CT and MRI of experimentally induced mesenteric ischemia in a porcine model

Objective : Our goal was to assess the value of CT and MRI for the detection of bowel wall changes in experimentally induced mesenteric ischemia. Methods : In 18 female pigs, a percutaneous embolization of the superior mesenteric artery was performed with buthyl-2-cyanoacrylate and Lipiodol (1 :1) (experimental group). In six animals, only diagnostic imaging and histologic evaluation were performed (control group). CT was carried out 3, 6, and 12 h after occlusion. Incremental CT (1 s scan time, 5 mm slice thickness, 7 mm increment, 120 kV/290 mAs) and spiral CT (slice thickness 5 mm, pitch 1.5, 120 kV/165 mA) were performed pre and post contrast injection (Somatom Plus/Siemens). Serial CT was carried out after intravenous contrast injection (1 ml/kg, 2 ml/s). MRI (Magnetom 1.5 T ; Siemens) was performed with T1 (pre and post 0.01 mmol/kg Gd-DTPA ; Magnevist ; Schering, Germany), T2, and proton density images in axial orientation. Slice thickness was 3 mm and slice gap I mm. Additionally, a T1-weighted GE sequence (multislice FLASH 2D) was obtained in dynamic technique (before and 30, 60, and 90 s after contrast agent injection) with a slice thickness of 5 mm. Biometrical monitoring included blood pressure, heart frequency, blood cell count, electrolyte status, blood gas analysis, and determination of serum lactate. Image evaluation included morphological analysis and determination of the enhancement pattern. Histological specimens were obtained and analyzed according to the Chiu classification. Results : The histologic workup of the specimen 3, 6, and 12 h after vascular occlusion revealed an average Chiu state 3, 4, and 5. On CT, the bowel wall had a thickness of 4.7 mm on average in the ischemic segments. There was a significant difference from the control group (average 3 mm). Free intraperitoneal fluid and intramural gas were seen after 12 h of ischemia in 80%. In ischemic bowel segments, no mural enhancement was seen. Normal segments and the bowel of the control animals showed an enhancement of 34 HU on average (SD = 3.1 HU ; p < 0.01). In MRI, S/N and C/N differed significantly between experimental and control groups in T1 and proton density images. In ischemic segments of all phases, the bowel wall did not show contrast enhancement. Healthy segments and bowel of control animals showed a significant enhancement (p < 0.01). Conclusion : Cross-sectional imaging has a high sensitivity for delineation of ischemic bowel wall segments. The enhancement pattern of the bowel wall enables detection of location, extent, and cause of a acute arterial mesenteric ischemia with high accuracy in an early phase.