Delayed portal vein thrombosis after experimental radiofrequency ablation near the main portal vein

Abstract

Background: Portal venous blood flow may protect adjacent tumour cells from thermal destruction with radiofrequency ablation (RFA). This study aimed to investigate the local effect of RFA on the main portal vein branch, and the completeness of cellular ablation in its vicinity, with or without a Pringle manoeuvre using a porcine model. Methods: This was an in vivo study on 23 domestic pigs. RFA using a cooled-tip electrode was performed 5 mm from the left main portal vein branch under ultrasonographic guidance for 12 min with (n = 10) or without (n = 10) a Pringle manoeuvre. Ten pigs were killed 4 h after the procedure to study the early effects of RFA and ten others were killed 1 week later to determine any delayed effect. As a control, sham operations with a Pringle manoeuvre for 12 min were performed on three pigs. The flow velocity changes of portal vein and hepatic artery were measured using Doppler ultrasonography, and the completeness of cellular ablation around the portal vein was assessed qualitatively by histochemical staining and quantitatively by measuring intracellular levels of adenosine 5′-triphosphate (ATP). Results: In the absence of the Pringle manoeuvre, there was no significant change in mean(s.d.) portal vein flow velocity before RFA (20.0(3.5) cm/s) and at 4 h (18.5(2.5) cm/s) (P = 0.210) and 1 week (19.5(2.2) cm/s) (P = 0.500) after the procedure. Gross and histological examination of the portal vein branches showed no damage without the Pringle manoeuvre. In all pigs that underwent RFA with a Pringle manoeuvre, the portal vein was occluded 1 week after the operation; histological examination of the affected portal vein showed severe thermal injury and associated venous thrombosis. The local effect of RFA on the hepatic artery was similar. With intact portal blood flow during RFA, complete ablation of liver tissue around the pedicle was demonstrated by histochemical staining and measurement of the intracellular ATP concentration. Conclusion: RFA was safe when applied close to the main portal vein branch without a Pringle manoeuvre, with complete cellular destruction. Use of the Pringle manoeuvre resulted in delayed portal vein and hepatic artery thrombosis and injury to the hepatic artery and bile duct.