Preoperative portal vein embolization: an audit of 84 patients.

Preoperative portal vein embolization (PVE) was performed in 84 patients before extensive liver resection for various diseases. By the criteria of liver volumetric determination, some patients were candidates for PVE, whereas others were not, even though the same surgical procedure, such as extended right lobectomy (ERL), was scheduled. PVE using gelatin sponge powder induced hypertrophy in the nonembolized lobe (0%-171%; median, 30%) and proportional atrophy in the embolized lobe in 2 weeks without eliciting any major inflammatory or necrotic reaction, as evidenced histologically and by the minimal elevations in the serum aspartate transaminase (AST) and alanine transaminase (ALT) values. Alterations in the total bilirubin level and prothrombin time were also insignificant and transient, indicating that hepatocyte functions were not impaired by PVE. Not all patients who undergo PVE proceed with the scheduled hepatic resection procedure, so it is a great advantage that gelatin sponge causes minimal damage compared with other embolizing materials such as cyanoacrylate and absolute ethanol, which have been reported to induce an inflammatory reaction or histological alteration. Our multiple regression analysis showed that three factors, diabetes mellitus, a high total bilirubin level at the time of PVE, and being male, each reduced the extent of hypertrophy in the nonembolized lobe (r2 =.30). By contrast, cholestasis appeared to accelerate the process of atrophy in the embolized lobe (r2 =.16). In conclusion, PVE by gelatin sponge powder is a safe and effective preoperative maneuver that induces hypertrophy of the section of the liver that will remain after partial hepatectomy.