Introduction

The visual input from the retina reaches primary visual cortex at the level of the highly myelinated stria of Gennari¹. Myelin content and the stria of Gennari represent a marker of input activity from the retina to the primary visual cortex in the calcarne fissure. Many conditions, such as brain stroke or tumors along the visual pathways, optic neuropathy, retinitis pigmentosa or macular degeneration may cause visual field defects or scotomas. In all of these conditions visual input does not reach primary visual cortex. In this study we hypothesize that myelin content in primary visual cortex co-varies with loss of visual input. We map myelin in participants with hemianopsia, a visual field defect consisting of an absolute visual scotoma limited to a single hemifield, and evaluate whether the clinical symptoms are reflected in cortical myelination in the occipital cortex, providing a proof of concept for an anatomical measure of visual field sensitivity. We introduce a novel method to assess visual field integrity, which has the potential to measure visual field integrity objectively, without the participant performing any task or even when the participant is non-collaborative. Using 7T MRI it is possible to visualize myelin distribution at a sub-millimeter resolution, and consequently visualize the stria of Gennari in living human participants^2,3,4,5. Our method builds upon modern neuroimaging techniques using a modified T1-w MPRAGE to enhance myelin visualization within gray matter, allowing to investigate the structural organization of human cerebral cortex at an unprecedent level of detail and visualizing myelination patterns over the cortical surface.

Method

T1-w images were acquired with a 3D-MPRAGE sequence that allowed enhanced visualization of intra-cortical myelin², with the following parameters: TD/TI: 6000/1200ms, adiabatic inversion, TR/TE: 8/3ms, flip angle: 8 degrees, voxel size = 500x500x500 μm³, FOV: 250x250x180 mm³, 360 sagittal slices, bandwidth 201Hz/px, number of excitations per inversion: 300, linear readout, acceleration using SENSE: 2.5 (anterior-posterior) x 2.5(right-left). Scan duration was 7.5 min (4 to 5 scans were acquired for each participant). A proton density (PD) scan was also acquired and used to correct T1-w images for B1 inhomogeneities; PD scan parameters: 3D turbo-field echo, TR/TE: 6/3 ms, FOV: 250x250x180 mm, voxel size: 1x1x1 mm³, flip angle 1 degree, 180 sagittal slices, BW: 253 Hz/px, SENSE: 1.8(anterior-posterior) x 1.8(right-left), scan duration 48s.control participants: 8, hemianopsia patients: 8. Resulting T1-w volumes were motion corrected between scans, then averaged, and the gray matter was automatically segmented using in-house developed software (Figures 1&2). The primary visual cortex was manually selected from the hemisphere deprived from visual input and the healthy hemisphere from each patient, the same procedure was adopted for control participants. Intensity profiles were extracted sampling from white matter to gray matter surface using a volumetric layering approach, as described in the literature (Figures 1, 2)^3,4. Obtained profiles were tested against a bank of artificial profiles consisting of a linear trend superimposed (added) to a gaussian centered at 0.4, 0.5, 0.6 and 0.7 units of cortical depth. The similarity between each profile and the bank of profiles was assessed computing the R² from least squares (Figure 4). Each profile was assigned to the highest R² scoring bank profile.

Results

As expected, anatomical laminar profiles show a monotonic decrease in intensity starting from the white matter to the surface of the gray matter (Figure 3). A hyper-intensity in the middle of the cortical thickness could be detected on a significant portion of the laminar profiles derived from the hemisphere deprived from visual input as well as from the healthy hemispheres for patients (Figure 4), indicating the presence of a structure compatible with the stria of Gennari (Figure 5). The same could be observed in controls, for the left and right hemispheres. As shown previously in the literature, the exact location of the Stria of Gennari shifts slightly along the normalized cortical depth, following local curvature (not shown).

Discussion

Overall cortical myelination and the stria of Gennari was identified in healthy controls as we have shown previously². Results from the patient data are suggestive of a similar myelination content at a local and global scale between scotoma and healty portions of the visual cortex, confirming previous results obtained from congenitally blind participants⁵, showing the stria of Gennari in the calcarine fissure.

Conclusion

These results indicate that the stria of Gennari remains unaltered also in case of prolonged absence of visual input from the retina, showing that this specific myelinated band does not deteriorate in cortical locations corresponding to absolute visual scotomas, in hemianoptic patients.

Acknowledgements

No acknowledgement found.

References

1. Gennari, F., 1782. De Peculiari Structura Cerebri. Nonnullisque ejus morbis. Ex Regio Typographeo, Parma. 2. Fracasso, Alessio, et al. "Lines of Baillarger in vivo and ex vivo: Myelin contrast across lamina at 7T MRI and histology." NeuroImage 133 (2016): 163-175. 3. Waehnert, M. D., et al. "Anatomically motivated modeling of cortical laminae." Neuroimage 93 (2014): 210-220. 4. Waehnert, Miriam D., et al. "A subject-specific framework for in vivo myeloarchitectonic analysis using high resolution quantitative MRI." Neuroimage 125 (2016): 94-107. 5. Trampel, Robert, Derek VM Ott, and Robert Turner. "Do the congenitally blind have a stria of Gennari? First intracortical insights in vivo." Cerebral Cortex 21.9 (2011): 2075-2081. 6. ZS Saad, RC Reynolds (2012). SUMA. Neuroimage 62:768-773