Predictive factors for postoperative recurrence of hepatocellular carcinoma

Shuichi Okada, Kazuaki Shimada, Junji Yamamoto, Tadatoshi Takayama, Tomoo Kosuge, Susumu Yamasaki, Michiie Sakamoto, Setsuo Hirohashi

Research output: Contribution to journalArticlepeer-review

263 Citations (Scopus)


Background/Aims: The survival rate for hepatocellular carcinoma after hepatectomy remains unsatisfactorily low because of the high recurrence rate. The purpose of this study was to elucidate predictive factors for postoperative recurrence of hepatocellular carcinoma. Methods: Univariate analysis and a multiple regression model were used to retrospectively determine the factors potentially related to recurrence in 98 patients with hepatocellular carcinoma who had undergone curative hepatic resection. Results: DNA diploidy, absence of alcohol abuse, and absence of vascular invasion were determined to be independent favorable predictive factors using a multivariate analysis with Cox's proportional hazards model. A predictive index was developed using these three factors according to the following equation: 1.486 × (0 = DNA diploidy or 1 = DNA aneuploidy) + 1.138 × (0 = absence or 1 = presence of alcohol abuse) + 0.946 × (0 = absence or 1 = presence of vascular invasion). This index was used to classify the patients into two groups with good or poor prognosis. The median time for disease recurrence for the two groups was 24.1 and 9.2 months, respectively (P < 0.001). Conclusions: The results may be useful in the determination of treatment strategies and postoperative follow-up schedules and in the design and analysis of future clinical trials of therapy for hepatocellular carcinoma.

Original languageEnglish
Pages (from-to)1618-1624
Number of pages7
Issue number6
Publication statusPublished - 1994 Jun
Externally publishedYes

ASJC Scopus subject areas

  • Hepatology
  • Gastroenterology


Dive into the research topics of 'Predictive factors for postoperative recurrence of hepatocellular carcinoma'. Together they form a unique fingerprint.

Cite this