MRI and histopathologic study of a novel cholesterol-fed rabbit model of xanthogranuloma

To develop a rabbit model of xanthogranuloma based on supplementation of dietary cholesterol. The aim of this study was to analyze the xanthogranulomatous lesions using magnetic resonance imaging (MRI) and histological examination.

Rabbits were fed a low-level cholesterol (CH) diet (n = 10) or normal chow (n = 5) for 24 months. In vivo brain imaging was performed on a 3T MR system using fast imaging employing steady state acquisition, susceptibility-weighted imaging, spoiled gradient recalled, T1 -weighted inversion recovery imaging and T1 relaxometry, PD-weighted and T2 -weighted spin-echo imaging and T2 relaxometry, iterative decomposition of water and fat with echo asymmetry and least-squares estimation, ultrashort TE MRI (UTE-MRI), and T2* relaxometry. MR images were evaluated using a Likert scale for lesion presence and quantitative analysis of lesion size, ventricular volume, and T1 , T2 , and T2* values of lesions was performed. After imaging, brain specimens were examined using histological methods.

In vivo MRI revealed that 6 of 10 CH-fed rabbits developed lesions in the choroid plexus. Region-of-interest analysis showed that for CH-fed rabbits the mean lesion volume was 8.5 ± 2.6 mm(3) and the volume of the lateral ventricle was significantly increased compared to controls (P < 0.01). The lesions showed significantly shorter mean T2 values (35 ± 12 msec, P < 0.001), longer mean T1 values (1581 ± 146 msec, P < 0.05), and shorter T2* values (22 ± 13 msec, P < 0.001) compared to adjacent brain structures. The ultrashort T2* components were visible using UTE-MRI. Histopathologic evaluation of lesions demonstrated features of human xanthogranuloma.

Rabbits fed a low-level CH diet develop sizable intraventricular masses that have similar histopathological features as human xanthogranuloma. Multiparametric MRI techniques were able to provide information about the complex composition of these lesions. J. Magn. Reson. Imaging 2016;44:673-682.