Xenon when inhaled into the lungs, dissolves in the pulmonary bloodstream and is transported to distal organs (heart, brain, kidney and liver). Upon reaching the cerebral vasculature, xenon passively diffuses into the brain tissues. ¹²⁹Xe exhibits a wide chemical-shift range, providing contrast for different compartments of the brain (grey-matter, white-matter, cerebrospinal fluid and blood)^1-3. Exploiting these properties, we demonstrate the feasibility of using hyperpolarized ¹²⁹Xe for in vivo evaluation of cerebral perfusion and, using tracer kinetic analysis, we enable quantitative measurement of the permeability of the blood-brain barrier.

A bandpass RF birdcage coil (8-leg) tuned to 35.35 MHz (3.0 T ¹²⁹Xe Larmor frequency) was constructed in-house with capacitors 220 pF on end-rings and 75 pF on legs. The RF coil for 1.5 T was as described previously⁴. MR of hyperpolarized ¹²⁹Xe dissolved in the human head was performed at 1.5 T and 3.0 T on GE Signa HDx and Philips Ingenia systems respectively. ¹²⁹Xe gas was hyperpolarized to ~25% by spin exchange optical pumping⁵.

High resolution spectroscopy of hyperpolarized ¹²⁹Xe dissolved in the human head was performed under breath-hold, commencing 4 s after inhalation of a 1 L gas dose. Acquisition parameters for both 1.5 T and 3.0 T were; flip angle = 45°, centre frequency = 198 ppm downfield, receiver bandwidth = 136 ppm, spectral resolution = 0.033 ppm, number of spectra = 10 and TR = 2 s. Spectra were averaged without line broadening filters.

At 1.5 T, 3 sagittal 2D gradient-echo images of the human brain were acquired at 8 s, 16 s and 24 s after inhalation of 1 L xenon gas. Imaging parameters were; flip angle = 12°, centre frequency as above, TE = 1.7 ms, TR = 34 ms, bandwidth = 226 ppm, FOV = 22 cm, slice thickness = 50 mm, matrix = 32 x 32. Acquired images were averaged.

At 1.5 T, a “brain uptake saturation and recovery” (BUSAR) spectroscopy pulse sequence was developed to monitor the movement of hyperpolarized ¹²⁹Xe between cerebral blood and grey-matter. The sequence, shown in Figure 1(a), was initialized by two 90° RF saturation pulses. Each FID in the time series was acquired after a time-gap of 4 s with a 90° RF pulse (pulse duration = 500 μs) followed by two 90° RF saturation pulses. By de-convolving the time course of signal from grey-matter with the input red blood cell signal, we arrive at the relative quantity of xenon transferred between these two compartments. The de-convolved signal was fitted with a custom-derived two-compartment tracer kinetic model shown in Figure 1(b). F_E in the tracer kinetic model is the fraction of xenon atoms transferred from grey-matter to cerebral blood which estimates the permeability and surface area of the blood-brain barrier.

A comparison of high resolution spectroscopy of hyperpolarized ¹²⁹Xe dissolved in the human head at 1.5 T and 3.0 T is shown in Figure 2. Both the spectra exhibited similar features with 5 distinct peaks at 187 ppm, 192 ppm, 195 ppm, 199 ppm and 217 ppm associated with ¹²⁹Xe dissolved in soft muscular tissue, white-matter, grey-matter, cerebrospinal/interstitial fluid and red blood cells^1-4. The sagittal image of hyperpolarized ¹²⁹Xe dissolved in the human brain is shown in Figure 3. A typical spectroscopic time series acquisition from the BUSAR sequence is shown in Figure 4(a). The time course of signal from grey-matter in Figure 4(c) was de-convolved with signal from red blood cells in Figure 4(d). The de-convolved signal was fitted with the tracer kinetic model as shown in Figure 4(b) using F_E = 0.388. Which implies, 38.8% of xenon atoms dissolved in grey-matter is transferred to cerebral blood within time-interval TR = 4 s.Chemical shifts of the resonances observed in the high resolution spectra of ¹²⁹Xe dissolved in the human head at both 1.5 T and 3.0 T are consistent with previous studies^1-3, however this is the first time side-by-side comparison have been made in the same subject. The image of hyperpolarized ¹²⁹Xe dissolved in the human brain indicates cerebral perfusion. The SNR of the image is limited by the quantity of ¹²⁹Xe that is transported and diffused into the brain tissue before the polarization decays, but demonstrates the feasibility of direct brain imaging with inhaled xenon.This is the first time the feasibility of using hyperpolarized ¹²⁹Xe to estimate the intrinsic physiological characteristics of the blood-brain barrier such as permeability and surface area has been investigated. We have demonstrated the feasibility of imaging the human brain at both clinical field strengths with inhaled hyperpolarized ¹²⁹Xe.