宫颈上皮内瘤样病变阴道微生态及其危险因素预测模型

Vaginal microecology of cervical intraepithelial neoplasia and the predictive model of its risk factors

目的探讨宫颈上皮内瘤样病变(CIN)阴道微生态及其危险因素,并构建预测模型,分析其预测价值。方法回顾性分析2021年12月-2022年12月永康市第一人民医院收治的宫颈病变患者120例的临床资料,根据宫颈病变程度分为宫颈癌组(25例,宫颈活检确诊)、高级别组(35例,CIN 2~3级)、低级别组(60例,CIN 1级),另选130名同期健康体检者作为对照组;收集临床资料,检测阴道微生态指标,统计人乳头瘤病毒(HPV)E6/E7 mRNA阳性情况,采用Logistic回归分析CIN的危险因素,并构建预测模型;采用受试者工作特征(ROC)曲线分析预测模型对CIN程度的预测价值。结果阴道分泌物清洁度Ⅲ~Ⅳ级、过氧化氢(H_(2)O_(2))<2μmol/L、酸碱度(pH)>4.5患者占比:宫颈癌组>高级别组>低级别组>对照组(P<0.05);HPV E6/E7 mRNA阳性率在对照组、低级别组、高级别组、宫颈癌组四组中呈升高趋势(P<0.05);阴道分泌物清洁度Ⅲ~Ⅳ级、pH>4.5、高危HPV感染是CIN的危险因素(OR=2.452、2.782、2.710、2.447,P<0.05);构建预测模型:Logit(P)=-10.398+阴道分泌物清洁度Ⅲ~Ⅳ级×1.023+pH>4.5×0.997+高危HPV感染×0.895;ROC曲线分析结果显示,当Logit(P)>9.78时,曲线下面积(AUC)值为0.916,95%CI为0.852~0.959,敏感度为80.00%,特异度为91.58%。结论CIN程度的风险因素包括阴道分泌物清洁度Ⅲ~Ⅳ级、pH>4.5、高危HPV感染等,据此构建预测模型对CIN程度具有较好的预测价值。

OBJECTIVE To investigate the vaginal microecology of cervical intraepithelial neoplasia(CIN)and its risk factors, and to construct a prediction model to explore its predictive value. METHODS The clinical data of 120 patients with cervical lesions admitted to Yongkang City First People′s Hospital from Dec.2021 to Dec.2022 were retrospectively analyzed, and the patients were divided into cervical cancer group(25 cases, confirmed by cervical biopsy), high-grade group(35 cases, CIN 2-3 grade) and low-grade group(60 cases, CIN 1 grade) according to the degree of cervical lesions, another 130 healthy persons underwent physical examination during the same period were selected as control group. The clinical data were collected, the vaginal microecological indexes were detected, the positive rate of human papillomavirus(HPV) E6/E7 mRNA was counted, the risk factors of CIN were analyzed by logistic regression, and the prediction model was constructed.Receiver operating characteristic(ROC) curve was used to analyze the predictive value of the predictive model for CIN. RESULTS The proportion of patients with cleanliness grade Ⅲ-Ⅳ of vaginal discharge, hydrogen peroxide(H_(2)O_(2)<2 μmol/L and pH>4.5 were as follows: cervical cancer group high grade group low grade group control group(P0.05);the positive rate of HPV E6/E7 mRNA showed a trend of gradual increase in the control group, low grade group, high grade group and cervical cancer group(P0.05).Cleanliness grade Ⅲ-Ⅳ of vaginal secretion, pH4.5 and high risk HPV infection were the risk factors of CIN(OR=2.452, 2.782, 2.710, 2.447, P0.05).The predictive model was established as follows: Logit(P)=-10.398+ cleanliness grade Ⅲ-Ⅳ ×1.023+pH4.5×0.997+ high-risk HPV infection ×0.895;ROC curve analysis showed that when Logit(P)9.78, the area under the curve(AUC) value was 0.916, with 95%CI of 0.852-0.959, a sensitivity of 80.00%, and a specificity of 91.58%. CONCLUSION The risk factors for the degree of CIN included cleanliness of vaginal secretion grade Ⅲ-Ⅳ, pH4.5, high-risk HPV infection, etc.The prediction model constructed on the basis of the above indicators had a good predictive value for the degree of CIN.