Liver resection for multiple colorectal metastases - Influence of parenchymal involvement and total tumor volume, vs number or location, on long-term survival

Hypothesis: Multiple and/or bilateral liver metastases are not absolute contraindications to surgical resection. Design: Retrospective analysis. Setting: University department of surgery and transplantation. Patients and Intervention: A total of 245 curative liver resections for colorectal metastases were divided into 3 groups: M1, single lesions; M2, multiple unilobar, and M3, multiple bilobar. Main Outcome Measures: Univariate and multivariate analysis using several prognostic factors was performed to distinguish variables affecting long-term survival. Results: Overall operative mortality was 0.8%. Overall 5-year survival was 34%. On multivariate analysis, only the percentage of hepatic involvement by the tumor significantly affected prognosis. By replacing this variable with the total tumor volume, this latter variable was the only independent predictor of survival. Patients with multiple metastases and total tumor volume less than 125 cm(3) had a significantly better outcome than patients with single nodules and total tumor volume more than 380 cm(3). Operative mortality and morbidity were comparable among groups M1, M2, and M3. The 5-year survival was 41%, 17%, and 34%, respectively (group M1 vs M2, P = .05; group M1 vs M3 and group M2 vs M3, not significant). The 5-year survival was 41% and 23% in patients with single and multiple lesions, respectively and was 35% and 32% in patients with unilobar and bilobar lesions, respectively. Conclusions: A better outcome in patients with small single lesions was shown. In patients with multiple and/or bilateral metastases, an acceptable 5-year survival superior to 20% was obtained by surgical approach. The total volume of metastases, not number and location, seems to be the strongest predictor of survival.