Glaucomas keep being defined as a chronic progressive neuropathy with excavation on the optic nerve disc and subsequent changes in the visual field. This formulation does not, however, correspond to the most recent knowledge and has to be revised. Taking a more up-todate approach, glaucoma is defined as a disease in which progressive loss of ganglion cells of the retina and their axons demonstrates itself by changes in the visual field and atrophy of the optic nerve with excavation. However, even this definition emphasising damage to ganglion cells prior to its axons is not complete as it does not indicate damage to the ganglion cells of the subcortical and cortical centres of the brain.

Nowadays, there are a number of studies proving damage to the corpus geniculatum laterale and the visual cortex both in experimental animals [1-8].

In our paper, we tried to prove damage to the visual cortex in hightension glaucoma patients in different stages of the disease.

Group of patients and methods

Nine patients with different stages of high-tension glaucoma were enrolled in the group (3 females aged 41-65 and 6 males aged 40-73). This group was compared with a group of eight healthy persons (3 females aged 23-46 and 5 males aged 23-65).

Functional MR imaging

All measurements of functional MR imaging (fMRI) were performed using the Philips Achieva TX SERIES with the magnetic field of 3 Tesla. A standard SENSE RF 8-channel head coil was used for the scanning. Optical stimulation for fMRI was carried out using the Eloquence (InVivo) commercial stimulation system.

FMRI was measured by the BOLD method using a gradient-echo EPI sequence with the following parameters: TE=30 ms, TR=3 s, flip angle 90°. The measured volume consisted of 39 continual slices of 2 mm in thickness and the size of the measured voxel (spatial resolution) was 2 x 2 x 2 mm (FOV=208 x 208 mm, matrix 104 x 104, reconstruction matrix 128 x 128, SENSE factor of 1.8). FMRI was performed in all the subjects. An alternation of a black and white checkerboard (Figure 1) was shown to the subjects during the fMRI acquisition. The alternation was in the form of colour inversion with the frequency of 2 Hz. During the resting phase, the subjects were shown a static cross hair placed in the middle of the visual field. Each measurement consisted of a five30-second blocks of active phase periods (10 dynamic scans) and five resting periods of the same length. Altogether, each measurement consisted of 100 dynamic scans and lasted 5 minutes.

Figure 1: illustrates the black and white (BW) checkerboard used for the stimulation. During this stimulation the checkerboard was alternated, with its inversion at the frequency of 2 Hz.

Figure 2: Resulting fMRI images of a selected healthy subject - male aged 65. a) projection of activations in the axial, sagittal and coronal directions, b) orthogonal slices of the selected regions.

FMRI assessment was performed using SPM8 software. During the pre-processing, the data were motion corrected (realigned), corrected for slice timing, smoothed the Gauss filter with FWHM of 6 x 6 x 6 mm and finally normalized into the MNI_152 space. The subject-level statistic was created using a general linear model with the canonical hemodynamic response function (HRF) applied to the stimulation paradigm. The statistical maps were thresholded at the level of p=0.05 with FWE correction. The group statistic was carried out using the unpaired t-test with the uncorrected threshold at p=0.005 and the minimum number of continuous voxels being 50 (8 subjects in each group).

Ophthalmological examination

The complex ophthalmological examination was supplemented by the visual field using the glaucoma program in the rapid threshold program mode (Medmont M700, visual acuity was corrected to short distance). The sum of sensitivities in the homolateral halves of the visual fields (ranging from 0-22 degrees) was compared to the range of contralateral activity of the visual cortex fMRI.

Statistical processing

Possible dependencies between the visual field changes and changes in the brain were assessed using correlation coefficients arranged in a correlation matrix. Since the data normality requirement in some of the parameters was not fulfilled and distant values sometimes occurred in the group, the non-parametric Spearman’s rank correlation coefficient was applied.

The value of the correlation coefficient can be used to distinguish weak (|R|<0.3), median (0.3<|R|<0.8) and strong (|R|>0.8) dependency (correlation).

The following table present the arithmetic average, median and standard deviation of the measured parameters for the control group and the patients (visual field and fMRI).

The resulting correlation coefficient between the right half of the visual field and fMRI activation extent on the left was 0.667 (p< 0.05). The correlation coefficient between the left halves of the visual fields and fMRI activation on the right was 0.767 (p< 0.016).

In two patients with the largest changes in the visual fields, neuron activation occurred in the region of the temporal lobes. This observation is proved by the patient whose fMRI finding is illustrated in (Figure 3). It can be suggested that cortex plasticity which may be superior to anatomical relationships enables even patients with substantial alteration of the visual functions to keep at least the rest of those functions. Examination of the visual field in the same patient is illustrated in (Figure 4).

Figure 3: Example of fMRI in a patient with secondary glaucoma (male aged 58). A significant drop in neuron activity can be observed. a) projection of activations in the axial and coronal directions, b) orthogonal slices in the selected region.

Figure 4: Top - right visual field, bottom - left visual field of the same patient. On the left is sensitivity in the visual field shaded in grey, on the right is the sensitivity drop compared to a normal visual field.
No = an absolute scotoma, negative numerical values illustrate the drop in dB.

The group statistic results can be seen in (Figsures 5 and 6). Compared to the patients, the activation in the controls is more significant only in a narrow region around the medial centre of the occipital gyrus (coordinates in MNI (-6,-100,-3)) and a small region on the lateral surface of the inferior occipital gyrus on the right (39,- 66,-13), and also in the frontal lobe (gyrus frontalis inferior (-53,7,13)), (Figure 5).

Figure 5: illustrates the regions with statistically high activation in the control group in comparison to the patients, with stimulation using a black and white checkerboard (8 subjects in each group. a) projection of activations in the axial, sagittal and coronal directions, b) orthogonal slices of the selected region.

The most significant reaction in patients, when compared to the controls, can be seen bilaterally in the temporal lobes (middle temporal gyrus (-53,-58,22), (-50,-20,-16) and (48,-54,22)), in the right parietal lobe (54,-58,32) and in the medial cingulum and paracingular gyrus (-6,-47,35), (Figure 6).

Figure 6: illustrates regions with statistically higher activation in patients when compared with the control group during the stimulation using the black and white checkerboard.

There are not many studies dealing with functional magnetic resonance for glaucoma [9,10], however, all of them proved changes in the visual cortex region.

The aim of this study with patients with different changes in the visual field was not only to demonstrate the damage to the central nervous system (CNS) associated with glaucoma, but also the extent of functional changes in the visual cortex depending on the changes in the visual field.

The animal model of glaucoma suggests that retinal ganglion cells have an adverse effect on V1 cells owing to transsynaptic degeneration. There are not many reports of this fact occurring in humans.

Transsynaptic degeneration may play a role in the animal glaucoma model as well [11,12]. The death of retinal ganglion cells may quickly trigger a cascade of actions along the retino-cortical pathways with neurochemical [13], metabolic [2,14], functional [15] and neuropathological consequences for the corpus geniculatum laterale and V1, which may result in neuron loss and their final shrinking [16-19].

It must be emphasised that the measurement of neuron activity changes using fMRI does not provide direct evidence of glaucomalinked neurodegeneration in CGL or V1. Functional changes in the neuron activity associated with neuronal disease can be independent of the structural changes of neurons/axons in the CNS and vice versa. For example, the current fMRI studies found that the neuronal activity in the cortical representation of the fovea centralis persists despite a clear macular retinal pathology with a loss of foveal vision [20,21].

Cell death in glaucoma is not an important limit of transsynaptic degeneration associated with the glaucoma eye. There is a decrease in the number and size of M and P cells receiving inputs from the nonglaucoma eye in the primate model [19].

Changes in visual fields in glaucoma result from the loss of ganglion cells and their axons in the whole vision system. Zhang et al. [10] were surprised to find a negative correlation between the changes in visual fields (evaluating PSD) and the fMRI activations. This negative correlation means that if PSD values increase with the visual field changes, the fMRI activation extent decreases. In our group, where was compared the sum of sensitivities in the homolateral halves of visual fields, the value of sensitivities in the visual field as well as the fMRI activation extent decrease with the disease progression, i.e. was established a positive correlation. The correlation coefficient between the right half of the visual field and the fMRI activations on the left was 0.667 (p< 0.05). The correlation coefficient between the left half of the visual field and the fMRI activations on the right was 0.767 (p< 0.016).

On the basis of our results, we believe that the drop in the fMRI activation is not caused by a vascularization disorder in the respective region but is associated with changes in the neurovascular link during oxygen extraction caused by a reduced number of surviving neurons. This so-called transsynaptic degeneration is also advocated by Duncan et al. [9].

The authors proved that the progression of glaucoma disease corresponds with the functional changes in the visual cortex. The loss of ganglion cells of the striate cortex probably results in the interconnection of optical radiation with the functional ganglion cells of the temporal lobe.