GlucoCEST matches 18F-FDG PET on a pulse focus ultrasound induced traumatic brain injury
Tsang-Wei Tu1, Zsofia I. Kovacs1, George Z. Papadakis1, Neekita Jikaria1, William Reid1, Dima Hammoud 1, and Joseph A. Frank1

1Radiology and Imaging Sciences, National Institute of Health, Bethesda, MD, United States

Synopsis

18F-FDG positron emission tomography (PET) is used to non-invasively measure the glucose metabolism in the brain. However PET imaging is also limited on the longitudinal monitoring of glucose due low spatial and anatomical resolution. This study compares the glucoCEST and 18F-FDG PET in detecting the glucose concentration in a new traumatic brain injury model using MRI guided pulsed focus ultrasound. Our data show that the glucoCEST could deliver comparable results with the 18F-FDG PET results in detecting the event of hypo-metabolism in the traumatized brain with greater higher image resolution as compared to PET scans.

PURPOSE

Traumatic brain injury (TBI) results in an instant perturbation to the cerebral glucose metabolism causing cognitive deficits significantly. 18F-FDG positron emission tomography (PET) is the major molecular imaging modality to non-invasively measure the metabolic changes in brain. The innate drawbacks of PET, including low resolution images, radiation dosimetry limitation and increased FDG accumulation, hinder longitudinal usage on the monitoring of glucose levels. Gluocse chemical exchange saturation transfer (glucoseCEST) MRI has been shown useful for mapping the glucose concentration non-invasively without the need of radioisotope. In this study, we performed in vivo glucoCEST and 18F-FDG PET on a new traumatic brain injury model generated by MRI guided pulsed focus ultrasound (MRIgpFUS) for monitoring the metabolism in TBI. Our preliminary results indicate that the glucoCEST could be an alternative imaging modality to measure glucose distribution non-invasively in TBI.

Methods

Rat was undergone MRI guided pFUS + Microbubble (MB, Definity®, Lantheus Medical Imaging, Inc., Billerica, MA) in the striatum and the contralateral hippocampus. 0.3 MPa acoustic pressure was applied in 10 ms burst length and 1% duty cycle (9 focal points, 120 sec/9 focal points – striatum) using a single-element spherical pFUS transducer (center frequency: 589.636 kHz; focal number: 0.8; active diameter: 7.5 cm; FUS Instruments, Toronto, Ontario, Canada). During the sonication, T2, T2* and Gd-enhanced T1-weighted images were obtained by 3.0 T MRI (Philips, Amsterdam, Netherlands). T2, T2*, diffusion tensor imaging (DTI) and glucoCEST imaging was performed on 9.4 T MRI (Bruker, Billerica, MA) at baseline and 7 days-post-injury. Parameters for DTI: 3D spin echo EPI; TR/TE 700ms/37ms; b-value 800s/mm2 with 17 encoding directions; voxel size 200 (μm, isotropic). Diffusion weighted images were corrected for B0 susceptibility induced EPI distortion, eddy current distortions, and motion distortion with b-matrix reorientation using Tortoise.1 CEST data were corrected for B0/B1 field inhomogeneity and WASSR water referencing.2 Parameter for glucoCEST: 2D fast spin echo with (MT) and without (M0) magnetization transfer (MT) pulses (TR/TE 3.5s/11.5ms; in plane resolution: 200µm, thickness: 0.8mm; MT pulse: 1.5μT, 1s). The MT offset frequences (Δω) were set from -2kHz to +2kHz with 100Hz stepping to detect the proton metabolites of glucose (1.2ppm, 2.1ppm, 2.9ppm).3 Fractional anistropy (DTI-FA) and the asymmetry of magnetization transfer ratio (MTRasym) were derived for mapping structural injury and glucose metabolism.

Results

After pFUS+MB, Gd-enhanced T1-weighted image detected the opening of blood brain barrier at striatum and the contralateral hippocampus, where the T2, and T2* weighted images showed the hyperintense edema and hypointense voxels consistent with micro-hemorrhage (Fig 1). Compared to the baseline, clear focal TBI pattern was seen on the T2 and T2* weighted images at 7 days following pFUS injury (Fig 2A, B). Decrease of DTI-FA was detected in the external capsule (arrowhead in Fig 2C) demonstrating the pattern of axonal injury. The MTRasym maps of glucoCEST showed loss of contrast (yellow in Fig 2D) at pFUS targets in the straitum and contralateral hippocampus. The similar pattern of contrast loss was also shown in the 18F-FDG PET imaging indicating the decease of glucose uptake following pFUS TBI (Fig 2E).

Discussion

18F-FDG PET is currently the only non-invasively imaging to measure glucose uptake in brain. Our preliminary data showed that glucoCEST detected comparable results showing hypo-metabolism in brain following MRIgpFUS+MB. In contrast to known FDG uptake mechanisms in the brain from 18F-FDG PET, glucoCEST contrast was detects the proton exchange from both the intracellular or extracellular glucose levels within the parenchyma. The current glucoCEST results parallelled with 18F-FDG PET indicating that glucoCEST might also detect hypo-metabolism and decreased glucose uptake in the traumatized brain. Moreover, higher image resolution compared to PET imaging. Further experiments are currently undergoing to investigate the correlation between the changes of glucoCEST and 18F-FDG PET metabolic maps.

Acknowledgements

No acknowledgement found.

References

[1] Irfanoglu, Med Image Comput Comput Assist Interv, 2008; 11:1041-22. [2] Kim, MRM 2009 Jun;61(6):1441-50. [3] Walker-Samuel, Nat.Med, 2013; 19:1067-72.

Figures

Fig. 1 pFUS generated focal TBI. Clear Gd-enhanced contrast is seen in the T1W image after pFUS induced TBI (A). Immediately after pFUS, T2W image shows hyperintense lesion suggesting edema (B). Micro-hemorrhage is also seen in multiple locations at the site of pFUS (C).

Fig. 2 Multivariate MRIs and FDG-PET acquired between baseline and day-7 following pFUS TBI. After pFUS, clear TBI lesions and micro-hemorrhage are seen in T2W image (A) and T2* map (B). DTI-FA shows destruction of external capsule (C). GlucoCEST detects similar pattern of decreased glucose uptake as shown in PET image (E).



Proc. Intl. Soc. Mag. Reson. Med. 24 (2016)
1211