Shingo Shimada1, Toshiya Kamiyama2, Hideki Yokoo2, Tatsuya Orimo2, Kenji Wakayama2, Takahiro Einama2, Tatsuhiko Kakisaka3, Hirofumi Kamachi2, Akinobu Taketomi2. 1. Department of Gastroenterological Surgery I, Hokkaido University Graduate School of Medicine, North15-West7, Kita-Ku, Sapporo, Hokkaido, 060-8638, Japan. shingoshimada1979@true.ocn.ne.jp. 2. Department of Gastroenterological Surgery I, Hokkaido University Graduate School of Medicine, North15-West7, Kita-Ku, Sapporo, Hokkaido, 060-8638, Japan. 3. Department of Surgery, Sapporo Kousei Hospital, Sapporo, Japan.
Abstract
BACKGROUND: The aims of this study were to investigate predictive factors for microscopic portal venous invasion (mPVI) in hepatocellular carcinoma (HCC) and whether anatomical liver resection (ALR) was useful in such cases. METHODS: We analyzed 852 patients with HCC without macroscopic portal venous invasion who were treated at our hospital between January 1990 and May 2014. These patients were stratified into a microscopic portal venous invasion group (mPVI group; n = 153) and non-microscopic portal venous invasion group (NmPVI group; n = 699). RESULTS: PIVKA-II ≥100 mAU/ml, a tumor size ≥5 cm, a confluent lesion, and poor differentiation were found to be independent risk factors for mPVI. Among the mPVI group who had single HCC under 5 cm, serum albumin level <4.0 g/dl, PIVKA-II ≥100 mAU/ml, a positive surgical margin, and non-ALR (NALR) were independent unfavorable prognostic factors for overall survival (OS). PIVKA-II ≥100 mAU/ml, a positive surgical margin and NALR were independent unfavorable prognostic factors for relapse-free survival (RFS). ALR was significantly favorable factor for both OS and RFS of the mPVI group who had single HCC under 5 cm. CONCLUSIONS: Even if no portal venous invasion is detectable in HCC patients preoperatively, a PIVKA-II ≥100 mAU/ml, tumor size ≥5 cm, and a confluent lesion indicate a high risk of mPVI. ALR should be considered for the patients with these characteristics because it is a favorable prognostic factor in these cases with mPVI.
BACKGROUND: The aims of this study were to investigate predictive factors for microscopic portal venous invasion (mPVI) in hepatocellular carcinoma (HCC) and whether anatomical liver resection (ALR) was useful in such cases. METHODS: We analyzed 852 patients with HCC without macroscopic portal venous invasion who were treated at our hospital between January 1990 and May 2014. These patients were stratified into a microscopic portal venous invasion group (mPVI group; n = 153) and non-microscopic portal venous invasion group (NmPVI group; n = 699). RESULTS: PIVKA-II ≥100 mAU/ml, a tumor size ≥5 cm, a confluent lesion, and poor differentiation were found to be independent risk factors for mPVI. Among the mPVI group who had single HCC under 5 cm, serum albumin level <4.0 g/dl, PIVKA-II ≥100 mAU/ml, a positive surgical margin, and non-ALR (NALR) were independent unfavorable prognostic factors for overall survival (OS). PIVKA-II ≥100 mAU/ml, a positive surgical margin and NALR were independent unfavorable prognostic factors for relapse-free survival (RFS). ALR was significantly favorable factor for both OS and RFS of the mPVI group who had single HCC under 5 cm. CONCLUSIONS: Even if no portal venous invasion is detectable in HCC patients preoperatively, a PIVKA-II ≥100 mAU/ml, tumor size ≥5 cm, and a confluent lesion indicate a high risk of mPVI. ALR should be considered for the patients with these characteristics because it is a favorable prognostic factor in these cases with mPVI.