Purpose: Cardiac imaging with PET/CT allows measurement of coronary artery calcium (CAC), myocardial perfusion and coronary vascular function. We investigated whether the combined assessment of regional CAC score, ischemic total perfusion deficit (ITPD) and quantitative coronary vascular function would further improve the diagnostic accuracy of PET/CT in predicting obstructive coronary artery disease (CAD).
Methods: We analyzed 113 patients with suspected CAD referred to 82Rb PET/CT myocardial perfusion imaging with available coronary angiographic data. Obstructive CAD was defined as ≥75% stenosis. The receiver operating characteristic area under curve (AUC) was applied to evaluate the ability of CAC score, ITPD, hyperemic myocardial blood flow (MBF) and coronary flow reserve (CFR) to identify CAD.
Results: Vessels with obstructive CAD (71 vessels) had higher ITPD (4.6 ± 6.2 vs. 0.6 ± 1.3) and lower hyperemic MBF (1.01 ± 0.5 vs. 1.75 ± 0.6 ml/min/g) and CFR (1.56 ± 0.6 vs. 2.38 ± 0.7; all p < 0.001) than those without. In prediction of per-vessel CAD, the AUCs for the models including CAC/ITPD/hyperemic MBF (0.869) and CAC/ITPD/CFR (0.875) were higher (both p < 0.01) than for the model including CAC/ITPD (0.790). Compared with CAC/ITPD, continuous net reclassification improvement was 0.69 (95% bootstrap confidence interval, CI, 0.365-1.088) for the CAC/ITPD/hyperemic MBF model and 0.99 (95% bootstrap CI 0.64-1.26) for the CAC/ITPD/CFR model.
Conclusion: Hyperemic MBF and CFR provide incremental information about the presence of CAD over CAC score and perfusion imaging parameters. The combined use of CAC, myocardial perfusion imaging and quantitative coronary vascular function in may help predict more accurately the presence of obstructive CAD.
Keywords: Coronary artery calcium; Coronary artery disease; Coronary flow reserve; Hybrid PET/CT; Myocardial perfusion.