Quantitative Perfusion Analysis for Transcatheter Intraarterial Perfusion MR Imaging

Introduction: Transcatheter arterial embolization (TAE) and chemoembolization (TACE) are established treatment methods for unresectable liver tumor. TRanscatheter Intraarterial Perfusion (TRIP)-MRI is an intra-procedural technique to monitor liver tumor perfusion changes during TAE [1] and TACE [2]. However, previous developed TRIP-MRI approaches either used semi-quantitative perfusion analyses which have poorly-defined links to blood flow, or used a peak gradient method [3] which can oversimplify the description of contrast tracer kinetics, to calculate blood flow. In this study, we presented a potentially superior quantitative TRIP-MRI perfusion analysis approach, and evaluated its efficacy in a gel perfusion phantom and in rabbits with VX2 liver tumors during TAE. Methods: All experiments were performed using a 1.5T clinical MRI scanner (Siemens Magnetom Espree). In phantom studies, we used a chromatography column packed with Sephadex gel as a perfusion phantom. The phantom flow rate was adjusted from 24 mL/min to 2 mL/min at a 2 mL/min interval, and TRIP-MRI measurement was performed at each flow rate. In animal studies, we surgically implanted VX2 carcinoma into the left liver lobe of 12 rabbits. 3 weeks after implantation, we catheterized each rabbit under angiographic guidance to super-selectively deliver 40-120 μm embolic microspheres to liver tumors. After raabits were transfer to MRI scanner, TRIP-MRI measurements were performed before and after TAE. A quantitative TRIP-MRI measurement involved 3D B1 mapping using catalyzed double-angle method (60°/120°) [4], baseline 3D R10 mapping using GRE variable flip angle method (2°, 9°, 15°,19°), and dynamic 3D R1 mapping using dynamic GRE sequence at 15° flip angle after intraarterial injection of Gd-DTPA contrast agent. Other dynamic imaging parameters included: phantom studies: TR/TE = 5/1.62 ms, 320×160×40 mm FOV, 1.2 sec sampling rate; animal studies: TR/TE = 6/1.62 ms, 200×113×40 mm FOV, 1.6 sec sampling rate. With B1 calibration and baseline R10 map, an R1 map time series and further contrast concentration map series were derived from each TRIP-MR image series [3]. We applied the commonly used modified Kety model to describe contrast tracer pharmacokinetics [5] (Eq. 1). Given that the super-selective transcatheter bolus injection temporarily suppress antegrade blood flow and control the maximum vascular contrast agent concentration at the catheter tip immediately proximal to the tumor tissues, we estimated Cp(t) using prior information about the bolus injection parameters (Eq. 2).