Physiological Correlates and Predictors of Functional Recovery After Chiasmal Decompression

Abstract

Background The intrinsic abilities and limits of the nervous system to repair itself after damage may be assessed using a model of optic chiasmal compression, prior to and following a corrective surgical procedure. Methods Visual fields (VFs), multi-focal visual evoked potentials (mfVEP), retinal nerve fiber layer (RNFL) thickness and diffusion tensor imaging (DTI) were used to evaluate a patient before and after removal of a meningioma compressing the chiasm. Normally-sighted individuals served as controls. The advantage of each modality to document visual function and predict post-operative outcome (two years follow up) was evaluated. Results Post-surgery visual recovery was best explained by critical mass of normally-conducting fibers, and not associated with average conduction amplitudes. Recovered VF was observed in quadrants in which more than 50% of fibers were identified, characterized by intact mfVEP latencies but severely reduced amplitudes. Recovery was evident despite additional reduction of RNFL thickness and abnormal optic tract diffusivity. The critical mass of normally conducting fibers was also the best prognostic indicator for functional outcome two years later. Conclusion Our results highlight the ability of the remaining normally conductive axons to predict visual recovery following decompression of the optic chiasm. The redundancy in anterior visual pathways may be explained, neuro-anatomically, by overlapping receptive fields.

Comparison of multiple tractography methods for reconstruction of the retinogeniculate visual pathway using diffusion MRI

“…Table 1 gives a summary of existing tractography-based RGVP studies to describe the current state of the art in terms of the fiber tracking methods being used, among which the standard diffusion tensor imaging (DTI) tractography and more advanced methods such as those using constrained spherical deconvolution (CSD) and generalized Q-sampling imaging (GQI) are most widely used. While these studies have shown highly promising T A B L E 1 Summary of published tractography studies of the RGVP and its subregions, organized according to the fiber reconstruction methods employed and the anatomical region studied OT Backner et al (2018), Frezzotti, Giorgio, Toto, De Leucio, andDe Stefano (2016), Glick-Shames, Backner, Bick, Raz, and Levin (2019), Hoffmann et al (2012), Levin, Dumoulin, Winawer, Dougherty, and Wandell (2010), Malania, Konrad, Jägle, Werner, and Greenlee (2017), Maleki, Becerra, Upadhyay, Burstein, and Borsook (2012), Ogawa et al (2014, Raz, Bick, Klistorner, et al (2015), Raz and Levin (2014) RGVP tracking performance, most have focused on certain RGVP subregions of interest, for example, the optic nerve, the optic tract, and/or the RGVP fibers within the optic chiasm (as summarized in et al, 2020a;Zhang, Xie, et al, 2020). A comprehensive quantitative and visual comparison is performed using dMRI data from the Human Connectome Project (HCP; Glasser et al, 2013;Van Essen et al, 2013).…”

Comparison of multiple tractography methods for reconstruction of the retinogeniculate visual pathway using diffusion MRI

“…In our study, we used the single-shell b=1000 s/mm 2 data, consisting of 90 DW images and 18 baseline images, to perform fiber tracking using each compared tractography method (see Section 2.2 for details). We chose single-shell b=1000 s/mm 2 data because it is similar to clinical acquisition protocols as applied in many previous RGVP-related studies (Altintaş et al, 2017; Attyé et al, 2018; Backner et al, 2018; Burton et al, 2018; de Blank et al, 2013; Frezzotti et al, 2016; Glick-Shames et al, 2019; Hofer et al, 2010; Hofstetter et al, 2019; Jacquesson, Frindel, et al, 2019; Jin et al, 2019; Lober et al, 2012; Malania et al, 2017; Maleki et al, 2012; Ogawa et al, 2014; Noa Raz et al, 2015; Noa Raz & Levin, 2014; Schmidt et al, 2018; Wu et al, 2019). Furthermore, single-shell b=1000 s/mm 2 data has been shown in our previous study to be more effective for identification of cranial nerves than higher b values (Xie et al, 2020a).…”

Using Deep Learning for the Classification of Images Generated by Multifocal Visual Evoked Potential

“…This technique dismisses the subjectivity of patients; thus, it is beginning to show promise in evaluating patients with compressive neuropathy ( 2 ). Several studies have shown that mfVEP can predict visual outcome in these patients ( 3 , 4 ).…”