Neurodegenerative diseases (NDDs) are a group of disorders that impair memory, cognition, and motor function. They share common features: strong association with aging, aggregation of abnormal proteins, and a slow, irreversible course. Representative diseases include Alzheimer’s disease (AD), Parkinson’s disease (PD), amyotrophic lateral sclerosis (ALS), Huntington’s disease, and frontotemporal dementia.
These diseases are characterized by insidious onset, with neuropathological changes progressing years before clinical symptoms appear. This makes early diagnosis difficult and leads to delays in treatment. It is not uncommon for the clinical picture to remain vague until a substantial amount of neural tissue has been irreversibly lost.
The retina is an extension of the central nervous system (CNS) and is the only site that can be observed non-invasively. Embryologically, the retina and optic nerve develop from the diencephalon and remain connected to the CNS via the optic nerve after birth. Anatomically, the layered structure of the retina and the blood-retinal barrier are similar to those of the CNS, and the composition of the neurovascular unit (NVU) is also shared. 1)
Optical coherence tomography (OCT) is a non-invasive imaging technique that uses reflected light to create high-resolution two- and three-dimensional images of retinal structures. Because it can objectively and quantitatively assess the thickness of various retinal layers, it is attracting attention as a tool for detecting structural damage in the visual pathway. 1) Postmortem tissue studies have shown reduced retinal thickness in AD, PD, and ALS, and OCT is a powerful means of capturing these changes in living individuals.
OCT angiography (OCTA), a subset of OCT, images the integrity of retinal and optic nerve vasculature, primarily assessing blood flow perfusion. It can depict changes at the capillary level (5–15 µm) without the use of contrast agents, and abnormalities have been reported in PD, Huntington’s disease, ALS, AD, and multiple sclerosis. 1)
The retina shares embryological, anatomical, and physiological features with the brain. Just as neurodegenerative diseases damage neurons in the brain, they also cause degeneration of retinal neurons. OCT enables non-invasive quantitative measurement of each retinal layer, and is being studied as a biomarker that reflects neurodegeneration in the brain. 1)
Visual symptoms associated with neurodegenerative diseases vary by condition, but common symptoms include the following.
The main retinal changes detected by OCT are shown by disease.
Alzheimer's Disease
Peripapillary RNFL thinning: Observed in all quadrants, most prominent in the superior quadrant. Standardized mean difference (SMD) = −0.67. 1)
Macular GC-IPL thinning: SMD = −0.46. May be more sensitive than peripapillary RNFL for assessing neurodegeneration in AD. 1)
Retinal vascular changes: Sparse vascular network and increased vascular tortuosity are observed compared to the control group. 1)
Parkinson's disease
Thinning of RNFL in all quadrants: Meta-analysis has confirmed thinning in all quadrants in PD patients. 1)
Inner retinal layer (IRL) thinning: Parafoveal IRL is reduced by approximately 15% compared to the control group. This is thought to result from the loss of dopaminergic amacrine cells. 1)
Changes in foveal pit: Thinning and widening of the foveal pit have been reported as quantifiable features of PD. 1)
In PSP (progressive supranuclear palsy), the ONL (outer nuclear layer):OPL (outer plexiform layer) ratio has been reported to distinguish PSP (<5.03) from MSA (>5.03) with 88% sensitivity and 91% specificity. 1) In ALS (amyotrophic lateral sclerosis), thinning of the peripapillary RNFL has also been reported, but the number of studies is still limited.
OCT has high detection power for damage to RGCs and their axons due to optic pathway lesions, and is useful as an objective and quantitative evaluation method. cpRNFL thickness and macular inner retinal layer thickness are analyzed using probability displays compared to the built-in age-matched normal eye database, but since normal thickness varies greatly among individuals, evaluation by actual measured values and comparison with the fellow eye are also important.
At present, it is difficult to use for severity assessment. Some studies show a correlation between RNFL thinning and disease progression, but there are also reports that failed to show a correlation between MMSE (Mini-Mental State Examination) scores and RNFL thickness in AD patients. 1) The use of OCT for severity assessment is a topic for future research.
Neurodegenerative diseases are multifactorial and share the following common features.
Regarding degeneration in the retina, it remains unclear whether it is primary neurodegeneration (degeneration within the retina itself) or secondary retrograde degeneration due to loss of neurons in the brain. Two mechanisms have been proposed in AD: 1)
In the retina of AD patients, amyloid beta deposition is increased compared to controls, with a tendency to cluster around blood vessels, particularly in the superior peripheral quadrant. 1) Retinal amyloid beta accumulation may occur earlier than in the brain and is thought to increase with disease progression. 1) However, these findings are inconsistent across studies, and the existence of retinal pathology in AD remains controversial. 1)
Retinal evaluation using OCT plays a central role in biomarker research for neurodegenerative diseases. The main evaluation parameters and measurement methods are described below.
The main OCT evaluation parameters are shown in the table below.
| Evaluation Parameter | Abbreviation | Main Target Diseases |
|---|---|---|
| Peripapillary retinal nerve fiber layer thickness | cpRNFL | AD, PD, ALS |
| Macular ganglion cell-inner plexiform layer complex | GC-IPL | AD, PD |
| Inner retinal layer | IRL | PD |
| Outer nuclear layer to outer plexiform layer ratio | ONL:OPL ratio | PSP, MSA differentiation |
SD-OCT (spectral-domain OCT) is the current standard method. Measurement of macular GC-IPL may have higher sensitivity for evaluating neurodegeneration in AD than peripapillary RNFL, due to the anatomical feature that this region contains more than 50% of the total RGC volume. 1)
The measurement reproducibility of peripapillary RNFL by SD-OCT is high (inter-test ICC=0.927, CoV=3.83%), and even higher reproducibility has been shown for GC-IPL (ICC=0.968, CoV=1.91%). 1)
Swept-source OCT (SS-OCT) is a new-generation technology, and some studies have shown retinal thinning in PD as well as increased choroidal thickness. 1)
OCTA visualizes retinal capillaries without contrast agents and can quantify the following parameters.
In the neuro-ophthalmology field, OCTA can evaluate the radial peripapillary capillaries (RPC) around the optic disc, capturing dilation, tortuosity, and decreased vessel density in the superficial retinal vessels. Decreased vessel density is observed in areas corresponding to nerve fiber layer defects (NFLD) seen on fundus examination.
The usefulness of OCT in differentiating PSP from MSA has also been investigated. When using the ONL:OPL ratio as an indicator, it has been reported that PSP (ratio <5.03) and MSA (ratio >5.03) can be distinguished with a sensitivity of 88% and specificity of 91%. 1)
At present, a definitive diagnosis cannot be made by OCT alone. OCT provides suggestive biomarkers for neurodegenerative diseases, but differentiation from retinal thinning due to other causes such as glaucoma or age-related changes is necessary. The diagnosis of neurodegenerative diseases is ultimately based on comprehensive clinical evaluation. 1)
Currently, OCT examination in neurodegenerative diseases is positioned as a research and monitoring tool, and no specific treatment based on OCT findings has been established. Treatment of each neurodegenerative disease is primarily managed by neurologists.
Ophthalmic management includes the following:
The similarities between the retina and brain are diverse. 1)
It is thought that the brain pathology of AD (amyloid beta plaques, tau neurofibrillary tangles, neuronal loss) impairs neural connections in the visual pathway, leading to retrograde degeneration in the optic nerve and retina. 1) However, in the posterior cortical atrophy variant of AD, where the occipital visual cortex is primarily affected, no difference in peripapillary RNFL thickness was detected compared to controls, suggesting that retrograde degeneration alone cannot explain all findings. 1)
Alternatively, a common pathology hypothesis has been proposed, in which amyloid beta, fibrillar tau, and neuroinflammation occur simultaneously in the brain and retina. Reactive gliosis (RNFL thickening) sometimes seen in the inner retina in early AD may precede RNFL thinning or mask subtle RNFL thinning on OCT. 1)
Moreover, most retinal amyloid beta deposits are found in the GC-IPL, and some cluster around blood vessels. Retinal amyloid beta is qualitatively similar to brain amyloid beta plaques, which are defining lesions of AD, and is detected only in minimal amounts in non-AD controls. 1)
Dopamine is an important neurotransmitter in the retina, and dopaminergic amacrine cells are present in the inner nuclear layer and inner plexiform layer. Dopamine receptors have been identified in retinal pigment epithelial cells, photoreceptors, Müller cells, bipolar cells, horizontal cells, and ganglion cells. 1)
Dopamine modulates the receptive fields of the ganglion cell layer to provide spatial contrast sensitivity and color vision, and is also involved in light adaptation and melatonin production regulation. Postmortem analyses have shown that the eyes of PD patients have reduced dopamine content compared to controls, 1) which provides a basis for structural and functional changes in the retina.
The thickness of the parafoveal inner retinal layer (IRL) is reduced by approximately 15% in PD patients compared to controls, and it is hypothesized that this is directly due to the loss of dopaminergic cells. 1)
MCI is an intermediate stage between normal age-related memory problems and dementia, and amnestic MCI has a high risk of progression to AD. OCT has been shown to potentially help differentiate MCI from AD. MCI shows peripapillary RNFL loss but not as pronounced as in AD, and AD affects the ganglion cell layer-inner plexiform layer (GCC-IPL) complex more strongly than MCI. 1)
Longitudinal studies have shown that thinning of RNFL and GC-IPL is associated with future cognitive decline, and OCT is expected to become a non-invasive biomarker for predicting preclinical AD. 1)
The introduction of next-generation computational technology (AI) into retinal image analysis is progressing. AI technology is expected to be applied to screening and risk stratification of retinal imaging, which has advantages over brain imaging techniques such as being non-invasive, relatively low-cost, and available at primary care facilities. 1)
Spund et al. reported thinning and widening of the foveal pit in PD patients, suggesting that this change may provide a quantifiable feature of PD. 1) Research is ongoing to determine whether mathematically characterizing foveal pit changes can serve as a tool for PD diagnosis and progression prediction.
SD-OCT-based retinal layer assessment is considered a promising aid for differentiating PSP from PD. Significant differences have been found in PSP compared to both age-matched controls and PD patients, and it is expected to become a future diagnostic aid. 1)
Zou et al. showed that in the eyes of PD patients, compared to healthy age-matched controls, there is a decrease in RNFL thickness in the temporal quadrant, total macular volume, macular retinal thickness, and GCL-IPL thickness. OCTA data revealed reduced vessel length density (VLD) in the central, inner, and full regions, reduced vessel perfusion density (VPD) in the full region, and a reduced FAZ circularity index. 1)
Large-scale OCTA studies on ALS and Huntington’s disease are still lacking, and future research is anticipated. 1)
At present, this is still at the research stage, and a definitive diagnosis using OCT alone is difficult. However, longitudinal studies have shown that OCT biomarkers (RNFL and GC-IPL thinning) are associated with future risk of cognitive decline, and combined with AI technology, they are expected to potentially become screening tools in community healthcare settings in the future. 1)
