3T magnetic resonance imaging accurately depicts radiofrequency ablation zones in a blood-perfused bovine liver model.

PURPOSE: To determine if noncontrast T1-weighted (T1W) images from 3T magnetic resonance (MR) imaging accurately depict radiofrequency (RF) ablation zones as determined macroscopically and microscopically in a blood-perfused bovine liver model. MATERIALS AND METHODS: Three-dimensional (3D) gradient-recalled echo (GRE) T1W images were obtained on a 3T MR imaging scanner after RF ablations (n = 14) of in vitro blood-perfused bovine livers. The resulting central hypointense and peripheral hyperintense signal regions were measured and compared with the inner tan and outer red zones of the gross specimen. Corresponding ablated hepatic tissue samples were examined microscopically and stained with nicotinamide adenine dinucleotide phosphate (NADPH) to assess for the presence or absence of NADPH diaphorase activity. Bootstrap two-sample hypothesis tests were used to compare MR imaging, gross, and histopathologic measurements. RESULTS: The MR imaging inner ablation zone had a mean radius of 0.80 cm (range 0.33-1.14 cm); the inner zone plus the outer ablation zone had a mean radius of 1.40 cm (range 1.01-1.74 cm). Comparison of the measurements of the inner ablation zone on MR imaging versus the gross specimen showed equivalence (95% confidence interval [CI] -0.122 cm, 0.223 cm). Comparison of the measurements of the outer ablation zone on MR imaging versus the gross and histologic specimens also showed equivalence (95% CI -0.095 cm, 0.244 cm, and -0.146 cm, 0.142 cm). CONCLUSIONS: Noncontrast 3D GRE T1W 3T MR imaging accurately depicts the RF ablation zones in a blood-perfused bovine liver model and can be used as a noninvasive means to assess the 3D morphologic characteristics of RF ablation lesions in the model.

In vitro blood-perfused bovine liver model: a physiologic model for evaluation of the performance of radiofrequency ablation devices.

PURPOSE: To describe an in vitro blood-perfused bovine liver model for the testing of radiofrequency (RF) ablation devices and compare the performance of a specific RF ablation device in the model relative to three other biologic models. MATERIALS AND METHODS: Fresh bovine livers were used to create three in vitro models: blood-perfused, Krebs-Henseleit (KH) solution-perfused, and nonperfused. The perfused models were connected to a heart-lung machine via the portal vein and perfused with heparinized autologous blood or KH solution under physiologic conditions. Six swine were used as in vivo liver models. A cluster electrode and RF ablation system was operated in impedance mode for 12 minutes in all models. Ablated livers were sectioned, with long- and short-axis measurements of the ablations obtained, and statistical analysis was performed. RESULTS: A total of 39, 23, 17, and 12 ablations were performed in 14, 6, 5, and 6 blood-perfused bovine livers, KH solution-perfused bovine livers, nonperfused bovine livers, and in vivo porcine livers, respectively. On cut specimens, the average diameters of ablation zones were 4.00 cm (95% CI, 3.88-4.13) in blood-perfused livers, 4.34 cm (95% CI, 4.14-4.50) in KH solution-perfused livers, 4.67 cm (95% CI, 4.50-4.83) in nonperfused livers, and 3.56 cm (95% CI, 3.26-3.83) in in vivo porcine livers. In all models, the ablation zone diameters were normally distributed. CONCLUSIONS: In the in vitro blood-perfused bovine liver model, the size of ablations produced by an RF ablation device are closer in size to those seen in porcine liver in vivo compared with the lesions produced in KH solution-perfused or nonperfused bovine liver.