影响表浅性膀胱癌患者术后复发的多因素分析及预测研究  

Multivariate analysis and predictive study on recurrent factors of non-muscle-invasive bladder carcinoma after operation

作　　者：李文琦[1] 赵文超[2] 牛天力[2] 郑明华[2] 张炜[3] 

机构地区：[1]泰州市人民医院质控中心,江苏泰州225300 [2]泰州市人民医院泌尿外科 [3]江苏省人民医院泌尿外科

出　　处：《临床泌尿外科杂志》2015年第10期923-926,共4页Journal of Clinical Urology

摘　　要：目的：探讨影响表浅性膀胱癌患者术后复发的影响因素并建立复发的预测模型。方法：采用回顾性研究,选取2012年6月1日-2014年6月31日期间在泰州市人民医院行表浅性膀胱癌手术患者153例,收集临床病理资料和免疫酶标结果并随访其复发状况。采用Log-rank时序检验进行单因素分析,Cox比例风险模型进行多因素分析,并建立预测指数（PI）方程。结果：经Cox多因素模型分析有意义的因素有肿瘤pTNM分期、Ki-67、p53基因和接受膀胱灌注治疗（P〈0.05）。建立PI方程ROC曲线下面积为0.833,灵敏度为75%,特异度为98%,约登指数为73%。结论：影响表浅性膀胱癌患者术后复发的危险因素有肿瘤pTNM分期、Ki-67、p53基因,保护因素有接受膀胱灌注治疗;影响表浅性膀胱癌术后患者复发的PI方程预测准确性高,可预测大部分表浅性膀胱癌术后患者的复发情况。Objective：To explore the influencing factors relating to the recurrence of non-muscle-invasive bladder carcinoma,and to establish a prediction model for patients＇ recurrence.Method：From June 2012 to June 2014,153 cases with non-muscle-invasive bladder cancer in Taizhou People＇s Hospital were selected in the study and clinical/pathological factors were collected.Log-rank test and proportional-hazards regression model（Cox model）were used for univariate and multivariate analysis.A prognostic index（PI）basing on the Cox＇s relative hazards regression model was constructed.Result：Tumor pTNM stage,Ki-67,p53 and adjuvant chemotherapy of bladder irrigation were independent prognostic factors of non-muscle-invasive bladder cancer.An area under the ROC curve was 0.833.The sensitivity was 75%and the specificity was 98%.Moreover,Youden index was 73%.Conclusion：Risk factors of non-muscle-invasive bladder carcinoma are tumor pTNM stage,Ki-67,and p53,while chemotherapy of bladder irrigation is a protective factor.Individual recurrent rate was evaluated basing on the prognostic PI model,which may provide valuable information for physicians in predicting the long-term recurrent rate of patients with non-muscle-invasive bladder cancer.

关 键 词：表浅性膀胱癌 复发 预测 COX模型 

分 类 号：R737.14[医药卫生—肿瘤]