Normal saline as a natural intravascular contrast agent for dynamic perfusion-weighted MRI of the brain: Proof of concept at 1.5T

作者：Jara Hernan^*; Mian Asim; Sakai Osamu; Anderson Stephan W; Horn Mitchel J; Norbash Alexander M; Soto Jorge A

来源：Journal of Magnetic Resonance Imaging, 2016, 44(6): 1580-1591.

DOI：10.1002/jmri.25291

• MR perfusion
• normal saline
• transient hemodilution

PurposeGadolinium-based contrast agents have associated risks. Normal saline (NS) is a nontoxic sodium chloride water solution that can significantly increase the magnetic resonance imaging (MRI) relaxation times of blood via transient hemodilution (THD). The purpose of this pilot study was to test in vivo in the head the potential of normal saline as a safer, exogenous perfusion contrast agent.
Materials and MethodsThis Health Insurance Portability and Accountability Act (HIPAA)-compliant prospective study was approved by the local Institutional Review Board (IRB): 12 patients were scanned with T-1-weighted inversion recovery turbo spin echo pulse sequence at 1.5T. The dynamic inversion recovery pulse sequence was run before, during, and after the NS injection for up to 5 minutes: 100 ml of NS was power-injected via antecubital veins at 3-4 ml/s. Images were processed to map maximum enhancement area-under-the-curve, time-to-peak, and mean-transit-time. These maps were used to identify the areas showing significant NS injection-related signal and to generate enhancement time curves. Hardware and pulse sequence stability were studied via phantom experimentation. Main features of the time curves were tested against theoretical modeling of THD signal effects using inversion recovery pulse sequences. Pearson correlation coefficient (R) mapping was used to differentiate genuine THD effects from motion confounders and noise.
ResultsThe scans of 8 out of 12 patients showed NS injection-related effects that correlate in magnitude with tissue type (gray matter approximate to 15% and white matter approximate to 3%). Motion artifacts prevented ascertaining NS signal effects in the remaining four patients. Positive and negative time curves were observed in vivo and this dual THD signal polarity was also observed in the theoretical simulations. R-histograms that were approximately constant in the range 0.1 < |R| < 0.8 and leading to correlation fractions of F-corr(|R| > 0.5) = 0.45 and 0.59 were found to represent scans with genuine THD signal effects.
ConclusionA measurable perfusion effect in brain tissue was demonstrated in vivo using NS as an injectable intravascular contrast agent. J. Magn. Reson. Imaging 2016;44:1580-1591.

• 出版日期2016-12