Robin A de Graaf1, Arjan D Hendriks2, Dennis W Klomp2, Chathura Kumaragamage1, Dimitri Welting2, Catalina S Arteaga de Castro2, Peter B Brown1, Scott McIntyre1, Terence W Nixon1, Jeanine J Prompers2, and Henk M De Feyter1
1Yale University, New Haven, CT, United States, 2University Medical Center Utrecht, Utrecht, Netherlands
Synopsis
Deuterium
metabolic imaging (DMI) is a novel, MR-based method to spatially map
metabolism. DMI has been shown to provide robust and sensitive maps of cerebral
glucose metabolism in healthy volunteers and patients with brain tumors. Here
the magnetic field dependence of DMI in terms of sensitivity and resolution is
investigated. Using RF coils sized for animal and human studies on magnetic fields
ranging from 4T to 11.7T, a supralinear
magnetic field dependence of the DMI sensitivity is established. The increased
sensitivity of DMI at 7T compared to 4T allows the acquisition of human brain
DMI at a 1 mL spatial resolution.
Introduction
Deuterium
metabolic imaging (DMI) is a novel MR-based method to spatially map metabolism [1].
The technical simplicity and robustness of DMI, together with the range and
affordability of relevant deuterated substrates gives DMI high clinical
potential. The applicability of DMI can be further cemented by maximizing its
detection sensitivity. DMI is based on the detection of deuterium (2H),
a quadrupolar nucleus with a small gyromagnetic ratio. The low 2H
Larmor frequency, together with the field independent T1 and T2
relaxation constants, predicts a near-quadratic magnetic field dependence of the
SNR. Here the magnetic field dependence of DMI is investigated for a range of
MR systems and RF coils in order to understand the possibilities on clinical MR
systems and predict the potential gains achievable at 7T.Methods
All studies were performed on 4T (Bruker), 7T
(Philips) and 11.7T (Bruker) MR systems with single-turn RF surface coils tuned
for 2H (26.2 MHz, 45.9 MHz and 76.7 MHz at 4T, 7T and 11.7T,
respectively) and sized for animal (Ø 30 mm, 4T and 11.7 T) or human (Ø 80 mm,
4T and 7T) studies. MR system noise figures and loaded and unloaded coil Q
values were measured. 2H MRS was
performed with a pulse-acquire pulse sequence (TR = 333 ms) extended with
phase-encoding gradients for 3D DMI (11 x 11 x 11 matrix over 220 x 220 x 220
mm for human studies and 13 x 13 x 13 matrix over 39 x 39 x 39 mm for animal
studies). SNR was determined as the natural abundance 2H water
signal integral (0.2 kHz integration width) divided by the RMS spectral noise
level. The magnetic field dependence of the 2H SNR was modeled as
SNR ~ B0n, whereby n = [1 … 2]. The spatial and spectral resolution
of DMI on human brain at 4T and 7T was investigated between 50 and 115 min
following oral administration of [6,6’-2H2]-glucose (0.75
g/kg).Results
Fig.
1 shows the magnetic field dependence of the 2H natural abundance
water signal on rat brain post-mortem. For identical-sized RF coils (Ø 30 mm)
the SNR at 11.7T is ~6.5 - 8.0x higher than at 4T, corresponding to a
near-quadratic magnetic field dependence (B0n, with n ~1.7 - 1.9).
Fig.
2 shows the magnetic field dependence of the 2H natural abundance
water signal on human brain in vivo.
The SNR at 7T is ~2.3x higher than at 4T (n = 3), corresponding to a
supralinear magnetic field dependence (B0n, with n ~ 1.5). The average line widths
in the 8 mL DMI volumes at 4T and 7T were 11.0 +/- 2.6 Hz (or 0.42 +/- 0.1 ppm)
and 16.3 +/- 2.3 Hz (or 0.36 +/- 0.05 ppm), respectively.
Fig.
3 summarizes DMI data acquired on human brain following the oral administration
of [6,6’-2H2]-glucose. At a nominal spatial resolution of
8 mL (20 x 20 x 20 mm) the DMI data at 4T and 7T appear comparable, although
the spectral resolution at 7T is clearly higher than at 4T. At a 1 mL resolution
the 4T DMI data become unreliable (Fig. 3E/F), whereas the 2.3x SNR increase at
7T leads to a recognizable 2H spectral signature (Fig. 3G/H).Conclusions
The
sensitivity of deuterium metabolic imaging (DMI) has a strong dependence on the
magnetic field strength with near-quadratic (n ~ 1.7 - 1.9) and supralinear (n ~
1.5) dependencies for RF coils suitable for animal and human studies,
respectively. The near-quadratic field dependence and limited coil loading indicate
that RF coil noise dominates the overall noise level for small coils. For
human-sized RF coils the sample noise becomes more important, thereby pushing the
magnetic field dependence into the supralinear regime. In addition to the
increased sensitivity, higher magnetic fields greatly benefit from the
increased spectral dispersion and resolution due to the favorable T2
and T2* relaxation times. With optimized, high-density phased array
receivers, the acquisition of whole-brain DMI at a nominal spatial resolution
of 1 mL is a realistic goal at 7T.Acknowledgements
This
research was supported by NIH grant R01- EB014861 and the James S. McDonnell Foundation.References
[1] H. M. De Feyter, K. L. Behar, Z. A.
Corbin, R. K. Fulbright, P. B. Brown, S. McIntyre, T. W. Nixon, D. L. Rothman,
R. A. de Graaf, Deuterium metabolic imaging (DMI) for MRI-based 3D mapping of
metabolism in vivo, Sci. Adv. 2018; 4 : eaat7314