Click here to


Are you sure ?

Yes, do it No, cancel

Assessing Myocardial Perfusion After Cardiac Irradiation Using Dynamic Contrast Enhanced Hybrid PET/MRI

O Chau1*, O El-Sherif2 , M Mouawad3 , F Prato4 , S Gaede5 , (1) London Regional Cancer Centre, London, ON, (2) Mayo Clinic, Rochester, MN, (3) Western University, London, ON, (4) Lawson Health Research Institute, London, ON, (5) London Regional Cancer Program, London, ON


(Sunday, 7/14/2019) 3:00 PM - 3:30 PM

Room: Exhibit Hall | Forum 5

Purpose: To determine the optimal curve fitting method for the Toft’s model to accurately quantify and detect radiation-induced changes in myocardial blood flow (MBF) parameter, Ktrans, measured with dual bolus Gd-DTPA based Dynamic Contrast Enhanced (DCE)-MRI.

Methods: Five canines received simultaneous N-13 Ammonia PET and dual bolus Gd-DTPA DCE-MR imaging at baseline, 1-week, 1, 3, 6, and 12-months after external beam irradiation targeted at the left anterior descending artery. Rest and adenosine-induced stress scans were conducted utilizing a fast gradient echo sequence on a Siemens 3T hybrid PET/MR scanner. The myocardium was contoured on the DCE-MRI dataset following a 17-segment-model. Since MBF = Ktrans/E, where E is the extraction fraction, the left ventricle was selected as the arterial input function (AIF). The myocardial tissue curves were fitted in MATLAB using Toft’s model for both dual bolus and separate bolus techniques. Ktrans from blood plasma to extracellular/extravascular space was then calculated. N-13 Ammonia PET defined MBF was also calculated using 1-compartment model in FlowQuant (Ottawa Heart Institute, Canada). Paired t-tests were used to compare MBF between DCE-MRI and N-13 PET for each curve fitting technique.

Results: Statistical significance was observed between MBF obtained by separate bolus DCE-MR curve fitting technique and N-13 PET (p<0.05) in all artery supplied regions under both rest and stress conditions. This indicated an AIF signal saturation effect. However, MBF obtained by the dual bolus DCE-MR technique was not statistically different than N-13 PET defined MBF (p>0.05). Also, DCE-MR Ktrans rest and stress data trended differently after the 1-month timepoint.

Conclusion: Dual bolus curve fitting using Toft’s model to determine myocardial Ktrans was necessary to accurately compute MBF with GD-DTPA MR imaging. However, we were not able to observe the same post-irradiation trends when comparing rest to stress MBF with both N13-PET and DCE-MR imaging.


Gd-DTPA, Myocardial Perfusion, Modeling


IM/TH- Image Analysis (Single modality or Multi-modality): Quantitative imaging

Contact Email