Congenital blindness (CB) refers to a group of diseases or conditions occurring in the neonatal period or infancy that result in permanent blindness later in life if left untreated (1). In China, the incidence of blindness in children under the age of 5 is <0.10% or about 70000 children in this age range (2). The three most common causes of blindness in Chinese children are cataract, retinal dystrophy, and optic nerve hypoplasia (3). The development of vision is a complicated neural process. Visual stimulation begins at photoreceptors and then travels to the brain's visual centers via a series of pathways and connections (4). Normal visual function requires the optics of the eye and the retinal/brain neural networks to function well. Studies have found that visual deprivation before the critical period is likely to cause functional reorganization of the brain (5) with the visual cortex or visual-related area becoming more involved in non-visual information processing such as auditory, tactile, or cognitive tasks (6). In individuals with CB, the visual cortex may process non-visual information through cross-modal plasticity (7, 8). In addition, many studies have reported brain anatomical changes in patients with CB (9). For instance, CB patients may have a thicker occipital cortex than sighted individuals (10) and significantly reduced lateral geniculate nucleus (LGN) volume (11). Therefore, CB may be closely associated with these brain dysfunctions or anatomical abnormalities.

Static functional magnetic resonance imaging (fMRI) is a reliable and informative method to study brain activity in a resting state and has been widely applied in CB patients. The regional homogeneity (ReHo) method has proven a reliable and sensitive resting-state fMRI data analysis method since it was introduced in 2004 (12). Each voxel's Kendall coefficient of concordance is obtained by analyzing the consistency between its blood oxygen level-dependent signals and that of 26 adjacent voxels in a given time period. Higher ReHo value indicates greater consistency. We have applied the ReHo method to evaluate the regional consistency of brain activity in patients with eye disease or conditions including acute eye pain (13), corneal ulcer (14), diabetic optic neuropathy (15), diabetic retinopathy (16), dry eye (17), dysthyroid optic neuropathy (18), late monocular blindness (19), optic neuritis (20), post-ophthalmectomy (21), retinal detachment (22), retinal vein occlusion (23), and strabismus (24, 25). In the present research, we use fMRI and ReHo analysis intending to determine whether spontaneous brain activity is normal in CB, and thus to better understand these patients.

Functional MRI may be used to measure cerebral activity and simultaneously determine the active location. This provides an effective and reliable means by which to understand the brain activity anomalies in cognition-related disorders or diseases. Resting-state fMRI does not require patients to perform specific tasks during scans, so it can reduce the influence of motor neural activity which occurs in a task-based fMRI scan. In addition, ReHo analysis has been used to reveal abnormal brain activity in ophthalmology (Table 5).

In contrast to the HCs, patients in the CB group showed significantly decreased mean ReHo values in the RMFGorb, bilateral MOG, and right posterior dorsolateral part of superior frontal gyrus (RSFGdl_p) (Figure 5).

The middle frontal gyrus (MFG) is located in the middle of the frontal lobe. The dorsolateral prefrontal cortex lies in the MFG which may be related to depression (31). The cortical regions associated with visual fixation of peripheral targets are known as the frontal eye fields (FEF) and lie in a region around the junction of the MFG and the precentral gyrus. The FEF is thought to be associated with the control of eye movements and visual attention since unilateral FEF impairment is associated with both eyes deviating to the ipsilateral side. CB patients are not able to consciously move their eyes and lack the vestibulo-ocular reflex while eye movement is practically normal in those with acquired blindness (32, 33). Since visual input is absent from birth in CB patients, their voluntary eye movements may be compromised. The MFG has also been found to be involved in cognitive function and emotion processing. Previous studies indicate that the right MFG may play a part in contingency awareness (34) and reorienting attention (35), that right MFG gray matter volume is reduced in late-life depression and remitting depression patients (31, 36), and that activity of the right orbital part of MFG is reduced in patients with depression and anxiety (37). It is known that visual impairment and difficult living conditions will have a notable psychological impact on CB patients. Anomalies may occur in the brain regions associated with emotional processing, and the fact that 14 of 23 CB patients in our study showed borderline or abnormal states of anxiety or depression is consistent with this possibility. The abnormal changes at RMFGorb may help us to detect anxiety or depression in CB patients earlier. Other studies have also found that CB patients are indeed prone to suffer from depression and anxiety (38–40). We found a significantly decreased mean ReHo value at the RMFGorb in CB patients compared with HCs and the ReHo value of RMFGorb in CB patients was negatively correlated with their anxiety and depression scores. These findings may be related to the impaired voluntary eye movements in CB patients due to their visual deprivation and may be associated with the psychiatric changes such as depression and anxiety that occur in CB patients (Figure 6).

The occipital lobe is located at the posterior part of the cerebrum and is known as the vision and visual processing center (41). The primary visual cortex (also known as the striate cortex, V1) is at the medial and posterior part of the occipital lobe within the calcarine sulcus in each hemisphere. Goodale and Milner first established the classic hypothesis of two visual pathways (42). They believed the primary visual cortex projects information to two pathways, namely the dorsal/localization or “where” stream, mainly processing motion or stereoscopic depth visual information, and the ventral/identification pathway or “what” stream, mainly dealing with colors, shapes and other detailed visual information. However, some researchers have found that the two pathways may not be so clearly distinguished (43, 44). Congenitally blind patients lack visual experience, which suggests that their visual pathway does not develop normally. Previous studies have found that people with early-onset blindness may have brain anatomical changes such as decreased white matter volume in part of the visual pathway and decreased gray matter volume in the visual cortex (45). Despite the anatomical changes, less functional connectivity within the occipital lobe has also been found in early-onset blindness (46). Three gyri (the superior, the middle, and the inferior occipital gyrus) are located along the lateral portion of the occipital lobe (47). Studies have shown that the occipital cortex of people with CB is engaged in other forms of sensory processing, since the occipital cortices, the lingual-gyrus (Brodmann area 18), parts of the cuneus, and the MOG were activated more in early blindness than in sighted control subjects undertaking auditory and tactile tasks (48, 49). The right MOG is part of the dorsal/localization stream and shows increased activity during spatial tactile and auditory tasks in people with early-onset blindness. However, the bilateral MOG showed significantly decreased ReHo values in CB compared to HCs in our resting-state fMRI test, implying that spatial tactile and auditory task-related areas may be inhibited in the resting state.

The superior frontal gyrus (SFG), also known as the marginal gyrus, is located in the superior part of the prefrontal cortex (50) and is involved in higher cognitive functions such as working memory (51). The audio-spatial working memory of CB patients is compromised since visual experience supports spatial memory (52). Significantly reduced ReHo values have been demonstrated in the right dorsolateral part of Brodmann area 8, located at the posterior dorsolateral part of SFG (SFGdl_p), in CB patients (53). This area includes the FEF, known to be related to voluntary eye movements, visual search, and attention control (54, 55). The right and left FEF have slightly different functions, the right FEF playing a role in signal integration (56). The SFGdl_p also has a connection with brain regions that are associated with cognitive control, such as the posterior parietal cortex and dorsolateral prefrontal cortex (57). While FEF may be related to stronger endogenous orienting of non-visual attention (such as verbal cues) in CB patients (58), ReHo is decreased in the right SFGdl_p of CB patients compared with HCs.

However, the mean ReHo values of the CB group in the left paracentral lobule (PCL), right insula, and bilateral thalamus were significantly higher compared with those in HCs (Figure 5).

The PCL is on the medial side of each cerebral hemisphere and is responsible for controlling motor innervations and processing sensory information of the contralateral lower limbs. The left side upper postcentral gyrus (Brodmann area 3) is the precise area with increased ReHo value. The postcentral gyrus is located at the lateral parietal lobe and is also known as the primary somatosensory cortex (59). All of the CB patients and HCs in the present study were right-handed. Increased ReHo of the postcentral gyrus contralateral to the dominant hand in CB patients indicates that they may have more sensitive tactile perception than controls. Several behavioral studies also support the concept of superior tactile abilities in early-blind persons (60, 61). One previous study also found increased activity in the postcentral gyrus in CB subjects compared with sighted individuals on a tactile task (62), consistent with our suggestion.

The insular cortex folds deep within the lateral sulcus in each hemisphere of the brain. The insulae are generally thought to be involved in multimodal sensory processing, taste, self-awareness, motor control, homeostasis, auditory perception, salience, emotions, and other functions. People with CB are thought to be highly sensitive to salient non-visual stimuli. Enhanced intra- and inter-network connectivity such as salience network and frontoparietal network in CB may optimize their abilities to use non-visual stimuli and their attention control, which may help them to adapt to their environment (63). The anterior insula (AI) is believed to be part of the salience network that detects and facilitates higher cognitive processing of the most relevant stimuli in a multitude of forms of sensory information. The functional connectivity between the dorsal AI and the dorsal/localization (where) stream and between the ventral AI and the ventral/identification (what) stream is enhanced in CB (64). This may support our finding that the bilateral insulae are more activated in CB in the resting state. The AI cortex is also believed to be engaged in emotional cognition (empathy), interoceptive awareness (hunger, thirst) and may play a role in conveying this information to consciousness (65). fMRI results show that the AI is correlated with the ability to feel one's own heartbeat and emotion elicitation (66). Increased activity and gray matter volume of the right AI in normal subjects are correlated with higher accuracy of interoceptive sense (measuring own heartbeat). In addition, more negative feelings (higher scores on the questionnaire of the Hamilton Anxiety Scale) are correlated with their interoceptive task accuracy (67). Whether these abilities in CB people are superior remains to be studied in the future. In our study, the increased ReHo at the right insula in CB patients may indicate that they have better interoceptive awareness during the resting state compared to sighted individuals.

The thalamus comprises a pair of large oval gray matter nuclei located in the dorsal part of the diencephalon, just beside the third ventricle. The left and right thalami are connected by a gray matter mass. The thalamus plays roles ranging from sensation relay, regulation of memory and emotion, consciousness, and others (68). The nuclei in the thalamus are separated by a Y-shaped white matter plate into anterior, lateral, and medial groups. Various sensory pathways (except olfactory) replace neurons at the thalamus and then project to the cerebral cortex. For instance, the LGN (part of the thalamus) is the most important relay station in the visual pathway. The retinal ganglion cell gathers visual information from the retina and projects it to the LGN which then projects fibers to the primary visual cortex (area V1). Notably, the LGN not only receives neural projections from the retina, but also has extensive interconnections with the thalamus, brainstem, and visual cortex. Brain functional and structural changes occur after early visual deprivation. One study found that the mean volumes of structures along the visual pathway (including LGN) were reduced by 50% in CB patients (69). Although the LGN may have impaired function and reduced size in CB, volumes of other relay nuclei such as the medial geniculate nucleus (MGN), ventrolateral nucleus (VL), and ventral posterior nucleus (VP) were not affected (70). In addition, studies have found that tactile stimulation in people with CB can stimulate occipital cortical activities. This phenomenon is most likely accomplished by the thalamocortical pathway (71). In animal studies, the deprivation of hamsters' visual systems at birth can lead to retinal projections to the auditory thalamus (72) and the primary visual cortices of enucleated opossums receive projections from nuclei other than LGN, including MGN (auditory), VP (sensation), VL (motor), and anterior nuclear groups (alertness, learning) (73). Human studies have shown that in CB patients the occipital cortex is responsive to auditory stimuli and the superior colliculus is recruited by the auditory system (74). The increased ReHo values of the bilateral thalamus in CB patients probably reflect the structural and functional variations due to early visual deprivation.

Pearson correlation analysis demonstrated no linear correlation between the duration of CB and the mean ReHo values in the significantly altered brain regions.

ROC analysis in our study demonstrates the potential diagnostic value of mean ReHo values (Z-score) of the brain regions found to differ between CB patients and HCs (Figure 3). The AUC of each of these regions was >0.85, indicating effective accuracy in the identification of CB. The AUC of RSFGdl and LPCL was as high as 0.977 and 0.955, respectively. However, these CB patients had been blind for some years (mean age 14.80 ± 2.03 years, with minimum 10.8 years old and maximum 17.8 years old), and it remains unclear whether these positive results hold in newborn blind patients. But ReHo analysis do have the potential to be used for legal or medical proof of CB grown-ups since it is an objective examination.

Patients with CB in our study can only be prevented from the cause at early stages. For these patients who have been CB for years, there is currently no good way to restore sight. Future treatments for patients with CB, such as the use of artificial prosthesis (75), may reverse the abnormalities in these altered brain regions. And the ReHo analysis could help in tracking the treatment effect and assess the function restoration.

Our experimental results found that patients with CB can have regional abnormal brain activities at a resting state. These abnormal neural activities are likely to be associated with the potential regional structural changes, functional reorganization, and even psychological impact on CB patients (Table 6). These abnormal brain regions suggest the possible location of primary or secondary brain damage in CB, which may help to study the pathogenesis and development of CB in the future. ReHo analysis can record changes of regional brain activities non-invasively. It is of great significance use for clinical follow-up and research, and has great application value in the clinic to discover abnormalities related to brain function. Moreover, our ROC result shows ReHo analysis has the potential to be a safe and effective approach to identify CB for medical or legal purpose. It can also be used in diagnosing of patients with CB with complex causes or for whom vision examination is not available. However, our study does have some limitations. Firstly, the sample size of our study was relatively small. Individual differences may have an impact on our result, so a larger population study is needed to prove our findings. In addition, the 15-min scanning time may result in small unavoidable movements which may undermine the accuracy of our analysis. Also, because some of the CB patients didn't have intact medical records, their histories are based on the subjective description of their parents which might be not accurate. In the future, we will minimize the possible influencing factors to obtain more reliable results.

The datasets used and/or analyzed during the current study are available from the corresponding author on reasonable request.

The studies involving human participants were reviewed and approved by Medical Ethics Committee of the First Affiliated Hospital of Nanchang University. The patients/participants provided their written informed consent to participate in this study.

J-JH and NJ were major contributors, both conceived and designed the experiments, analyzed the data, wrote, and revised the manuscript. JC, PY, MK, and S-HX recruited the patients and healthy controls for the study, performed the MRI experiments, and helped finish the questionnaires. JZ, HW, and QL collected and treated the data. YS designed the study and offered financial support. All authors read and approved the final manuscript.

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.