摘要
目的探讨基于列线图模型预测子宫内膜异位症(endometriosis,EMT)恶变的价值。方法回顾性选取2015年1月至2020年6月在湖南中医药高等专科学校附属第一医院就诊并进行手术治疗,且术后病理诊断为EMT相关性卵巢癌的患者45例和卵巢EMT囊肿无恶变的患者312例,根据EMT是否癌变将研究对象分为癌变组与非癌变组。观察并记录研究对象一般资料、实验室检查和影像学检查结果,并比较两组患者数据的差异。采用Logistic回归判定恶变的独立危险因素,构建列线图预测模型,绘制受试者工作曲线(receiver operating characteristic,ROC)评估列线图模型的预测能力。结果人附睾蛋白4(human epididymal protein 4,HE4),卵巢癌风险评估指数(Risk of Ovarian Malignancy Algorithm,ROMA),超声影像学检查的囊肿实性成分、囊壁乳头、血流信号、囊壁增厚、囊肿最大径,绝经情况,年龄,月经异常和病程是EMT患者恶变的独立危险因素。基于这些危险因素构建的列线图模型拟合效果良好,ROC曲线下面积高达0.982,预测能力优良。结论卵巢癌的肿瘤标志物HE4对EMT恶变的预测能力很强,超声影像学检查也有很好的辅助作用。基于本研究发现的独立危险因素构建的列线图模型可作为量化工具用于EMT患者恶变的预测,有助于术前治疗方案的制定,提高患者预后水平。
Objective To explore the value of predicting endometriosis(EMT)associated malignant cancer based on the nomogram model.Methods 45 cases of patients with EMT⁃associated ovarian cancer and 312 cases of ovarian EMT cysts without malignant cancer were retrospectively selected from January 2015 to June 2020 in the First Affiliated Hospital of Hunan College of Traditional Chinese Medicine and underwent surgical treatment,according to weather was EMT carcinogenesis divided the research subjects into the cancerous group and non⁃cancerous group.Observed and recorded the general information,physical symptoms,laboratory examinations,and imaging examinations of the research subjects,and compared the difference between the two groups of patients.Logistic regression was used to determine the independent risk factors of malignant cancer,a nomogram prediction model was constructed,and receiver operating characteristic(ROC)was drawn to evaluate the predictive ability of the nomogram model.Results Human epididymal protein 4(HE4),ovarian cancer risk assessment index(Risk of Ovarian Malignancy Algorithm,ROMA),solid components of the cyst,papilla of cyst wall,blood flow signal,cyst wall thickening,menopause,the maximum diameter of the cyst,age,menstrual abnormalities and course of EMT were the independent risk factors for malignant cancer.The nomogram model was constructed based on these risk factors and had a good fitting effect.The area under the ROC of the nomogram model was as high as 0.982,which meant the predictive ability was excellent.Conclusions HE4,a tumor marker of ovarian cancer,had a strong ability to predict malignant carcinogenesis of EMT.Ultrasound imaging examination also had a good auxiliary ability.The nomogram model constructed based on the independent risk factors found in this study can be used as a quantitative tool for the prediction of malignant cancer in EMT patients,which could be helpful for the formulation of preoperative treatment plans and improves the prognosis of patients.
作者
言娜
邹丽莎
汤洁
张伶俐
YAN Na;ZOU Li-sha;TANG Jie;ZHANG Ling-li(Department of Gynecology,the First Affiliated Hospital of Hunan College of Traditional Chinese Medicine,Hu'nan Zhuzhou 412000,China)
出处
《循证医学》
2022年第4期230-236,242,共8页
The Journal of Evidence-Based Medicine