Relating excitatory and inhibitory neurochemicals to visual perception: A magnetic resonance study of occipital cortex between migraine events

Abstract

Certain perceptual measures have been proposed as indirect assays of brain neurochemical status in people with migraine. One such measure is binocular rivalry, however, previous studies have not measured rivalry characteristics and brain neurochemistry together in people with migraine. This study compared spectroscopy-measured levels of GABA and Glx (glutamine and glutamate complex) in visual cortex between 16 people with migraine and 16 non-headache controls, and assessed whether the concentration of these neurochemicals explains, at least partially, inter-individual variability in binocular rivalry perceptual measures. Mean Glx level was significantly reduced in migraineurs relative to controls, whereas mean occipital GABA levels were similar between groups. Neither GABA levels, nor Glx levels correlated with rivalry percept duration. Our results thus suggest that the previously suggested relationship between rivalry percept duration and GABAergic inhibitory neurotransmitter concentration in visual cortex is not strong enough to enable rivalry percept duration to be reliably assumed to be a surrogate for GABA concentration, at least in the context of healthy individuals and those that experience migraine.

Fig 1. Example magnetic resonance spectroscopy spectrum and stimulus sequence for the binocular rivalry task.

(a) An example spectrum measured from the visual cortex of a migraine participant. Frequency and phase correction are performed on the pre-aligned raw spectrum (red) using Gannet to generate a post-align spectrum (blue). Gannet then generates best fit curves of the (b) Glx (c) GABA peaks on the post-align spectrum (blue). Glx and GABA concentrations are defined as the area under the fitted curves (red). Inset figure illustrates the voxel placement at the occipital cortex. (d) Stimulus sequence used to assess binocular rivalry. A green grating oriented at 135° was shown to the left eye while a red grating oriented at 45° was shown to the right eye throughout each 90s test block. Observers perceived switches between the red and green grating within the 90s test blocks.

Fig 2. Group averaged GABA and Glx levels.

Individual measures (small symbols, jittered along the x-axis) and group averages (large symbols; error bars represent 95% confidence intervals) of occipital (a) GABA+/water, (b) Edited-Glx/water, and (c) GABA/Glx ratio, obtained from the visual cortex of controls (filled) and migraineurs (unfilled) Migraine with aura are represented by squares, and migraine without aura by triangles.

Fig 3. Group averaged rivalry percept duration.

Individual measures (small symbols, jittered along the x-axis) and group averages (large symbols; error bars represent 95% confidence intervals) of median percept duration (a) in controls (filled) and migraineurs (unfilled). Migraine with aura are represented by squares, and migraine without aura by triangles.

Fig 4. Relationship between median percept duration for the binocular rivalry task and the GABA+/water in visual cortex.

Data is shown for each group separately but pooled to determine the correlation. Migraine with aura are represented by squares, and migraine without aura by triangles. Controls are filled circles.