Clinical Validation of the Accuracy of Absolute Myocardial Blood Flow Quantification with Dual-Source CT Using O-Water PET.

Purpose: To determine the fitting equation that can correct for the underestimation of myocardial blood flow (MBF) measurement by using dynamic CT perfusion (CTP) with dual-source CT (MBF), using MBF with oxygen 15-labeled water (O-water) PET (MBF) as a reference, and to determine the accuracy of corrected MBF (MBF) compared with MBF in a separate set of participants.

Materials And Methods: In this prospective study (reference no. 2466), 34 participants (mean age, 70 years ± 8 [standard deviation]; 27 men) known or suspected to have coronary artery disease underwent dynamic stress CTP and stress O-water PET between January 2014 and December 2018. The participants were randomly assigned to either a pilot group ( = 17), to determine the fitting equation on the basis of the generalized Renkin-Crone model that can explain the relation between MBF and MBF, or to a validation group ( = 17), to validate MBF compared with MBF. The agreement between MBF and MBF was evaluated by intraclass correlation and Bland-Altman analysis.

Results: In the pilot group, MBF was lower than MBF (1.24 mL/min/g ± 0.28 vs 2.51 mL/min/g ± 0.89, < .001) at the segment level. The relationship between MBF and MBF was represented as MBF = MBF × {1-exp[-(0.11 × MBF + 1.54)/MBF]}. In the validation group, MBF was 2.66 mL/min/g ± 1.93, and MBF was 2.68 mL/min/g ± 1.87 at the vessel level. MBF showed an excellent agreement with MBF (intraclass correlation coefficient = 0.93 [95% CI: 0.87, 0.96]). The measurement bias of MBF and MBF was -0.02 mL/min/g ± 0.74.

Conclusion: Underestimation of MBF by CT was successfully corrected with a correction method that was based on contrast kinetics in the myocardium. CT, CT-Perfusion, PET, Cardiac, Heart © RSNA, 2021.