术前多模态超声参数对乳腺良性结节微波消融术后疗效的预测价值OACSTPCD
Predictive Value of Preoperative Multimodal Ultrasound Parameters for Curative Effect on Benign Breast Nodules after Microwave Ablation
目的:探讨术前多模态超声参数对乳腺良性结节微波消融术后疗效的预测价值.方法:选取2019年1月至2022年9月在本院择期行微波消融术的乳腺良性结节患者175例(共210个结节).根据术后3个月的乳腺结节缩小率和是否消融完全,将患者分为有效组(n=143)和无效组(n=32).收集患者临床资料,并于术前采用多模态超声(常规超声、超声造影和超声弹性成像)检查记录结节最大直径(Dmax)、峰值强度(PI)、达峰时间(TTP)、杨氏模量最大弹性值(Emax)和杨氏模量平均弹性值(Emean).采用单因素和多因素Logistic回归分析影响乳腺良性结节患者微波消融术后疗效的独立危险因素,并建立风险预测模型,受试者操作特征(ROC)曲线检测模型的预测效能.结果:有效组和无效组患者在术前Dmax、BMI、PI、TTP、Emax和Emean比较上有明显差异(P＜0.05),在年龄、婚育史和合并基础病等比较上无明显差异(P＞0.05).Logistic回归分析显示,术前Dmax、PI和Emax均为影响乳腺良性结节患者微波消融术后疗效的独立危险因素.风险预测模型概率P=1/(1+e-25.568-4.911×Dmax+0.541×PI+0.604×Emax),Hosmer-Lemeshow x2=6.235(P=0.621),ROC分析显示,Logistic回归模型预测乳腺良性结节患者微波消融术后疗效的AUC为0.793,95％CI为0.712～0.874.结论:术前Dmax、PI和Emax均为影响乳腺良性结节患者微波消融术后疗效的独立危险因素,基于多模态超声参数构建Logistic回归模型可以有效预测乳腺良性结节患者微波消融术后疗效,并通过该风险预测模型进行有效干预.
Objective:To explore the predictive value of preoperative multimodal ultrasound parameters for curative effect on benign breast nodules after microwave ablation.Methods:A total of 175 patients with benign breast nodules(210 nodules)undergoing elective microwave ablation in the hospital were enrolled between January 2019 and September 2022.According to reduction rate of breast nodules at 3 months after surgery and presence or absence of complete ablation,they were divided into effective group(m=143)and ineffective group(n=32).The clinical data of patients were collected.Before surgery,maximum diameter(Dmax),peak intensity(PI),time to peak(TTP),maximum elasticity of Young's modulus(Emax)and mean elasticity of Young's modulus(Emean)of nodules were recorded by multimodal ultrasound(routine ultrasound,contrast-enhanced ultrasound,ultrasonic elastography).The independent risk factors affecting the curative effect of microwave ablation in patients with benign breast nodules were analyzed by univariate and multivariate Logistic regression analysis.The risk prediction model was constructed,and its predictive efficiency was detected by receiver operating characteristic(ROC)curves.Results:Before surgery,there were significant differences in Dmax,BMI,PI,TTP,Emax and Emean between effective group and ineffective group(P＜0.05),but there was no significant difference in age,marital history or underlying diseases(P＞0.05).Logistic regression analysis showed that preoperative Dmax,PI and Emax were independent risk factors affecting the curative effect of microwave ablation.The probability by risk prediction model was as follow:P=1/(1+e-25.568-4 911×Dmax+0.541×PI+0.604×Emax),Hosmer-Lemeshow x2=6.235,P=0.621.ROC curves analysis showed that AUC and 95％CI of Logistic regression model for predicting curative effect of microwave ablation were 0.793 and 0.712-0.874,respectively.Conclusion:Preoperative Dmax,PI and Emax are all independent risk factors affecting the curative effect of microwave ablation in patients with benign breast nodules.Logistic regression model based on multimodal ultrasound parameters can effectively predict the curative effect of microwave ablation,and effective intervention can be taken by the risk prediction model.
刘晓璐;刘景萍;王谦;徐娟
唐山市人民医院超声科,河北 唐山 063000
多模态超声乳腺良性结节微波消融术疗效危险因素预测模型
multimodal ultrasoundbenign breast nodulemicrowave ablationcurative effectrisk factorprediction model
《影像科学与光化学》 2024 (002)
141-147 / 7
河北省医学科学研究重点课题(20181232)
评论