4차원 위상변화 혈류측정법에서 가돌리늄의 영향

Purpose：Gadolinium contrast agent is used for time resolved 3D phase contrast (4D flow) due to lack of velocity to noise ratio (VNR). However, there is no quantification analysis of VNR improvement with gadolinium concentration and potential phase error induced by local inhomogeneity. This study evaluated gadolinium effect on 4D flow using Signal to noise ratio (SNR) measurement related to VNR and R2* estimation caused by phase error. Materials and Methods：Flow circuit system was composed of 3D printed tricuspid valve model, 5L working fluid (100% water), and water pump (DPW44-12, DAEHWA electric, Korea)(Fig. 1). A gadolinium contrast agent(meglumine) was diluted into 5 different volumes (0, 3.5, 7, 10.5, 14mL) with 4.8L working fluid prepared in plastic reservoir. The working fluid was circulated through the flow circuit system at a constant flow (80cm/s). 4D flow experiments were performed on 3T clinical scanner(Ingenia CX, Philips, Netherlands) with 32-channel torso coil. The scan parameters were as follows: TR=10.6ms, FOV=200mm x 100mm, Voxel size= 1.5mm x 1.5mm x1.5mm, Temporal resolution= 30, VENC= 150cm/s, Acquisition time= 2min 30sec. Flip angles were 5/10/15/20° for Ernst angle due to relaxivity of gadolinium agent. Since SNR is proportional to VNR, when VENC and flow velocity are fixed, SNR was measured. TE were 2.3/4.6/6.9/10.2ms for estimation of R2* change caused by potential phase error. NEMA’s method was used for SNR measurement and nonlinear square curve fitting method was used for R2* estimation. All post processing and statistical analysis (one-way ANOVA) were performed in MATLAB 2019a (Mathworks Inc.,Natick, MA, USA). Results：As a result of SNR measurement, SNR were dramatically increased from 0mL to 3.5mL in all flip angles (FA 5°-116%, FA 10°-77%, FA 15°-65%, and FA 20°-65%). After 3.5mL, the results of SNR were slightly increased in FA 5°, and remains stable in FA 10°, 15°, 20°(p<0.05) as shown in Fig. 2. R2* values were minimally risen in each volume (0mmol=0.0769, 1.75mmol=0.0793, 3.5mmol=0.0813, 4.75mmol=0.0826, and 7mmol=0.0892). Conclusion：This study demonstrated 2-fold SNR improvement in 4D flow using a small amount of gadolinium contrast agent. Even though gadolinium shortened T1 relaxation time, increase in signal intensity was limited due to optimal Ernst angle. Therefore, additional gadolinium contrast agent is not needed. However, as results of phase error, it is considered that gadolinium affected potential phase error, because R2* value changed a little. The reason of a small variance of R2* value seems to be phase error in bipolar gradient encoding. With phase error, investigators could not evaluate the flow parameter in 4D flow with gadolinium contrast agent. Thus, a small amount of gadolinium should not be used for 4D flow imaging.