Elsevier

NeuroImage

Volume 49, Issue 1, 1 January 2010, Pages 488-497
NeuroImage

Quantitative imaging of spontaneous neuromagnetic activity for assessing cerebral ischemia using sLORETA-qm

https://doi.org/10.1016/j.neuroimage.2009.07.039Get rights and content

Abstract

To image cerebral neural activity in ischemic areas, we proposed a novel technique to analyze spontaneous neuromagnetic fields based on standardized low-resolution brain electromagnetic tomography modified for a quantifiable method (sLORETA-qm). Using a 160-channel whole-head-type magnetoencephalographic system, cerebral magnetic fields were obtained pre- and postoperatively from 5 patients with unilateral internal carotid artery occlusive disease and 16 age-matched healthy volunteers. For quantitative imaging, voxel-based time-averaged intensities of slow waves in 4 frequency bands (0.3–2 Hz, 2–4 Hz, 4–6 Hz and 6–8 Hz) were obtained by the proposed technique based on sLORETA-qm. Positron emission tomography with 15O gas inhalation (15O-PET) was also performed in these patients to evaluate cerebral blood flow and metabolism. In all 5 patients, slow waves in every frequency band were distributed in the area of cerebrovascular insufficiency, as confirmed by 15O-PET preoperatively. In 4 patients, slow-wave intensities in theta bands (4–6 Hz, 6–8 Hz) decreased postoperatively along with improvements in cerebral blood flow and metabolism, whereas delta bands (0.3–2 Hz, 2–4 Hz) showed no significant differences between pre- and postoperatively. One patient with deterioration of cerebral infarction after surgery showed marked increases in slow-wave intensities in delta bands (0.3–2 Hz, 2–4 Hz) postoperatively, with distribution close to the infarct region. The proposed quantitative imaging of spontaneous neuromagnetic fields enabled clear visualization and alternations of cerebral neural conditions in the ischemic area. This technique may offer a novel, non-invasive method for identifying cerebral ischemia, although further studies in a larger number of patients are warranted.

Introduction

Imaging of the ischemic penumbra (Astrup et al., 1981, Furlan et al., 1996), in which cerebral blood flow is reduced but neurons have not yet become necrotic, is currently a key area of interest in the preoperative investigation of cerebrovascular disease (Heiss, 2000, Srinivasan et al., 2006). Assessments of this area have been based on clinical symptoms (e.g., transient ischemic attack), anatomical findings (from computed tomography (CT) and magnetic resonance imaging (MRI)), and measurement of cerebral blood flow and metabolism (single photon emission computed tomography (SPECT) and positron emission tomography (PET)).

Recently, diffusion-weighted imaging (DWI) and perfusion-weighted imaging (PWI) have been applied for the evaluation of ischemic cerebrovascular disorders, and the area of diffusion–perfusion (DWI–PWI) mismatch is thought to suggest the area of reversible ischemia (ischemic penumbra) (Prosser et al., 2005, Røhl et al., 2001). CT perfusion imaging (CTP) has also been applied for evaluating the ischemic area in the acute stage of stroke (Murphy et al., 2006, Wintermark et al., 2006). Electroencephalography (EEG), a type of electrophysiological examination, is known to be capable of demonstrating slow waves corresponding to areas of cerebral ischemia (Faught, 1993, Faught et al., 1988, Fernandez-Bouzas et al., 2000). However, due to the markedly low spatial resolution caused by the presence of the scalp, skull and cerebrospinal fluid between the brain and electrodes, EEG has had minimal clinical applications in the age of MRI. In contrast, magnetoencephalography (MEG), which is beginning to be used more frequently, enables direct capture of cerebral neural activities and resolves the problems associated with the low spatial resolution of EEG (Higuchi et al., 1998).

Studies on slow-wave distributions as assessed by MEG have been conducted, and increased slow-wave intensity has been confirmed around cerebral infarctions (Baayen et al., 2003, Butz et al., 2004, Kamada et al., 1997, Vieth, 1990). The possibility of slow-wave activity as a reversible reaction to cerebral ischemia has also been reported (Stippich et al., 2000). However, in the past, slow-wave analysis was performed by comparing raw data or estimating equivalent current dipoles (ECDs), which are unsuited to estimating spontaneous slow-wave activities with multiple sources beyond a certain volume of the brain. Images obtained using these methods have therefore not always been clear. Previously, we presented an imaging technique for slow-wave analysis using spatial filtration (Sakamoto et al., 2008). However, to ensure that technique was strictly a qualitative test, not quantitative, absolute values of slow-wave intensities were not determined and precise alternations of neural activity could not be evaluated. If slow-wave activities corresponding to the cerebral ischemic area could be detected not only with high spatial resolution, but also in terms of absolute intensity by MEG, this might provide clues to identifying the ischemic penumbra.

In the present study, we propose a method of spontaneous neuromagnetic analysis based on standardized low-resolution brain electromagnetic tomography modified for a quantifiable method (sLORETA-qm) (Terakawa et al., 2008), and evaluate the clinical usefulness of this method by analyzing slow-wave activity in patients with cerebral ischemic disease.

Section snippets

Subjects

Subjects comprised 5 male patients with a mean age of 71 years (range, 60–70 years). Clinical profile and characteristics of patients are summarized in Table 1. Underlying pathology comprised unilateral internal cerebral artery (ICA) stenosis in 4 patients and unilateral ICA occlusion in 1 patient. Carotid artery stenting (CAS) was performed for 3 patients and superficial temporal artery–middle cerebral artery (STA–MCA) anastomosis for 2 patients for recovery of cerebrovascular insufficiency.

Quantitative frequency analysis of spontaneous cerebral magnetic fields in healthy controls

Detailed results of investigation are shown in Table 2. In every frequency band, slow waves were distributed in the area of frontal and/or parieto-occipital vascular vulnerability, which is thought to be recognized in elderly people (Busse et al., 1956, Silverman et al., 1955). Normal ranges of slow-wave intensities were determined as mean ± standard deviation.

Results of investigation with 15O-PET and MEG in patients

Detailed results of investigation using 15O-PET and MEG are shown in Table 3.

In the PET study, decreased CBF and increased OEF in the

Quantitative imaging of spontaneous neuromagnetic fields using sLORETA-qm

To image neuromagnetic fields arising from the human brain, the characteristics of neuronal activity in various clinical situations must be taken into account. The ECD technique is a conventional source estimation method that has generally been used for the analysis of neuromagnetic activities, but this technique is only reasonable for evaluating neuronal activity recognized in a relatively localized area of cerebrum, as evoked activity or spike activity (Baumagartner et al., 2000, Bundo et

Conclusion

Our method of quantitatively imaging cerebral magnetic fields, based on sLORETA-qm, clearly demonstrated preoperative distributions of spontaneous neuromagnetic activities and postoperative alternations in accordance with cerebrovascular conditions. This imaging technique, depicting the natural neuronal activity of the cerebrum, may offer a novel, non-invasive method for identifying areas of cerebral ischemia.

Acknowledgments

We confirm that we have read the Journal's position on issues involved in ethical publication and affirm that this report is consistent with those guidelines.

We are grateful to the members of Hokuto Hospital for help with the present study. In particular, we would like to thank H. Yamahata, N. Matsuyama, S.Yamauchi, M. Nakasato, A. Hashiguchi and G. Odake of the Department of Neurosurgery, N. Kato, H. Ishihara, D. Sasaki, T. Shimoji and D. Yamamoto of the PET Center and T. Okada, S. Nishio, A.

References (51)

  • AstrupJ. et al.

    Thresholds in cerebral ischemia — the ischemic penumbra

    Stroke

    (1981)
  • BaayenJ.C. et al.

    Localization of slow wave activity in patients with tumor-associated epilepsy

    Brain Topogr.

    (2003)
  • BaumagartnerC. et al.

    Neuromagnetic recordings in temporal lobe epilepsy

    J. Clin. Neurophysiol.

    (2000)
  • BundoM. et al.

    Changes of neural activity correlate with the severity of cortical ischemia in patients with unilateral major cerebral artery occlusion

    Stroke

    (2002)
  • BusseE.W. et al.

    Psychological functioning of aged individuals with normal and abnormal electroencephalograms. I. A study of non-hospitalized community volunteers

    J. Nerv. Ment. Dis.

    (1956)
  • FaughtE.

    Current role of electroencephalography in cerebral ischemia

    Stroke

    (1993)
  • FaughtE. et al.

    Patterns of EEG frequency content during experimental transient ischaemia in subhuman primates

    Neurol. Res.

    (1988)
  • Fernandez-BouzasA. et al.

    Sources of abnormal EEG activity in brain infarctions

    Clin. Electroencephalogr.

    (2000)
  • FrackowiakR.S. et al.

    Quantitative measurement of regional cerebral blood flow and oxygen metabolism in man using 15O and positron emission tomography: theory, procedure, and normal values

    J. Comput. Assist. Tomogr.

    (1980)
  • FurlanM. et al.

    Spontaneous neurological recovery after stroke and the fate of the ischemic penumbra

    Ann. Neurol.

    (1996)
  • GloorP. et al.

    Brain lesions that produce delta waves in the EEG

    Neurology

    (1977)
  • GotoT. et al.

    Usefulness of somatosensory evoked magnetic field dipole measurements by magnetoencephalography for assessing spinal cord function

    J. Neurosurg.

    (2002)
  • HeissW.D.

    Ischemic penumbra: evidence from functional imaging in man

    J. Cereb. Blood Flow Metab.

    (2000)
  • HiguchiM. et al.

    System integration and trade-offs of SQUID system for biomagnetic applications

    Applied Superconductivity

    (1998)
  • HuangM.X. et al.

    A sensor-weighted overlappingsphere head model and exhaustive head model comparison for MEG

    Phys. Med. Biol.

    (1999)
  • Cited by (10)

    • SLORETA-qm for interictal MEG epileptic spike analysis: Comparison of location and quantity with equivalent dipole estimation

      2012, Clinical Neurophysiology
      Citation Excerpt :

      In Fig. 1, adopted voxels are shown in pink. Previous reports have described how to obtain quantitative multiple source estimation using sLORETA-qm (Sakamoto et al., 2010; Terakawa et al., 2008). In short, the source reconstruction was performed by sLORETA at each time point as a first step (Pascual-Marqui, 2002).

    • Effects of daily levels of fatigue and acutely induced fatigue on the visual evoked magnetic response

      2012, Brain Research
      Citation Excerpt :

      The mean magnetic signal in the pre-stimulus time period (from 0 to 500 ms before the start of each visual stimulus) was subtracted in each channel to remove the baseline shift of MEG data. The mean magnetic intensities of VEF's were evaluated using sLORETA-qm (Sakamoto et al., 2010; Terakawa et al., 2008). The mean magnetic intensities were log-transformed (ln) for normalization (Terakawa et al., 2008).

    • Cerebrovascular diseases

      2016, Clinical Applications of Magnetoencephalography
    View all citing articles on Scopus
    View full text