人乳头瘤病毒阳性与阴道微生态相关性及临床预测模型建立  

Relationship between HPV positivity and vaginal microecology and establishment of clinical prediction model

作　　者：汪辉 蔺茹[2] 倪亚莉[1] 刘亚萍 邵亚雯 贺清[1] 武苗苗 杨杰[1] WANG Hui;LIN Ru;NI Yali;LIU Yaping;SHAO Yawen;HE Qing;WU Miaomiao;YANG Jie(Gansu Provincial Maternal and Child Health Care Hospital,Gansu Provincial Central Hospital,Lanzhou,Gansu 730050,China)

机构地区：[1]甘肃省妇幼保健院(甘肃省中心医院)生殖医学中心,甘肃兰州730050 [2]甘肃省妇幼保健院(甘肃省中心医院)宫颈癌防治中心,甘肃兰州730050

出　　处：《社区医学杂志》2023年第23期1250-1254,共5页Journal Of Community Medicine

基　　金：甘肃省中医药科研课题(GZKP-2021-23)。

摘　　要：目的探讨阴道微生态与人乳头瘤病毒(HPV)阳性相关性及其诊断预测的临床意义。方法选取2021-01-01-2021-03-31于甘肃省妇幼保健院行阴道微生态及HPV检测的1791例患者为研究对象,根据检测结果分为HPV阳性组(n=535)、HPV阴性组(n=1256)。分析2组阴道微生态的差异并建立阴道微生态与HPV阳性的临床诊断预测模型,根据该模型的预测因子构建列线图。通过受试者工作特征曲线(ROC)、HL检验、临床决策曲线(DCA),分别评估模型的区分度、校准度、临床实用度。结果与HPV阴性组比较,多孕多产、阴道微生态失衡(如细菌性阴道病、外阴阴道假丝酵母菌病、混合性阴道炎)、H_(2)O_(2)浓度、唾液酸苷酶、白细胞酯酶、Β-葡萄糖醛酸苷酶、菌群多样性与密集度的改变,均增加HPV感染风险,差异有统计学意义,均P<0.05。通过logistic回归分析筛选出细菌性阴道病、混合性阴道炎、白细胞酯酶、菌群密集度作为预测因子,纳入临床诊断预测模型并制作列线图。ROC曲线分析结果显示,该模型的ROC曲线下面积为0.68(95%CI为0.66~0.71),表明该模型具有较好的区分度;HL检验P>0.05,表明校准曲线拟合较好,说明预测概率与实际概率的一致性较好;DCA分析结果表明,该模型在一定范围内有较大临床实用度。结论阴道微生态平衡对于预防HPV感染起积极作用。本研究建立临床预测模型具有较好的区分度、校准度、实用度,有利于针对阴道微生态失衡及HPV阳性患者制定个体化治疗方案。Objective To investigate the relationship between vaginal microecology and human papillomavirus(HPV)positivity and and its clinical significance in diagnosis and prediction.MethodssA total of 179l patientswho underwent vaginal microbiota and HPV testing at Gansu Maternal and Child Health Hospital from January 1,2021 to March 31,2021 were selected as the research subjects.They were divided into HPV positive group(n=535)and HPV negative group(n=1256)based on the test results.We analyzed the differences in vaginal microbiota between two groups and established a clinical diagnostic prediction model for vaginal microbiota and HPV positivity.Based on the predictive factors of this model,we constructed a column chart.We evaluated the discriminability,calibration,and clinical practicality of the model through the receiver operating characteristic curve(ROC),hosmer-lemeshow(HL)test,and decision curve analysis(DCA).Results Compared with the HPV negative group,the general data of the two groups had no statistical significance in terms of age.Compared with the HPV negative group,the following factors,including multiple pregnancies,prolificacy,vaginal microecology imbalance,such as bacterial vaginitis,vulvovaginal candidiasis,mixed vaginitis,H_(2)O_(2) concentration,sialic acid glycosidase,leukocyte esterase,B-glucuronic acid glycosidase,and flora diversity and density increased the risk of HPV infection,and the differences were statistically significant(all P<0.05).Bacterial vaginitis,mixed vaginitis,leukocyte esterase,and bacterial population density were selected as predictive factors via logistic regression analysis,and were included in the clinical diagnosis prediction model and nomographs were made.The ROC analysis results showed that,the area under the ROC curve of the model was 0.68(95%CI:0.66-0.71),which indicated that the model had good discrimination.P>0.05 tested by HL indicated that the good fitting of the calibration curve,which meant that the predicted probability was in good agreement with the actual probabili

关 键 词：阴道微生态 人乳头瘤病毒 临床预测模型 

分 类 号：R711.3[医药卫生—妇产科学] R737.33[医药卫生—临床医学]