs (which were lower, but higher with PET/CTA) were 53 % (39 - 66%) for CTA, 53 % (40 - 66%) for PET and 85 % (73 - 97%) for PET/CTA. In six patients, CTA analysis was hampered by the presence of severe calcifications. However, with the addition of the PET data, all six patients were correctly categorized. Conclusion: Cardiac quantitative hybrid PET/CTA imaging has better diagnostic accuracy than CTA alone and PET alone. CTA has a suboptimal PPV, suggesting that hybrid PET/CTA imaging should be used to assess the significance of coronary stenoses diagnosed by CTA.