Past, present and future role of retinal imaging in neurodegenerative disease

Abstract

Retinal imaging technology is rapidly advancing and can provide ever-increasing amounts of information about the structure, function and molecular composition of retinal tissue in humans in vivo. Most importantly, this information can be obtained rapidly, non-invasively and in many cases using Food and Drug Administration-approved devices that are commercially available. Technologies such as optical coherence tomography have dramatically changed our understanding of retinal disease and in many cases have significantly improved their clinical management. Since the retina is an extension of the brain and shares a common embryological origin with the central nervous system, there has also been intense interest in leveraging the expanding armamentarium of retinal imaging technology to understand, diagnose and monitor neurological diseases. This is particularly appealing because of the high spatial resolution, relatively low-cost and wide availability of retinal imaging modalities such as fundus photography or OCT compared to brain imaging modalities such as magnetic resonance imaging or positron emission tomography. The purpose of this article is to review and synthesize current research about retinal imaging in neurodegenerative disease by providing examples from the literature and elaborating on limitations, challenges and future directions. We begin by providing a general background of the most relevant retinal imaging modalities to ensure that the reader has a foundation on which to understand the clinical studies that are subsequently discussed. We then review the application and results of retinal imaging methodologies to several prevalent neurodegenerative diseases where extensive work has been done including sporadic late onset Alzheimer's Disease, Parkinson's Disease and Huntington's Disease. We also discuss Autosomal Dominant Alzheimer's Disease and cerebrovascular small vessel disease, where the application of retinal imaging holds promise but data is currently scarce. Although cerebrovascular disease is not generally considered a neurodegenerative process, it is both a confounder and contributor to neurodegenerative disease processes that requires more attention. Finally, we discuss ongoing efforts to overcome the limitations in the field and unmet clinical and scientific needs.

Keywords: Alzheimer's disease; Cerebral small vessel disease; Huntington's disease; Imaging; Parkinson's disease; Retina.

Fig. 1.

Examples of retinal imaging modalities from a 65 year old female illustrate commonly used methods for evaluation of retinal disease and retinal changes in neurodegenerative diseases. (A) Color fundus photograph illustrating the macula, optic disc and retinal arteries and veins. (B) Digital collage of color fundus images of the same subject demonstrating 60-90° field of view that includes the peripheral retina outside the vascular arcades. (C) Optical coherence tomography angiogram of the parafoveal area illustrating the capillaries in the area and the foveal avascular zone. Red and green pseudocoloring represent the depth of retinal capillaries in the superficial and deep retinal layers, respectively. (D) Short wave fundus autofluorescence image of the macula.

Fig. 2.

Ultra-Widefield fundus photograph and optical coherence tomography angiogram from a human subject. (A) Optos™ pseudocolor ultra-widefield fundus photograph illustrates the peripheral retina where traditional color fundus imaging does not typically reach. (B) Widefield optical coherence tomography angiogram (Zeiss PlexElite™) demonstrates non-invasive imaging of retinal arteries, veins and capillaries beyond the arcades in the same subject.

Fig. 3.

Example of retinal layer and choroidal identification on enhanced depth imaging optical coherence tomography (EDI-OCT) compared to histology. (A) OCT image depicts retinal layers relative to the corresponding retinal layers on (B) histology in a representative control micrograph stained with hematoxylin and eosin within the macula. (Reproduced under the Creative Commons Attribution-NonCommercial-NoDerivatives (CC BY-NC-ND) 4.0 http://creativecommons.org/licenses/by-nc-nd/4.0/ International License from Asanad S, Ross-Cisneros FN, Nassisi M, Barron E, Karanjia R, Sadun AA. The retina in Alzheimer’s disease: histomorphometric analysis of an ophthalmologic biomarker. Invest Ophthalmol Vis Sci. 2019; 60:1491–1500. https://doi.org/10.1167/iovs.18-25966).

Fig. 4.

Retinal histopathology of AD and control subjects. Light microscopy depicts (A) supero-temporal RNFL (black arrows) in control and (B) AD postmortem tissue. Qualitative assessment of the supero-temporal RGCL, INL, and ONL (marked by red boxes) in (C) representative control and (D) subject with AD depicts supero-temporal RGCL, INL, and ONL thinning most pronounced in the macular region of the subject with AD. All stains are hematoxylin and eosin. (Reproduced under the Creative Commons Attribution-NonCommereial-NoDerivatives (CC BY-NC-ND) 4.0 International License from Asanad S, Ross-Cisneros FN, Nassisi M, Barron E, Karanjia R, Sadun AA. The retina in Alzheimer’s disease: histomorphometric analysis of an ophthalmologic biomarker. Invest Ophthalmol Vis Sci. 2019; 60:1491-1500. https://doi.org/10.1167/iovs.18-25966).

Fig. 5.

Peripapillary retinal nerve fiber layer thickness is reduced in subjects with preclinical AD. (A) The thicknesses of the retinal nerve fiber layer (temporal, superior, nasal, inferior) was measured using OCT in the regions outlined in black. (B) Depicts least-squares mean (95% CI) total retinal nerve fiber layer thickness adjusted for side and region between cognitively healthy controls (blue) and cognitively healthy participants with pathologic CSF Aβ₄₂/Tau levels (red). (Adapted from Asanad S, Fantini M, Sultan W, Nassisi M, Felix CM, Wu J et al. (2020) Retinal nerve fiber layer thickness predicts CSF amyloid/tau before cognitive decline. PLoS ONE 15(5): e0232785. https://doi.org/10.1371/journal.pone.0232785 under the terms of the Creative Commons Attribution License http://creativecommons.org/licenses/by/4.0/).

Fig. 6.

Color fundus photographs and positron emission computed tomograms (PET) demonstrating clinical correlation of retinal and brain imaging findings in (A,B) 74 year old subject without dementia and (C,D) 69 year old subject without dementia. (A) Color fundus photograph of the macula demonstrates numerous drusen (area 7.35 mm²) and (B) F-AV45 (florbetapir) PET yields positive standard uptake values (SUVR) of 1.15 (values greater than or equal to 1.10 are considered positive). (C,D) Similar images for a 69 year old female subject without dementia. (C) This subject had minimal drusen on fundus images and (D) similarly negative SUVR ratio of 1.04. SUVR values indicate the ratio of cortical to cerebellar A-beta accumulation. (Adapted with permission from Shoda C et al., Journal of Alzheimer’s Disease 62 (2018) 239–245 (doi 10.3233/JAD-170956).

Fig. 7.

Qualitative assessment of the choroid in controls and subjects with Alzheimer’s Disease. Representative light microscopy revealed qualitative (A) superonasal choroidal thinning in patients with AD relative to (B) controls. Representative light microscopy revealed qualitative (C) superotemporal macular choroidal thickening with increased vascularity in patients with AD relative to (D) controls. All stains are hematoxylin and eosin. Abbreviation: AD, Alzheimer’s disease. (Reproduced under the Creative Commons Attribution-NonCommercial-NoDerivatives (CC BY-NC-ND) 4.0 http://creativecommons.org/licenses/by-nc-nd/4.0/ International License from Asanad et al., Alzheimer’s and Dementia: Diagnosis, Assessment and Disease Monitoring 11 (2019a) 775–783; doi.org/10.1016/j.dadm.2019.09.005).

Fig. 8.

Representative MRI findings in cerebral small vessel disease.

Fig. 9.

Optical coherence tomography angiography (OCTA) images and quantification of retinal capillary changes in a subject with cognitive impairment and control. The figure shows 3 × 3mm² parafoveal OCTA images of two subjects with CDR-SOB scores = 0 and CDR-SOB = 1, and ages 76 and 73 years, respectively. Neither subject has diabetes but both have hypertension. Images on the second and third columns show the corresponding skeletonized images and pseudo-colored maps of the capillary density. The subject with CDR-SOB = 0 has more localized areas of higher capillary density than the age and medical condition similar subject with CDR-SOB >0. (Reproduced under the Creative Commons Attribution-NonCommercial-NoDerivatives (CC BY-NC-ND) 4.0 http://creativecommons.org/licenses/by-nc-nd/4.0/ International License from Ashimatey et al., Alzheimer’s and Dementia: Diagnosis, Assessment and Disease Monitoring in press 2020).

Fig. 10.

Representative 3D registration and Jacobian maps of OCT image volumes in two subjects with diabetic retinopathy using non-linear registration. Images are a cut through the volume. (A) OCT volume of the one subject with visible diabetic macular edema and alterations in the shape of the foveal pit. (B) Cut through the normative atlas image demonstrating retinal structure of a healthy subject. (C) Color-coded Jacobian map demonstrating magnitude of localized contraction and expansion in diseased subject versus healthy subject independent of layer segmentation. (D) OCT volume of another subject with visible tissue loss. (E) Cut through the atlas image which represents the retinal structure of a healthy subject. (F) Color-coded Jacobian map demonstrating magnitude of localized contraction and expansion in diseased subject versus healthy subject independent of layer segmentation. Red indicates areas of expansion (edema). Blue indicates areas of contraction (tissue loss).

Fig. 11.

New Fig 12 Demonstration of the 3D OCTA surface reconstruction and the subsequent shape and Reeb analysis that may be used for future quantitative analyses of retinal vessels. (A) The original 3D OCTA volume is processed to reconstruct (B) the 3D vessels surface representation. Using (B), we can perform intrinsic shape analysis for (C) large and small vessels classification, and (D) Reeb graph analysis to quantify valuable 3D vessel geometry and topology.

Fig. 12.

New Fig 11 Example of OCTA images from a normal subject before and after preprocessing algorithm for image registration and optimization for further analysis. (A) Original 3D-OCTA image volume. (B) Enhanced OCTA volume obtained by applying 3D curvelet denoising on previous panel. (C) OCTA vesselness map generated by computing the optimally oriented flux response of previous panel. (D) OCTA binary vessel mask obtained by applying Otsu’s global thresholding on previous panel. (E–H) Selected enface views of images in above panels.