Abstract
OBJECTIVES: Recent studies of margin-related recurrence have raised questions on the necessity of ensuring wide resection margins in the resection of colorectal liver metastases. The aim of the current study was to determine whether resection margins of 10 mm provide a survival benefit over narrower resection margins. METHODS: A total of 425 laparoscopic liver resections were carried out in 351 procedures performed in 317 patients between August 1998 and April 2012. Primary laparoscopic liver resections for colorectal metastases were included in the study. Two-stage resections, procedures accompanied by concomitant liver ablations and one case of perioperative mortality were excluded. A total of 155 eligible patients were classified into four groups according to resection margin width: Group 1, margins of < 1 mm [n = 33, including 17 patients with positive margins (Group 1a)]; Group 2, margins of 1 mm to < 3 mm (n = 31); Group 3, margins of ≥ 3 mm to <10 mm (n = 55), and Group 4, margins of ≥ 10 mm (n = 36). Perioperative and survival data were compared across the groups. Median follow-up was 31 months (range: 2-136 months). RESULTS: Perioperative outcomes were similar in all groups. Unfavourable intraoperative incidents occurred in 9.7% of procedures (including 3.2% of conversions). Postoperative complications developed in 11.0% of patients. Recurrence in the resection bed developed in three (1.9%) patients, including two (6.1%) patients in Group 1. Rates of actuarial 5-year overall, disease-free and recurrence-free survival were 49%, 41% and 33%, respectively. Median survival was 65 months. Margin status had no significant impact on patient survival. The Basingstoke Predictive Index (BPI) generally underestimated survival. This underestimation was especially marked in Group 1 when postoperative BPI was applied. CONCLUSIONS: Patients with margins of <1 mm achieved survival comparable with that in patients with margins of ≥ 10 mm. When modern surgical equipment that generates an additional coagulation zone is applied, the association between resection margin and survival may not be apparent. Further studies in this field are required. Postoperative BPI, which includes margin status among the core factors predicting postoperative survival, seems to be less precise than preoperative BPI.