Results of curative gastrectomy for carcinoma.

BACKGROUND: In the western literature, controversy exists regarding the operative management of carcinoma of the stomach. Still debated are the extent of gastric resection, the role of splenectomy, and the significance of lymphadenectomy. STUDY DESIGN: We performed a retrospective study of 646 consecutive patients who underwent curative gastrectomy for carcinoma of the stomach from 1950 to 1989. A multivariate analysis of nine clinicopathologic and treatment-related variables, including type of gastrectomy, splenectomy, and extent of lymphadenectomy was done to assess the relative influence of these variables on survival. RESULTS: The following variables emerged as independent predictors of death: serosal involvement, distant lymph node metastases, and total gastrectomy. Splenectomy and lymphadenectomy were not related to prognosis. However, a separate analysis on defined subsets of patients revealed that radical lymphadenectomy (removal of N2 nodes) provided a survival benefit in patients with T1-2 N0-1 disease. CONCLUSIONS: These findings suggest that although prognosis of carcinoma of the stomach is mainly determined by the stage of disease at time of operation, the choice of the surgeon can affect survival. Total gastrectomy should be avoided when it is not necessary for proximal tumor location. Subtotal gastrectomy in distally located tumors seems to be the best option. Radical lymph node dissection should be performed when nodal involvement is expected to be limited or absent.