Analysis of risk factors and predictive model construction of chronic vaginitis complicated with vulvar intraepithelial neoplasia

Abstract:

Objective：To explore the risk factors of vulvar intraepithelial neoplasia（VIN） in patients with vaginitis and construct a predictive model. Methods：Using retrospective analysis，clinical data of 368 patients with chronic vaginitis and suspected VIN admitted to the gynecological clinic of the First Affiliated Hospital of Kunming Medical University from February 2015 to February 2020 were collected in the study. Patients were divided into VIN group（n=46） and non-VIN group（n=322） according to the diagnosis of VIN based on histopathological examination. Single factor analysis and multivariate logistic regression analysis were performed on the clinical data of the two groups when they came to the hospital，and a predictive model was built. Sensitivity，specificity，area under the receiver operating characteristic（ROC） curve and prediction accuracy were used to evaluate the effectiveness of the model. Results：The comparison of the age of VIN and non-VIN groups，human papillomavirus（HPV） infection，interleukin 17（IL-17），ratio of neu－trophils to lymphocytes（NLR），macrophage colony stimulating factor（M-CSF） and serum albumin（Alb） indicators showed statistical differences（P<0.05）. After multi-factor logistic regression analysis，it was found that，HPV infection（OR=29.254，95%CI=4.124-153.574），IL-17（OR=4.604，95%CI=2.209-97.683），NLR（OR=2.835，95%CI=1.357-5.405），M-CSF（OR=2.361，95%CI=1.135-3.975） and Alb（OR=1.099，95%CI=1.015-1.263） were risk factors for chronic vaginitis complicated with VIN（P<0.05）. According to the risk factors，the expression equation of the predictive model was derived：Prob=1/（e-Y），Y=40.507－3.376×HPV infection－1.527×IL-17－1.042×NLR－0.859×M-CSF－0.095×Alb. After verification，the sensitivity of the model was 86.90%，the specificity was 85.70%，and the accuracy was 85.90%. Conclusion：The risk factors for VIN in patients with chronic vaginitis include HPV infection，high expression of IL-17，high expression of NLR，high expression of M-CSF and low expression of Alb，thereby building a prediction model，with good distinguishing ability，which can effectively assess the risk of vaginitis complicated by VIN.