Resection is the gold standard in the treatment of liver metastases from colorectal cancer. An internal cooled radiofrequency electrode was described to achieve tissue coagulation to a greater margin width. The aim of this study is to determinate if a RF-assisted transection device (RFAT) has any effect on local hepatic recurrence (LHER) compared to conventional technologies. A study population of 103 patients who had undergone a hepatic surgical resection was retrospectively analysed. Patients were classified into two groups according to the device used: a RF-assisted device (RFAT group; n = 45) and standard conventional devices (control group; n = 58). LHER was defined as any growing or enhancing tumour in the margin of hepatic resection during follow-up. Cox proportional models were constructed and variables were eliminated only if p > 0.20 to protect against residual confounding. To assess the stability of Cox's regression model and its internal validity, a bootstrap investigation was also performed. Baseline and operative characteristics were similar in both groups. With a mean follow-up of 28.5 months (range 2-106), in patients with positive margins, we demonstrated 0% of LHER in RFAT vs. 27% in control group (p = 0.032). In the multivariate analysis five factors demonstrated significant influence on the final model of LHER: RFAT group, size of the largest metastases, number of resected metastases, positive margin and usage of Pringle-manoeuvre. This study suggests that parenchymal transection using a RFAT able to create deep thermal lesions may reduce LHER especially in case of margin invasion during transection.