Department of Radiology, Kagoshima University Graduate School of Medical and Dental Sciences, 8-35-1 Sakuragaoka, Kagoshima, 890-8544, Japan.
Sci Rep. 2021 Feb 2;11(1):2780. doi: 10.1038/s41598-021-82455-2.
The purpose of this study is to determine whether multiparametric non-contrast MR imaging including diffusion-weighted imaging (DWI), arterial spin labeling (ASL), and amide proton transfer (APT) weighted imaging can help differentiate malignant from benign salivary gland lesions. The study population consisted of 42 patients, with 31 benign and 11 malignant salivary gland lesions. All patients were evaluated using DWI, three-dimensional pseudo-continuous ASL, and APT-weighted imaging on 3 T MR imaging before treatment. Apparent diffusion coefficient (ADC), tumor blood flow (TBF), and APT-related signal intensity (APTSI) values within the lesion were compared between the malignant and benign lesions by Mann-Whitney U test. For each parameter, optimal cutoff values were chosen using a threshold criterion that maximized the Youden index for predicting malignant lesions. The performance of ADC, TBF, APTSI, individually and combined, was evaluated in terms of diagnostic ability for malignant lesions. Diagnostic performance was compared by McNemar test. APTSI was significantly higher in malignant lesions (2.18 ± 0.89%) than in benign lesions (1.57 ± 1.09%, p = 0.047). There was no significant difference in ADC or TBF between benign and malignant lesions (p = 0.155 and 0.498, respectively). The accuracy of ADC, TBF, and APTSI for diagnosing malignant lesions was 47.6%, 50.0%, and 66.7%, respectively; whereas the accuracy of the three parameters combined was 85.7%, which was significantly higher than that of each parameter alone (p = 0.001, 0.001, and 0.008, respectively). Therefore, the combination of ADC, TBF, and APTSI can help differentiate malignant from benign salivary gland lesions.
本研究旨在确定多参数非对比磁共振成像(包括弥散加权成像[DWI]、动脉自旋标记[ASL]和酰胺质子转移[APT]加权成像)是否有助于区分良恶性唾液腺病变。研究人群包括 42 例患者,其中 31 例为良性,11 例为恶性唾液腺病变。所有患者在治疗前均于 3TMR 成像上接受 DWI、三维伪连续 ASL 和 APT 加权成像检查。采用 Mann-Whitney U 检验比较良恶性病变间病变内表观弥散系数(ADC)、肿瘤血流(TBF)和 APT 相关信号强度(APTSI)值。对于每个参数,均采用以 Youden 指数最大为标准的阈值准则选择最佳截断值,以预测恶性病变。分别评估 ADC、TBF、APTSI 以及它们的组合对恶性病变的诊断能力。采用 McNemar 检验比较诊断性能。恶性病变的 APTSI 值明显高于良性病变(2.18±0.89%比 1.57±1.09%,p=0.047)。良性和恶性病变间 ADC 或 TBF 无显著差异(p=0.155 和 0.498)。ADC、TBF 和 APTSI 诊断恶性病变的准确率分别为 47.6%、50.0%和 66.7%;而三者联合的准确率为 85.7%,明显高于各参数单独的准确率(p=0.001、0.001 和 0.008)。因此,ADC、TBF 和 APTSI 的联合有助于区分良恶性唾液腺病变。