Background
Major depressive disorder (MDD), affecting thousands of millions of people all around the world [
1], is one of the most debilitating disease with the symptoms of low mood, declined interests and impaired cognition [
2], along with somatic ones, such as headache [
3‐
7] and insomnia [
7]. Major depressive disorder was reported to be one of the five leading causes of years lived with disability (YLDs) in 2016 globally, resulting in 34.1 million of total YLDs [
8].
Several lines of evidences show that there exist visual abnormities in patients with major depressive disorder. Photophobia, perceived dimness, anomalous preattentive processing of visual information and self-reported visual function loss were found to exists in patients with major depressive disorder [
9‐
14]. Using the CogState battery (CSB) Chinese language version, a sensitive cognitive assessment instrument, an impairment in the visual, working, and verbal memory was found in first-episode, drug-free MDD patients in a Chinese population [
15]. Longer visual search time was needed when detecting a target calling for difficult attentive search strategy in patients with MDDs than in controls [
16]. Retinal contrast gain, visual contrast sensitivity, and visually evoked potentials (VEPs) amplitudes were found to decrease significantly in patients with MDD, by utilizing the method of pattern electroretinogram (PERG), subjective visual contrast test and VEPs recordings [
17‐
20]. The normalization of decreased contrast gain after anti-depressant treatment was also reported, the therapies being varied markedly [
21]. Developmentally and anatomically, the retina is thought to be an extension of the CNS and thus much of knowledge and discoveries obtained from eye research could be applied to the CNS [
22].
Apart from the methods of PERG recording, VEPs recording and subjective visual contrast test, pattern glare test is another tool to explore vision abnormities. Some people can perceive visual perceptual distortions and discomfort, manifesting as the symptoms of eyestrain, headaches and glare, as well as illusions of shapes, colors, and motion, when viewing repetitive striped patterns [
23,
24]. The extent of these experiences varies according to the character of the pattern and individual susceptibility [
25]. High-contrast striped patterns, with spatial frequency around 3 cycles/degree and with equal width and spacing, tend to generate maximal effect [
25]. The phenomenon depicted above has been named ‘pattern glare’ [
26]. The term ‘Meares–Irlen syndrome’ or ‘visual stress’ was also used to describe the symptoms generated by pattern glare [
27]. The neural mechanism underlying pattern glare is widely considered as cortical origin, that is, cortical hyper-excitability or poor cortical inhibition generated by deficiency of inhibitory mechanisms that are insufficient to contain the overexcited conditions [
25,
27‐
33]. An increase of blood oxygenation in the visual cortex, evidenced by the fMRI BOLD signal, was found when subjects responded to square-wave gratings [
34]. Jason J. Braithwaite and colleagues concluded that heightened levels of pattern glare can reflect increasement of cortical hyperexcitability associated with some possible abnormal experiences in some nonclinical populations [
35]. Pattern glare has been assessed using The Wilkins and Evans Pattern Glare Test, which was first performed by Wilkins and Evans to identify individuals susceptible to visual stress by means of pattern related questionnaire, by counting scores for the number of visual illusions and discomfort induced by viewing three high-contrast gratings of spatial frequencies 0.5 cycles per degree (cpd), 3 cpd, and 12 cpd [
27]. The application of pattern glare test in a great variety of neurological conditions, such as synaesthesia, autism, myalgic encephalomyelitis, multiple sclerosis, stroke, and reading ensued [
36‐
43]. To the best of our knowledge, there is no paper reporting the association between major depressive disorder and pattern glare.
Inspired by the above-mentioned evidences that patients with major depressive disorder display various vision abnormities and pattern glare-related symptoms, such as headache and photophobia, we presume high level of pattern glare in the disease, which is assessed by using the pattern glare test. Meanwhile, the current way to diagnose major depressive disorder, depending mainly on the self-reporting of symptoms and mental evaluation, accompanies with the chance of misdiagnosis, which calls for a reliable biomarker with good accuracy for identifying the disease [
44]. We attempted to explore possible diagnostic value of pattern glare test, as a potential biomarker, in major depressive disorder.
Discussion
We demonstrated that there exists high level of pattern glare in major depressive disorder. The MDDs group scored higher in the mid and high frequencies but not low frequency pattern gratings than the HCs group. Previous studies have found that specifically the mid-SF pattern that would induce most distortions. The mid-SF score and mid-high difference are the two most distinctive indications of pattern glare [
23,
36,
43]. Significantly negative correlation between mid-high difference and age in HCs group was found. Our performance of ROC analysis using pattern glare score of mid-SF showed the score possesses limited value of identifying the depressed and the healthy.
Several lines of evidences reported decreased GABA levels in occipital cortex, anterior cingulate cortex (a prefrontal cortical region) and the dorsomedial/dorsal anterolateral prefrontal ROI in MDD subjects and even in recovered depressed patients [
51‐
55].
The association between altered GABA levels and vision abnormity has also been reported frequently. Visual deficit-poor vision-was found to be rendered by loss of dendritic cell factor 1 through the GABA system in mouse primary visual cortex [
56]. In the early stage of type 2 diabetes, occipital cortical GABA has been reported to be a novel predictor of visual psychophysical performance, that is, speed and achromatic discrimination thresholds [
57]. The application of GABA and its agonist to senescent macaques lead to improved orientation and direction, an enhanced ability to signal visual stimuli, combined with decreased visual responsiveness and spontaneous activity [
58]. Increased visual cortical GABA levels was found to be correlated with longer percept durations [
59]. Remarkably reduced occipital GABA concentrations has been reported to be possibly responsible for the visual problem in first-episode, unmedicated MDD [
60].
These findings thrown some light on interpreting pattern glare via the GABA system. The symptoms of pattern glare are more obvious under binocular than monocular conditions, GABA being the neurotransmitter whose interneurons are responsible for the combination of the input from the two eyes in the cortex, which suggest a role of GABA in the mechanism of pattern glare [
24,
27,
61]. Additionally, the decreased GABA levels in MDD lead to deficient cortical inhibition or cortical hyper-excitability and thus high level of pattern glare, cortical hyper-excitability being deemed to be the mechanism of pattern glare.
Whether psychiatric stress or other psychiatric conditions, such as anxiety or schizophrenia, plays a role in increasing pattern glare score could be concerned. The patients in our study were free of other psychiatric symptoms. It is well worth involving more psychiatric diseases, such as anxiety disorder, schizophrenia and obsessive-compulsive disorder, in the future study of pattern glare.
The result of significantly negative correlation between mid-high SF difference and age in HCs group showed an overall decreasing level of pattern glare with age, which is consistent with previous finding that there existed a significant inverse correlation between age and the pattern glare score for the 3 cpd. The result indicated increased cortical inhibition with age, agreeing with a recent finding that visual cortical GABA levels increased in older adults [
59]. Primary cortex synchrony, contributing to cortical hyperexcitability, was reported to get losing with age in monkey, which leads to a decreasing level of pattern glare with age [
62,
63].
Many optical changes in the eye with age including lens changes, increased aberrations and senile pupillary miosis can result in a reduction of the contrast sensitivity function; these changes may give rise to lower level of light reaching the retina and might have an influence on the results of the pattern glare test with age [
27,
64]. The loss of sensory acuity with age may also be responsible for the decreased pattern glare scores [
27]. However, the MDDs group displayed no negative correlation between age and pattern glare scores. Since the age between the MDDs and the HCs did not differ significantly, this result indicated relatively reduced cortical inhibition in MDDs group compared to the HCs group, which is consistent to our assumption.
The ROC analysis of pattern 2 displayed limited clinical diagnostic value of identifying the depressed and the healthy, the area under the curve being not so large. But we hold that the test deserves further exploration, for it is rather easy to implement, but may represent a new potential biomarker of diagnosing MDD. Replication of the test with large sample size is warranted to explore whether it is possible to find the diagnostic value of pattern 2 more powerful.
Our study has two limitations. First, though there were no significant differences in pattern glare score between first episode mediation-free subjects and recurrent medication-ongoing patients, it is unclear whether the effect of disease episode numbers on pattern glare scores and the effect of medicine can influence each other. Three groups, including healthy controls, first-episode medication-free subjects and the subjects after medication therapy can be set to explore the effects of medicine on pattern glare score in future study. Second, our study was limited by a small sample size. The results need to be verified in future study for it is the first time the pattern glare test has been applied to major depressive disorder.
Publisher’s Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.