目的 探讨动态对比增强磁共振成像(dynamic contrast-enhanced MRI,DCE-MRI)技术在甲状腺相关性眼病(thyroid-associated ophthalmopathy,TAO)临床活动性分期中的价值.材料与方法 前瞻性纳入TAO患者56例,其中活动期组(37例,74只眼),非活动期组(19例,38只眼).计算并比较两组间眼外肌的半定量参数[达峰时间(time to peak,TTP)、时间-信号强度曲线下面积(area under the time-signal intensity curve,AUC)和时间-信号强度曲线最大斜率值(maximum enhancement slope,Slopemax)]以及定量参数[容积转运常数(volume transfer constant,Ktrans)、反流速率常数(the rate constant,Kep)和血管外细胞外间隙容积分数(fractional volume of the extravascular-extracellular space,Ve)]的最小值、平均值和最大值.采用多因素logistic回归筛选TAO活动性分期的独立影响因素.采用受试者工作特征(receiver operating characteristic,ROC)曲线分析显著差异参数及其联合模型在TAO分期中的效能.结果 活动期组眼外肌的TTP均值、AUC最大值、AUC均值、Ve最大值和Ve均值均高于非活动期组(P<0.05).AUC最大值和Ve均值是TAO活动性分期的独立影响因素(AUC最大值,P=0.030;Ve均值,P=0.014).采用AUC最大值和Ve均值预测活动期TAO的ROC曲线下面积分别为0.689和0.673.联合AUC最大值和Ve均值可进一步提高对TAO活动性分期的诊断效能(ROC曲线下面积0.731).结论 眼外肌DCE-MRI半定量和定量参数均可辅助TAO活动性分期.联合AUC最大值和Ve均值可以有效评估TAO活动性分期.
Value of dynamic contrast-enhanced MRI in evaluating the microcirculation of extraocular muscle and stage of thyroid-associated ophthalmopathy
Objective:To assess the value of dynamic contrast-enhanced magnetic resonance imaging (DCE-MRI) for staging thyroid-associated ophthalmopathy (TAO). Materials and Methods:We prospectively enrolled 56 TAO patients,and divided them into active group (37 patients with 74 eyes) and inactive group (19 patients with 38 eyes). The minimum,mean,and maximum values of semi-quantitative[time to peak (TTP),area under the time-signal intensity curve (AUC),maximum enhancement slope (Slopemax)]and quantitative[volume transfer constant (Ktrans),the rate constant (Kep),fractional volume of the extravascular-extracellular space (Ve)]parameters were calculated and compared between groups. Multivariate logistic regression analysis was applied to identify the independent imaging indicators of active TAO. Receiver operating characteristic (ROC) curve analysis was used to evaluate the performance of the identified significant imaging parameters and their combination. Results:Active patients showed significantly higher mean TTP,mean and maximum AUC,mean and maximum Ve than inactive patients (P<0.05). Maximum AUC and mean Ve were found to be independent variables for determining the disease activity (P=0.030,0.014,respectively). The area under ROC curve of active TAO was 0.689 and 0.673,respectively,using the maximum AUC and the mean Ve values. Combination of two parameters could determine the active TAO with optimal performance (area under the ROC curve 0.731). Conclusions:DCE-MRI-derived semi-quantitative and quantitative parameters are all useful for determining the activity of TAO. The model combining maximum AUC and mean Ve can effectively help to stage the patients with TAO.