Influence of Amplitude-related Perfusion Parameters in the Parotid Glands by Non-fat-saturated Dynamic Contrast-enhanced Magnetic Resonance Imaging

目的: 探討因脂肪飽和技術的施用與否，所造成腮腺動態對比增強微灌流磁振造影參數差異之成因。 方法: 本研究實驗分為三部分: 一、分析兩組臨床病患(脂肪飽和技術與非脂肪飽和技術各18人)腮腺的動態對比增強微灌流磁振造影參數；二、製作三組不同脂肪比例的假體，每組各分為六種對比劑濃度，各用脂肪飽和技術與非脂肪飽和技術擷取影像分析；三、招募九名健康受試者，各接受兩種不同微灌流造影掃描。 將使用脂肪飽和技術及非脂肪飽和技術兩種方法的微灌流參數利用T檢定比較兩組之間差異，並利用線性迴歸觀察脂肪含量與參數之間的關係，統計結果最後用Bonferroni修正多重比較。 結果: 病患資料統計分析顯示，非使用脂肪飽和技術的病人，A(5.08±2.95 a.u.)、PE (34.44±12.48%)以及斜率(1.08±0.60%/s)三個參數，顯著低於使用脂肪飽和技術的病人(8.90±4.03 a.u., 74.55±13.79%, 以及1.79±0.85%/s)。仿體實驗結果顯示信號增強的比例正比於對比劑的濃度，並且非使用脂肪飽和技術的影像信號增強比例顯著低於使用脂肪飽和技術。受試者也是在非使用脂肪飽和技術的參數顯著的低於使用脂肪飽和技術。脂肪飽和技術造成的參數顯著差異，則可以利用圈選脂肪含量較少的組織，做為標準化的基準強度來降低差異。 結論: 富含脂肪的組織如腮腺，動態對比增強微灌流磁振造影特徵會受到脂肪含量影響，使用脂肪飽和技術的可以降低脂肪含量對於參數的影響。選擇脂肪含量較少的組織，做為標準化的基準強度也可以降低影響。Purpose: To investigate the discrepancy of perfusion parameters of the parotid gland acquired by fat-saturated (FS) versus non-fat-saturated (NFS) dynamic contrast-enhanced magnetic resonance imaging (DCE-MRI). Materials and Methods: Approved by a local institutional review board with written informed consent obtained, this study consisted of three parts. First, a retrospective study analyzed DCE-MRI data previously acquired using NFS (18 patients) or FS scans (18 patients). Second, a phantom study simulated the signal enhancements in the presence of Gd contrast agent at 6 different concentrations and 3 different fat contents. Finally, a prospective study recruited 9 healthy volunteers to investigate the influence of fat suppression on perfusion quantification on the same subjects. T tests and linear regression analysis were used for statistical analysis with Bonferroni correction applied for multiple comparisons. Results: Patients undergoing NFS DCE-MR scan showed significantly lower parameter A (5.08±2.95 a.u.), peak enhancement (PE) (34.44±12.48%), and slope (1.08±0.60%/s) as compared to 8.90±4.03 a.u., 74.55±13.79%, and 1.79±0.85%/s, respectively, in those with FS scan (all P<0.0167). Phantom study showed that the relative signal enhancement was proportional to the dose of gadolinium contrast agent and was higher in FS scan than in NFS scan. Volunteer study showed significantly lower parameter A (6.75±2.38 a.u.), PE (42.12±14.87%), and slope (1.43±0.54%/s) in NFS scan as compared to 17.63±8.56 a.u., 104.22±25.15%, and 3.68±1.67%/s, respectively, in those with FS scan (all P<0.005). These perfusion parameter differences were remedied by using skeletal muscles and pure water as reference on in vivo and phantom studies, respectively. Conclusion: DCE-MRI perfusion characterization is affected by the use of FS on fat-containing tissues such as parotid glands. The use of fat saturation is important to reduce the influence of parotid fat content on perfusion quantification. The selection of a relatively fat-free tissue as baseline is a simple and effective method to reduce bias from fat content in DCE MRI of the parotid glands