Intravoxel Incoherent Motion and Dynamic Contrast-Enhanced Magnetic Resonance Imaging to Early Detect Tissue Injury and Microcirculation Alteration in Hepatic Injury Induced by Intestinal Ischemia-Reperfusion in a Rat Model

Abstract

Background: Intravoxel incoherent motion (IVIM) can provide quantitative information about water diffusion and perfusion that can be used to evaluate hepatic injury, but it has not been studied in hepatic injury induced by intestinal ischemia-reperfusion (IIR). Dynamic contrast-enhanced (DCE) magnetic resonance imaging (MRI) can provide perfusion data, but it is unclear whether it can provide useful information for assessing hepatic injury induced by IIR.

Purpose: To examine whether IVIM and DCE-MRI can detect early IIR-induced hepatic changes, and to evaluate the relationship between IVIM and DCE-derived parameters and biochemical indicators and histological scores.

Study type: Prospective pre-clinical study.

Population: Forty-two male Sprague-Dawley rats.

Field strength/sequence: IVIM-diffusion-weighted imaging (DWI) using diffusion-weighted echo-planar imaging sequence and DCE-MRI using fast spoiled gradient recalled-based sequence at 3.0 T.

Assessment: All rats were randomly divided into the control group (Sham), the simple ischemia group, the ischemia-reperfusion (IR) group (IR1h, IR2h, IR3h, and IR4h) in a model of secondary hepatic injury caused by IIR, and IIR was induced by clamping the superior mesenteric artery for 60 minutes and then removing the vascular clamp. Advanced Workstation (AW) 4.6 was used to calculate the imaging parameters (apparent diffusion coefficient [ADC], true diffusion coefficient [D], perfusion-related diffusion [D^* ] and volume fraction [f]) of IVIM. OmniKinetics (OK) software was used to calculate the DCE imaging parameters (K^trans , K_ep , and V_e ). Alanine aminotransferase (ALT) and aspartate aminotransferase (AST) were analyzed with an automatic biochemical analyzer. Superoxide dismutase (SOD) activity was assessed using the nitro-blue tetrazolium method. Malondialdehyde (MDA) was determined by thiobarbituric acid colorimetry. Histopathology was performed with hematoxylin and eosin staining.

Statistical tests: One-way analysis of variance (ANOVA) and Bonferroni post-hoc tests were used to analyze the imaging parameters and biochemical indicators among the different groups. Pearson correlation analysis was applied to determine the correlation between imaging parameters and biochemical indicators or histological score.

Results: ALT and MDA reached peak levels at IR4h, while SOD reached the minimum level at IR4h (all P < 0.05). ADC, D, D^* , and f gradually decreased as reperfusion continued, and K^trans and V_e gradually increased (all P < 0.05). The degrees of change for f and V_e were greater than those of other imaging parameters at IR1h (all P < 0.05). All groups showed good correlation between imaging parameters and SOD and MDA (r[ADC] = 0.615, -0.666, r[D] = 0.493, -0.612, r[D^* ] = 0.607, -0.647, r[f] = 0.637, -0.682, r[K^trans ] = -0.522, 0.500, r[V_e ] = -0.590, 0.665, respectively; all P < 0.05). However, the IR groups showed poor or no correlation between the imaging parameters and SOD and MDA (P [K^trans and MDA] = 0.050, P [D and SOD] = 0.125, P [the remaining imaging parameters] < 0.05). All groups showed a positive correlation between histological score and K^trans and V_e (r = 0.775, 0.874, all P < 0.05), and a negative correlation between histological score and ADC, D, f, and D^* (r = -0.739, -0.821, -0.868, -0.841, respectively; all P < 0.05). For the IR groups, there was a positive correlation between histological score and K^trans and V_e (r = 0.747, 0.802, all P < 0.05), and a negative correlation between histological score and ADC, D, f, and D^* (r = -0.567, -0.712, -0.715, -0.779, respectively; all P < 0.05).

Data conclusion: The combined application of IVIM and DCE-MRI has the potential to be used as an imaging tool for monitoring IIR-induced hepatic histopathology.

Level of evidence: 1 TECHNICAL EFFICACY STAGE: 2.

Keywords: dynamic contrast-enhanced MRI; intestinal ischemia-reperfusion; intravoxel incoherent motion; secondary hepatic injury.

FIGURE 1

Demonstration of intravoxel incoherent motion (IVIM) and dynamic contrast‐enhanced (DCE) magnetic resonance imaging. (a) Axial T₂WI image of the liver. (b) Corresponding ADC map from monoexponential fitting with 9 b‐values. (c–e) Parametric maps of IVIM (D ^*, D, and f) from biexponential fitting with 9 b‐values. (f) Coronal T₂WI image of the liver. (g–i) Parametric maps of DCE obtained from improved the Tofts–Kermode two‐compartment model.

FIGURE 2

Changes in tissue diffusion and perfusion studied by intravoxel incoherent motion (IVIM) and dynamic contrast‐enhanced (DCE) parameters. (a) The relative change in ADC (black line), D (red line), D ^* (green line), and f (yellow line) at sham, 1 hour (I1h), 2 hours (IR1h), 3 hours (IR2h), 4 hours (IR3h), and 5 hours (IR4h). (b) The relative change in K ^trans (yellow line) and V _e (red line) at sham, 1 hour (I1h), 2 hours (IR1h), 3 hours (IR2h), 4 hours (IR3h), and 5 hours (IR4h).

FIGURE 3

Changes in diffusion and perfusion in tissue studied by corresponding parameters in intravoxel incoherent motion (IVIM) and dynamic contrast‐enhanced (DCE) magnetic resonance imaging. (a–f) Scatter plots showing the correlations between histological score and ADC value (r = −0.739, P < 0.001), D value (r = −0.821, P < 0.001), D ^* value (r = −0.841, P < 0.001), f value (r = −0.868, P < 0.001), K ^trans value (r = 0.775, P < 0.001), and V _e value (r = 0.874, P < 0.001) for all groups. (j–l) Scatter plots showing the correlation between histological score and ADC value (r = −0.567, P < 0.001), D value (r = −0.712, P < 0.001), D ^* value (r = −0.779, P < 0.001), f value (r = −0.715, P < 0.001), K ^trans value (r = 0.747, P < 0.001), and V _e value (r = 0.802, P < 0.001) for ischemia–reperfusion groups.

FIGURE 4

Representative histological liver sections (H&E stain; magnification, ×200) at different time points during the course of IIR. (a) Normal liver structure in the Sham group. (b) In the I1h group, a small number of proliferating hepatocytes can be observed. (c) In the IR1h group, pyknosis is apparent in some hepatocytes. (d) In the IR2h group, hyperemia was observed in the hepatic sinus with surrounding cloudy and swollen hepatocytes. (e) In the IR3h group, the number of hepatocytes with pyknosis increased significantly, and congestion was obvious in the hepatic tissues. (f) In the IR4h group, significant degeneration and necrosis of hepatocytes is observed, in addition to disappearance of the hepatic boundary plate and destruction of the sinus cord structure.