HPV16/18型阳性且TCT正常患者发生HSIL+风险预测模型的构建

Construction of a Risk Prediction Model for HSIL+in Patients with Positive HPV16/18 and Normal TCT

目的探究HPV16/18阳性且薄层细胞学检测(thinprep cytologic test,TCT)正常患者发生宫颈高级别上皮内病变及以上(high-GRADE squamous intraepithelial lesion or worse,HSIL+)病变的风险因素并构建风险预测模型,以期为临床医生提供一种转诊阴道镜及宫颈活检的辅助评估工具。方法收集2021年6月至2023年6月就诊于内蒙古自治区人民医院门诊,HPV16/18型阳性且TCT正常并行阴道镜+宫颈活组织病理学检查的患者,共207例,通过Logistic回归分析筛选独立风险因素,构建预测模型,建立连线图,并评估模型的预测性能、符合度、准确度及临床实用性。结果(1)单因素卡方检验显示,BMI(χ^(2)=12.13,P<0.001)、吸烟(χ^(2)=15.28,P<0.001)、阴道微生态(χ^(2)=10.074,P=0.002)、感染时间(χ^(2)=82.444,P<0.001)、阴道状态(χ^(2)=66.458,P<0.001)为HPV16/18阳性且TCT正常患者发生HSIL+的独立风险因素;(2)风险预测模型的AUC为0.94(95%CI=0.89~0.98);(3)列线图模型进行内部验证后,其一致性指数(C-index)为0.937。结论本研究基于BMI、阴道微生态、吸烟、感染时间及阴道状态等影响因素建立的预测模型预测HPV16/18型阳性且TCT正常患者发生HSIL+具有较好的区分度、准确度及临床实用性,可为临床医生面对HPV16/18型阳性且TCT正常患者是否行阴道镜+宫颈活组织病理学检查提供参考意见。

Objective To explore the risk factors for high-grade Squamous intraepithelial lesion or worse lesions(HSIL+)in patients with HPV16/18 positive and Thinprep cytologic test(TCT)normal,and to construct a risk prediction model.To provide clinicians with an adjunctive assessment tool for referral colposcopy and cervical biopsy.Methods A total of 207 patients who were positive for HPV16/18 and normal TCT and colposcopy+cervical biopsy pathology were collected from June 2021 to June 2023,and the independent risk factors were screened by logistic regression analysis,the prediction model was constructed,the connection graph was established,and the predictive performance,compliance,accuracy and clinical practicability of the model were evaluated.Results(1)One-way chi-square test showed BMI(χ^(2)=12.13,P<0.001),Smoking(χ^(2)=15.28,P<0.001),vaginal microecology(χ^(2)=10.074,P=0.002),Time of infection(χ^(2)=82.444,P<0.001),Vaginal state(χ^(2)=66.458,P<0.001)was an independent risk factor for HSIL+in patients with positive HPV16/18 and normal TCT;(2)The AUC of the risk prediction model was 0.94(95%CI=0.89-0.98);(3)the consistency index(C-index)of the nomogram model was 0.937 after internal verification.Conclusion The prediction model established based on BMI,vaginal microecology,smoking,infection time and vaginal status in this study predicts the occurrence of HSIL+in patients with positive HPV16/18 and normal TCT,which has good discrimination,accuracy and clinical practicability,which can provide a reference for clinicians facing HPV16/18 positive and normal TCT patients with colposcopy+cervical biopsy.