Predictors of Occlusion of Hepatic Blood Vessels after Irreversible Electroporation of Liver Tumors
What factors predict occlusion of the portal or hepatic veins following irreversible electroporation (IRE) for liver tumors?
Vessel occlusoin can be predicted based on vessel size (<=4 mm) and location (within the ablation zone)
Masashi Tamura et al. Predictors of Occlusion of Hepatic Blood Vessels after Irreversible Electroportaion of Liver Tumors. Journal of Vascular and Interventional Radiology. 2020: 31; 2033-2042.
Figure 1. A 76-year old man with colorectal cancer metastasis in segment 8. (a) CT before the procedure shows the metastatic liver tumor in segment 8 from colorectal cancer (arrowhead). Three portal vein branches were running near the tumor measure 5 mm (vessel A), 3.5 mm (vessel B), and 5 mm (vessel C) from theventral side. (b) CT after the procedure shows teh ablation zone as a nonenhanced liver area. The 3 portal vein branches were judged as adjacent (vessel A: arrow A), within (vessel B: arrow B), and within (vessel C: arrow C). (c) On follow-up CT, although vessel A (arrow A) and vessel C (arrow C) remained patent, vessel B was occluded.
Unlike the more common modalities for percutaneous ablation, irreversible electroporation (IRE), is predominantly non-thermal. This avoids susceptibility to heat sink effects which can inhibit microwave ablation (MWA) and radiofrequency ablation (RFA1. While it has been suspected that IRE would not affect vessel patency, unlike MWA and RFA), this has not been fully studied.
This study evaluated 39 patients who underwent IRE to hepatic lesions and evaluated 33 portal veins and 64 hepatic veins that were within or adjacent to the ablation zone. On follow up imaging, 37% of portal veins and 27% of hepatic veins were occluded. Vessels <= 4mm showed a significantly higher rate of occlusion compared to vessels > 4 mm (72.7% versus 18.1% for portal veins and 54.8% vs 0% for hepatic veins). Additionally, vessels within the ablation zone were occluded at a significantly higher frequency than adjacent vessels (55.6.7% versus 13.3% for portal veins and 45.4% vs 6.4% for hepatic veins).
When choosing an ablation modality, common thinking is that IRE is preferred in locations where heat sink could prohibit a full ablation and where vessel damage needs to be minimized. While IRE does not suffer limitations of heat sink, potentially, we have minimized the possible affects IRE may have on vessels in and around the ablation margin. As this study demonstartes, venous occlusion is predictable using size and location compared to the ablation cavity. Given the clear difference in rate of occlusion based on size, central lesions are likely still great candidates for IRE where the portal veins and/or hepatic veins are likely to be larger than 4 mm, however potential venous occlusion should not be completely discounted.
David M Mauro, MD
Department of Radiology, Vascular and Interventional Radiology
University of North Carolina