Elsevier

NeuroImage

Volume 28, Issue 2, 1 November 2005, Pages 500-506
NeuroImage

Technical Note
Concurrent CBF and CMRGlc changes during human brain activation by combined fMRI–PET scanning

https://doi.org/10.1016/j.neuroimage.2005.06.040Get rights and content

Abstract

A novel approach for concurrent measurement of regional cerebral blood flow (CBF) and regional cerebral metabolic rate for glucose consumption (CMRGlc) in humans is proposed and validated in normal subjects during visual stimulation. 18F-labeled fluorodeoxyglucose was administered during the measurement of CBF by continuous arterial spin labeled magnetic resonance imaging (MRI). Subsequent positron emission tomographic (PET) scanning demonstrated the distribution of labeled deoxyglucose during the MRI acquisition. An excellent concordance between regional CBF and regional CMRGlc during visual stimulation was found, consistent with previously published PET findings. Although initially validated using a brief, non-quantitative protocol, this approach can provide quantitative CBF and CMRGlc, with a broad range of potential applications in functional physiology and pathophysiology.

Introduction

Functional brain imaging has contributed to a greater understanding of regional brain function at rest, during normal sensorimotor and cognitive function, and in disease states. At present, most functional imaging methods measure changes in regional cerebral blood flow (CBF) or metabolism that are coupled to changes in regional brain function. Positron emission tomography (PET) methods are capable of measuring CBF, cerebral blood volume (CBV), cerebral glucose metabolism (CMRGlc), and cerebral oxygen metabolism (CMRO2). These measurements can be made sequentially, but not concurrently. Magnetic resonance imaging (MRI) methods are capable of measuring CBF, CBV, and a complex interaction between blood flow, blood volume, and oxygen utilization termed blood oxygenation level dependent (BOLD) contrast (Ogawa et al., 1992). CBF is measured in MRI using either dynamic susceptibility (Belliveau et al., 1991) contrast or arterial spin labeling (Detre et al., 1992). Arterial spin labeled perfusion MRI utilizes magnetically labeled arterial blood water as an endogenous flow tracer that is directly analogous to 15O–H2O used in PET measurements of cerebral blood flow. Magnetic resonance spectroscopy (MRS) can also be used to measure regional brain metabolism by detecting the fate of exogenously administered isotopic substrates (Zhu et al., 2001, Morris and Bachelard, 2003, Shulman et al., 2004).

In pathological conditions such as stroke, epilepsy, brain tumors, and degenerative diseases, resting alterations in regional blood flow and metabolism detected by PET and MRI have contributed to clinical diagnosis and management. In particular, regional glucose utilization measured by FDG-PET is now an accepted diagnostic test for brain tumor recurrence (increased utilization), lateralization of temporal lobe epilepsy (ipsilateral interictal hypometabolism), and Alzheimer's disease (temporoparietal hypometabolism). Functional MRI (fMRI) with BOLD contrast is readily detected with relatively high spatial and temporal resolution, and is widely used for detecting changes in regional brain activation in response to sensorimotor and cognitive tasks, but does not provide a robust resting measure. Over the past decade, there has been a marked expansion in the use of functional imaging in basic and clinical neuroscience due to the ease and widespread availability of fMRI, which does not require intravenous access, arterial sampling, radioactive isotopes, or a cyclotron.

While a tight coupling between regional neural activity and changes in blood flow and metabolism has been recognized for over a century (Roy and Sherrington, 1890), the exact mechanism for this coupling remains uncertain. Seminal work in the 1980s using multimodal PET scanning demonstrated changes in blood flow, glucose utilization, and oxygen consumption during functional activation in humans (Fox and Raichle, 1986, Fox et al., 1988). Over the ensuing decades, there has been a continuing effort to gain a better understanding of the physiology of functional activation and the relationship between the measurable parameters in functional neuroimaging and neural activity. Measurements of brain metabolism at rest and during task activation in animal models have demonstrated close correlations between regional brain activity and oxidative metabolism (Smith et al., 2002) and BOLD signal changes (Logothetis et al., 2001), though the cellular compartmentalization of these phenomena are still being investigated (Magistretti and Pellerin, 1999, Kasischke et al., 2004). fMRI measurements of oxygen metabolism require calibrated approaches that are primarily applicable to detecting changes with task activation, and add complexity to investigations in human subjects. Hemodynamic effects are much more readily quantified both at rest and with functional activation using MRI, but their relationship to metabolic substrate utilization is less well characterized, particularly in pathological states.

Here, we report the development and initial validation of a method for concurrent, multimodal measurement of CBF and glucose utilization. In this approach, FDG is administered to subjects during the acquisition of ASL perfusion MRI data. Because FDG is trapped in the brain during the course of its metabolism and has a half-life of 109 min, subsequent PET scanning reflects glucose utilization in the minutes following administration. Initial feasibility is demonstrated using a brief, semi-quantitative protocol during photic stimulation; however, fully quantitative studies are also possible. Task activation used an extremely well-characterized paradigm of alternating checkerboard visual stimulation at 8 Hz. The scanning protocol began with BOLD fMRI using a blocked paradigm with 1-min blocks of visual stimulation alternating with 1 min of fixation. Subsequently, ASL perfusion MRI was measured during 10 min of fixation. Finally, a 10-min ASL perfusion MRI scan was carried out during visual stimulation. At the beginning of this acquisition, the subject was injected with FDG. Once the 10-min scan was completed, subjects were removed from the MRI scanner and moved to the PET scanner to measure regional FDG accumulation, which was carried out from 30 to 60 min following the injection. To assess CMRGlc changes with visual stimulation, a resting PET scan was carried out on a separate day using a similar stimulation protocol but outside the MRI scanner. Arterial blood samples were not acquired for this component of the study, but could easily be performed during the fMRI. Arterial activity curves could then be generated to obtain absolute quantification of the cerebral glucose metabolism.

Section snippets

Overview of protocol

All subjects provided informed consent using a protocol approved by the institutional review board, the radiation safety committee, and the oversight committee for MRI studies. On the day of the fMRI study, subjects first had an intravenous catheter inserted and were then brought to the MRI scanner and placed in the magnet. An initial anatomical scan was performed for coregistration between the fMRI and PET scans. The subject then underwent BOLD imaging while having their eyes open looking at a

Results

Five subjects completed the scanning protocol. Group results and individual activation within a region of interest (ROI) in primary visual cortex were compared for each modality following coregistration and normalization of global intensities. Fig. 1 shows a histogram plot of normalized values for a representative single brain slice at Z = 0. The figure shows that there is a strong correlation between normalized CBF and CMRGlc values and that they are similarly distributed (although the CBF

Discussion

These results demonstrate the feasibility of obtaining concurrent CBF and CMRGlc using a combined fMRI–PET approach. The close agreement in the magnitude and spatial localization of CBF and CMRGlc in visual cortex during visual stimulation measured using this approach is in agreement with previously published PET results (Fox and Raichle, 1986). Differences in the magnitude of CBF and CMRGlc changes between the two studies are likely attributable to differences in the regions over which these

Acknowledgments

This research was supported in part by the following grants: NIH-NS045839, NIH-RR02305, and a grant from the Counter Drug Technology Assessment Center. We would also like to thank the MRI and PET Center personnel for their assistance with this study.

References (25)

  • O.H. Grohn et al.

    Assessment of brain tissue viability in acute ischemic stroke by BOLD MRI

    NMR Biomed.

    (2001)
  • E.J. Hoffman et al.

    Assessment of accuracy of PET utilizing a 3-D phantom to simulate the activity distribution of [18F]fluorodeoxyglucose uptake in the human brain

    J. Cereb. Blood Flow Metab.

    (1991)
  • Cited by (49)

    • Biophysically based method to deconvolve spatiotemporal neurovascular signals from fMRI data

      2018, Journal of Neuroscience Methods
      Citation Excerpt :

      To interpret and exploit the BOLD signal, uncovering the dynamics of its underlying physiological processes, i.e., neural activity, astrocytic dynamics, cerebral blood flow (CBF), cerebral blood volume (CBV), and deoxygenated hemoglobin (dHb) concentration, is crucial. The conventional way to study these processes is through other neuroimaging modalities such as electroencephalography (EEG), positron emission tomography (PET), invasive optical imaging, arterial spin labeling (ASL), vascular space occupancy (VASO), near infrared spectroscopy (NIRS), or diffuse optical tomography (DOT) (Feng et al., 2004; Newberg et al., 2005; Hillman et al., 2007; Zhang et al., 2007; Talagala et al., 2004; Boas et al., 2001). These modalities have a variety of spatial and temporal resolutions, each of which reveals particular structural and functional features of the brain.

    • Vascular-metabolic and GABAergic Inhibitory Correlates of Neural Variability Modulation. A Combined fMRI and PET Study

      2018, Neuroscience
      Citation Excerpt :

      BOLD fMRI was used to quantify TV while [18F]-fluoro-deoxyglucose PET (FDG-PET) was used as a measure of regional glucose, and as such energy, consumption (rMRGlu). Regional cerebral blood flow (rCBF), as measured with arterial spin labeling (ASL), was used as an additional proxy measure of energy consumption due to the known coupling between rCBF and glucose uptake during both rest and task states (Cha et al., 2013; Galazzo et al., 2016; Newberg et al., 2005). This additional measure was used to circumvent issues relating to scanning participants in two different states with PET.

    • Qualitative agreement and diagnostic performance of arterial spin labelling MRI and FDG PET-CT in suspected early-stage dementia: Comparison of arterial spin labelling MRI and FDG PET-CT in suspected dementia

      2017, Clinical Imaging
      Citation Excerpt :

      There are relatively few studies that directly compare FDG PET(-CT) and ASL-MRI. 2 in healthy subjects [26,27] and one in neurologically asymptomatic oncological patients [28] showed overall good correlation but with substantial regional variability [27]. Scarce, mostly (semi)quantitative and less visual comparative studies in dementia showed also a fairly good correlation [29–34].

    • Dynamic functional imaging of brain glucose utilization using fPET-FDG

      2014, NeuroImage
      Citation Excerpt :

      The mean percent increase in glucose utilization derived from fPET-FDG for our three subjects was 25% for the full-field checkerboard, 26% for the left hemi-field checkerboard and 28% for the right hemi-field checkerboard (Fig. 3). The absolute changes in FDG utilization as measured by fPET-FDG are consistent with previous studies measuring single response in a two-scan paradigm (Newberg et al., 2005). fPET-FDG provides a simple method to observe brain glucose utilization changes dynamically in a single imaging session.

    View all citing articles on Scopus
    View full text