BACKGROUND AND OBJECTIVES: Free perforation and major bleeding in patients with gastric cancer are rare but serious conditions with potentially dangerous effects. To clarify the clinicopathologic characteristics of patients with these conditions and to determine the optimum management, we reviewed 16 cases of perforation and 13 cases of major bleeding in patients with gastric cancer who required emergency surgery. METHODS: We compared the clinical and histologic features of the patients with perforation and those with bleeding. Cox's multivariate regression analysis was used to compare survival rates between patients who underwent single-step surgery or a two-step radical procedure, between patients with stage I or II and stage III or IV cancer, between patients who underwent complete (R0) and incomplete (R1 or R2) resection, and between patients with bleeding and those with perforation. RESULTS: Many of the patients had advanced disease. There were no significant differences in clinicopathologic findings or survival between patients with gastric perforation and those with major bleeding. Patients who had major bleeding tended to have larger cancers. In the univariate analysis, gastrectomy (vs. no gastrectomy), R0 (vs. R1 or R2) resection, and lower stage (vs. higher stage) were highly correlated with improved survival time. CONCLUSIONS: Overall, patients with gastric cancers who underwent emergency gastrectomy had a poor prognosis, but it was better than that of patients who could not have gastrectomy because of the prXesence of advanced cancer. However, the survival rate was excellent in patients with early-stage cancer who underwent complete (R0) resection. We recommend complete resection when possible. Copyright 2002 Wiley-Liss, Inc.
BACKGROUND AND OBJECTIVES: Free perforation and major bleeding in patients with gastric cancer are rare but serious conditions with potentially dangerous effects. To clarify the clinicopathologic characteristics of patients with these conditions and to determine the optimum management, we reviewed 16 cases of perforation and 13 cases of major bleeding in patients with gastric cancer who required emergency surgery. METHODS: We compared the clinical and histologic features of the patients with perforation and those with bleeding. Cox's multivariate regression analysis was used to compare survival rates between patients who underwent single-step surgery or a two-step radical procedure, between patients with stage I or II and stage III or IV cancer, between patients who underwent complete (R0) and incomplete (R1 or R2) resection, and between patients with bleeding and those with perforation. RESULTS: Many of the patients had advanced disease. There were no significant differences in clinicopathologic findings or survival between patients with gastric perforation and those with major bleeding. Patients who had major bleeding tended to have larger cancers. In the univariate analysis, gastrectomy (vs. no gastrectomy), R0 (vs. R1 or R2) resection, and lower stage (vs. higher stage) were highly correlated with improved survival time. CONCLUSIONS: Overall, patients with gastric cancers who underwent emergency gastrectomy had a poor prognosis, but it was better than that of patients who could not have gastrectomy because of the prXesence of advanced cancer. However, the survival rate was excellent in patients with early-stage cancer who underwent complete (R0) resection. We recommend complete resection when possible. Copyright 2002 Wiley-Liss, Inc.
Authors: Reid M Wainess; Justin B Dimick; Gilbert R Upchurch; John A Cowan; Michael W Mulholland Journal: J Gastrointest Surg Date: 2003-11 Impact factor: 3.452