Eye Changes May Identify Parkinson's Disease

January 07, 2021

Researchers have identified changes in the eye in patients with Parkinson's disease that can be seen with noninvasive, inexpensive imaging equipment, raising hopes that this could in future become a method for the early diagnosis of the condition.

The team led by eye specialists from Duke University Medical Center, Durham, North Carolina, report decreased retinal microvascular perfusion and structural alterations in the choroid compared with findings in cognitively healthy control individuals.

Their findings were published online in JAMA Ophthalmology on December 23.

Cason Robbins

"Our group has previously reported retinal changes in patients with Alzheimer's and mild cognitive impairment and we are now extending this work to patients with Parkinson's disease," lead author, Cason Robbins, BS, a medical student at Duke University, told Medscape Medical News.

"It has been shown previously that Parkinson's is associated with increased vascular disease in the brain. It could be that we are observing the same changes in the vasculature in the retina. We have seen similar changes in the retinal blood vessels in Alzheimer's but there is more retinal thinning in Alzheimer's disease," Robbins commented.

"We think that in both Parkinson's and Alzheimer's, the changes in the retina reflect dying neurons similar to what is happening in the brain, but we can see it more easily in the eye."

Co-senior author Dilraj Grewal, MD, Duke University School of Medicine, added: "We are struggling to manage the epidemic of Alzheimer's and Parkinson's patients in terms of accurate early diagnoses and monitoring progression. Using the eye is a huge opportunity to help with this. The eye is the window to the brain and can be used to detect early changes in the brain."

For the current cross-sectional study, the researchers used optical coherence tomography (OCT) and OCT angiography to compared the structure and microvasculature of the retina and choroid in eyes of 69 Parkinson's patients and 137 age- and sex-matched cognitively healthy control individuals with no history of tremor or evidence of motor dysfunction. Individuals with other conditions that might affect the eye (such as diabetes, glaucoma, or other dementias) were excluded.

Results showed that the eyes of Parkinson's disease patients had 2% to 3% lower superficial capillary plexus vessel density and perfusion density, 9% to 10% higher total choroidal and choroidal luminal area, and 1% lower choroidal vascularity compared with eyes from age- and sex-matched cognitively normal controls.

"Area under the curve" (AUC) analyses suggested these differences would not be sufficient on their own to diagnose Parkinson's disease with values of 0.5 to 0.7.

Robbins noted that AUC analyses show how much overlap there is with controls. "Some Parkinson's disease eyes look like controls and vice versa if we just look at one value. We are seeing very small changes and there are other conditions that could explain these changes so there is a gray area in the middle between the two groups. For a definite biomarker to diagnose Parkinson's we would be looking for an AUC of 0.9 or 0.95," he said.

However, he added: "The fact that we have shown some differences is still useful, but it doesn't give us the whole picture on its own. We may need other parameters — possibly from other retinal observations or from clinical symptoms. In future we can combine various measurements, or we might need more sophisticated imaging techniques."

The researchers are hopeful that in future such changes in the eye might be able to identify Parkinson's disease before symptoms develop so that patients could be given preventive medication.

"We know Parkinson's affects blood vessels in the brain and the smallest blood vessels that go to the retina are likely to be the first affected. And could be the earliest evidence that something is wrong with the nerve cells serving the retina. In future, it might be the case that looking for these changes indicative of early Parkinson's or Alzheimer's could be included in an annual eye check."

Still, Grewal added, "We have a long way to go before that situation. It won't be a replacement for a neurological exam — but it could be an adjunct."

The other co-senior author, Sharon Fekrat, MD, Duke University School of Medicine, noted that for this study, they excluded patients with other conditions, "as we wanted a clean dataset. We now need more studies with much larger numbers of participants and a more diverse range of patients with different comorbidities," she said. "The challenge will be to see if we can differentiate Parkinson's disease in patients with other comorbidities such as diabetes, which can also affect the eyes."

In an accompanying editorial, Jonathan B. Lin, MD, Harvard Medical School, Boston, Massachusetts, and Rajendra S. Apte, MD, Washington University, St Louis, Missouri, say: "Taken together, these findings support the overarching idea that structural and microvascular changes in the retina and choroid may reflect or be associated with the underlying pathology in Parkinson's disease. These interesting findings suggest that retinal biomarkers may have utility in improving our diagnostic algorithms for Parkinson's disease."

The editorialists suggest that future studies should investigate whether these retinal and choroidal biomarkers differ in various subtypes of Parkinson's, as they demonstrate heterogeneity in disease presentation, response of motor symptoms to dopaminergic agents, and rate of disease progression. "If there is indeed a difference based on Parkinson's disease subtype, these findings would have important implications for improving our ability to counsel patients regarding prognosis," they note.

"Regardless of whether these changes reflect a retinal manifestation of the pathophysiology of PD vs underlying cerebral vasculopathy (or both), these findings suggest that OCT and OCTA may be a valuable addition to our armamentarium for Parkinson's disease diagnosis," they conclude. "Although these biomarkers are not yet ready for clinical practice given the likely need to use them in conjunction with other diagnostic tools, they provide a foundation for future studies to investigate the possibility."

JAMA Ophthalmol. Published online December 23, 2020. Abstract, Editorial

For more Medscape Neurology news, join us on Facebook and Twitter.


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.