Cox Regression Analysis of Factors Influencing Postoperative Survival Rate of Gastric Cancer Patients
HAN Fang-hai,ZHAN Wen-hua,ZHANG Zhao-da,HE Yu-long,ZHOU Xiang
2006-01-01
Abstract:Objective To study the influence of clinicopathologic characteristics and surgical treatment of gastric cancer on patients' survival rate.Methods From Apr.1994 to Aug.2005, the data of 759 gastric cancer patients concerning surgical treatment, pathological diagnosis and outcome were collected. Retrospective analysis of the results was made, 3-year and 5-year survival rates were calculated by Kaplan-Meier curve method, univariate analysis was done through Log-rank and multiple factors comparison through Cox regression analysis, and follow-up duration was 4-131 months.Results Single factor analysis indicated that age,tumor location,diameter of tumor, Borrmann type, type of histology, TNM stage, depth of infiltration, lymph node metastasis, liver metastasis, peritoneal dissemination, blood of transfusion during operation, extent of the radical cure of the tumor and excision techniques were significantly influential factors for the prognosis of patients. Cox regression analysis showed that tumor location, diameter of tumor,depth of infiltration, lymph node metastasis,liver metastasis, TNM stage, peritoneal dissemination, blood transfusion during operation, extend of the radical cure of the tumor and excision techniques were independent factors influencing the postoperative survival rate.Conclusion Independent factors influencing the postoperative survival rate include tumor location, diameter of tumor, lymph node metastasis, infiltration depth of the tumor, pathological classification, liver metastasis, peritoneal dissemination, and TNM stage, extent of the radical cure of the tumor, lymphanodectomy techniques and blood transfusion during operation are also important factors.