The aim of the study was to analyse the prognostic factors for long-term outcome of liver resections for metastases from colorectal cancer. The retrospective analysis included 297 liver resections for colorectal carcinoma liver metastases. The following prognostic factors were considered: age, gender, stage and grade of differentiation of the primary tumour, node metastases, site of the primary colorectal cancer, number and diameter of the hepatic lesions, time interval from primary cancer to liver metastases, preoperative CEA level, adjuvant chemotherapy after hepatic resection, type of hepatic resection, use of intraoperative ultrasound and portal triad clamping, blood loss and transfusions, postoperative complications and hospital stay, tumour-free surgical margins, clinical risk score (as defined by the Memorial Sloan-Kettering Cancer Centre group, MSKCC-CRS). Overall survival rates were estimated according to the Kaplan-Meier method and were compared at univariate analysis using the log-rank test. Multivariate analysis was performed including significant variables at univariate analysis using the Cox regression model. Differences were considered significant at p < 0.05. The 1, 3, 5 and 10-year overall survival rates were 90.6%, 51%, 27.5%, and 16.9%, respectively. The univariate analysis revealed a statistically significant difference (p < 0.05) in overall survival in relation to: grade of differentiation of the primary cancer (5-year survival of grades G1-G2 vs grades G3-G4: 30.7% vs 14.4%, p = 0.0016), preoperative CEA level > 5 and > 200 ng/ml (5-year survival of CEA < 5 ng/ml vs CEA > 5 ng/ml: 51.1% vs 15.5%, p = 0.0016; 5-year survival of CEA < 200 ng/ml vs CEA > 200 ng/ml: 27.9% vs 17.4%, p = 0.0001), diameter of major lesions > 5 cm (5-year survival of diameter < or = 5 cm vs > 5 cm: 30.0% vs 18.8%, p = 0.0074), disease-free interval between primary tumour and liver metastases longer than 12 months (5-year survival of patients with disease-free interval < or = 12 months vs > 12 months: 23.0% vs 36.1%, p = 0.042), high MSKCC-CRS (5-year survival of MKSCC-CRS 0-1-2 vs 3-4-5: 36.4% vs 1 6.3%, p = 0.017). The multivariate analysis showed three independent negative prognostic factors: G3-G4 primary cancer, CEA level > 5 ng/ml, and high MSKCC-CRS class. No single prognostic factor turned out to be associated with such disappointing outcomes after hepatic surgery for colorectal liver metastases as to permit the identification of specific subgroups of patients to be excluded on principle from undergoing liver resection. However, in the presence of a number of specific prognostic factors (G3-G4 grade of differentiation of the primary tumour, preoperative CEA level > 5 ng/ml, high MSKCC-CRS) enrolment of the patient in trials exploring new diagnostic tools or new adjuvant treatments may be suggested to improve the preoperative staging of the disease and reduce the incidence of tumour recurrence after liver resection.