女性附件扭转坏死影响因素分析及列线图预测模型构建  

Analysis of influencing factors for torsion necrosis of female adnexa and establishment of nomogram prediction model

作　　者：王晓玲 王清蕊 黄士荷 张金峰 Wang Xiaoling;Wang Qingrui;Huang Shihe;Zhang Jinfeng(Department of Gynaecology,Zhuhai Maternal and Child Health Hospital;Zhuhai Campus of Zunyi Medical University,Zhuhai Guangdong 519400,P.R.China)

机构地区：[1]珠海市妇幼保健院妇科,广东珠海519400 [2]遵义医科大学珠海校区,广东珠海519400

出　　处：《中国计划生育和妇产科》2024年第12期102-107,共6页Chinese Journal of Family Planning & Gynecotokology

基　　金：2022年度珠海市医学科研项目(项目编号:2220009000173);2019年度四川医院管理和发展研究中心项目(项目编号:SCYG2019-70);遵义市红花岗区科技计划项目(项目编号:遵红科合字[2021]06号)。

摘　　要：目的建立女性附件扭转坏死影响因素的列线图预测模型,为早期有效预测附件扭转情况提供参考。方法回顾性分析2010年1月至2020年12月珠海市妇幼保健院收治的174例附件扭转患者的临床资料,根据其病理结果分为坏死组和未坏死组。收集并比较两组的人口学资料、临床资料、实验室检查、附件情况,通过单因素分析筛选变量后,进行二元Logistic回归分析确定最终纳入模型的独立影响因素,根据Logistic回归结果建立列线图模型,并对模型进行验证。结果两组腹痛、盆腔积液、发热、生理阶段、术前血常规白细胞计数和中性粒细胞计数、扭转具体部位、扭转颜色、扭转圈数,差异均有统计学意义(P<0.05)。二元Logistic回归结果显示:盆腔积液、生理阶段中月经初潮前和绝经期、输卵管扭转,扭转颜色红润为扭转坏死发生的独立影响因素(P<0.05)。以R软件建立列线图模型,ROC法评估列线图模型对扭转坏死的预测效能,AUC为0.853(95%CI:0.781-0.924)。校准曲线结果显示:预测曲线与标准曲线基本拟合,H-L拟合优度检验结果显示χ2=9.041,P=0.171。结论女性附件扭转坏死的发生与盆腔积液、生理阶段中月经初潮前和绝经期、输卵管扭转,扭转颜色红润等因素有关,采用上述指标建立的列线图模型可用于扭转坏死的预测且具有较高的准确度与区分度。Objective To establish a predictive model using line charts to identify factors influencing adnexal torsion necrosis in women,providing insights for early and accurate prediction of adnexal torsion.Methods A retrospective analysis was conducted on 174 cases of adnexal torsion admitted to Zhuhai Maternal and Child Health Hospital between January 2010 and December 2020.Patients were categorized into necrotic and non-necrotic groups based on pathological findings.Demographic,clinical data,laboratory data and adnexal condition were collected and compared between the two groups.Univariate analysis was utilized to screen variables,followed by binary Logistic regression to determine independent predictors included in the model.A line chart model was developed based on these predictors and validated accordingly.Results Statistically significant differences were observed between the two groups in terms of abdominal pain,pelvic effusion,fever,menstrual phase,preoperative white blood cell count,preoperative neutrophil count,specific location and color of torsion,and number of torsion circles(P<0.05).Binary Logistic regression revealed pelvic effusion,premenarche,menopause,tubal torsion and the torsional color to rosy as independent predictors for torsion necrosis(P<0.05).Using R software,the line chart model demonstrated a predictive efficacy with an AUC of 0.853(95%CI:0.781-0.924).Calibration curve analysis indicated good fit between the predicted and standard curves,with a nonsignificant H-L goodness-of-fit test result(χ~2=9.041,P=0.171).Conclusion Factors such as pelvic effusion,premenarche,menopause,tubal torsion,and torsion color significantly influence the occurrence of adnexal torsion necrosis in women.The established line chart model offers a highly accurate and discriminative approach for predicting torsion necrosis.

关 键 词：附件扭转 坏死 风险预测 列线图 

分 类 号：R711.2[医药卫生—妇产科学]