University of Groningen Neuro-imaging of visual field defects Boucard, Christine IMPORTANT NOTE: You are advised to consult the publisher's version (publisher's PDF) if you wish to cite from it. Please check the document version below. Document Version Publisher's PDF, also known as Version of record Publication date: 2006 Link to publication in University of Groningen/UMCG research database Citation for published version (APA): Boucard, C. (2006). Neuro-imaging of visual field defects. s.n. Copyright Other than for strictly personal use, it is not permitted to download or to forward/distribute the text or part of it without the consent of the author(s) and/or copyright holder(s), unless the work is under an open content license (like Creative Commons). Take-down policy If you believe that this document breaches copyright please contact us providing details, and we will remove access to the work immediately and investigate your claim. Downloaded from the University of Groningen/UMCG research database (Pure): http://www.rug.nl/research/portal. For technical reasons the number of authors shown on this cover page is limited to 10 maximum. Download date: 08-03-2019
Visual Stimulation, 1H-MR Spectroscopy and fmri of the Human Visual Pathways 107 CHAPTER 5 Visual field defects and the metabolic brain Part 2 Visual Stimulation, 1 H-MR Spectroscopy and fmri of the Human Visual Pathways Authors: Christine C. Boucard Jop P. Mostert Frans W. Cornelissen Jacques H.A. de Keyser Matthijs Oudkerk Paul E. Sijens Published in: European Radiology 2005 Jan;15(1):47-52
108 Chapter 5 Abstract The purpose was to assess changes in lactate content and other brain metabolites under visual stimulation in optical chiasm, optic radiations and occipital cortex using multiple voxel MR spectroscopy (MRS). 1 H chemical shift imaging (CSI) examinations of transverse planes centered to include the above structures were performed in four subjects at an echo time of 135 ms. Functional MRI (fmri) was used to confirm the presence of activity in the visual cortex during the visual stimulation. Spectral maps of optical chiasm were of poor quality due to field disturbances caused by nearby large blood vessels and/or eye movements. The optic radiations and the occipital lobe did not show any significant MR spectral change upon visual stimulation, i.e., the peak areas of inositol, choline, creatine, glutamate and N-acetylaspartate were not affected. Reproducible lactate signals were not observed. fmri confirmed the presence of strong activations in stimulated visual cortex. Prolonged visual stimulation did not cause significant changes in MR spectra. Any signal observed near the 1.33 ppm resonance frequency of the lactate methyl-group was artefactual, originating from lipid signals from outside the volume of interest (VOI). Previous claims about changes in lactate levels in the visual cortex upon visual stimulation may have been based on such erroneous observations.
Visual Stimulation, 1H-MR Spectroscopy and fmri of the Human Visual Pathways 109 Introduction According to the astrocyte-neuron lactate shuttle hypothesis, lactate is formed in the astrocyte, subsequently transferred to the mitochondria of the neuron, and serves there as the main fuel for oxidative metabolism (1). In line with this hypothesis, several singlevoxel MR spectroscopy (MRS) studies on the effect of visual stimulation on brain metabolism, reported that lactate signals increased in the occipital part of the brain upon visual stimulation (2-5). Other MRS investigators, however, did either not observe such increased lactate signals at all (6,7) or found only an extremely short-lived increase followed by a decline (8). In all published examples, the CH3-lactate signals detected near 1.33 ppm appear to be small. In addition, signals may potentially have been contaminated with -(CH2)n- lipid signals, arising from the fatty tissue between brain and skull that is very close to the posterior part of the MRS voxel, that resonate at 1.30 ppm. In the present MRS study, multiple voxel chemical shift imaging (CSI) was used to assess any metabolic changes in the visual pathway during visual stimulation. The visual pathway runs from the retina through the optical chiasm to the lateral geniculate nucleus. From there the optic radiations project to the visual cortex. Multiple voxel CSI of transverse planes centered on optical chiasm, optic radiations and visual cortex allowed for direct comparison of any spectral changes observed in the main structures of the optical pathways with results in brain areas outside theses regions of interest. fmri was used to confirm the presence of activity in visual cortex during similar visual stimulation. Materials and methods MR examinations of four healthy volunteers were performed at the Department of Radiology of the University Hospital Groningen at a field strength of 1.5 T using the standard head coil of a Siemens Magnetom Vision MR scanner (Siemens AG, Erlangen,
110 Chapter 5 Germany). Age of volunteers was 21, 25, 26 and 40 years. Informed consent was obtained after the nature of the procedures had been fully explained. In all subjects MRS was preceded by the acquisition of T2 weighted MRI scans in order to position the volumes of interest on the anatomical structures of interest. Automated hybrid PRESS (point resolved spectroscopy) 2D-CSI measurements with a repetition time (TR) of 1500 ms and an echo time of 135ms (SE, double spin echo) were performed. Hybrid-CSI includes pre-selection of a VOI that is located within the brain to prevent the strong interference from subcutaneous fat and is smaller than the phase-encode field of view (FOV) that must be large enough to prevent wraparound artefacts (9). CSI 16x16 phase encoding of a transverse FOV of 16x16 cm2 was thus combined with VOIs of dimensions allowing for optimal measurement of optical chiasm, optic radiations and of the visual cortex. Automated localized multiple angle projection (MAP) shimming resulted in water peak line widths of less than 8 Hz in the VOI. Excitation with 2.56 ms sinc-hanning shaped RF pulses preceded by 25.6 ms Gaussian shaped RF pulses for chemical shift selective excitation (CHESS) and subsequent spoiling of the resultant water signal, was followed by collection of the second spin echo using 1024 data points and a spectral width of 500 Hz. All 16x16 2D-CSI measurements were 1 acquisition per phase encoded step with 4 prescans and TR's of 1500 ms (acquisition time 7 min). Time domain data were multiplied with a Gaussian function (centre 0 ms, half width 256 ms), 2D-Fourier transformed, phase and baseline corrected and quantified by means of frequency domain curve fitting with the assumption of Gaussian line shapes, using the standard "Numaris-3" software package provided with the MR system. Sixth order polynomial lines with a 0-4.3 ppm calculation range were used for baseline correction. In the curve fitting the number of peaks fitted included the chemical shift ranges restricted to 3.4-3.6 ppm for inositol 3.1-3.3 ppm for choline (Cho), 2.9-3.1 for creatine (Cr), 2.2-2.4 for glutamate (Glu), 1.9-2.1 for N-acetyl aspartate (NAA), and 1.2-1.5 ppm for lactate (Lac), and their line widths and peak intensities unrestricted. Using standard postprocessing protocols the raw data were thus processed automatically, allowing for operator-independent quantifications. Metabolite concentrations were compared between a visual stimulation condition and a base line condition during which subjects had their eyes closed. During visual stimulation subjects viewed an 8 Hz flickering high-
Visual Stimulation, 1H-MR Spectroscopy and fmri of the Human Visual Pathways 111 contrast dartboard pattern (pattern size about 15 deg diameter, 0.5 deg checks, central fixation cross present). Such patterns are known to strongly activate visual cortex (10). Stimulation and control blocks lasted typically 14 minutes and were repeated twice for each of the four subjects. fmri experiment: One subject (F.W.C.) was tested immediately after the MRS experiment. During the block design fmri experiment, a baseline condition (blank screen with only a central fixation cross present) was alternated with a visual stimulation condition (flickering dartboard pattern identical to the one reported above). Blocks lasted 30 s. and the sequence was presented 12 times. fmri data were acquired using a T2* weighted gradient recalled echo planar imaging sequence. Technical data for the measurements were: TE 60 ms, TR 3080 ms, flip angle 90, 26 slices in one volume, matrix 64x64. Field of view 240 mm. Voxel size:3.75x3.75x3 mm. fmri data analysis was performed with SPM99 software (SPM99; Wellcome Department Imaging Neuroscience, London, UK. http://www.fil.ion.ucl.ac.uk/spm/spm99.html). The EPI functional volumes were matched to the first volume to eliminate movement artefacts (SPM: realignment) and spatially smoothed (SPM: smooth, gaussian kernel, FWHM of 5 mm). An fmri block design analysis procedure (t-test, p>0.05, corrected for multiple comparisons) was used to compare stimulation with baseline activation in order to assess brain activity induced by our stimulation. Results Optical chiasm and visual cortex. The spectral map of a transverse slice with a FOV of 6x11x1 cm3 that was angulated to include optical chiasm as well as occipital brain tissue is shown in Fig.1. Three baseline CSI measurements were performed as well as two measurements acquired during optical stimulation. The two anterior rows in which the optical chiasm is located, show poor spectra due to field interference caused by nearby large blood vessels and/or eye movement. The quality of the rest of the spectra is negatively affected by the selection of a FOV that did not allow for an entirely
112 Chapter 5 successful shimming procedure. Thus, upon optical stimulation no significant change in any metabolite level could be observed inside or outside the occipital lobe. Figure 1. The CSI (TE135/TR1500) spectral map of a transverse slice with a FOV of 6x11x1 cm3 that was angulated to include optical chiasm as well as occipital brain tissue (visual cortex). Optic radiations and visual cortex. The next step was to examine a 9x7x2 cm3 VOI including the optic radiations and the occipital lobe. Spectral quality is much better (Fig.2). Again the protocol included three baseline and two stimulation measurements. Over the total number of quantified voxels Cho, Cr and NAA peak areas showed mean percent standard deviations of 11.3, 8.2 and 4.4 respectively (baseline). Table 1 shows the peak areas of these metabolites for the optic radiation and occipital voxels included
Visual Stimulation, 1H-MR Spectroscopy and fmri of the Human Visual Pathways 113 in the spectral map. The changes in these compounds and those in inositol, Glu and lactate were not significant. (The apparent lack of inositol in the stimulated spectrum of Fig.2 rather reflects a signal-to noise ratio inadequate for accurate detection of this particular compound than any change as a result of visual stimulation). Figure.2. CSI (TE135/TR1500) spectral map a 9x7x2 cm3 VOI including the optic radiations and the occipital. The spectra shown are from the same optic radiation voxel with (middle right) and without optic stimulation (lower right). Single voxel occipital: visual cortex. In the above examinations, lactate was not significantly present in any voxel, irrespective of whether it was part of the visual pathways or not. We therefore included a single voxel TE135/TR1500 MRS examination
114 Chapter 5 of a 3x3x3 cm3 volume measured six times without and ten times with visual stimulation. Measurements of 128 acquisitions resulted in a time resolution of 3:12 min. The spectra shown have a (slightly) better signal-to-noise ratio and a lower resolution between peaks as expected (Fig.3). However, during both presence and absence of visual stimulation no significant levels of lactate were present. The peaks labelled Lac in Fig. 3 appear at the wrong frequency, and represent noise rather than lactate. The Cho, Cr and NAA peak areas showed mean percent standard deviations of 7.2, 5.7 and 4.5 respectively (baseline) and upon stimulation Cho, Cr and NAA showed mean percent changes of -6.4, 0.3 and 2.0 (not significant). Glutamate (Glu) and inositol signals were not observed. Figure 3. Single voxel (TE135/TR1500) spectra of 3x3x3 cm3 volume with (upper right) and without optical stimulation (lower right).
Visual Stimulation, 1H-MR Spectroscopy and fmri of the Human Visual Pathways 115 CSI occipital: visual cortex. The final experiment was a CSI with a VOI narrower than in the third study (5x7x2cm3) in order to cover as much as possible area of the occipital lobe. The protocol included four baseline and four visual stimulation MRS measurements. Over the total number of quantified voxels, Cho, Cr and NAA peak areas showed mean percent standard deviations of 10.0, 10.6 and 4.3 respectively (baseline). Upon stimulation Cho, Cr and NAA showed mean percent changes of 7.3, 5.7 and 0.4 (not significant) and these observations did not differ significantly between occipital lobe (Table 1, last line) and other areas. Glu and inositol were not observed (peak areas equal to zero). In the most posterior row of spectra, a signal that could have a lactate contribution was seen twice, once in a baseline and once in a stimulation CSI (Fig.5). Figure 4. CSI (TE135/TR1500) spectral map a 5x7x2 cm3 VOI including the occipital lobe. The spectra shown are from the same occipital voxel with (middle left) and without optic stimulation (middle right).
116 Chapter 5 Figure 5. Posterior parts of CSI (TE135/TR1500) spectral maps from the same examination as Fig.4. In two of eight maps, once with (upper right) and once without stimulation (lower right) one voxel shows artifact signal from the adipose tissue between brain and skull that one could easily mistake for lactate signal The fourth MRS experiment was immediately followed by fmri. Strong activations were observed throughout the occipital lobe upon presentation of our stimulation patterns (Fig.6). The activity in the visual cortex covered both occipital lobes, and extended from the occipital poles (were the central visual field representations are known to be situated) into the interhemispheric sulcus (peripheral visual field representation). The maxima of activation thus corresponded with the MRS ROI used for this subject.
Visual Stimulation, 1H-MR Spectroscopy and fmri of the Human Visual Pathways 117 Figure 6. fmri pattern of activity in visual cortex obtained applying the comparison stimulation > baseline in SPM (p>0.05 corrected). The fmri stimulation consisted of 12 series of 30 sec. of central fixation cross, followed by 30 sec. of an 8 Hz flickering dartboard pattern. Stimulated / Control Cho Cr NAA Examination Fig.2: Optic rad. (2 voxels) 109 ± 11 98 ± 6 102 ± 8 Occipital (2 voxels) 92 ± 12 108 ± 6 95 ± 4 Examination Fig.3: Occipital (1 voxel) 94 ± 10 100 ± 8 102 ± 5 Examination Fig.4: Occipital (4 voxels) 108 ± 15 101 ± 12 99 ± 5 Table 1. Metabolite peak areas in visually stimulated brain relative to the corresponding areas before stimulation (% with SD)
118 Chapter 5 Discussion In this study, we assessed changes in lactate content and other brain metabolites under visual stimulation in optical chiasm, optic radiations and occipital cortex. Our results indicate an absence of any significant changes in any of the studied metabolites as a result of prolonged visual stimulation. We conclude that visual stimulation that does result in strong fmri activation does not cause significant changes in MR spectra, at least for visual stimulations lasting up to 14 min (and a time resolution of 7 min in our CSI experiments and 3:12 min in our single voxel study). Our observations are thus in agreement with those of others who did not observe lactate increases (6,7) or only very early and brief changes in lactate level (increases reversed after 12 sec after the onset of visual stimulation) (8). Our failure to detect true lactate signals does not reflect inadequate sensitivity of the MRI equipment used in this study; to the contrary, in terms of signal-to-noise ratio and resolution the spectra shown here (Fig.2-5) are not inferior to those published by others. When one considers the entire spectral map with inclusion of the signals from outside the VOI, it is easily visualized that the lactate represents outof-phase lipid signals originating from the fatty tissue between brain and skull (fig.5). We suggest that claims about increased lactate levels made in several publications (2-5) may have been based on such artefacts. With the use of higher field MRS equipment (3T or higher) it might still be possible to achieve a sensitivity for MRS to visual stimulation that approaches that of fmri. Our conclusion is that in the visual pathways running from the retina through the optical chiasm and the lateral geniculate nucleus to the visual cortex, the lactate level remains very low (<0.5 mm level, the detection limit at 1.5T MRS), even after checkerboard stimulation.
Visual Stimulation, 1H-MR Spectroscopy and fmri of the Human Visual Pathways 119 References 1. Magistretti PJ, Pellerin L, Rothman DL, Shulman RG (1999) Energy on demand. Science 283:496-7. 2. Prichard J, Rothman D, Novotny E, Petroff O, Kuwabara T, Avison M, Howseman A, Hanstock C, Shulman R (1991), Lactate rise detected by 1H NMR in human visual cortex during physiologic stimulation. Proc Natl Acad Sci USA 88:5829-5831. 3. Sappey-Marinier D, Calabrese G, Fein G, Hugg JW, Biggins C, Weiner MW (1992) Effect of photic stimulation on human visual cortex lactate and phosphates using 1H and 31P magnetic resonance spectroscopy. J Cereb Blood Flow Metab 12:584-592. 4. Kuwabara T, Wanatabe H, Tanaka K, Tsuji S, Ohkubo M, Ito T, Sakai K, Yuasa T (1994), Mitochondrial encephalopathy: elevated visual cortex lactate unresponsive to photic stimulation a localized 1H MRS study. Neurology 44:557-559. 5. Frahm J, Krüger G, Merboldt KD, Hänicke W, Kleinschmidt A (1996), Dynamic uncoupling and recoupling of perfusion and oxidative metabolism during focal brain activation. Magn Reson Med 35:143-148. 6. Merboldt K-D, Bruhn H, Haenicke W, Michaelis T, Frahm J (1992). Decrease of glucose in the human visual cortex during photic stimulation. Magn Reson Med 25:187-194. 7. Etta A, Fischer-Klein C, Chemelli A, Daxer A, Felber S. Nuclear magnetic resonance spectroscopy (1994) Principles and applications in neuro-opthalmology. Int Opthalmology 18:171-181. 8. Mangia S, Garreffa G, Bianciardi M, Giove F, Di Salle F, Maraviglia B (2003), The aerobic brain: lactate decrease at the onset of neural activity. Neuroscience 118:7-10. 9. Sijens PE, van den Bent MJ, Nowak PJCM, van Dijk P, Oudkerk M (1997), 1H Chemical shift imaging reveals loss of brain tumor choline signal after administration of Gd-contrast agent. Magn Reson Med 37:222-225. 10. Wandell BA (1999), Computational neuroimaging of human visual cortex. Annu Rev Neurosci 22:145-173
120 Chapter 6