Article Text
Abstract
Aims To investigate the potential morphological alterations of grey and white matter in monocular amblyopic children using voxel-based morphometry (VBM) and diffusion tensor imaging (DTI).
Methods A total of 20 monocular amblyopic children and 20 age-matched controls were recruited. Whole-brain MRI scans were performed after a series of ophthalmologic exams. The imaging data were processed and two-sample t-tests were employed to identify group differences in grey matter volume (GMV), white matter volume (WMV) and fractional anisotropy (FA).
Results After image screening, there were 12 amblyopic participants and 15 normal controls qualified for the VBM analyses. For DTI analysis, 14 amblyopes and 14 controls were included. Compared to the normal controls, reduced GMVs were observed in the left inferior occipital gyrus, the bilateral parahippocampal gyrus and the left supramarginal/postcentral gyrus in the monocular amblyopic group, with the lingual gyrus presenting augmented GMV. Meanwhile, WMVs reduced in the left calcarine, the bilateral inferior frontal and the right precuneus areas, and growth in the WMVs was seen in the right cuneus, right middle occipital and left orbital frontal areas. Diminished FA values in optic radiation and increased FA in the left middle occipital area and right precuneus were detected in amblyopic patients.
Conclusions In monocular amblyopia, cortices related to spatial vision underwent volume loss, which provided neuroanatomical evidence of stereoscopic defects. Additionally, white matter development was also hindered due to visual defects in amblyopes. Growth in the GMVs, WMVs and FA in the occipital lobe and precuneus may reflect a compensation effect by the unaffected eye in monocular amblyopia.
- Visual (cerebral) Cortex
- Visual pathway
- Child health (paediatrics)
- Imaging
Statistics from Altmetric.com
Introduction
Amblyopia is defined as reduced visual acuity that cannot be corrected by refractive means and cannot be attributed to obvious structural or pathological ocular anomalies.1 It is commonly associated with visual deprivation, anisometropia or strabismus.2
Although often regarded as a developmental disorder resulting from anomalous binocular visual input early in life, the mechanisms underlying amblyopia are still under investigation.3 Animal experiments with artificially induced amblyopia have shown that the physiological basis for amblyopia is mainly located at the cortical cortex and the lateral geniculate nucleus, and less in the retina.4 Fewer foveal V1 neurons may be driven by the amblyopic eye than the fellow fixing eye.5 Recently, positron emission tomography and functional MRI (fMRI) have made the direct investigation of neural activity in human possible. Consistently, functional activation of the calcarine cortex is reduced in the amblyopic eyes compared with the nonamblyopic eyes.6 ,7 In the structural MRI studies, reduced grey matter volume (GMV) was found in the visual cortical regions.8 Multiple extrastriate areas are also involved, such as parietal–occipital areas, the ventral temporal cortex and the cuneus.8 ,9 It is important to notice that white matter (WM) is also essential for normal function. Efficient brain function requires integration of information from segregated regions and thus depends on the structural properties of the connecting network. Up to now, there has been limited study regarding the changes in WM related to amblyopia. Xie et al10 selectively investigated the optic radiation in amblyopic children by diffusion tensor imaging (DTI) and tractography. The anatomical changes in the grey matter (GM) are not limited to the striate cortex, so WM changes may also involve areas besides the optic radiation. Voxel-based whole-brain analysis may reveal underlying morphological changes associated with amblyopia.
It should also be noted that most of the current studies of amblyopia recruited subjects with mixed aetiologies. However, different types of amblyopia may be associated with different brain structural changes. Take monocular amblyopia, for example, where the binocular process is seriously interrupted and defects in spatial vision are especially prominent. Additionally, with visual acuity in one eye being in the normal range, the unaffected eye may show a certain degree of compensatory effects and the brain may be modified in a specific way. Therefore, we mainly focused on monocular amblyopic children and made a comprehensive evaluation of GM and WM in the presence of amblyopia using voxel-based morphometry (VBM) and DTI. VBM provides whole brain analysis of intergroup differences in GMV and WM volume (WMV) in a standard space. DTI can provide structural and functional information about the WM quantitatively by measuring fractional anisotropy (FA). A combination of these techniques may be helpful to elucidate global brain changes that are peculiar to monocular amblyopia and reveal the possible reasons for the condition.
Methods
Subjects
A total of 40 right-handed Chinese children participated in this study. Among of them, 20 children had amblyopia (15 boys and 5 girls aged 4–15 years) and 20 subjects had normal sight (10 boys and 10 girls aged 5–14 years). The subjects’ parents provided informed consent, and the ethics committee at Tianjin Medical University approved all study protocols. The amblyopic children all had monocular amblyopia (12 had right monocular amblyopia and eight had left monocular amblyopia) and were recruited by physician referral from the ophthalmology service at Tianjin Eye Hospital. Initially, they completed a series of ophthalmologic exams, such as tests of visual acuity and ocular motility, a fundus exam and so on. After the diagnosis of monocular amblyopia was confirmed (visual acuity of amblyopic eye ≤0.6, visual acuity for unaffected eye ≥0.9), they were referred to MRI scanning prior to treatment. The controls were recruited from volunteers and children who underwent MRI examination for other purposes unrelated to vision problems. All the participants were confirmed to have no structural head disease and to be free of neurological conditions.
MRI data acquisition
All MRI scans were performed using a Signa HDx 1.5 Tesla MR Scanner (GE Medical Systems) in Tianjin Medical University General Hospital. Head motion was minimised with foam padding provided by the manufacturer around the head, and noise was attenuated with earplugs. The importance of head immobility was stressed to each subject. Conventional axial T2-weighted images were obtained previously to rule out the presence of any detectable lesions in their brains. Axial three-dimensional T1 -weighted magnetic resonance images were obtained using a fast spoiled gradient recalled acquisition in steady state, gradient-recalled echo sequence with the following parameters: echo time (TE)=1.836 ms; time of repetition (TR)=8.636 ms; time of inversion (TI)=450 ms; flip angle=20°; slice thickness=1 mm; slice gap=0 mm; field of view (FOV)=24 cm×24 cm; matrix=256×224; slice number=124; number of excitation (NEX)=1. The scan lasted 5 min and 38 s.
Diffusion-weighted data were acquired with an echo planar imaging sequence. The diffusion-sensitive gradients were applied along 13 non-collinear directions with a b-value of 1000 s/mm2, together with an acquisition without diffusion weighting (b=0). The acquisition parameters were as follows: TR=7000 ms, TE=105.7 ms, matrix=128×128, FOV= 24 cm×24 cm, slice thickness=4 mm, slice gap=0 , NEX=3. A total of 378 contiguous axial slices were acquired, and the scan time was 5 min and 8 s.
Data processing
VBM analysis
The VBM analysis was performed with Statistical Parametric Mapping V8 (SPM8) (http://www.fil.ion.ucl.ac.uk/spm/software/spm8). T1-weighted structural MR images were segmented into GM, WM and cerebrospinal fluid using the standard unified segmentation model in SPM8. Non-linear warping of GM and WM images was performed to the standard GM and WM template in Montreal Neurological Institute (MNI) space with 1.5 mm cubic resolution. The GMV and WMV of each voxel were obtained through modulation by multiplying the GM and WM concentration map by the non-linear determinants derived from the spatial normalisation step. Finally, to compensate for residuals between-subject anatomical differences, the GMV and WMV images were smoothed with a full width at half maximum kernel of 4 mm full width at half-maximum. In effect, the analysis of modulated data tests for regional differences in the absolute volume of the brain and removes the confounding effect of variance in individual brain sizes. After spatial pre-processing, the smoothed, modulated and normalised GMV and WMV maps were used for statistical analysis.
A voxel-based two-sample t-test was employed to test the difference between monocular amblyopia and normal vision groups with a threshold of p<0.001 (uncorrected) at a cluster level of 25 voxels. Gender and age were entered as covariates of no interest.
DTI analysis
Each subject's DTI data were pre-processed and FA values were calculated using DTI software (Johns Hopkins University, Baltimore, Maryland, USA). FA is a ratio of diffusion coefficients without a unit and its value ranges from 0 to 1, where 0 represents isotropic diffusion and 1 represents highly anisotropic diffusion. The resulting FA maps were first spatially normalised using SPM8. Briefly, using the affine and nonlinear spatial normalisation algorithm, the FA map of each participant was spatially normalised to the echo planar imaging template using the normalisation parameters derived from the normalisation of the b=0 image. Finally, the normalised FA map image was re-sampled to a 2×2×2 mm3 voxel and spatially smoothed using a Gaussian kernel of 4 mm full width at half-maximum.
Comparisons between monocular amblyopia and normal vision groups were performed using statistical parametric maps of FA. A two-sample t-test was carried out with gender and age as covariates. The threshold was set at p<0.001 (uncorrected) and a cluster size of ≥10 voxels.
Results
Visual and demographic data
Among the participant pool, some participants were excluded due to poor image quality or excessive head motions. Then, based on the gender and age matching criteria, data from 12 amblyopic participants and 15 normal controls were qualified to use in the VBM analysis. For DTI analysis, 14 amblyopic participants and 14 controls were included. The visual acuity information and demographic data of the subjects are shown in table 1.
Between-group differences in GMV and WMV
Compared to the normal-sighted group, reduced GMVs were detected in the left inferior occipital gyrus, the bilateral parahippocampal gyrus and the left supramarginal/postcentral gyrus in the monocular amblyopic group (table 2, figure 1). In contrast, areas of significant WM decline were seen in the left calcarine and the bilateral inferior frontal and right precuneus areas (table 3, figure 2). An interesting phenomenon was that increases in GMV and WMV were observed. WMVs increased in the right cuneus, middle occipital and left orbital frontal areas, with GMV elevated in the lingual gyrus.
Between group differences in FA values
Diminished FA values were observed in the optic radiation in the monocular amblyopic children compared to the control group. In addition, another two areas showed increased FA in amblyopic subjects, including the right precuneus and left middle occipital area (table 4, figure 3).
Discussion
This study indicated that GM and WM are associated with morphological changes in monocular amblyopic children using a voxel-based analysis method. Moreover, these modifications were not limited to decreasing cortices, but also increased GMV, WMV and FA in the visual and related cortices.
Significant reductions in GMVs were found in the inferior occipital gyrus, the bilateral parahippocampal gyrus and the supramarginal/postcentral gyrus in our study, which may be caused by the subjects’ history of abnormal visual experience in the critical developmental period. Previous studies about amblyopia have suggested that besides the primary visual cortex, there are deficits at the higher levels of the visual pathway as well.8 ,9 ,11 This statement is supported by this study,where cortices related tospatial vision were found to undergo morphological changes in patients with monocular amblyopia. The parahippocampal cortex is a critical component of orientation and navigation,12 and is particularly concerned with the spatial properties of a scene.13 Doeller and Kaplan14 suggested that the parahippocampal cortex transforms external visual information into well-defined spatial features, which may be associated with its anatomical location. Parahippocampal cortex lies at the interface between the spatial representational system and the visual system, receives strong projections from the visual and visual-related cortex, and, in turn, provides input into the entorhinal cortex, subiculum and the CA1.15 Some other studies have shown that the parahippocampal cortex is linked with visual areas that are traditionally associated with the dorsal visual pathway.16 Furthermore, the supramarginal/postcentral gyrus also belongs to the dorsal visual system, which is functionally specialised for spatial vision. In monocular amblyopic patients, stereoscopic vision was obviously impaired, and the cortical abnormalities in these areas may well be the anatomical basis is anblyopia.
A significant part of the amblyopic defects may be due to anomalous interactions between cells in disparate brain regions.17 Xie et al10 found more voxels in the posterior optic radiations of normal children than in the amblyopic children, and this was confirmed in our study from the perspective of FA. FA reflects the angle of cellular structures within the fibre tracts and therefore reflects the fibres’ structural integrity.18 Some factors, such as myelination, cell-packing density, fibre diameter and directional coherence, can influence FA. Decreased FA in the optic radiation indicated that the optic radiation lost volume and underwent changes in cellular components or local microstructural texture. Meanwhile, reduced WMVs were seen in the left calcarine areas, which implied WM variations in the primary visual area as well as GMV changes. Visual defects in amblyopic patients might hinder WM from maturation. GM and WM interact with each other, and the function of GM depends on the information transmitted by axons, which is the main component of WM. Previously, GMV reductions have been considered to be due to less stimulation from the outside world. Undoubtedly, this is correlated with less information flow in the related fibre tracts. Long-term depression will weaken the synaptic strength,19 especially in the critical developmental period. Perhaps the underdevelopment of WM, in turn, prolongs neural transmission and impairs the connection between the visual and visual-related cortices.
An interesting phenomenon is that the right precuneus showed increased FA but with lowered WMV. Considering factors influencing FA, ordered packing of fibre tracts may be the reason. Additionally, WMV increased in the right cuneus, the middle occipital area and the left orbital frontal cortex (OFC), but had no changes in FA, which may be related to the growth in nerve fibres. Meanwhile, elevated GMV was seen in the lingual gyrus. We suggest that these modifications were due to the compensation effect. Similarly, Chan et al20 found greater GMV in strabismic adults relative to normal controls at the frontal lobe and subcortical regions. In a study of monocularly deprived young mice, there was an initial decrease in the stimulation of the deprived eye followed by potentiation of the responsiveness to stimulation of the non-deprived eye.21 In this study, the amblyopic subjects all had monocular amblyopia. Since the weaker eye could not achieve normal vision, visual information processing would rely more on the unaffected eye. Accordingly, the related visual information flow arose and the GM received more stimulus than their counterparts related to the amblyopic eye. Intensive practice could lead to measurable DTI changes22 and result in increased FA. The precuneus is involved in attentive tracking, visuo-spatial imagery and spatially guided behaviours.23 It has been reported that the precuneus may be involved in the generation of the spatial information that is necessary for imagined whole-body movements.24 As discussed above, in monocular amblyopia, stereoscopic vision is affected and increased FA in the precuneus may compensate for this defect.
Furthermore, one may postulate that cortical defects may occur in the amblyopic side and compensation may exist in the unaffected side. However, visual cortices are organised into ocular dominance columns, which are stripe-like and alternate between the left and right eye.25 The visual information from the nasal visual field of each retina crosses over to the opposite side of the brain via the optic nerve at the optic chiasm. The temporal visual field information, on the other hand, is transmitted to the same side. Therefore, in the unilateral hemisphere, there are both ipsilateral and contralateral visual cortical representations. Unfortunately, however, the resolution of our MRI images was not high enough to differentiate between the ocular dominance columns. Using high spatial resolution (0.5×0.5×3 mm3) fMRI, Goodyear et al26 reported smaller pixel numbers for the ocular dominance map corresponding to the amblyopic eye than the unaffected eye. Further studies with higher-resolution anatomical images will be conducted in the future.
In conclusion, this study extended our understanding of neural relationships with amblyopia. Using whole-brain analysis, we detected declined and augmented FA, GMV and WMV, indicating that monocular amblyopia has some special brain modification characteristics. At first, the cortices related to spatial vision, such as areas pertaining to the dorsal visual pathway, underwent volume loss, which provided direct neuroanatomical evidence for spatial vision defects, one of the most prominent impairments in monocular amblyopia. Then abnormalities in the optic radiation were confirmed from the microscopic perspective, and WM in the primary visual area was also hindered from development. Finally, a growth in the GMV, WMV and FA in the occipital lobe and precuneus may reveal a compensatory effect for visual defects by the unaffected eye in monocular amblyopia, which requires further confirmation with high-resolution images.
Acknowledgments
This study was supported by the Applied Basic Research Programs of Science and Technology Commission Foundation of Tianjin, People's Republic of China (grant no. 10JCYBJC10700).
References
Footnotes
-
Contributors Qian L was responsible for acquisition, analysis and interpretation of data; subject screening; drafting the article and final approval of the version to be published. Qingji L was responsible for acquisition and analysis of the data, drafting the article and final approval of the version to be published. MG was responsible for experimental design, analysis and interpretation of data, revising the manuscript critically for important intellectual content and final approval of the version to be published. Qingji L was responsible for subject recruitment and ophthalmologic examination of the subjects, revising the article for important intellectual content and final approval of the version to be published. CC was responsible for taking case histories and physical examination of the subjects, revising the article for some important clinical content and final approval of the version to be published. XY was responsible for acquisition of the data, revising the article and final approval of the version to be published.
-
Competing interests None.
-
Patient consent Obtained.
-
Provenance and peer review Not commissioned; externally peer reviewed.
Linked Articles
- Corrections