摘要
目的 分析围绝经期异常子宫出血患者发生子宫内膜病变的危险因素,并构建列线图预测模型。方法 选取2018年10月至2021年10月在上海交通大学医学院附属松江医院治疗围绝经期异常子宫出血的260例患者作为研究对象,收集其临床资料,使用电脑随机分组法将182例(70%)作为建模组,剩余78例(30%)作为验证组,根据患者是否合并子宫内膜病变分为子宫内膜正常组及子宫内膜病变组。以单因素分析及多因素Logistic回归分析影响建模组围绝经期异常子宫出血患者发生子宫内膜病变的危险因素,并使用R软件构建列线图预测模型。再以拟合度曲线及ROC曲线评估该模型对建模组、验证组围绝经期异常子宫出血患者发生子宫内膜病变的预测有效性及区分度。结果 单因素分析结果显示,年龄、体质量指数、肿瘤标记物阳性、合并糖尿病、出血时间为围绝经期异常子宫出血患者发生子宫内膜病变的影响因素(P<0.05),身高、孕产次、初潮年龄、家族肿瘤史、宫内节育器、既往治疗史(妇科手术、激素治疗)、既往病史(高血压、过敏)、子宫内膜厚度、吸烟史及饮酒史等为无关因素(P>0.05);将单因素分析具有差异的因素纳入多因素Logistic回归分析模型,结果显示,体质量指数较大、年龄较大及合并糖尿病为围绝经期异常子宫出血患者发生子宫内膜病变的独立危险因素(P<0.05);使用R软件构建列线图预测模型,以拟合度曲线评估围绝经期异常子宫出血患者发生子宫内膜病变的列线图预测模型的预测有效性,建模组χ~2=6.524,P=0.427;验证组χ~2=6.178,P=0.378。以ROC曲线评估该模型的区分度,结果显示,建模组ROC曲线下面积为0.893,特异性、敏感性分别为90.10%、76.50%;验证组ROC曲线下面积为0.802,敏感性、特异性分别为83.21%、64.71%。结论 体质量指数较大、年龄较大及合并糖尿病为围绝经期异常子宫出血患者发生子宫内膜病变的独立危险因素,以此构建的预测模型具有较好的预测能力及区分度,可为临床早期预测围绝经期异常子宫出血患者发生子宫内膜病变的发生提供参考。
Objective To analyze the risk factors of endometrial lesions in patients with abnormal uterine bleeding during perimenopausal period, and construct a nomogram prediction model.Methods A total of 260 patients with abnormal uterine bleeding during perimenopausal period in Songjiang Hospital Affiliated to Shanghai Jiao Tong University School of Medicine from October 2018 to October 2021 were selected as the research objects.The clinical data of patients were collected, using a computerized randomization method, 182 cases(70%) were used as the modeling group, and the remaining 78 cases(30%) were used as the verification group.According to whether the patients were combined with endometrial lesions, they were divided into normal endometrial group and endometrial lesions group.Univariate analysis and multivariate Logistic regression analysis were used to analyze the risk factors of endometrial lesions in patients with perimenopausal abnormal uterine bleeding in the modeling group, and R software was used to construct a nomogram prediction model.Then, the fit curve and ROC curve were used to evaluate the effectiveness and discrimination of model in predicting the occurrence of endometrial lesions in patients with perimenopausal abnormal uterine bleeding in the modeling group and the verification group.Results Univariate analysis showed that age, body mass index, positive tumor markers, diabetes mellitus, and bleeding time were the influencing factors for endometrial lesions in patients with abnormal uterine bleeding during perimenopause period(P<0.05).and the height, times of pregnancy and delivery, age at menarche, family tumor history, intrauterine device, previous treatment history(gynecological surgery, hormone therapy),previous medical history(hypertension, allergies),endometrial thickness, smoking history and drinking history were not related(P>0.05);the factors with differences in univariate analysis were incorporated into the multivariate logistic regression analysis model, the results showed that larger body mass index, older age, and diabetes mellitus were independent risk factors affecting endometrial lesions in patients with abnormal uterine bleeding during perimenopausal period(P<0.05);R software was used to construct a nomogram prediction model, and the predictive effectiveness of the nomogram prediction model for endometrial lesions in patients with abnormal uterine bleeding during perimenopausal period was evaluated with a degree of fit curve, the results showed that modeling group χ~2=6.524,P=0.427;verification group χ~2=6.178,P=0.378.The degree of discrimination of the model was evaluated by the ROC curve, and the results showed that the area under the ROC curve of the modeling group was 0.893,and the specificity and sensitivity were 90.10% and 76.50%,respectively;the area under the ROC curve of the verification group was 0.802,and the sensitivity and specificity were 83.21% and 64.71%,respectively.Conclusion Larger body mass index, older age, and diabetes mellitus are independent risk factors affecting endometrial lesions in patients with perimenopausal abnormal uterine bleeding.On this basis, the predictive model has good predictive ability and discrimination, can provide a reference for early clinical prediction of the occurrence of endometrial lesions in patients with abnormal uterine bleeding during perimenopausal period.
作者
于静静
崔敏
Yu Jingjing;Cui Min(Department of Obstetrics and Gynecology,Songjang Hospital Affiliated to Shanghai Jiao Tong University School of Medicine(preparatory stage),Shanghai 201699,P.R.China)
出处
《中国计划生育和妇产科》
2022年第9期84-88,共5页
Chinese Journal of Family Planning & Gynecotokology
关键词
子宫内膜病变
围绝经期
异常子宫出血
危险因素
列线图预测模型
endometrial lesions
perimenopausal period
abnormal uterine bleeding
risk factors
nomogram prediction model
