Abstract

BACKGROUND--No nodal grouping category of gastric cancer has been universally accepted for the grading of the effectiveness of therapeutic regimens. AIMS--To establish an appropriate nodal grouping as a forecaster of distant disease and test its validity as a determinant in survival. PATIENTS--Five hundred and ten patients who underwent curative resections for gastric cancer were studied. METHODS--Retrospectively analyse the prognostic significance of the number of metastatic lymph nodes. RESULTS--A total of 17 176 lymph nodes with an average of 34 per specimen were removed, of which 2811 (16%) showed metastases. Among the 510 patients, 287 (56%) had lymph node metastases, with an average of 9.8 per metastatic case. The survival of all patients was related to their nodal status, an abrupt decrease in survival was seen between 0 and 1 and 4 compared with 5 or more modes while little difference in survival existed among 1, 2, 3, and 4, and among 5, 6, 7, and 8 positive nodes. Multivariate analysis showed that the number of positive nodes (1-4, 5-8 versus > or = 9; relative risk 2.2) and depth of cancer invasion (three levels; relative risk 1.9) were independently correlated with survival. The current nodal stage was not a prognostic factor. CONCLUSIONS--Gastric cancer patients with 0, 1 to 4, 5 to 8, and > 9 positive nodes may represent four appropriate prognostic groups and should be adopted for classification of nodal stage in gastric cancer.