A multimodal approach towards macular function assessment

Content of research project: The aim of this project is to clarify the relation between characteristics of retinal diseases, as determined by conventional diagnostic methods used in ophthalmology, and impairment of visual perception, which is mediated by specific areas in the grey matter of the brain (the visual cortex). Mapping of the function of the visual cortex based on functional magnetic resonance imaging (fMRI) is an ideal method for obtaining detailed information about the functional relation between activity of localized retinal areas and the corresponding regions of the visual cortex (retinotopic information). In previous studies, we demonstrated the ability of current fMRI methodology to assess the size of simulated visual field defects in healthy individuals and showed that central visual field defects down to a radius of a 2.35 visual angle can be detected reliably. In a subsequent clinical pilot study, we described fMRI-based mapping of the visual cortex in patients with retinal diseases and addressed the relation between fMRI data and results obtained by conventional assessment, including visual field measurements (microperimetry, MP) and imaging of the retinal structure. Central retinal regions with reduced function and structural alterations correlated well with fMRI-based maps showing reduced cortical activity. Hypotheses: Application and validation of this method in a larger patient population including 20 patients with Stargardt disease (STGD), the most common inherited retinal degeneration, 20 patients with dry age-related macular degeneration (AMD) and 20 healthy control patients with simulated scotomata, including assessment of data reproducibility, might prove fMRI to be a valuable adjunct to conventional ophthalmological investigation in retinal diseases. Assessment of patients with acute retinal diseases (20 with branch retinal artery occlusions and 20 with macular holes) over one year compared with patients with chronic, long-standing retinal scotomata (STGD and dry AMD) might explain the possible development of cortical plasticity as an adaptive mechanism of the visual cortex to compensate for retinal disease. Methods: fMRI provides retinotopic information (coverage maps) that is then correlated with the results of conventional ophthalmic testing including MP, visual acuity and contrast sensitivity testing, reading performance, optical coherence tomography and autofluorescence imaging. Explanation indicating what is new and/or special about the project: The present project applies a novel approach for linking retinal function assessed with MP and structural data to fMRI-based maps of cortical function in a representative patient population with acute and chronic retinal diseases. The project seeks to contribute to best practice methods for using fMRI to assess retinal diseases both for documentation of the natural course and during therapy in a study setting.

The aim of this project was to clarify the relation between characteristics of retinal diseases, as determined by conventional diagnostic methods used in ophthalmology, and impairment of visual perception, which is mediated by specific areas in the grey matter of the brain (the visual cortex). Mapping of the function of the visual cortex based on functional magnetic resonance imaging (fMRI) is an ideal method for obtaining detailed information about the functional relation between activity of localized retinal areas and the corresponding regions of the visual cortex (retinotopic information). We conducted this study on the only ultra-high-field fMRI, which has a magnetic field strength of 7 Tesla. Compared to conventional 3 Tesla MRIs, this allows for increased sensitivity and specificity of the imaging data acquired. As patients with retinal diseases often have difficulties to maintain stable fixation, we implemented an innovative eyetracking system that recorded gaze direction in real-time in order to compensate gaze instabilities in fMRI data analysis. In the scope of this project, we investigated the feasibility of fMRI to detect artificial scotoma in healthy subjects and retinal scotoma of patients with Stargards' disease, dry age-related macular degeneration (AMD), vascular occlusions and acute macular holes. Mapping of the visual cortex was compared to conventional morphological (optical coherence tomography) and functional (microperimetry) assessment of the retina. Reduced function and structural alterations of central retinal areas are precisely represented in fMRI based mapping of the visual cortex as reduced activity of topographically corresponding areas. No investigation has so far been conducted regarding the effects of stimulus choice (wedge-ring/bar stimuli) on the results of fMRI-based mapping of the visual cortex. Data obtained from healthy subjects and patients with dry AMD show significant differences between stimuli in the size of receptive fields in the visual cortex and their distribution in the visual field. In addition, the number of neuronal activation centres in the central visual field and the size of receptive fields is smaller when using wedge-ring stimuli compared to bar stimuli. These differences may be exploited by combining different stimuli to enhance mapping of the visual cortex. In order to assess the reliability of fMRI-based mapping of the visual cortex, the test-retest reproducibility was assessed in patients with Stargards' disease and healthy subjects. For this purpose, comparisons of four sessions at weekly intervals were conducted. These showed a robust reproducibility of both intersessional and intrasessional runs. A novel real-world oriented approach for the correlation of conventional assessment of retinal disease with fMRI-based mapping of the visual cortex was established in a representative patient population. This project demonstrated the feasibility of fMRI is a promising method for objectively documenting the natural course and therapeutic response of retinal disease in a clinical study setting.