Altered spontaneous brain activity in children with deprivation amblyopia: a resting-state functional magnetic resonance imaging study

Abstract

Background

To investigate the alterations in spontaneous brain activity and the similarities and differences between monocular deprivation amblyopia and binocular deprivation amblyopia.

Methods

Twenty children with binocular deprivation amblyopia, 26 children with monocular deprivation amblyopia and 20 healthy controls underwent resting-state functional magnetic resonance imaging. The evaluation of altered spontaneous brain activity was conducted using fractional amplitude of low-frequency fluctuations (fALFF). One-way analysis of variance was employed to analyze fALFF values among the three groups. Additionally, the relationship between fALFF values and best corrected visual acuity (BCVA) was analyzed via correlation analysis.

Results

Compared to healthy controls, children with binocular deprivation amblyopia presented increased fALFF values in the left medial superior frontal gyrus, left middle frontal gyrus, left anterior cingulate cortex, left postcentral gyrus and bilateral precentral gyrus, and decreased fALFF values in the right fusiform gyrus. Compared to healthy controls, children with monocular deprivation amblyopia presented increased fALFF values in the right lingual gyrus, right superior frontal gyrus, right middle frontal gyrus, left superior temporal gyrus, triangular part of the left inferior frontal gyrus and bilateral middle temporal gyrus, and decreased fALFF values in the right precuneus. Compared with monocular deprivation amblyopia, fALFF values of binocular deprivation amblyopia were decreased in the triangular part of the left inferior frontal gyrus, right lingual gyrus and right cuneus, and increased in the left precentral gyrus and left postcentral gyrus. No significant correlations were found between the fALFF values of identified regions and the BCVA of amblyopic eyes for either type of amblyopia.

Conclusions

Children with deprivation amblyopia presented alterations in spontaneous activity in multiple brain regions, and these alterations differed between monocular amblyopia and binocular amblyopia. These abnormal spontaneous activities may reflect dysfunctions and compensation related to amblyopia.

Introduction

Amblyopia is a common neurodevelopmental disorder resulting from abnormal visual experiences during the sensitive period of visual development. Infantile cataract, which blocks the stimulation of light to the macula at the development stage, will lead to severe deprivation amblyopia if not detected or treated promptly. Studies based on different techniques have revealed anatomical and functional abnormalities in the eyes and brain of patients with amblyopia [1,2,3,4], but the neural mechanism of deprivation amblyopia is still lacking from the perspective of brain function and structure. Resting-state functional magnetic resonance imaging (rs-fMRI) is a widely used noninvasive neuroimaging technique that is especially suitable for children, unconscious patients and psychiatric patients who cannot accomplish specific tasks well [5]. The fractional amplitude of low-frequency fluctuations (fALFF), which is the ratio of the power spectrum in low-frequency to that in the whole frequency range, has greater sensitivity and specificity and can effectively reveal local neural activity in the brain [6].

The purpose of this study was threefold: first, to investigate the alterations in spontaneous brain activity in children with deprivation amblyopia by using rs-fMRI and fALFF analysis; second, to analyze the similarities and differences in brain functional changes between monocular amblyopia and binocular amblyopia; and third, to explore the relationship between brain activity and visual acuity in amblyopia, which may provide more insights into the neural mechanism of amblyopia and be conducive to improving clinical classification and better prediction of treatment response.

Materials and methods

Subjects

This study was carried out in accordance with the Declaration of Helsinki and approved by the Ethics Committee of our hospital. All participating children and their guardians were informed of the purpose and procedure of this study, and written informed consent was obtained from their legal guardians. All children received rs-fMRI scans and detailed ophthalmic examinations, including optometry, best corrected visual acuity (BCVA), intraocular pressure, slit lamp examination, dilated fundus examination, and examinations of strabismus and ocular motility. Snellen visual acuity was converted to the logarithm of the minimum angle of resolution (LogMAR).

A total of 46 children with untreated infantile cortical cataract were diagnosed with deprivation amblyopia and divided into two groups: 20 children in the binocular deprivation amblyopia group and 26 children in the monocular deprivation amblyopia group. The inclusion criteria for the deprivation amblyopia groups were as follows: (1) aged between 5 and 14 years with unilateral or bilateral infantile cortical cataract; (2) BCVA falling below the lower range limit of acuity standards at the same age; and (3) right-handed. The exclusion criteria were as follows: (1) complicated with other eye diseases (such as high ametropia, anisometropia, strabismus, glaucoma, corneal disease, ptosis or fundus disease); (2) previous or current neuropsychiatric diseases; (3) previous history of ocular or neuropsychiatric trauma or surgery; (4) any prior treatment for amblyopia; (5) indistinct fundus; and (6) contraindications to MRI.

Twenty age- and gender-matched children were included in the healthy control group. The inclusion criteria were as follows: (1) aged between 5 and 14 years; (2) BCVA was ≤ 0.0 logMAR and differed by no more than two lines in two eyes; and (3) right-handed. The exclusion criteria were as follows: (1) previous history of ocular or neuropsychiatric diseases, trauma or surgery; and (2) contraindications to MRI.

MRI data acquisition

MRI data were obtained via a 3.0 Tesla MR scanner (Magnetom Prisma, Siemens, Germany). Horizontal T1-weighted and T2-weighted images were routinely scanned to exclude brain lesions. Then, rs-fMRI scans were performed using an echo planar imaging sequence with the following scan parameters: repetition time = 1000 ms, echo time = 30 ms, flip angle = 70°, matrix = 110 × 110, field of view = 220 × 220 mm², slice thickness = 2.2 mm, and total volume = 400. All children were asked to lay supine, keep their eyes closed, remain awake and move as little as possible during scanning. Tight but comfortable foam pads were used to minimize head motion, lights were turned off, and earplugs were used to reduce scanner noise. All these MRI scans were performed by the same skilled radiologist.

Data preprocessing

The fMRI images were conventionally preprocessed using Data Processing and Analysis for Brain Imaging (DPABI) software (http://rfmri.org/dpabi) on the MATLAB R2018b platform. The DICOM format was converted to the NIFTI format for analysis, and the first 10 time points were discarded for MRI signals to reach a steady state. Time correction was performed to eliminate the influence caused by the different scanning time of each layer of images in each volume. Subjects were excluded if their head movement was more than 2 mm translation in any axis and 2° rotation in any axis during scanning. The data were then spatially normalized to the Montreal Neurological Institute (MNI) template (resampling voxel size = 3 × 3×3 mm³) and smoothed with a 6-mm full width at half maximum to reduce spatial noise.

fALFF analysis

The linear trend was removed after preprocessing. The time series of each voxel were transformed to a frequency domain with a fast Fourier transform, and the power spectrum was then obtained without bandpass filtering. The fALFF value was calculated by the ratio of the power spectrum in the low-frequency range (0.01–0.08 Hz) to that in the whole frequency range (0–0.25 Hz). The fALFF value of each voxel was then divided by the global mean fALFF value for each subject to obtain a normalized fALFF value and reduce the global effects of variability. All procedures were performed via DPABI software.

Statistical analysis

SPSS 23.0 software (iBM Corp., Armonk, NY, USA) was used for the statistical analysis of the clinical data. Measurement data were expressed as mean ± standard deviation or median with interquartile ranges. The differences in age and years of education among the three groups were analyzed by one-way analysis of variance (ANOVA). The difference in the duration of illness between the monocular amblyopia group and the binocular deprivation amblyopia group was analyzed via a two-sample t-test. BCVA among the three groups was analyzed by a nonparametric Kruskal‒Wallis test. The difference in the age of onset between the monocular amblyopia group and the binocular deprivation amblyopia group was analyzed by the Mann–Whitney U test. The difference in gender among the three groups was analyzed by the Chi-square test. P < 0.05 was considered to be statistically significant.

With DPABI software, one-way ANOVA and LSD post hoc tests were conducted to compare the differences in fALFF values among the three groups using Gaussian random field correction (voxel significance: P < 0.01, cluster significance: P < 0.05, two-tailed) with a cluster size of at least 47 voxels. The fALFF values in each significant cluster were extracted for each subject, and linear correlation analyses between the fALFF values of the identified regions and the BCVA of amblyopic eyes were performed in the monocular and binocular deprivation amblyopia groups.

Results

Demographic characteristics

No significant differences were observed in age, gender, or years of education among the three groups. No significant differences in the duration of illness or age of onset were observed between the monocular and binocular deprivation amblyopia groups. The difference in the BCVA of amblyopic or nondominant eye was statistically significant among the three groups (P < 0.001). The demographic characteristics of all the subjects are summarized in Table 1.

Table 1 Demographic characteristics of all subjects

fALFF differences among the three groups

Compared with the healthy controls, children with binocular deprivation amblyopia presented increased fALFF values in the left medial superior frontal gyrus (SFG), left middle frontal gyrus (MFG), left anterior cingulate cortex (ACC), left postcentral gyrus (PostCG) and bilateral precentral gyrus (PreCG) (all P < 0.05). The regions that presented decreased fALFF values were distributed mainly in the right fusiform gyrus (FFG) (P < 0.05). These results are shown in Table 2; Fig. 1.

Table 2 Brain regions with significant fALFF values between children with binocular deprivation amblyopia and normal controls

Fig. 1

Anatomical distribution of altered fALFF values between children with binocular deprivation amblyopia and normal controls. Hot color represents greater fALFF values in the binocular deprivation amblyopia group than in the normal controls, while cool color represents lower fALFF values

Compared with the healthy controls, children with monocular deprivation amblyopia presented increased fALFF values in the right lingual gyrus (LG), right SFG, right MFG, left superior temporal gyrus (STG), triangle part of the left inferior frontal gyrus (IFGtri) and bilateral middle temporal gyrus (MTG) (all P < 0.05). The regions that presented decreased fALFF values were distributed mainly in the right precuneus (Pcun) (P < 0.05). These results are shown in Table 3; Fig. 2.

Table 3 Brain regions with significant fALFF values between children with monocular deprivation amblyopia and normal controls

Fig. 2

Anatomical distribution of altered fALFF values between children with monocular deprivation amblyopia and normal controls. Hot color represents greater fALFF values in the monocular deprivation amblyopia group than in the normal controls, while cool color represents lower fALFF values

Compared to children with monocular deprivation amblyopia, the fALFF values of children with binocular deprivation amblyopia were lower in the left IFGtri, right LG and right cuneus (CUN) (all P < 0.05). The regions showing increased fALFF values in the binocular deprivation amblyopia group included the left PreCG and left PostCG (both P < 0.05). These results are shown in Table 4; Fig. 3.

Table 4 Brain regions with significant fALFF values between monocular and binocular deprivation amblyopia

Fig. 3

Anatomical distribution of altered fALFF values between monocular and binocular deprivation amblyopia. Hot color represents greater fALFF values in the binocular deprivation amblyopia group than in the monocular deprivation amblyopia group, while cool color represents lower fALFF values

Correlations between fALFF values and BCVA in amblyopia

No significant correlations were found between the fALFF values of identified regions and the BCVA of amblyopic eyes for either type of amblyopia by using Pearson correlation coefficient (all P > 0.05).

Discussion

Blood oxygenation level-dependent fMRI is a noninvasive, intuitive, high-resolution neuroimaging technique for mapping brain function, while rs-fMRI provides a reliable measure of brain activity and connectivity at baseline. The fALFF is an effective index for measuring alterations in spontaneous brain activity and has been successfully applied in various ocular diseases [7,8,9]. To date, the understanding of the neural mechanism underlying amblyopia is limited due to inconsistent findings, and studies focusing on the brain function of deprivation amblyopia are still lacking[9,10,11].To the best of our knowledge, this study represents the first application of rs-fMRI and fALFF to investigate spontaneous brain activity in children with deprivation amblyopia and to compare the differences between monocular amblyopia and binocular amblyopia. Our results revealed abnormal spontaneous activities in multiple brain regions in both monocular and binocular deprivation amblyopia, and notable differences in the alterations of spontaneous brain activity were detected between monocular amblyopia and binocular amblyopia. Additionally, no significant correlations were found between the fALFF values of the identified abnormal regions and the BCVA of amblyopic eyes in either monocular or binocular deprivation amblyopia.

In the present study, compared to the normal control group, we detected increased fALFF values in the SFG and MFG of the left hemisphere in the binocular deprivation group, as well as increased fALFF values in the SFG and MFG of the right hemisphere in the monocular deprivation group. The SFG is involved in higher-level information processing, including sequences of visual-guided saccades, eye‒hand coordination [12], cognitive control and emotion regulation-related processes [13]. The frontal eye field (FEF) located in the posterior MFG is regarded as not only a motor area for saccades and head movements, but also a critical region for spatial attention [14]. The binocular deprivation amblyopia group showed increased fALFF values in bilateral PreCG when compared to the normal control group, and increased fALFF values in the left PreCG when compared to the monocular deprivation amblyopia group. The PreCG constitutes part of the primary motor cortex where neural activities reflect internal processes involved in planning and executing volitional movements. A new study with structural MRI and task-state fMRI highlighted an outstanding role of the PreCG in flexible oculomotor control [15]. Min et al. also reported increased ALFF values in the right SFG and bilateral PreCG in strabismus with amblyopia patients [12]. Increased fALFF values in these brain regions represented enhanced spontaneous brain activities and may imply a potential compensatory mechanism and functional reorganization in ocular movement control in deprivation amblyopia.

The IFG in the left hemisphere (dominant hemisphere of right-handed people) includes the motor language center—Broca's area, which is a crucial brain area for linguistic computations and is closely related to speech processing and complex syntax [16]. The opercular and triangular parts of the left IFG are also regarded as grammar centers [17]. The STG contains the auditory speech center—Wernicke's area, the key site for speech coding and processing [18]. In fact, the left IFG and STG are co-activated within meaningful word conditions [16]. Studies have demonstrated bidirectional neural activity between Broca's area and Wernicke’s area during verbal interaction [19] and visual‒auditory interactions in auditory functional areas [20]. On the basis of these findings, the increased fALFF values in the left STG and left IFGtri in the monocular deprivation amblyopia group compared to the normal control group may indicate that visual deprivation could cause functional compensatory enhancements in the auditory and language systems. Furthermore, compared with the binocular deprivation amblyopia group, the monocular deprivation amblyopia group presented increased fALFF values in the left IFGtri, suggesting more obviously enhanced brain activities for speech processing in patients with monocular deprivation amblyopia.

The PostCG is the primary sensory cortex. There are functional connections and cross-modal integration between visual regions and other regions associated with somatosensory, kinesthetic and auditory senses [21,22,23]. Compared with the normal control group and monocular deprivation amblyopia group, the fALFF values of the binocular deprivation amblyopia group were increased in the left PostCG, suggesting that patients with binocular deprivation amblyopia may experience paresthesia. Tang et al. [24] reported increased ALFF values in the left STG in patients with anisometropic amblyopia. Lin et al. [25] discovered increased ReHo values in the left STG, bilateral PostCG and bilateral PreCG in anisometropic amblyopia. However, Yu et al. [23] and Wen et al. [26] found weakened functional connections between the bilateral primary visual area and PostCG in patients with early or late blindness. These findings suggest that early deprivation of a single visual sensory pattern may lead to functional changes between deprived functional areas and other sensory-related brain regions.

Previous studies have demonstrated that functional connections and cross-modal integration emerge between visual regions and other areas, such as the frontal, temporal, and parietal lobes, which are associated with somatosensory, kinaesthetic, and auditory processing through cross-modal plasticity [21, 27]. There is a temporary projection of many brain areas into the visual cortex at birth, but these cross-modal connections are heavily pruned to establish normal visual neural circuits during normal visual development. However, for the visually impaired children, reduced visual experience may lead to abnormal synaptic pruning, resulting in excess cross-modal connections with compensatory plasticity that gradually drive the visual cortex to integrate and process information from other sensory modes. Patients with amblyopia have dysfunction in spatial localization and global motion perception. Our study found increased brain activity in the SFG, MFG, IFG, STG, PreCG, and PostCG. This suggests that the integration of these sensory and motor channels based on cross-modal plasticity is essential for maintaining spatial perception, orientation, visuomotor integration, and motor perception in patients with amblyopia.

The FFG, integral to face recognition within the ventral visual pathway, exhibited decreased fALFF values on its right side among children with binocular deprivation amblyopia in our study. Dai et al. [2] found that children and young adults with amblyopia presented significantly decreased effective connectivity from the right FFG to the right calcarine fissure. Wang et al. [28] also reported decreased short-range functional connectivity density in the left inferior temporal gyrus and FFG. These findings suggest that the inadequate capability of object recognition in patients with deprivation amblyopia may be related to impaired FFG and ventral pathway functions. However, Dai et al. [11] found decreased ALFF values in the left MTG, left SFG, bilateral MFG and bilateral PreCG and increased ALFF values in the right FFG, which were not consistent with our findings. This inconsistency was possibly because their subjects were a mixture of anisometropic and ametropic amblyopia, while the subjects in our study were only deprivation amblyopia, and the ALFF analysis they used was more easily disturbed by noise than the fALFF analysis we used.

The Pcun, part of the posterior parietal cortex (PPC), is involved in complex integrated functions, including visuospatial imaging, sensorimotor function and consciousness [29]. There is evidence of decreased ALFF values [30] and ReHo values [25] in the Pcun in patients with amblyopia. Consistent with these prior findings, we detected decreased fALFF values in the right Pcun in the monocular deprivation amblyopia group, suggesting impairments in somatosensory integration, visuomotor coordination and cognition in amblyopic children. The MTG and PPC represent advanced functional regions of the dorsal visual pathway that are crucial for visuomotor analysis, spatial position, behavior prediction and execution [31]. Decreased fALFF values in the right Pcun and increased fALFF values in the bilateral MTG in monocular deprivation amblyopia suggest dysfunctions in the dorsal visual pathway, and this view aligns with the findings of Ding et al. [32], Simmers et al. [33] and Thompson et al. [34]. Deficits of the dorsal visual pathway typically manifest distinct disorders in the detection of motion, object locations and eye‒hand coordination, whereas deficits of the ventral visual pathway reflect abnormities in processing shape, color, material properties of objects or faces, and stereoscopic depth perception. In our current study, we revealed dysfunctions in the ventral visual pathway in patients with binocular deprivation amblyopia while dysfunctions in the dorsal visual pathway in patients with monocular deprivation amblyopia, suggesting differential patterns of visual pathway impairment between these two forms of amblyopia.

The default mode network (DMN) is the core functional network responsible for maintaining resting state of the brain nervous system. DMN mainly includes the posterior cingulate cortex, medial prefrontal cortex, inferior parietal lobule and temporal cortex [35]. Our results revealed increased fALFF values across extensive areas of frontal and temporal lobes in both monocular and binocular deprivation amblyopia, indicating an aberrant activation mode of the DMN. DMN is associated with maintaining arousal and monitoring external stimuli [36], and this high level of activation may facilitate monitor and adaptation to blurred surroundings in amblyopic patients. The ACC is a subregion of the cingulate gyrus and is involved in reward–punishment processing and emotion [37]. Shao et al. [38] observed increased ReHo values in the bilateral ACC in strabismus and amblyopia patients. Similarly, we found increased fALFF values in the left ACC in patients with binocular deprivation amblyopia. Given the fact that amblyopia can affect children's psychosocial well-being, and perceptual learning and training can effectively improve the visual function of children [39, 40], our findings hint that enhanced neural activity ACC may be related to learning and emotion regulation based on the reward‒punishment feedback effect.

The CUN and LG are separated by calcarine fissures where surrounding cortical structures constitute primary visual cortex (V1). V1 is the primary site for visual information processing and is considered the main damage site of amblyopia. The functional abnormalities of V1 have been widely explored in amblyopia. Barnes et al. [41] observed reduced fMRI activation in V1 within the amblyopic passband in patients with strabismic amblyopia. Shao et al. [38] and Yang et al. [42] reported decreased ReHo values in V1 in patients with strabismic amblyopia. In contrast, the monocular deprivation amblyopia group presented increased fALFF values in the right LG when comparing with the normal control group and increased fALFF values in the right LG and right CUN when comparing with the binocular deprivation amblyopia group. The increased fALFF values in V1, indicating increased brain activity, may reflect visual cortical plasticity and compensation for the visual impairments caused by amblyopia.

The observed differences may be attributed to the varying types of amblyopia, diverse research methods, and different age groups of the subjects. The visual cortex exhibits significant plasticity during the critical period, but this neuroplasticity diminishes with age. In this research, children with deprivation amblyopia were aged between 5 and 14 years, with most still within the critical period of visual development; thus, abnormal visual experiences (such as visual deprivation) can affect defects and compensation of visual-related cortices through neurodevelopmental plasticity. Owing to interference from the opacified crystals, light stimulation into the amblyopic eye was reduced, prompting increased visual input from the retina to the visual cortex via the contralateral eye. No similar situation was found in binocular deprivation amblyopia, possibly because there was little difference in binocular vision, and the phenomenon of interocular imbalance and competitive inhibition was not obvious.

This study revealed, in agreement with the study of Min et al. [12], that there was no linear correlation between fALFF values in brain regions with significant differences and visual acuity in the amblyopic eye. Yang et al. [42] reported no correlation between ReHo values and visual acuity in patients with strabismic amblyopia. Task-based fMRI studies also revealed no association between visual acuity and the activation of brain regions in patients with amblyopia [43]. These results suggest that the visual loss associated with amblyopia may not correlate directly with functional defects in brain regions. This is because visual acuity is a subjective indicator that reflects the function of macular fovea, whereas amblyopia is a neurodevelopmental disorder characterized by functional and structural changes in the brain. Currently, diagnosing and grading the severity of amblyopia based solely on visual acuity may be inadequate for assessing the extent of brain impairment. Additionally, determining the endpoint of treatment based exclusively on visual acuity could contribute to the recurrence of amblyopia. In the future, the fALFF value could serve as a novel biomarker for assessing alterations in cortical function as well as therapeutic outcomes in amblyopic patients, providing deeper insights into the neural mechanism of amblyopia and facilitating improved clinical classification and better prediction of treatment response.

The present study had several limitations. First, the sample size for each group was relatively small because of the difficulty in recruiting children and the poor cooperation of some children during rs-fMRI scans. Future research should aim for larger sample sizes with the sample size calculated using G*Power software (http://www.gpower.hhu.de/). Second, this study was cross-sectional and conducted prior to treatment; thus, longitudinal studies comparing functional changes in brain regions before and after amblyopia treatment are needed. Finally, this was only a preliminary study and cannot comprehensively describe the full-scale brain activity alterations associated with amblyopia. Employing multimodal imaging techniques will be essential for more accurately delineating the neural mechanisms underlying amblyopia. These limitations are all aspects of improvement in our future studies.

Conclusion

In conclusion, the current study demonstrates distinct alterations of spontaneous brain activity in multiple brain regions in children with deprivation amblyopia. Additionally, these alterations differ between monocular and binocular deprivation amblyopia, thereby enhancing our understanding of amblyopia. These abnormal spontaneous activities in specific brain regions may account for functional damage and compensation in amblyopia and help elucidate the neuropathogenesis of amblyopia.