Altered spontaneous activity in the frontal gyrus in dry eye: a resting-state functional MRI study

This study investigated neurologic changes in patients with dry eye (DE) by functional magnetic resonance imaging (fMRI) and to used regional homogeneity (ReHo) analysis to clarify the relationship between these changes and clinical features of DE. A total of 28 patients with DE and 28 matched healthy control (HC) subjects (10 males and 18 females in each group) were enrolled. fMRI scans were performed in both groups. We carried out ReHo analysis to assess differences in neural activity between the 2 groups, and receiver operating characteristic curve (ROC) analysis was performed to evaluate the performance of ReHo values of specific brain areas in distinguishing DE patients from HCs. The relationship between average ReHo values and clinical characteristics was assessed by correlation analysis. ReHo values of the middle frontal gyrus, inferior frontal gyrus, and superior frontal gyrus were significantly lower in DE patients compared to HCs. The ROC analysis showed that ReHo value had high accuracy in distinguishing between DE patients and HCs (P < 0.0001). The ReHo values of the middle frontal gyrus and dorsolateral superior frontal gyrus were correlated to disease duration (P < 0.05). Symptoms of ocular surface injury in DE patients are associated with dysfunction in specific brain regions, which may underlie the cognitive impairment, psychiatric symptoms, and depressive mood observed in DE patients. The decreased ReHo values of some brain gyri in this study may provide a reference for clinical diagnosis and determination of treatment efficacy.

www.nature.com/scientificreports/ areas involved in human cognition and the localization and quantification of neuronal activity. Functional (f)MRI is the main modality used to evaluate the functional state of the brain 9 . Compared to traditional MRI technology, fMRI has higher spatial resolution and allows detailed visualization of brain microstructure and changes thereof. fMRI comprises diffusion weighted imaging (DWI) and diffusion tensor imaging (DTI). The former reveals spatial information on tissues and can be used to detect early morphologic and physiologic abnormalities based on changes in tissue water content, while the latter is the only technique that allows 3-dimensional (3D) analysis of white matter fiber bundles based on the direction of diffusion of water molecules. It was reported that gray matter volume in the occipital and parietal eye fields was significantly reduced in patients with strabismus compared to healthy subjects 10 . Additionally, functional abnormalities in multiple brain regions have been reported in patients with eye diseases 11 . However, there have been no studies to date using fMRI to investigate DE. MRI signals reflect functional networks in the central nervous system 12 ; changes in the resting state signal can provide a basis for predicting and diagnosing diseases including primary insomnia 13 , obstructive sleep apnea 14,15 , and sleep deprivation 16 . Analysis of regional homogeneity (ReHo) in resting-state fMRI is a sensitive and reliable method for measuring changes in neural activity 17 . An advantage of ReHo is that it does not require a seed region to be specified, and can therefore be used to detect changes in any brain region 18 . ReHo analysis has been used to investigate the pathogenesis of epilepsy 19 , primary insomnia 13 , Parkinson disease 20 , depression 21 , and other neurologic diseases; and ophthalmologic studies have applied this method to optic neuritis 22 and comitant strabismus 11 . In the present work, we explored changes in brain activity in DE and examined their relationship to clinical features by fMRI and ReHo analysis.

Materials and methods
Subjects. Patients with moderate or severe DE (N = 28; 18 women and 10 men) were recruited at the Ophthalmology Department of the First Affiliated Hospital of Nanchang University Hospital. The following criteria were used to diagnose: dry eyes; foreign body sensation, burning sensation, cracking and itching, photophobia, blurred vision, visual fatigue, and other annoying visual symptoms; variable conjunctival and corneal staining; tear breakup time < 10 s, Schirmer I test < 10 mm/5 min.
Inclusion criteria for DE patients were as follows: (1) met the diagnostic criteria for lacrimal dry eye; (2) age 20-75 years old; (3) not treated with any drugs or had discontinued treatment for > 2 weeks prior to enrollment; and (4) provided informed consent to participate in the study. Exclusion criteria were as follows: (1) conjunctival scarring, atresia of the lacrimal gland opening, or complete atrophy of accessory lacrimal glands; (2) other obvious conjunctiva, cornea, or iris lesions; (3) pregnant or lactating women; (4) suspected or confirmed history of drug abuse (Fig. 1).
We also recruited 28 healthy control (HC) subjects (18 women and 10 men) who were matched to DE patients in terms of age, sex, and other demographic parameters. The HCs met the following criteria: (1) no abnormalities in brain parenchyma observed by MRI; (2) no eye diseases, visual impairment, or corrected vision (visual acuity [VA] > 1.0); (3) normal mental health and no abnormalities upon neurologic examination; and (4) no contraindications for MRI. This study was authorized by the ethics committee of the First Affiliated Hospital of Nanchang University Hospital, and the protocol met the requirements of the Declaration of Helsinki and conformed to the principles of medical ethics. All volunteers participated voluntarily and were informed of the purpose, methods, and potential risks of the study. All participants signed an informed consent form.

MRI parameters.
MRI was performed with a 3-Tesla magnetic resonance scanner (Magnetom Trio; Siemens, Munich, Germany). Subjects were instructed to keep their eyes closed but remain awake and relaxed until the end  Statistical analysis. Differences in demographic and clinical data between DE patients and HCs were evaluated with the independent samples t-test using SPSS v20.0 software (SPSS, Chicago, IL, USA) differences in ReHo values between SA and HC subjects were evaluated using two-sample t-tests in REST software (State Key laboratory of cognitive neuroscience and learning, Beijing Normal University, Beijing, China). At the voxel level, the statistical threshold was set to P < 0.05, and for multiple comparisons using Gaussian random field theory voxels, thresholds of P < 0.001 and cluster size of > 40 voxels (AlphaSim-corrected) were employed. We speculated that differences in ReHo values can serve as a biomarker for diagnosing DE and identifying any associated neurologic abnormalities, and tested this hypothesis by receiver operating characteristic (ROC) curve analysis. Accuracy was considered low or high if the area under the ROC curve (AUC) was 0.5-0.7 and 0.7-0.9, respectively. For all tests, statistical significance was set at P < 0.05.
Brain-behavior correlation analysis. REST

Demographic information and visual measurements.
There were no significant differences in age (P = 0.839), height (P = 0.668), body weight (P = 0.724), body mass index (P = 0.912), or best-corrected same-eye VA between DE patients and HCs (Table 1).

ReHo differences.
ReHo values of the triangular part of the inferior frontal gyrus (IFG), middle frontal gyrus (MFG), and dorsolateral superior frontal gyrus (SFG) were significantly lower in DE patients than in HC subjects ( Fig. 2 and Table 2). The average ReHo values of the 2 groups are shown in Fig. 3A. The values of the MFG and SFG showed an inverse correlation with disease duration. We found that there was a correlation between patient ReHo values and progression duration, and middle and Superior beta gyrus Dorsolateral ReHo values were correlated with progression duration, which was statistically significant. (P < 0.05; Table 3, Fig. 4).
Compared with HC, the ReHo values of the inferior frontal gyrus triangularpart, middle frontal gyrus, and superior frontal gyrus dorsolateral areas of DE patients were significantly reduced.    (Fig. 4), indicating that the ReHo values of these brain areas have high good accuracy in distinguishing DE patients from HCs. See Fig. 3B.

Discussion
DE is a disease in which tear film instability or ocular surface damage caused by changes in the quantity and quality of tear fluid or abnormal tear fluid dynamics results in ocular discomfort and visual dysfunction. As an early complication after corneal refractive surgery, DE can usually be cured 6-9 months after surgery. DE has a multifactorial etiology and is often accompanied by mental or neurologic disorders such as depression, anxiety, stress, posttraumatic stress disorder, and sleep disorders. Drugs used to treat mental illnesses can also affect DE along with neuropathic pain, chronic pain syndrome, peripheral neuropathy, and central nervous system diseases 23 . Thus, DE is likely related to nervous system and brain dysfunction.
Resting-state functional magnetic resonance imaging can provide insight into brain abnormalities in disease states 24 . The ReHo method can reveal abnormal activities in specific brain regions 25 . At present, the ReHo method has been successfully applied to a variety of craniocerebral injuries, neurogenic diseases and ophthalmological diseases, with great potential for development ( Table 4). As far as we know, the current study is the first to use ReHo technology to evaluate resting brain activity in DE patients.
Compared with HC, the ReHo values of the inferior frontal gyrus triangularpart, middle frontal gyrus, and superior frontal gyrus dorsolateral areas of DE patients were significantly reduced. Figure 5 shows the abnormal brain areas.
The MFG is located in the frontal lobe of the cerebral cortex between the suprafrontal and subfrontal sulci; the posterior part of the MFG processes information related to the movement of the head and eyes. The MFG has been implicated in response to stress 33 and cognitive function 34 , and we previously showed that it is more active during task performance 35 . A study using DTI to analyze structural changes in the brain of patients with ophthalmologic diseases showed that the diffusion coefficient of the MFG was significantly lower in these individuals than in control subjects 36 .
The IFG is mainly involved in language and voluntary movement and is implicated in the neurologic effects associated with ocular surface damage in DE patients. We found that the ReHo value of the IFG in DE patients was significantly lower than that of HCs, which is consistent with our previous research 5 .
The SFG is linked to depression and plays an important role in cognition and attention 37 . Abnormalities in the SFG have been observed in patients with depression [38][39][40] . Interestingly, depression has been reported in patients with DE and other corneal diseases 41,42 . We speculate that this is related to a decrease in the ReHo value of the SFG, which was observed in our study. Our research results show that the clinical symptoms of DED patients are indeed related to brain dysfunction, which is consistent with the study of Alexandra et al. Table 3. Pearson correlations analysis. ReHo regional homogeneity, SD standard deviation.

Brain regions
ReHo value (mean ± SD) Duration (years) (mean ± SD) r-value P-value  www.nature.com/scientificreports/ Therefore, we speculate that DE may cause brain dysfunction. (Fig. 6, Table 5). This study also used the ROC curve, and the results showed that the AUC of each brain area was greater than 0.7. These REHO values can be used to supplement the results of statistical analysis, but the ReHo values cannot provide diagnostic capabilities.
This study had some limitations. Firstly, the sample size was small; additional studies are needed in a larger population in order to validate our findings. Second, the length of the scan time and small body movements may affect the scan results; these individual differences may undermine the accuracy of our analysis.
In addition, the literature shows that there is no spatial smoothing in data preprocessing to improve the reliability of ReHo 43 , and the 6 × 6 × 6 mm smoothing method we use may have certain errors.
Nonetheless, the results of our study demonstrate that DE patients exhibit abnormal neural activity in the frontal gyrus, which may be related to DE pathogenesis. The ReHo values of the IFG, MFG, and SFG can be used to assist clinicians in identifying DE and help judge the prognosis.