The aim of our analysis was to assess retrospectively the effect on local relapse, overall survival (OS) and disease-free survival (DFS) of a limited or an extended lymphadenectomy in radically resected gastric cancer patients. This study was performed in order to identify a subgroup of patients possibly not benefiting from a therapeutic approach such as chemoradiation therapy. We divided our patients into two groups according to lymphadenectomy type: group A for limited (<25 resected lymph nodes) and group B for extended (>25 resected lymph nodes) lymph nodes resection. A total of 418 patients were analysed: tumour stage at diagnosis was pT2-3 pN1-3 M0 in 339 patients and pT3 N0 M0 in 79 patients. Median age at diagnosis was 68 years (range 30-92 years). A total of 306 patients (73.2%) were in group A and 112 (26.8%) in group B. The median survival time (OS) for patients in groups A and B was 58.8 and 84.8 months, respectively (P=0.0371); median DFS was 28.8 months in group A and 59.9 months in group B (P=0.0027). At multivariate analysis, extension within the gastric wall, nodal involvement and the number of resected lymph nodes appeared to affect both OS and DFS. An inadequate lymph nodes resection can affect survival and result in a higher incidence of local relapse, making the latter group of patients optimal candidates for adjuvant chemoradiation.