尿脱落细胞学735例临床病理分析

Urinary exfoliated cytology:a clinical pathological study of 735 cases

目的 回顾性分析尿脱落细胞学与膀胱镜活检病理诊断的符合率,探讨尿脱落细胞病理诊断特点及影响因素。方法 筛选2010年11月～2014年7月在北京大学首钢医院病理科先后接受尿脱落细胞学检查和膀胱镜活检病例735例,尿脱落细胞行巴氏染色,活检标本行HE染色。将所有病例按照尿脱落细胞学检查结果分为三组：阴性组（未发现恶性肿瘤细胞和非典型细胞）、可疑组（发现非典型细胞）和阳性组（发现恶性肿瘤细胞）,并与对应活检结果进行对比,用ROC曲线分析、Cochran-Armitage趋势检验和Logistic回归统计方法分析尿脱落细胞诊断结果的敏感性和特异性、组间关系以及与患者年龄和性别的相关性。结果 735例中,男性551例,女性184例,年龄28～91岁,中位年龄69岁。细胞学阳性组187例,其中184例恶性,3例为假阳性;可疑组186例,其中67例恶性,119例为良性反应性改变;阴性组362例,其中90例恶性。细胞学诊断ROC曲线下面积AUC（95%CI）为0.800（0.767～0.834）,与组织病理学诊断结果比较差异有统计学意义（P〈0.001）。随着细胞学诊断阴性、可疑、阳性组级别的增加,癌的比率增加,Cochran-Armitage趋势检验差异有统计学意义（Z=15.83,P〈0.001）;以组织学诊断结果为因变量,年龄、性别、细胞学诊断分别为自变量,进行多因素Logistic回归分析,结果显示：年龄对癌的发生率有影响,每增加1岁,组织学确诊癌的风险提高1.04（1.03～1.05）倍,差异有统计学意义（P〈0.001）,性别差异无统计学意义（P=0.655）。结论 随着分组级别的增加,尿脱落细胞诊断阳性率增加;同时,患者年龄与癌的发生率呈正相关。掌握详细的临床资料能提高细胞学诊断的敏感性及准确性。

Purpose To explore the pathological feature of urinary exfoliated cell examination and influence factors by retrospectively comparing the coincidence of diagnosis between urinary exfoliated cell examination and histopathologic results of cystoscopic biopsy. Methods 735 patients underwent both urinary exfoliated cell examination and histopathologic biopsy of cystoseope evaluation from No- vember 2010 to July 2014 in Peking University Shougang Hospital were enrolled in this study. The urinary exfoliated cells were treated with Pap staining, while the histopathologic biopsy were dealt with HE staining. All cases were divided into three groups according to the diagnosis of urinary exfoliated cell examination ： negative group （ no cancer or atypical cell detected） , suspicious group （ atypical cell detected）and positive group （cancer cell detected）. These above diagnoses were confirmed with the histopathologic biopsy. ROC cupce analysis, Cochran-Armitage trend test and logistic regression analysis were performed to evaluate the sensitivity and the specificity of urinary exfoliated cell examination as well as the relationship between diagnoses with age and sex. Results The age range of 735 patients （551 male and 184 female） was 28 -9l years and the median age was 69 years. There were 187 patients in the positive group, including 184 malignant and 3 false-positive cases. The suspicious group, including 186 cases, consisted of 67 malignant, 119 benign reactive changes. Of all 362 cases in the negative group, malignant tumor was detected in 90 cases. For histologic diagnosis, the AUC of ROC（95% CI）was 0. 800 （0. 767 -0. 834） , displaying significant difference as compared to the histological pathological diagnostic results（P 〈 0. 001 ）. As the cyto-histologic diagnostic level elevated from negative, suspicious to positive, the results of Co- chran-Armitage trend test showed significant differences （Z = 15.83, P 〈 0. 001 ）. If standardized with the histopathologic biopsy re- sults, the AUC （ area under curve） of urinary exfoliated cell examination was 0. 800 （0. 767 -0. 834） in ROC curve analysis was sig- nificantly larger （ P 〈 0. 001 ）. Furthermore, we also found in Logistic regression that the incidence of cancer was 1.04 （ 1.03 - 1.05 ） times higher if aged one year older （ P 〈 0. 001 ） , while there was no significant relationship between the incidence and the sex （ P = 0. 655 ）. Conclusions The coincidence rate of urinary exfoliated cell examination increases with the malignant degree. A positive cor- relation is detected between age and the incidence of malignant tumor. Detailed clinical material can markedly improve the sensitivity and accuracy of cyto-histologic diagnosis.