Comparison of Intravoxel Incoherent Motion Parameters across MR Imagers and Field Strengths: Evaluation in Upper Abdominal Organs

Purpose: To determine the reproducibility of intravoxel incoherent motion (IVIM) parameters measured in upper abdominal organs with magnetic resonance (MR) imagers from different vendors and with different field strengths. Materials and Methods: This prospective study was approved by the independent ethics committees of Kanton Bern and Kanton Zurich, and signed informed consent was obtained from all participants. Abdominal diffusion-weighted images in 10 healthy men (mean age, 37 years 6 8 [standard deviation]) were acquired by using 1.5- and 3.0-T MR imagers from three different vendors. Two readers independently delineated regions of interest that were used to measure IVIM parameters (diffusion coefficient [D-t], perfusion fraction [F-p], and pseudodiffusion coefficient [D-p]) in the left and right lobes of the liver, and in the pancreas, spleen, renal cortex, and renal medulla. Measurement reproducibility between readers was assessed with intraclass correlation coefficients (ICCs). Variability across MR imagers was analyzed by using between- and within-subject coefficients of variation (CVs) and analysis of variance (ANOVA). Results: Between-reader reproducibility was high for D-t (ICC, 94.6%), intermediate for F-p (ICC, 81.7%), and low for D-p (ICC, 69.5%). Between- and within-subject CVs of D-t were relatively high (>20%) in the left lobe of the liver and relatively low (<10%) in the renal cortex and renal medulla. CVs generally exceeded 15% for F-p values and 20% for D-p. ANOVA indicated significant differences (P < .05) between MR imagers. Conclusion: IVIM parameters in the upper abdomen may differ substantially across MR imagers. (C) RSNA, 2015