Purpose. Splenectomy is sometimes performed simultaneously with curative gastrectomy for gastric carcinoma, especially when the tumor originates in the proximal one-third of the stomach or corpus, or when it invades the entire stomach, in an effort to remove metastatic lymph nodes at the splenic hilus and improve survival. However, splenectomy is not an innocent procedure and may cause increased morbidity and even mortality. Moreover, the long-term effect of splenectomy on survival is still controversial. The purpose of this study was to investigate the effect of simultaneous splenectomy on survival in patients with proximal tumors undergoing curative gastrectomy for gastric cancer. Methods. The effect of splenectomy on the 5-year survival of 61 patients with proximal gastric cancer (located in the cardia or the corpus) who underwent curative gastrectomy in our hospital between 1989 and 1993 was investigated retrospectively. Of these 61 patients, 38 (62.3%) underwent splenectomy and 23 (37.7%) did not. The relationship between the clinicohistopathological parameters and 5-year survival was retrospectively analyzed. Results. No significant differences were found in bivariate analysis between the survival of patients who underwent curative gastrectomy with and those who underwent curative gastrectomy without splenectomy (P = 0.984). Multivariate regression analysis indicated that only histological grade (P < 0.003) and lymph node metastasis (P < 0.001) were independent prognostic factors with or without splenectomy. Splenectomy itself was not an independent prognostic factor (P = 0.528). Conclusion. The findings of this retrospective study showed that simultaneous splenectomy had no effect on the survival of patients who underwent curative gastrectomy for gastric carcinoma. Thus, splenectomy may only be appropriate for patients with direct invasion of the spleen.