The purpose of this study is to evaluate the role of splenectomy in potentially curative total gastrectomy for proximal gastric carcinoma in terms of survival benefit and post-operative morbidity.

European clinical trials of gastrectomy showed that splenectomy is an important risk factor for post-operative morbidity and mortality. Retrospective comparisons suggested that splenectomy is associated with poor long term survival. However, Japanese studies revealed that 20 - 30 % of patients with non-early carcinoma in the proximal stomach have nodal metastasis in the splenic hilum and therefore, pancreas-preserving splenectomy is part of the standard operation in specialized centers where splenectomy is not considered a risk factor for operative mortality. There have been no prospective randomized trials to evaluate the survival benefit of splenectomy in total gastrectomy for proximal gastric cancer.

Comparison: Total gastrectomy with pancreas-preserving splenectomy versus total gastrectomy without splenectomy

Inclusion Criteria:

Preoperatively

• Histologically proven adenocarcinoma
• T2 or deeper lesion in the upper third of the stomach without involvement of the greater curvature or esophageal invasion, irrespective of the primary tumor location or existence of multiple foci
• No distant metastasis, not linitis plastica ('Borrmann 4'), not stump carcinoma, no prior treatment for 364 Randomized trial for splenectomy gastric cancer
• Sufficient organ function
• Written informed consent.

Intra-operatively

• T2/T3/T4 and N0/N1/N2, no tumor on the greater curvature, no direct invasion of the pancreas or spleen, negative peritoneal lavage cytology
• No apparent nodal metastasis in the splenic hilum or along the splenic artery

Exclusion Criteria:

• Liver cirrhosis or portal hypertension
• Idiopathic thrombocytopenic purpura
• Severe pulmonary dysfunction
• Synchronous or metachronous (within 5 years) malignancy.