女性膀胱癌患者腹腔镜膀胱全切术与开放膀胱全切术的倾向性评分匹配比较
Propensity-matched comparison of laparoscopic and open radical cystectomy for female patients with bladder cancer

目的:比较女性膀胱癌患者腹腔镜膀胱全切术和开放膀胱全切术两种术式的围手术及肿瘤学预后差异。方法:回顾性分析2006—2017年于北京大学第一医院泌尿外科行根治性膀胱全切术的女性膀胱癌患者91例,将腹腔镜膀胱全切患者以1 :1的比例运用倾向性评分匹配(propensity score matching,PSM)方法与开放膀胱全切患者进行匹配,匹配因素包括年龄、体重指数(body mass index,BMI)、美国麻醉医师协会(American Society of Anesthesio-logists,ASA)评分、T分期及N分期,比较匹配前后两组患者的围手术及肿瘤学特征,应用Kaplan-Meier法比较匹配前后两组间总生存时间(overall survival,OS)、肿瘤特异生存时间(cancer specific survival,CSS)、无进展生存时间(progression free survival,PFS)。采用COX多因素回归分析校正匹配因素,进行敏感性分析。结果:共纳入行开放膀胱全切术患者65例,腹腔镜膀胱全切患者26例,中位随访时间为38个月(四分位距1869个月)。与开放膀胱全切组相比,腹腔镜膀胱全切组患者年龄更小(P<0.001),术前ASA评分更低(P=0.018),匹配后,22例开放膀胱全切患者与腹腔镜膀胱全切患者匹配成功。匹配前,腹腔镜膀胱全切组术中失血量更少(P=0.005),术中输血率更低(P<0.001),清扫淋巴结数目更多(P=0.035),围手术期并发症发生率较低(P=0.015),两组间OS(P=0.698)、CSS(P=0.942)、PFS(P=0.837)差异无统计学意义;匹配后,腹腔镜膀胱全切组术中失血量仍较开放膀胱全切组更少(P=0.009),术中输血率更低(P=0.001),围手术期并发症发生率较低(P=0.040),但清扫淋巴结数目两组间差异无统计学意义,两组间OS(P=0.432)、CSS(P=0.429)、PFS(P=0.284)差异亦无统计学意义。COX多因素回归分析显示,手术方式并不是女性膀胱癌患者OS(HR 1.134, 95%CI 0.335~3.835,P=0.839)、CSS(HR 1.051,95%CI 0.234~4.719,P=0.949)、PFS(HR 0.538,95%CI 0.138~2.095,P=0.371)的独立预测因子。结论:腹腔镜膀胱全切术术中出血量少、术中输血率低、围术期并发症少,但本研究并无证据表明女性膀胱癌患者接受腹腔镜膀胱全切术的预后优于开放膀胱全切术。
Objective: To compare the perioperative and oncologic outcomes of female patients recei-ving laparoscopic radical cystectomy (LRC) and open radical cystectomy (ORC).Methods: Retrospective review of 91 consecutive female patients with urothelial carcinoma of bladder undergoing radical cystectomy at a single academic institution from 2006 to 2017. Those female patients received open radical cystectomy were matched to the patients who underwent laparoscopic radical cystectomy by using propensity score matching in 1 :1 ratio. The matching factors included age, body mass index (BMI), American Society of Anesthesiologists (ASA) score, pathologic stage and pathologic nodal stage. The periope-ration and oncology characteristics were compared, and Kaplan-Meier method was used to analyze the overall survival (OS), cancer specific survival (CSS) and progression-free survival (PFS) estimates. Finally, we did a sensitive analysis by using multivariable COX regression of all the patients, adjusting for the matching factors.Results: There were 65 ORC and 26 LRC patients identified in this cohort with urothelial carcinoma of bladder, the median follow-up time was 38 months (interquartile range 18-69). The age (P<0.001) and ASA scores (P=0.018) were less for LRC before being matched. There were 22 LRC and 22 ORC patients matching successfully. Before being matched, the estimate blood loss (P=0.005), transfusion rate (P<0.001) and total complications rate (P=0.015) were less for LRC, and the