职业女性高危型HPV感染自然清除情况及其列线图预测模型的建立

Natural clearance of high-risk HPV infection in professional women and construction of a nomogram prediction model

目的建立职业女性高危型人乳头瘤病毒(HPV)感染自然清除情况及其列线图预测模型,并进行验证。方法选择2020年3月—2021年3月在某院进行职业体检且经检测确诊为高危型HPV感染并排除宫颈癌及宫颈上皮内瘤变的有固定正式职业的女性进行研究。以7∶3比例将患者分为模型组和验证组,模型组患者根据随访结果分为自然清除组与持续感染组,比较两组患者一般资料、生殖相关治疗、性伴侣相关情况及检查情况。以LASSO回归筛选职业女性高危型HPV感染自然清除的潜在因素后,采用多因素logistic回归,筛选出独立影响因素,根据多因素分析结果以R语言建立列线图模型并进行验证。结果共纳入病例329例,其中模型组230例,验证组99例,两组病例一般资料比较,差异均无统计学意义(均P>0.05)。模型组230例高危型HPV感染患者中,165例在随访结束转阴,自然清除率为71.74%。LASSO回归基础上进行多因素logistic回归分析显示,年龄、避孕方式、性伴侣数、性伴侣包皮过长、初始病毒载荷量、HPV感染类型及生殖道炎症均为职业女性高危型HPV感染自然清除的独立影响因素(均P<0.05)。受试者工作特征(ROC)曲线分析结果显示,模型组职业女性高危型HPV感染自然清除的曲线下面积(AUC)为0.834(95%CI:0.776~0.893),验证组AUC为0.817(95%CI:0.755~0.879)。H-L拟合优度检验结果显示列线图模型与理想模型差异无统计学意义(P>0.05)。校准曲线结果显示模型组与验证组预测曲线与标准曲线基本拟合,提示模型预测准确度较高。模型组决策曲线分析结果显示,当该列线图模型预测职业女性高危型HPV感染自然清除的概率阈值为0.15~0.95时,患者的净受益率>0。结论职业女性高危型HPV感染自然清除率较高,主要受患者年龄、避孕方式、性伴侣数等因素影响,本研究建立的列线图模型预测职业女性高危型HPV感染自然清除率具有较高的准确率与区分度。

Objective To construct and validate a nomogram prediction model for the high-risk human papillomavirus(HPV)infection and its natural clearance in professional women.Methods Women with regular professions,who underwent professional medical examination and were confirmed with high-risk HPV infection without cervical cancer and cervical epithelial neoplasia in a hospital from March 2020 to March 2021 were studied.Patients were divided into the model group and the validation group in a 7∶3 ratio.The model group were subdivided into the natural clearance group and the persistent infection group based on follow-up results.The general information,reproductive-related treatment,sexual partner-related information,and examination results of the two groups of patients were compared.Potential factors for natural clearance of high-risk HPV infection in professional women were screened out by LASSO regression.Independent influencing factors were screened out with multivariate logistic regression.Based on multivariate logistic regression results,a nomogram prediction model was constructed and validated using R programming language.Results A total of 329 cases were included,230 in the model group and 99 in the validation group.There was no statistically significant difference in general information between the two groups of patients(all P>0.05).Among the 230 high-risk HPV infection patients in the model group,165 turned negative at the end of follow-up,with a natural clearance rate of 71.74%.Based on LASSO regression analysis,multivariate logistic regression analysis showed that age,contraceptive method,number of sexual partners,excessive foreskin of sexual partners,initial viral load,HPV infection type,and reproductive tract inflammation were independent influencing factors for the natural clearance of high-risk HPV infection in professional women(all P<0.05).The receiver operating characteristic(ROC)curve analysis showed that the areas under the curve(AUC)of natural clea-rance of high-risk HPV infection in professional women in the model group and validation group were 0.834(95%CI:0.776-0.893)and 0.817(95%CI:0.755-0.879),respectively.H-L goodness-of-fit test result showed that the difference between the nomogram model and the ideal model was not statistically significant(P>0.05).The calibration curve results showed that the predicted curves of the model group and validation group were basically fit with the standard curve,indicating a high predictive accuracy of the model.The decision curve analysis results of the model group showed that when the probability threshold of natural clearance of high-risk HPV infection in professional women predicted by the nomogram model was 0.15-0.95,the net benefit rate of patients was>0.Conclusion The natural clearance rate of high-risk HPV infection in professional women is high,mainly influenced by factors such as age,contraceptive method,and number of sexual partners.The nomogram model constructed in this study has high accuracy and discrimination in predicting the natural clearance rate of high-risk HPV infection in professional women.