Prognostic impact of marginal resection for patients with solitary hepatocellular carcinoma: evidence from 570 hepatectomies

Abstract

Background: During resection of a hepatocellular carcinoma, surgeons encounter occasionally a situation where marginal resection is inevitable because of a close association between the hepatocellular carcinoma and major vasculature and/or underlying impaired liver function. We investigated the impact of marginal resection on recurrence-free survival after a resection of a solitary hepatocellular carcinoma.

Methods: The data of 570 patients who underwent macroscopically curative hepatectomy for a solitary hepatocellular carcinoma in our institution between 1990 and 2007 were analyzed. Marginal resection and non-marginal resection were defined as a cancer-negative surgical margin of ≤ 1 mm and a surgical margin of >1 mm, respectively. The macroscopic appearance of the hepatocellular carcinoma was classified as the simple nodular type or non-simple nodular type based on the classification of the Liver Cancer Study Group of Japan, and patients were categorized into 4 groups: group A, simple nodular type with cirrhosis; group B, simple nodular type without cirrhosis; group C, non-simple nodular type with cirrhosis; and group D, non-simple nodular type without cirrhosis.

Results: The surgical margins were diagnosed as cancer-positive in 31 patients, as marginal resection in 165 patients, and as non-marginal resection in 374 patients. The marginal resection group showed a better recurrence-free survival than the positive surgical margin group (P = .001), and also a worse recurrence-free survival than the non-marginal resection group (P = .003). In groups A, B, and C, the recurrence-free survival rates were similar between marginal resection and non-marginal resection patients (P = .458), while in group D, marginal resection was a significant poor prognostic factor of recurrence-free survival in both univariate and multivariate analyses.

Conclusion: Marginal resection is acceptable in group A, B, and C patients, because it did not negatively affect postoperative recurrence-free survival.