Analysis of risk factors and establishment of predictive model for detecting≥ CIN 2 in ASCUS patients
Objective To investigate the risk factors of pathological detection of ≥CIN 2 in patients with atypical squamous cells with unclear significance(ASCUS)in cervical cytology,and to construct a predictive model.Methods During January 2019 to May 2021,312 patients with ASCUS and high-risk HPV(HR-HPV)infected underwent colposcopy and cervical histological biopsy in Beijing Tiantan Hospital Affiliated to Capital Medical University were collected and analyzed as a training set.The age,HR-HPV infection status,the number of abortions and biopsy pathological results were analyzed;multivariate Logistic regression analysis was used to construct the prediction model of cervical lesions.Subsequently,316 ASCUS patients with HR-HPV collected from June 2021 to December 2022 were used as the validation set for model validation.Results Among the 312 patients in the training set,93 cases were≥CIN 2(29.8％),multivariate Logistic regression showed that HPV16 and HPV58 were independent risk factors for ≥ CIN 2 in ASCUS patients.The validation results showed that the model constructed with HPV16,HPV58,HPV33,HPV35,and HPV52 had better discrimination(AUC=0.731)in predicting CIN 2 and above lesions.Conclusion HPV16/58 are risk factors for ASCUS with≥CIN 2.By using predictive model,high-risk patients with ASCUS and HR-HPV infection can be screened,providing a certain reference for personalized treatment of patients.
万方数据
维普
