宫颈癌术后发生女性盆底功能障碍性疾病的风险预测模型构建  

Construction of risk prediction model for female pelvic floor dysfunction after cervical cancer surgery

作　　者：杨欣 孔敏 马燕[1] 王卓 王志芳 李妍[2] YANG Xin;KONG Min;MA Yan;WANG Zhuo;WANG Zhifang;LI Yan(Ningxia Medical University,Yinchuan 750004,China;General Hospital of Ningxia Medical University,Yinchuan 750004,China)

机构地区：[1]宁夏医科大学,宁夏银川750004 [2]宁夏医科大学总医院,宁夏银川750004

出　　处：《宁夏医学杂志》2024年第2期135-138,F0003,共5页Ningxia Medical Journal

基　　金：国家自然科学基金项目(82060275)。

摘　　要：目的建立宫颈癌患者手术治疗后发生女性盆底功能障碍性疾病(PFD)风险预测模型,为宫颈癌幸存者提高生活质量提供理论依据。方法收集确诊为宫颈癌且接受广泛子宫切除术+盆腔淋巴结清扫术±腹主动脉旁淋巴结清扫术±放化疗患者共195例为研究对象,对其术后1年内可能发生PFD的影响因素进行单因素分析,将单因素分析差异有统计学意义或单因素分析差异无统计学意义但专业上认为可能有意义的变量纳入多因素Logistic回归分析,建立宫颈癌患者术后盆底功能障碍性疾病的风险预测模型。绘制受试者工作特征曲线(ROC曲线)并计算曲线下面积AUC,校准曲线分别评估预测模型的区分度、一致性及临床有用性,综合评估该模型的预测效能。结果本研究构建的宫颈癌术后发生盆底功能障碍性疾病的风险预测模型的区分度较好(AUC=0.929,95%CI:0.882~0.976);校准曲线是斜率接近为1的虚线;内部验证显示验证后的ROC曲线下面积AUC约为0.902,灵敏度为94.9%,特异度为85.0%,最佳截断值0.785。结论本研究通过纳入危险因素如肿瘤直径≥4 cm、手术时长>90 min、术中出血量≥100 mL、留置导尿管天数≥7 d、化疗次数≥4次及放疗等,构建的宫颈癌术后发生盆底功能障碍性疾病的风险预测模型,可预测宫颈癌患者术后发生女性盆底功能障碍性疾病的风险并及时干预,从而提高癌症幸存者的生活质量。Objective To establish a risk prediction model of female pelvic floor dysfunction(PFD)in patients with cervical cancer after surgical treatment,and provide a theoretical basis of improving the quality of life for cervical cancer survivors.Methods A total of 195 patients with cervical cancer who underwent extensive hysterectomy,pelvic lymph node dissection,paraaortic lymph node dissection and chemoradiotherapy were collected and followed up for one year.Univariate analysis was conducted on the factors affecting the possible occurrence of PFD after surgery.Variables that had statistical significance in univariate analysis or had not statistical significance in univariate analysis but were considered to have potential significance in professional field were included in multivariate logistic regression analysis to establish a risk prediction model for postoperative pelvic floor dysfunction in patients with cervical cancer.The prediction model was presented in a nomogram.The receiver operating characteristic curve(ROC curve)was plotted and the area under the curve AUC and calibration curve were calculated to evaluate the differentiation,consistency and clinical usefulness of the prediction model,and comprehensively evaluate the prediction efficiency of the model.Results The risk prediction model of pelvic floor dysfunction after cervical cancer surgery was good discriminability(AUC=0.929,95%CI:0.882~0.976).The calibration curve was a dashed line with a slope approaching 1.Internal verification showed that the area under the ROC curve after verification was about 0.902,the sensitivity was 94.9%,the specificity was 85.0%,and the optimal cutoff value was 0.785.Conclusion This study constructed a risk prediction model for postoperative pelvic floor dysfunction in cervical cancer by incorporating risk factors included tumor diameter≥4 cm,operation duration>90 min,intraoperative blood loss≥100 mL,indwelling catheter days≥7 days,chemotherapy times≥4 times and radiotherapy.This model can predict the risk of postoperat

关 键 词：宫颈癌 女性盆底功能障碍性疾病 预测模型 

分 类 号：R737.33[医药卫生—肿瘤]