A Predictive Factor of Insufficient Liver Regeneration after Preoperative Portal Vein Embolization

• Published in: European surgical research Vol. 51; no. 3-4; pp. 118 - 128
• Main Authors: Mihara, K., Sugiura, T., Okamura, Y., Kanemoto, H., Mizuno, T., Moriguchi, M., Aramaki, T., Uesaka, K.
• Format: Journal Article
• Language: English
• Published: Basel, Switzerland S. Karger AG 01.01.2013
• Subjects: Aged; Embolization, Therapeutic; Female; Hepatectomy - methods; Humans; Invited Review; Liver Regeneration; Male; Middle Aged; Portal Vein; Preoperative Care; Extended hepatectomy; Biliary cancer; Postoperative liver failure; Portal vein embolization; Liver regeneration
• ISSN: 0014-312X; 1421-9921
• DOI: 10.1159/000356368

Background: Preoperative portal vein embolization (PVE) is performed to enhance the future remnant liver function (FRLF) and volume (FRLV). However, the volume of the nonembolized liver does not increase enough in some patients, which results in an insufficient FRLF. The aim of this study was to evaluate the predictors of insufficient FRLF after PVE for extended hepatectomy. Methods: This retrospective study included 172 patients (107 patients with cholangiocarcinoma, 40 patients with metastatic liver cancer and 25 patients with hepatocellular carcinoma) who underwent PVE before extended hepatectomy. The total liver function was evaluated by measuring the indocyanine green plasma clearance rate (KICG). Computed tomography volumetry was conducted to evaluate the total liver volume and FRLV. The KICG of the future remnant liver (remK) was calculated using the following formula: KICG × FRLV/total liver volume. The safety margin for hepatectomy was set at remK after PVE (post-PVE remK) ≥0.05. Results: One hundred and twenty-three patients with a post-PVE remK level of >0.05 underwent hepatectomy without postoperative liver failure [sufficient liver regeneration (SLR) group], and 9 patients with a post-PVE remK level of <0.05 did not due to insufficient FRLF [insufficient liver regeneration (ILR) group]. In the SLR group, the KICG values did not change after PVE (median, 0.144-0.146, p = 0.523); however, the %FRLV and remK increased significantly (35.0-44.3%, p < 0.001 and 0.0488-0.0610, p < 0001, respectively). In contrast, in the ILR group, the KICG values decreased significantly (0.128-0.108, p = 0.021) and the %FRLV increased marginally (27.4-32.6%, p = 0.051). As a result, the remK did not increase significantly (0.0351-0.0365, p = 0.213). A receiver operating characteristic curve demonstrated an remK value of 0.04 obtained before PVE (pre-PVE remK) to be the optimal cutoff point for defective liver regeneration. The univariate and multivariate analyses revealed that a pre-PVE remK value of <0.04 was a factor for ILR. It was also correlated with postoperative liver failure in the analysis of the patients who underwent hepatectomy. Conclusions: The patients in the ILR group did not achieve SLR after PVE due to a significant decrease in the KICG and an insufficient increase in %FRLV. A pre-PVE remK value of <0.04 is a useful predictor of insufficient regeneration of the nonembolized liver, even after PVE.