Qin Shi-Ze, Jiang Yu, Wang Yan-Lin, Liu Nian, Lin Zhen-Yang, Jia Qing, Fang Jie, Huang Xiao-Hua
Department of Radiology, Affiliated Hospital of North Sichuan Medical College, No. 1, Maoyuan South Road, Shunqing District, Nanchong, 637000, China.
School of Clinical Medicine, North Sichuan Medical College, No. 234, Fujiang Road, Shunqing District, Nanchong, 637000, China.
Abdom Radiol (NY). 2024 Jun;49(6):2017-2026. doi: 10.1007/s00261-023-03865-6. Epub 2023 Mar 13.
To predict the efficacy of high-intensity focused ultrasound (HIFU) ablation for uterine leiomyomas based on diffusion tensor imaging (DTI) indicators and imaging features.
Sixty-two patients with 85 uterine leiomyomas were consecutively enrolled in this retrospective study and underwent DTI scanning before HIFU treatment. Based on whether the non-perfused volume ratio (NPVR) was greater than 70%, all patients were assigned to sufficient ablation (NPVR ≥ 70%) or insufficient ablation (NPVR < 70%) groups. The selected DTI indicators and imaging features were incorporated to construct a combined model. The predictive performance of DTI indicators and the combined model were assessed using receiver operating characteristic (ROC) curves.
There were 42 leiomyomas in the sufficient ablation group (NPVR ≥ 70%) and 43 leiomyomas in the insufficient ablation group (NPVR < 70%). The fractional anisotropy (FA) and relative anisotropy (RA) values were higher in the sufficient ablation group than in the insufficient ablation group (p < 0.05). Conversely, the volume ratio (VR) and mean diffusivity (MD) values were lower in the sufficient ablation group than those in the insufficient ablation group (p < 0.05). Notably, the combined model composed of the RA and enhancement degree values had high predictive efficiency, with an AUC of 0.915. The combined model demonstrated higher predictive performance than FA and MD alone (p = 0.032 and p < 0.001, respectively) but showed no significant improvement compared with RA and VR (p > 0.05).
DTI indicators, especially the combined model incorporating DTI indicators and imaging features, can be a promising imaging tool to assist clinicians in predicting HIFU efficacy for uterine leiomyomas.
基于扩散张量成像（DTI）指标和影像特征预测高强度聚焦超声（HIFU）消融治疗子宫平滑肌瘤的疗效。
本回顾性研究连续纳入62例患有85个子宫平滑肌瘤的患者，在HIFU治疗前进行DTI扫描。根据无灌注体积比（NPVR）是否大于70%，将所有患者分为充分消融组（NPVR≥70%）和消融不足组（NPVR<70%）。纳入所选的DTI指标和影像特征构建联合模型。使用受试者工作特征（ROC）曲线评估DTI指标和联合模型的预测性能。
充分消融组（NPVR≥70%）有42个平滑肌瘤，消融不足组（NPVR<70%）有43个平滑肌瘤。充分消融组的分数各向异性（FA）和相对各向异性（RA）值高于消融不足组（p<0.05）。相反，充分消融组的体积比（VR）和平均扩散率（MD）值低于消融不足组（p<0.05）。值得注意的是，由RA和强化程度值组成的联合模型具有较高的预测效率，曲线下面积（AUC）为0.915。联合模型的预测性能高于单独的FA和MD（分别为p=0.032和p<0.001），但与RA和VR相比无显著改善（p>0.05）。
DTI指标，尤其是结合DTI指标和影像特征的联合模型，可能成为一种有前景的影像工具，帮助临床医生预测HIFU治疗子宫平滑肌瘤的疗效。
