Quantification of Gd-DTPA concentration in neuroimaging using T 1 3D MP-RAGE sequence at 3.0 T

Abstract

Purpose To propose a simple and accurate quantitative method based on the linear relationship between magnetic resonance (MR) signal enhancement (Δ SI= SI postcontrast− SI precontrast) and gadolinium-diethylenetriaminepentaacetic acid (Gd-DTPA) concentration ( C) by using T 1-weighted three-dimensional magnetization-prepared rapid acquisition gradient-echo (T 1 3D MP-RAGE) sequence for the in vivo measurement of Gd-DTPA concentration in real-time neuroimaging at 3.0 T. Methods Phantom experiment was carried out to study the linear fitting of signal intensity change vs. Gd-DTPA concentration (Δ SI– C) curve. A goodness-of-fit test was performed to compare the accuracy between the proposed method and the conventional method based on longitudinal relaxation rate ( R 1=1/ T 1) measurement. The influences on the goodness of fit ( R 2) and the signal-to-noise ratio (SNR) by sequence parameters were explored. Six human subjects with different brain tumors, who underwent a Gd-DTPA-enhanced MRI, were enrolled for in vivo application of the novel method. Results A good linear relationship between Δ SI and Gd-DTPA concentration existed over the concentration range of 0–1 mM ( R 2=0.985). The linearity of the Δ SI– C curve was as good as that of the 1/ T 1– C curve ( R 2=0.988). Concentrations calculated by both methods had a strong correlation ( R 2=0.920). An improved linearity of the Δ SI– C curve and an increased SNR can be achieved using sequences with a shorter inversion time (TI) and a higher flip angle. The concentration range of Gd-DTPA in human brain tumors was within the quantitative scope of 0–1 mM. Conclusions The proposed quantitative method based on Δ SI measurement is accurate and applicable for real-time neuroimaging at 3.0 T.