Assessment of liver ischemia reperfusion injury in mice using hepatic T2 mapping: Comparison with histopathology.

BACKGROUND: Liver ischemia reperfusion injury (IRI) occurs during liver surgery or transplantation resulting in an inflammatory response, tissue damage, and functional impairment of the organ. PURPOSE: To assess the feasibility of T2 mapping for noninvasive quantification of liver edema after partial liver IRI in mice. STUDY TYPE: Prospective, experimental study. ANIMAL MODEL: Partial liver IRI was induced in C57BL/6-mice by transient clamping of the left lateral and median liver lobes for 35 (n = 8), 45 (n = 6), 60 (n = 17), or 90 minutes (n = 5). For comparison, healthy C57BL/6-mice were examined as controls (n = 9). FIELD STRENGTH/SEQUENCE: Functional liver MRI was performed on a 7T scanner using a respiratory-triggered multiecho spin-echo sequence. ASSESSMENT: Healthy control mice and mice with partial liver IRI on day 1 after surgery, and additionally on day 7 in a subgroup with 60 minutes IRI (n = 8) were examined. Maps of T2 relaxation time of liver tissue were used to assess distribution, severity of tissue edema (mean T2 time), and the percentage of edematous liver tissue. STATISTICAL TEST: One-way analysis of variance (ANOVA) with Tukey's honest significant difference (HSD), paired t-tests, Pearson's test for correlation of MRI parameters with levels of liver enzymes, and histopathology, receiver operating characteristic (ROC) analysis. RESULTS: Significant tissue edema induced by liver IRI as compared to the control group was detected by increased mean T2 times in groups with 60 minutes (P < 0.001) and 90 minutes IRI (P < 0.001). The percentage of edematous liver tissue significantly increased with longer ischemia times (controls 3.4 ± 0.4%, 35 minutes 5.3 ± 0.6%, 45 minutes 23.3 ± 7.6%, 60 minutes 39.7 ± 3.6%, 90 minutes 51.3 ± 4.5%). Mean T2 times and the percentage of edematous liver tissue significantly correlated with elevation of liver enzymes (P < 0.001), histological evidence of liver injury (r = 0.80 and r = 0.82, P < 0.001), and neutrophil infiltration (r = 0.70 and r = 0.74, P < 0.001). In the subgroup with follow-up, the severity (P < 0.01) and extent of liver edema decreased significantly over time (P < 0.01). DATA CONCLUSION: T2 mapping allows quantification and follow-up of liver injury in mice. LEVEL OF EVIDENCE: 1 Technical Efficacy: Stage 2 J. Magn. Reson. Imaging 2018;48:1586-1594.

Gd-EOB-DTPA-enhanced MRI for quantitative assessment of liver organ damage after partial hepatic ischaemia reperfusion injury: correlation with histology and serum biomarkers of liver cell injury.

OBJECTIVE: To evaluate Gd-EOB-DTPA-enhanced MRI for quantitative assessment of liver organ damage after hepatic ischaemia reperfusion injury (IRI) in mice. METHODS: Partial hepatic IRI was induced in C57Bl/6 mice (n = 31) for 35, 45, 60 and 90 min. Gd-EOB-DTPA-enhanced MRI was performed 1 day after surgery using a 3D-FLASH sequence. A subgroup of n = 9 animals with 60 min IRI underwent follow-up with MRI and histology 7 days after IRI. The total liver volume was determined by manual segmentation of the entire liver. The volume of functional, contrast-enhanced liver parenchyma was quantified by a region growing algorithm (visual threshold) and an automated segmentation (Otsu's method). The percentages of functional, contrast-enhanced and damaged non-enhanced parenchyma were calculated according to these volumes. MRI data was correlated with serum liver enzyme concentrations and histologically quantified organ damage using periodic acid-Schiff (PAS) staining. RESULTS: The percentage of functional (contrasted) liver parenchyma decreased significantly with increasing ischaemia times (control, 94.4 ± 3.3%; 35 min IRI, 89.3 ± 4.1%; 45 min IRI, 87.9 ± 3.3%; 60 min IRI, 68 ± 10.5%, p < 0.001 vs. control; 90 min IRI, 55.9 ± 11.5%, p < 0.001 vs. control). The percentage of non-contrasted liver parenchyma correlated with histologically quantified liver organ damage (r = 0.637, p < 0.01) and serum liver enzyme elevations (AST r = 0.577, p < 0.01; ALT r = 0.536, p < 0.05). Follow-up MRI visualized recovery of functional liver parenchyma (71.5 ± 8.7% vs. 84 ± 2.1%, p < 0.05), consistent with less histological organ damage on day 7. CONCLUSION: We demonstrated the feasibility of Gd-EOB-DTPA-enhanced MRI for non-invasive quantification of damaged liver parenchyma following IRI in mice. This novel methodology may refine the characterization of liver disease and could have application in future studies targeting liver organ damage. KEY POINTS: • Prolonged ischaemia times in partial liver IRI increase liver organ damage. • Gd-EOB-DTPA-enhanced MRI at hepatobiliary phase identifies damaged liver volume after hepatic IRI. • Damaged liver parenchyma quantified with MRI correlates with histological liver damage. • Hepatobiliary phase Gd-EOB-DTPA-enhanced MRI enables non-invasive assessment of recovery from liver injury.