Schizophrenia — a severe mental-health condition that often involves psychosis — has long been an enigma. Before the early twentieth century, psychiatrists thought that it comprised several disorders, which were given labels such as ‘catatonic syndrome’ or ‘adolescent insanity’. And for much of the past century, researchers considered schizophrenia to be an illness of only the mind, and sometimes even attributed it to bad parenting. Although it has since been established that schizophrenia’s symptoms have biological origins, and some risk factors have been identified, its precise causes remain unclear, and its diagnosis is complicated and can be subjective.
To improve diagnosis and tracking of schizophrenia, researchers have been hunting for biomarkers — measurable physiological signals that can indicate a condition’s onset and progression. And in the past decade, research has begun to point to a promising source of such signals: the eye. For instance, the thickness of a person’s retina — the layer of light-sensitive tissue at the back of the eye — or the retina’s response to light could provide early signs that an individual is affected by or at risk of schizophrenia. “The retina is essentially a proxy for what’s happening in the brain,” says Steven Silverstein, a clinical psychiatrist at Rutgers University in Piscataway, New Jersey.
The potential to use the eye as a window on the brain goes beyond just schizophrenia. As understanding of the links between eye and brain health deepens, evidence is building that changes in the condition or function of the eye can hint at the presence and progression of neurological disorders, as well as brain injuries such as concussion. With that in mind, researchers are now taking advantage of improved imaging technology to develop simple, minimally invasive tools for examining the eye and vision to help screen for, diagnose and monitor neurological and mental-health conditions.
Vision is thought to require the use of about half of the brain’s neural pathways, says Laura Balcer, a neuro-ophthalmologist at New York University Langone Health in New York City. Anything that affects a person’s brain — whether it be a disease or a blow to the head — therefore has a strong chance of affecting their sight. Indeed, researchers have known for decades that certain neurological disorders can bring about changes in vision and eye motion.
Much of the work to assess neurological conditions through the eye has been done in the context of multiple sclerosis. In this disorder, the immune system disrupts communication in the central nervous system by attacking myelin, a fatty substance that forms a protective layer around nerve fibres. The process damages several parts of the brain, including those involved in pathways that are responsible for vision, and the optic nerve, which carries impulses between the eye and the brain.
Although people with multiple sclerosis often report having vision problems, standard eye tests — in which a person reads letters printed in black from a chart with a white background — typically fail to uncover any issues. So, in the early 2000s, Balcer and her collaborator Steven Galetta, also a neuro-ophthalmologist at Langone Health, helped to develop an alternative test that uses the Sloan low-contrast letter-acuity chart, in which a person has to identify grey letters on a white background — a task that is more difficult for people with multiple sclerosis, because they require greater levels of contrast than do people without the condition. The test has since become the leading tool for monitoring and studying vision problems associated with the condition.
Researchers are also developing tests that make use of other visual pathways to diagnose multiple sclerosis. The King-Devick test, developed by King-Devick technologies in Oakbrook Terrace, Illinois, assesses the brain pathways that control eye movement by requiring a person to read rapidly from a page of single-digit numbers. And a test known as the Mobile Universal Lexicon Evaluation System (MULES) asks participants to name images of objects as fast as they can to test pathways for colour perception, memory and object recognition.
These new tests can also help to diagnose concussion that is incurred while playing sport. Currently, a medical professional or athletic trainer has to assess a person for a range of symptoms, including dizziness, nausea and general confusion — a task that is often subjective. Simple tests of visual pathways, however, could provide an objective and standardized procedure for assessing concussion that can be performed anywhere — including at sporting events. In May 2018, US racing company NASCAR announced that it would begin to use the King-Devick test to assess drivers who are involved in crashes.
Mapping the eye
Neurological injuries and disorders can also cause physical damage to the retina and optic nerve that could give researchers an insight into the underlying condition. That’s because the brain and the eye are linked physically by the optic nerve. In fact, certain parts of the eye, including the retina, are essentially an extension of the brain, having developed from the same embryonic tissue.
Examining the retina has become much easier since the advent of a non-invasive imaging technique called optical coherence tomography (OCT), about 30 years ago. In OCT, the eye is scanned with a beam of light. The light that is reflected back is analysed to produce a 3D image of the retina. Used mainly to monitor and diagnose eye conditions such as glaucoma or age-related macular degeneration, OCT has now become a tool for studying, monitoring and potentially diagnosing neurological disorders.
The technique has revealed, in detail, how multiple sclerosis thins the retina and damages the optic nerve, leading to inflammation — a condition called optic neuritis. And by combining OCT with tests for low-contrast vision, researchers have uncovered a correlation between a thinning retina and vision problems in multiple sclerosis1.
Galetta and Balcer say that OCT-based observations of the optic nerve should be incorporated into standard criteria for diagnosing the condition. At present, such a diagnosis requires the identification of lesions in the central nervous system by magnetic resonance imaging (MRI). Although optic-nerve lesions are often present before symptoms appear, they are rarely used in diagnosis. The optic nerve is difficult to image by MRI because it is small and the eye moves constantly. OCT, however, boasts a resolution that is 1,000 times greater than that of MRI, and the technique is therefore more sensitive to optic-nerve lesions.
OCT also holds promise for identifying early signs of the neurological conditions Parkinson’s disease and Alzheimer’s disease, which could enable people to be treated before symptoms develop. In September, a team of researchers at Seoul National University used OCT to show that thinning of the retina was correlated with both an increasing severity of Parkinson’s disease and the death of neurons that produce the neurotransmitter dopamine2, the loss of which causes Parkinson’s disease.
In November, researchers in the United States combined OCT with angiography, a technique for imaging blood vessels, as part of efforts to find early biomarkers of Alzheimer’s disease. The team compared healthy retinas with those of people who had elevated levels of amyloid-β — a peptide linked to Alzheimer’s disease — but who did not yet exhibit symptoms of the condition. They found that the foveal avascular zone, a region of the retina that lacks blood vessels, was about one-third larger, on average, in people with elevated amyloid-β3.
The data are limited but promising, says Gregory Van Stavern, a neuro-ophthalmologist at Washington University in St. Louis, Missouri, who co-led the study. “We’re very enthusiastic to see if we can replicate it, and see if we can show this could be a really useful tool for screening down the line,” he says.
From eye to mind
Researchers are also looking to the eye for early biomarkers of neuropsychiatric disease, which could enable clinicians to intervene before their onset or thwart their progression. For instance, there is evidence that, in people with schizophrenia, the small veins, or venules, of the retina are wider and the retina is thinner4.
Perhaps the most promising tool in the short term for identifying people at risk of developing schizophrenia is electroretinography, a simple and minimally invasive test that measures the retina’s electrical response to light. That signal is captured using a small electrode that is attached to the cheek below the eye or placed under the upper or lower eyelid. “It feels like you have a hair in your eye for about a minute and then you don’t really notice it anymore,” Silverstein says.
A team of researchers led by clinical psychiatrist Michel Maziade at Laval University in Quebec City, Canada, has used electroretinography to identify how the response of light-sensitive cells in the retina called rods and cones changes in people at risk of or diagnosed with neuropsychiatric conditions such as schizophrenia and major depressive disorder5. In 2010, the researchers found that the rods of young people with a high genetic risk of developing schizophrenia respond more weakly to light than do those of young people without that risk6. Individuals considered to be at risk had one parent who had been diagnosed with bipolar disorder or schizophrenia, but showed no symptoms themselves, which suggests that electroretinography can help to identify people who might benefit from medical intervention before symptoms appear. A 2015 study with a larger population size found similar photoreceptor responses to light in 100 adults with schizophrenia7.
The data are impressive, says Anne Giersch, a neuroscientist at the University of Strasbourg in France. But they need to be replicated, she says, and further studies will be required to establish the relationships between the retina and the symptoms of schizophrenia before the technique can be used to help to determine whether treatment is necessary.
Nevertheless, Maziade is confident that electroretinography will soon enter the clinic. In 2015, he co-founded diaMentis, a personalized-medicine company in Quebec City that is developing software for electroretinography-signal analysis. Within three years, Maziade says, the company’s software should be advanced enough to help clinicians to make accurate diagnoses of schizophrenia. Further down the line, clinicians might even be able to use the technique to distinguish between neuropsychiatric disorders. In unpublished work, Maziade says, his team has shown that the electroretinography signals for schizophrenia and bipolar disorder are distinct.
As with any screening tool, care must be taken when using emerging techniques to minimize the chances of overdiagnosis or misdiagnosis. No single test such as an electroretinography examination can accurately predict the presence of neurological and mental-health conditions, Maziade says; people would still have to be evaluated in the context of their symptoms, genetics and exposure to environmental factors for clinicians to obtain reliably accurate diagnoses. For instance, if OCT were shown to be effective at detecting Alzheimer’s disease, such screening would probably be targeted at at-risk populations such as older people or those with a family history of the condition. “You have to decide what population you’re going to screen and what you’re going to do with that tool,” Van Stavern says.
But despite these concerns, eye tests have the potential to change how neurological and mental-health conditions are diagnosed, and Silverstein is excited to see this unusual idea move forward. “I’ve been studying vision and schizophrenia for 30 years,” he says. “Nobody ever thought of looking into the eye.”
This article is part of Nature Outlook: The eye, an editorially independent supplement.