AIM: The authors seek to assess whether the lymph node ratio (the ratio of positive nodes divided by the total number of retrieved nodes) could predict the risk of metachronous liver metastases. MATERIAL AND METHODS: A homogeneous group of 280 patients, followed-up for at least 5 years, was evaluated. In order to highlight the groups with the highest risk of metachronous liver metastases, patients were divided into four quartiles groups in relation to the LNR. RESULTS: The number of lymph nodes sampled in group “stage I” was significantly lower. Even if statistical significance between the global LNR and the development of liver metastases has not been reached, the subdivision into quartiles has made it possible to highlight that in the more advanced ratio groups, a higher incidence of metachronous liver metastases (p <0.028) was registered and was a different distribution of patients with or without liver metastasis in function of quartiles (P =0.01). DISCUSSION: The LNR has enabled us to prognosticate patients who are at greater risk of developing metachronous liver metastases. The lower lymph node sampling in the patients with less advanced staging (I) and in patients with nodenegative cancer (I+II) who developed liver metastases, leads us to believe that some patients have been understaged. CONCLUSION: We believe that the LNR, especially in cases of adequate lymph node sampling, is a useful gauge to better sub-stratify “node-positive” patients.