3125
Mapping brain oxygen metabolism with dual calibrated fMRI in pre-surgical evaluation of epilepsy: a case report comparison with FDG-PET.1Cardiff University Brain Research Imaging Centre, Cardiff, United Kingdom, 2Wales Research and Diagnostic PET Imaging Centre, Cardiff, United Kingdom, 3Welsh Epilepsy Unit, Department of Neurology, University Hospital of Wales, Cardiff, United Kingdom
Synopsis
Positron emission tomography (PET) with fluorodeoxyglucose (FDG) is a widely used approach to help identify putative epileptogenic areas in patients with epilepsy, as part of the epilepsy surgery evaluation. Epileptogenic areas typically show a regional reduction in glucose metabolism. Here we present a dual-calibrated fMRI method (acquiring BOLD and ASL CBF data simultaneously), which permits mapping of the cerebral metabolic rate of oxygen consumption noninvasively across grey matter. In this case report, we demonstrate close agreement between the two methods (dc-fMRI and PET-FDG) in localising a region of cerebral hypometabolism in epilepsy.
Introduction/Purpose
Quantification of metabolic dysfunction is a potentially important marker in the identification of pathology, diagnosis and treatment across a number of neurological conditions. In epilepsy, patients often undergo imaging investigations to localise the regions of brain that may be the epileptogenic focus. In the work-up for possible neurosurgery FDG-PET may be used to identify areas of reduced glucose metabolic consumption, indexed by standard uptake values (SUV)1. However, PET imaging is costly, invasive, requires the administration of radioactive tracers and may not be suitable for every patient. There has been a recent surge in interest to develop non-invasive MRI methods to detect regional alterations in cerebral oxygen metabolism (cerebral metabolic rate of oxygen consumption, CMRO2). Here, we present a single case report in a patient with epilepsy comparing FDG-PET to our dual-calibrated fMRI (dc-fMRI) method of measuring cerebral oxygen metabolism with the aim of comparing the localisation of areas of altered tissue metabolism.Methods
We evaluated one patient (female, 22years) with temporal lobe epilepsy, who previously had an FDG-PET scan as part of their clinical epilepsy pre-surgical investigation. The patient underwent the FDG-PET scan prior to the MRI session (13 months prior). The patient was instructed to follow the same preparation for the MRI as they did for the PET, including a 6 hour fast prior to the scan. In the current case report we adopted two approaches to estimate cerebral metabolism. First, we used an in-house built dual-excitation, PCASL based, dual-calibration method (dc-fMRI). This method involves simultaneously acquiring BOLD and CBF signals to permit quantitative mapping of oxygen metabolism (CMRO2) across grey matter.2,3 Scan parameters - TE1: 11ms, TR1: 3600ms, TE2: 30ms, TR2: 1100ms, slice thickness: 6mm, and GRAPPA acceleration factor 2, FOV read 220 (64x64 matrix), post label delay 1400ms, tag duration 1500ms. Total scan time for the PCASL sequence was 18 minutes, and included hypercapnic and hyperoxic respiratory challenge.2 Additional imaging included acquiring a phase-contrast flow scan for localizing neck vessels (for the ASL sequence), and a T1-MPRAGE (anatomical) for registration of functional images. Measurements were performed on a clinical 3T scanner (Prisma MAGNETOM, Siemens Healthcare, Erlangen, Germany). Second, we show a static PET-FDG image obtained 30 minutes post injection of 250 MBq of 18F-FDG (GE Discovery 690, GE Healthcare) in the same patient to visually compare to our MRI method. The data represents the SUV of the FDG across the brain. PET scan parameters include – 3.27mm slice thickness, 128x128 matrix size. The total scan duration for PET was 15 minutes (single bed position).Results
The PET-FDG scan results showed widespread unilateral hypometabolism indexed by SUV in the left hemisphere (temporal lobe), compared with the right. This result was in agreement, by visual inspection, with the dual-calibrated fMRI results, which also showed reduced oxygen metabolism in the same region (See Figure 1 for a single slice comparison between the PET-FDG SUV and the dc-fMRI data). We further demonstrate that the degree of observed CMRO2 asymmetry in the patient falls outside the normal range seen in healthy participants. We analysed data from an additional 21 healthy participants (age 34.57±5.87, 15 female) studied with the dc-fMRI protocol revealing a mean asymmetry index (AI) of 1.33 (SD=0.98). For the patient, an asymmetry index of 4.46 was observed, more than 2 standard deviations from the mean of the healthy participants data, suggesting significant asymmetry. Therefore, our results show mapping of CMRO2 with the dc-fMRI method is reasonably symmetrical between hemispheres in a healthy brain, across a group of healthy participants, but not for the patient.
Discussion/conclusions
Our findings reveal close agreement in areas of hypometabolism as measured with both FDG-PET and dc-fMRI. Compared to standard ASL methods, which have shown agreement in with FDG-PET in areas of hypoperfusion4, we suggest our dc-fMRI method for measuring oxygen metabolism may more closely reflect the metabolic alterations observed with PET. Results of the AI analysis highlight the degree of spatial sensitivity of our dc-fMRI method, and that the region of reduced oxygen metabolism observed for our patient in the case report reflects the reduced glucose metabolism revealed by PET. We suggest that the use of recently developed combined PET/MRI systems may provide a more detailed depiction of altered metabolism from relative changes in glucose and oxygen metabolism. While further investigation is needed with an increased sample size, we provide evidence of a proof of concept that MRI measures of oxygen metabolism can be comparable to PET-FDG measures of glucose metabolism.
Acknowledgements
We would like to acknowledge the EPSRC for their support at the early stages of this research. We would also like to thank Wellcome and MRC-CiC for their support in the continuation of this work.References
1Theodore, W. H., Dorwart, R., Holmes, M., Porter, R. J., & DiChiro, G. (1986). Neuroimaging in refractory partial seizures Comparison of PET, CT, and MRI. Neurology, 36(6), 750-750.
2Germuska, M., Merola, A., Murphy, K., Babic, A., Richmond, L., Khot, S., Hall, J.E. and Wise, R.G., 2016. A forward modelling approach for the estimation of oxygen extraction fraction by calibrated fMRI. NeuroImage, 139, pp.313-323.
3Schmithorst, V.J., Hernandez-Garcia, L., Vannest, J., Rajagopal, A., Lee, G., Holland, S.K., 2014. Optimized simultaneous ASL and BOLD functional imaging of the whole brain. J Magn Reson Imaging 39, 1104-1117.
4Galazzo, I. B., Mattoli, M. V., Pizzini, F. B., De Vita, E., Barnes, A., Duncan, J. S., ... & Groves, A. M. (2016). Cerebral metabolism and perfusion in MR-negative individuals with refractory focal epilepsy assessed by simultaneous acquisition of 18F-FDG PET and arterial spin labeling. NeuroImage: Clinical, 11, 648-657.
Figures
