Magnetization transfer MRI for quantitative assessment of intestinal fibrosis in Crohnʼs disease

Abstract

To develop a non-invasive method for detecting and quantifying intestinal fibrosis in patients with Crohn's disease using magnetization transfer (MT) magnetic resonance imaging (MRI). MT-MRI is more specific than traditional MRI for detecting tissue fibrosis by differentiating between free tissue water and water closely associated with macromolecules such as collagen. We hypothesize that changes in the MT signal profile can be used to differentiate and quantitatively characterize intestinal fibrosis, and will be more sensitive and specific for fibrosis than traditional MRIT1 and T2 relaxation times. Intestinal fibrosis was experimentally induced in Lewis strain rats. During laparotomy peptidoglycan-polysaccharide (PG-PS) was injected subserosally into the intestinal wall of the cecum, ileum, and distal Peyer's patches of four rats. Five control animals were similarly injected with human serum albumin (HSA). The PG-PS injected animals developed chronic intestinal inflammation, granulomas, and intense fibrosis 21 days after injection. Ex vivo intestinal tissue samples were scanned in a Varian Inova 2.0 T, 31 cm clear bore system equipped with Acustar S-180 actively shielded gradients and an MR probe. For magnetization transfer, a series of RF pulses were applied from -100 Hz to + 100 kHz off-resonance at three different RF power levels. In vivo MT-MRI images of a whole rat abdomen were also obtained. After undergoing MT-MRI scanning, intestinal histopathology determined the extent of tissue fibrosis. The standard MR imaging parameters, T1 and T2 relaxation times, as well as the MT data were recorded for both PG-PS and HSA samples. There was no difference between the T1 and T2 relaxation times. The MT data were analyzed and reduced into factors indicating the percent solid matter and overall tissue rigidity. The MT ratio (MTR) = 100*(1 -Msat/Mo) where Msat represents MT signal at 10 kHz off-resonance, and Mo represents MT signal at 100 kHz off-resonance. The mean MTR from PG-PS rats was 46.27 (SD 4.14), and control rats was 38.07 (SD 4.76). This demonstrated a statistically significant difference (p = 0.03) between fibrotic and non-fibrotic tissue. Histopathology confirmed the presence of fibrosis in the PG-PS rats, and lack of fibrosis in the control rats. (1) As expected, the data demonstrate that standard T1 and T2 relaxation times were not appreciably different. This confirms that traditional MRI is not sufficiently sensitive to detect intestinal fibrosis. (2) The MT data collected from rats with experimental Crohn's were statistically significantly different than control rats without fibrosis. This supports our hypothesis that MT-MRI can be used to distinguish intestinal fibrosis from non-fibrotic tissue. Furthermore, the MT profile provides a quantification of tissue fibrosis which will allow the clinician to follow disease progression. This technique has the significant advantage over currently available imaging modalities without exposing the subject to ionizing radiation.