摘要
目的通过分析卵巢子宫内膜异位症(EMT)患者在接受腹腔镜保守手术后复发的相关因素,构建一个临床预测模型,旨在为EMT腹腔镜保守手术后的患者提供复发筛查工具,有利于为患者提供个性化的治疗方案。方法对接受腹腔镜保守手术并经病理诊断为EMT的470名患者进行回顾性分析。根据复发的诊断标准,470名患者被分为复发组(100人)和未复发组(370人)。运用单因素、多因素logistic回归分析方法,筛选出导致EMT腹腔镜保守手术后复发的关键因素。使用R语言构建EMT腹腔镜保守手术后复发的预测模型,并绘制列线图、ROC曲线、校准图和临床决策曲线来评估模型的预测能力和临床效果。本研究使用随机拆分法,将470名EMT患者划分为训练集(330名)和验证集(140名)。结果产次、月经不规律、ASRM分期高、既往盆腔手术史、合并子宫腺肌症是EMT术后复发的相对危险因素(P<0.05)。本项研究构建的EMT患者腹腔镜保守手术后复发预测模型在ROC曲线下的面积(AUC=0.839,灵敏度为61.4%,特异度为90.4%)表现出了良好的区分能力;校准图展示了理论值与预测值的一致性,其斜率近似为1;决策曲线(DCA)具备一定临床应用价值。内部验证显示验证后的ROC曲线的下面积(AUC=0.813,灵敏度为76.7%,特异度为78.2%)。结论产次、月经不规律、ASRM分期高、既往盆腔手术史、合并子宫腺肌症是EMT术后复发的独立危险因素(P<0.05),手术年龄>35岁、术后妊娠及术后药物是保护因素。基于这些因素构建的列线图模型能更有效地进行卵巢子宫内膜异位症术后患者复发的预测。
Objective To construct a clinical predictive model by analyzing the relevant factors of recurrence in patients with ovarian endometriosis(EMT)after laparoscopic conservative surgery,in order to provide recurrence screening tools and personalized treatment plans for patients after laparoscopic conservative surgery.Method A retrospective analysis was conducted on 470 patients who underwent laparoscopic conservative surgery and were pathologically diagnosed with EMT.According to the diagnostic criteria for recurrence,470 patients were divided into recurrence group(100 cases)and non-recurrence group(370 cases).The univariate and multivariate logistic regression analysis methods were used to screen out the key factors that lead to recurrence after conservative surgery for EMT.R language was used to construct a predictive model for postoperative recurrence after conservative surgery for EMT,and draw column charts,ROC curves,calibration charts and clinical decision curves to evaluate the predictive ability and clinical effectiveness of the model.This study used a random split method to divide 470 patients into training set(330 cases)and validation set(140 cases).Result The parity,irregular menstruation,high ASRM staging,previous pelvic surgery history and concomitant adenomyosis were the relative risk factors for postoperative recurrence of EMT(P<0.05).The recurrence prediction model for patients undergoing laparoscopic conservative surgery constructed in this study showed good discriminative ability in the area under the ROC curve(AUC=0.839,sensitivity 61.4%,specificity 90.4%).The calibration chart showed the consistency between the theoretical and predicted values,with a slope of approximately 1.Decision curve(DCA)decisions had certain clinical application value.Internal validation showed the lower area of the validated ROC curve(AUC=0.813,sensitivity 76.7%,specificity 78.2%).Conclusion The parity,irregular menstruation,high ASRM staging,previous pelvic surgery history and concomitant adenomyosis are independent risk
作者
刘丽周
黑海波
闵玉玉
丁永慧
LIU Lizhou;HEI Haibo;MIN Yuyu;DING Yonghui(School of Clinical Medicine,Ningxia Medical University,Yinchuan 750004,China;General Hospital of Ningxia Medical University,Yinchuan 750004,China)
出处
《宁夏医学杂志》
CAS
2024年第2期113-116,共4页
Ningxia Medical Journal
