Resection of liver metastases from colorectal cancer: the real impact of the surgical margin.

AIMS The benefit of liver resection for metastatic colorectal cancer is now established. Nevertheless if the surgical margin on pre-operative imaging is considered to be less than 10 mm, this is considered an absolute contraindication to surgery by some, and a relative contraindication by others, so its real impact on prognosis is not clear. METHODS From 1984 to 1996, 269 patients underwent hepatectomy for liver metastases and were prospectively studied. The only two objectives of this surgery were to be curative (or achieve complete R0 resection), and to avoid mortality. Of the 269, 187 patients had surgical margins inferior to 10 mm. Sixty per cent had multiple liver metastases, and 37% had extrahepatic metastatic sites. Their clinical and pathological factors were specifically studied. RESULTS The crude 5-year survival of these 187 patients (including the 2% post-operative mortality) was 24.7%, and the disease-free survival was 18.8%. The surgical margin was 0 mm in 60 cases and was histologically invaded in 20 cases. The most important prognostic factor was whether the resection was considered palliative (R1-R2 resection according to UICC criteria) (P < 0.0001). When the cases with invaded margins were excluded, there was not prognostic difference between the 107 patients with a margin of 0-4 mm and the 143 patients with a margin greater than 4 mm. However, a surgical margin greater than 9 mm appears to be a second prognostic factor (P = 0.001), when these 187 patients are compared to others. The reasons behind this are that there is a close relationship between narrow margins and extensive disease (high number of metastases, bilateral localization and extended hepatectomy), and also an increased possibility of microscopic satellite lesions within 10 mm around the metastases. CONCLUSION The real prognostic impact of the surgical margin must not be overestimated. Hepatectomy for metastases can provide long-term survival in patients with supposed poor prognostic factors. Resection is justified so long as it is complete and with minimal risk. An experienced, specialized centre can be a prognostic determinant.