ADC最小值联合临床及影像学特征模型预测HIFU即刻消融子宫肌瘤疗效的价值

The Value of Minimum ADC Combined with Clinical and Imaging Models to Predict the Efficacy of Immediate HIFU Ablation of Uterine Leiomyomas

目的:探讨表观弥散系数(ADC)最小值联合临床及影像学特征构建模型预测高强度聚焦超声(HIFU)治疗子宫肌瘤疗效的临床价值。方法:收集323例确诊为子宫肌瘤并行HIFU治疗患者的临床及影像资料,依据肌瘤非灌注区体积比(NPVR)将其分为充分消融组(NPVR≥70%)和非充分消融组(NPVR<70%),其中充分消融组205例,非充分消融118例。由2名医生独立测量肌瘤ADC平均值、ADC最小值,并采用受试者工作特征(ROC)曲线比较两者的预测效能。通过单因素分析筛选出有意义的临床及影像学特征,采用二元logistic回归分别构建临床及影像学特征模型、ADC最小值联合临床及影像学特征模型。通过ROC曲线评估ADC最小值、临床及影像学特征和二者联合模型的预测效能。结果:ADC最小值[曲线下面积(AUC)=0.802]预测效能高于ADC平均值(AUC=0.730)。ADC最小值、临床及影像学特征和二者联合模型的AUC分别为0.802、0.760、0.829,DeLong检验结果显示二者联合模型与ADC最小值、临床及影像学特征之间的差异均有统计学意义(P<0.05),联合模型对HIFU即刻消融子宫肌瘤疗效的预测效能最佳,其AUC、灵敏度和特异度分别为0.829、84.4%和72.9%。结论:基于ADC最小值联合临床及影像学特征建立的模型对HIFU即刻消融子宫肌瘤疗效具有较好的预测效能,该模型可作为预测HIFU治疗子宫肌瘤疗效有前景的的影像学工具。

Purpose:To investigate the clinical significance of combining the minimum apparent diffusion coefficient(ADC)value with clinical and imaging features in predicting the efficacy of high-intensity focused ultrasound(HIFU)treatment for uterine leiomyomas.Methods:Clinical and imaging data from 323 patients diagnosed with uterine leiomyomas and treated with HIFU were collected.Based on the non-perfused volume ratio(NPVR)of fibroids,patients were divided into two groups:adequate ablation group(NPVR≥70%)consisting of 205 patients,and inadequate ablation group(NPVR<70%)consisting of 118 patients.The average ADC value and minimum ADC value of fibroids were independently measured by two physicians,and their predictive efficacy was compared using receiver operating characteristic(ROC)curve analysis.Significant clinical and imaging features were identified through univariate analysis,then the clinical and imaging features models and the minimum ADC value combined with clinical and imaging features models were constructed using binary logistic regression analysis.The ROC curve was used to assess the predictive performance of the minimum ADC value,clinical and imaging features and the combined models.Results:The predictive efficacy of the minimum ADC value[area under the curve(AUC)=0.802]was higher than that of the average ADC value(AUC=0.730).The AUC of the combination model was 0.829,which was with the best predictive efficiency for the efficacy of immediate HIFU ablation of uterine leiomyomas.The AUC,sensitivity and specificity were 0.829,84.4%and 72.9%,respectively.Conclusions:The model based on the minimum ADC value combined with clinical and imaging features has a preferable predictive efficiency for the efficacy of immediate HIFU ablation of uterine leiomyomas.This model can be used as a promising imaging tool for predicting the efficacy of HIFU treatment of uterine leiomyomas.