摘要
目的:探讨恶性风险模型联合三维能量超声模型对绝经后出血(PMB)女性子宫内膜癌(EC)的预测价值。方法:选择2016年7月至2019年4月在西安交通大学第一附属医院就诊的PMB女性为研究对象,根据病理检查结果分为子宫内膜良性病变组104例,EC组36例;另选择同期妇科查体正常的绝经后女性38例为对照组;选择2019年2~4月就诊的69例PMB患者为模型验证对象。以年龄、体质量指数(BMI)、子宫内膜厚度(ET)、反复阴道流血和糖尿病史建立恶性风险模型,以子宫内膜体积(V)、血管形成指数(VI)、血流指数(FI)以及血管形成-血流指数(VFI)构建三维能量超声模型。受试者工作特征(ROC)曲线分析不同预测方法的价值,并对恶性风险模型联合三维能量超声模型预测方法的有效性和重复性进行验证。结果:①EC组患者子宫内膜超声指标ET、V、VI、FI、VFI显著高于良性病变组和对照组,良性病变组显著高于对照组,差异均有统计学意义(P<0.05)。②超声指标中FI对EC的预测价值最大(AUC=0.86),其次为VI(AUC=0.84)、VFI(AUC=0.82),三维能量超声模型总分4~8分,截断评分值≥6分时对EC的预测价值最大(AUC=0.87)。③恶性风险模型总分0~9分,截断评分值≥4分时对EC的预测价值最大(AUC=0.80)。④两种模型联合应用预测价值(AUC=0.94)显著高于恶性风险模型(Z=3.126,P=0.039)和三维能量超声模型(Z=2.970,P=0.043);而恶性风险模型与三维能量超声模型比较,差异无统计学意义(Z=1.574,P=0.092)。⑤69例PMB验证对象采用联合模型进行预测验证,AUC=0.89,灵敏度为95.48%,特异度为83.97%,与模型建立时联合ROC曲线AUC比较,差异无统计学意义(Z=0.908,P=0.126),具有可重复性。结论:恶性风险模型联合三维能量超声模型对PMB女性EC有较好的预测价值,可提高诊断效果,但仍需多中心大样本研究进一步验证。
Objective:To probe the predictive value of malignant risk model combined with three-dimensional energy ultrasonic model for endometrial cancer(EC)in women with postmenopausal hemorrhage(PMB).Methods:Women with PMB in The First Affiliated Hospital of Xi′an Jiaotong University from July 2016 to April 2019 were selected,including 104 women with endometrial benign lesions and 36 women with EC according to the pathological examination results.38 postmenopausal women with normal gynecological examination in the same period were selected as the control group.69 PMB women admitted to our hospital from February to April 2019 were selected as the model verification objects.Age,body mass index(BMI),endometrial thickness(ET),repeated vaginal bleeding and history of diabetes were used to establish the malignant risk model.Endometrial volume(V),vascularization index(VI),flow index(FI)and vascularization flow index(VFI)were used to construct the three-dimensional energy ultrasonic model.The receiver operating characteristic(ROC)curve was used to analyze the value of different prediction methods,and the effectiveness and repeatability of the combined predictive method of malignant risk model and three-dimensional energy ultrasonic model was verified.Results:①The ET,V,VI,FI and VFI were significantly increased in EC group compared with benign lesion group and control group,which were significantly increased in benign lesion group compared with control group(P<0.05).②FI had the largest predictive value for EC(AUC=0.86)among the ultrasonic indicators,followed by VI(AUC=0.84),VFI(AUC=0.82).The total score of three-dimensional energy ultrasonic model is 4 to 8 points.The cutoff value of three-dimensional energy ultrasonic model was≥6 scores with the maximum predictive value(AUC=0.87).③The total score of malignant risk model is 0 to 9 points.The cutoff value of malignant risk model was≥4 scores with the maximum predictive value(AUC=0.80).④The predictive value of combining two models was significantly higher than that of the malignant risk model(Z=3.126,P=0.039)and the three-dimensional energy ultrasonic model(Z=2.970,P=0.043),and there was no statistically significant difference between the malignant risk model and the three-dimensional energy ultrasonic model(Z=1.574,P=0.092).⑤ 69 PMB women were verified by the combined model,the sensitivity was 95.48%,specificity was 83.97%,AUC=0.89.There was no statistically significant difference between the AUC under ROC curve when the two models were applied compared with the model established(Z=0.908,P=0.126),and with reproducibility.Conclusions:Malignant risk model combined with three-dimensional energy ultrasonic model has good predictive value for EC in women with PMB,and can increase the diagnosis effects,but multi-center large sample study is still needed to further verify.
作者
王丽
郭艳平
全诗敏
杨婷
赵娟
裴美丽
杨筱凤
WANG Li;GUO Yanping;QUAN Shimin(Department of Gynecology and Obstetrics,The First Affiliated Hospital of Xi’an Jiaotong University,Xi’an Shaanxi 710061,China)
出处
《实用妇产科杂志》
CAS
CSCD
北大核心
2019年第12期923-927,共5页
Journal of Practical Obstetrics and Gynecology
基金
陕西省卫生健康科研基金项目(编号:2018D054)
关键词
恶性风险模型
三维能量多普勒超声
绝经后出血
子宫内膜癌
预测价值
Malignant risk model
Three-dimensional energy doppler ultrasound
Postmenopausal bleeding
Endometrial cancer
Predictive value
