Introduction
The proposed contrast-enhanced first-pass perfusion cardiovascular MR (CMR) protocol integrates some recent technical MRI advances towards quantitative analysis of perfusion CMR images: fast multi-slice pulse sequence [1], robust saturation pulse [2], accurate dedicated AIF imaging [3], signal-to-concentration modeling [4], and higher SNR at 3T.
Purpose
To evaluate an integrated first-pass perfusion cardiovascular MR (CMR) protocol designed to determine absolute contrast-agent concentrations in blood and tissues.
Methods
A multi-slice saturation recovery (SR) pulse sequence with sequential SR time delays (TD) after a non-selective saturation pulse [2] was implemented at 3T (Fig.1). The rationale for this acquisition scheme was to acquire a dedicated arterial input function (AIF) image with a short TD (50ms) in the aortic root and short-axis myocardial images with longer TD values (~150-400ms), to allow for the different amounts of T1 shortening expected in blood and wall. First-pass perfusion CMR was performed in 7 volunteers (0.05mmol/kg, Gd-DTPA). A signal-to-concentration model was applied to calculate Gd-DTPA concentrations in blood and tissues [4,5]. A proton density-weighted (PDw) image was acquired in the first heartbeat, without the saturation pulse, in order to normalize the image signal, and obtain a theoretical signal-to-T1 relationship based on Bloch equation in the center of k-space. Gd-DTPA concentrations were calculated assuming: fast water exchange condition [6], longitudinal relaxivity r1=3.8L.mmol-1.s-1[7], and baseline T1 measured with a multi-point SR fit. TurboFLASH imaging parameters included: FOV=350mm×315mm, slice thickness=8mm, matrix=160×144, in-plane resolution=2.2mm×2.2mm, TE/TR=1.2/2.4ms, flip angle 10°, temporal resolution=114ms, tSENSEx3, centric k-space trajectory, and receiver bandwidth=1008Hz/pix. Total image acquisition time was 523ms for the acquisition of 4 slices, namely aortic root and SA base, mid, and apex levels, with respective TD values 50-164-278-393ms. Contours for the blood and left ventricle were drawn manually, and the myocardium was divided into 6 (base-mid) or 4 (apex) standard segments.
Figure 1.
Schematic diagram of the mulit-slice with sequential TD first-pass perfusion CMR pulse sequence
Results
Representative images at peak contrast in blood and myocardium are shown Fig.2, along with representative AIF and myocardial segment time-responses. The peak blood signal was not clipped in the short TD=50ms AIF images. Normalized signal in the myocardium increased along with TD; however similar [Gd-DTPA] were measured in all 3 short-axis images. First-pass perfusion peak Gd-DTPA concentrations were 3.95±0.080, 0.26±0.07mM in the blood and myocardium, respectively.
Figure 2.
Representative images at peak blood and peak myocardium concentrations.
Figure 3.
Representative [Gd-DTPA] times-curves in blood in the aortic root and in 6 cardiac segments in cardiac short-axis images.
Conclusions
The proposed integrated first-pass perfusion CMR protocol at 3T produced AIF and myocardial wall Gd-DTPA concentrations consistent with previously published results. Future work includes evaluation of the integrated protocol in cardiac patients.
References
- Nagel E, Circulation. 2003. pp. 432–437. [DOI] [PubMed]
- Kim D, Magn Reson Med. 2008. pp. 1368–1378. [DOI] [PMC free article] [PubMed]
- Kim D, Axel LJ. Magn Reson Imaging. 2006. pp. 81–86. [DOI] [PubMed]
- Cernicau A, Axel L. Acad Radiol. 2006. pp. 686–693. [DOI] [PubMed]
- Breton E, SCMR. 2010.
- Dobahue KM, J Magn Reson Imaging. 1997. pp. 102–110. [DOI] [PubMed]
- Rohrer M, invest Radiol. 2005. pp. 715–724. [DOI] [PubMed]



