宫颈上皮内瘤变患者宫颈锥切术后病变残留和复发的列线图预测模型构建与评估  

Construction and Evaluation of Nomogram Prediction Model for Residual and Recurrence in Patients with Cervical Intraepithelial Neoplasia After Cervical Conization

作　　者：李军英 冯相珍 LI Junying;FENG Xiangzhen(Puyang Maternal and Child Health Hospital,Puyang,457000)

机构地区：[1]河南省濮阳市妇幼保健院,457000

出　　处：《实用癌症杂志》2023年第5期834-838,共5页The Practical Journal of Cancer

摘　　要：目的分析宫颈上皮内瘤变患者宫颈锥切术后病变残留和复发的影响因素,以此构建列线图预测模型。方法选取325例经宫颈锥切术治疗的宫颈上皮内瘤变患者,随访后根据是否复发和残留将患者分为残留或复发组(48例)和未残留和复发组(277例)。比较2组患者术后液基薄层细胞学、病理结果、高危人乳头瘤病毒等基本资料及手术相关资料,以Logistic回归分析影响患者术后复发的危险因素,以R软件建立列线图模型,并评估其效果,绘制临床决策曲线(decision curve analysis,DCA)评估宫颈上皮内瘤变患者宫颈锥切术后病变残留和复发模型的临床预测价值。结果残留和复发组患者术前HSIL、HPVE6/E7mRNA阳性、病变累及腺体、手术切缘阳性人数占比与未残留和复发组比较均明显升高(P<0.05),且经Logistic回归分析结果显示,术前HSIL、手术切缘阳性、HPVE6/E7mRNA阳性、病变累及腺体为影响宫颈上皮内瘤病变患者残留和复发的独立危险因素(P<0.05);以ROC曲线下面积及H-L偏差度检验显示,模型区分度及准确度较好,DCA曲线显示该模型具有较高的临床应用价值。结论术前HSIL、HPVE6/E7mRNA阳性、手术切缘阳性、病变累及腺体为影响宫颈上皮内瘤变术后病变残留和复发的独立危险因素,以此建立的列线图模型准确度及区分度良好,能够有效甄别高风险人群,以利于及时预防、减少宫颈上皮内瘤病变残留和复发。Objective To analyze the influencing factors of residual and recurrence in patients with cervical intraepithelial neoplasia after cervical conization,and to construct a nomogram prediction model based on this.Methods A total of 325 patients with cervical intraepithelial neoplasia treated by cervical conization in our hospital were selected.After follow-up,patients were grouped into residual or recurrence group(48 cases)and non-residual and recurrence group(277 cases)according to whether there was recurrence and residual.The liquid-based thin-layer cytology,pathological results,high risk human papillomavirus and other basic information after operation were compared between 2 groups.The Logistic regression was applied to analyze the independent risk factors that affect the recurrence,and the Nomogram model was established by using R software and its effect was evaluated.The clinical decision curve analysis(DCA)was drawn to evaluate the clinical predictive value of the model for residual and recurrence of cervical intraepithelial neoplasia patients after cervical conization.Results The proportions of preoperative HSIL,HPVE6/E7mRNA positive,gland involvement and positive surgical margin in the recurrence group were obviously higher than those in the non-recurrence group(P<0.05).Logisitic regression analysis results showed that preoperative HSIL,HPVE6/E7mRNA positive,positive surgical margin and gland involvement were independent risk factors for residual and recurrence of cervical intraepithelial neoplasia(P<0.05);the area under the ROC curve and the H-L deviation test showed that the model had good discrimination and accuracy.The DCA curve showed that the model has high clinical application value.Conclusion The preoperative HSIL,HPVE6/E7mRNA positive,positive surgical margin and gland involvement are independent risk factors for residual and recurrence of cervical intraepithelial neoplasia.The nomogram model established on this basis has good accuracy and discrimination,can effectively screen high-risk population,an

关 键 词：预测模型 宫颈锥切术 宫颈上皮内瘤变 列线图 

分 类 号：R737.9[医药卫生—肿瘤]