Relationship between nodal stage and the number of dissected perigastric nodes in gastric cancer.

To evaluate the rationality of the current nodal staging system in gastric cancer, we retrospectively analyzed 152 patients with perigastric node involvement localized to a single station, in whom the route of metastasis to distant nodes was limited. No significant differences in pathology or survival were observed between patients with stage n1 and those with stage n2-3 nodal involvement, but the mean (standard deviation) number of perigastric nodes dissected was 22.6 (12.6) in those with stage nl involvement and 18.5 (9.5) in those with stage n2-3 involvement (P = 0.04). When perigastric node involvement was localized to station 3, the mean number of dissected station 3 nodes was 7.7 (4.2) in nl patients and 5.3 (2.8) in n2-3 patients (P = 0.04). This tendency was also observed in patients with perigastric node involvement limited to either station 1 (P = 0.08) or station 6 (P = 0.11). Thus, patients with fewer perigastric nodes may have more lymphatics that bypass perigastric nodes and empty directly into distant nodes, increasing the likelihood of skip metastases. The number of positive nodes, affected to a lesser degree by lymphatic distribution than the location of positive nodes, should be incorporated into the staging criteria.