Ablative Liver Partition and Portal Vein Embolization: Proof-of-Concept Testing in a Rabbit Model

To test the hypothesis that a modified approach to portal vein embolization (PVE)-termed ablative liver partition (ALP) and PVE (ALP-PVE)-is feasible and results in greater future liver remnant (FLR) growth compared with PVE alone in a rabbit model.

Eighteen rabbits (median weight, 2.7 kg) underwent PVE (n = 9) or ALP-PVE (n = 9). PVE to cranial liver lobes was performed with 100-300-μm microspheres and metallic coils; the caudal lobe was spared as the FLR. In the ALP-PVE cohort, a liver partition between cranial and caudal lobes was created by using microwave ablation (40 W, 1 min). Animals were euthanized and livers were harvested on postprocedure day 7. Caudal and cranial liver lobes were weighed after 4 weeks of oven drying. Ki-67 immunohistochemistry was used to quantify liver mitotic index. ALP-PVE feasibility was determined based on procedure technical success. Standardized FLR (sFLR; ie, FLR divided by whole liver weight) and mitotic index were compared between PVE and ALP-PVE groups by two-tailed independent-samples Mann-Whitney U test.

One PVE-group rabbit died during anesthesia induction and was excluded from technical success calculation. Eight of 8 (100%) and 8 of 9 rabbits (89%) underwent technically successful PVE and ALP-PVE, respectively. There was no difference in sex or weight distribution between groups. sFLR (0.32 vs 0.29; P = .022) and mitotic index (17.5% vs 6.2%; P = .051) were higher in ALP-PVE vs PVE caudal lobes when the first "learning-curve" case from each group was excluded.

ALP-PVE is feasible and may stimulate greater FLR growth compared with PVE in a rabbit model.