Effects of Visual Impairment and Its Restoration on Electroencephalogram During Walking in Aged Females
Visual decline is one of the most common manifestations of aging, significantly impacting motor abilities and increasing the risk of falls. The brain processes visual inputs through sensorimotor integration to make decisions for locomotor navigation. Understanding the effects of visual impairment and its rehabilitation on cerebral function during walking can provide valuable insights for developing strategies to protect motor abilities in the elderly through brain-level interventions. However, the cerebral mechanisms underlying these effects remain largely unexplored. Age-related cataract, a major cause of chronic vision deterioration, offers an ideal model for studying the impact of visual restoration on locomotion. This study aimed to analyze the effects of chronic visual impairment and its rehabilitation on sensorimotor integration during level walking in patients with age-related cataract, using electroencephalogram (EEG) recordings.
The study enrolled 14 female patients with age-related cataract, scheduled for consecutive cataract surgeries at the Department of Ophthalmology in Peking University Third Hospital. EEG signals during level walking were recorded using a portable EEG system before and 4 weeks after visual restoration. Walking speed was assessed using the Footscan system. Spectral power of the theta and alpha bands was analyzed with repeated-measures analysis of variance, considering Assignment (rest and walking), Phase (preoperative and postoperative), and Electrode sites (F3, Fz, F4, O1, and O2) as within-subject factors.
The results revealed that theta band power significantly decreased after visual restoration compared to the visual impairment state. Specifically, theta activity was notably reduced during walking, while theta power at rest remained stable. Changes in walking speed were correlated with alterations in theta power at electrode sites of O1 and O2. Alpha band power remained stable during walking and was unaffected by visual status. These findings suggest that chronic visual impairment from age-related cataract triggers enhanced cerebral activation of sensorimotor integration to compensate for visual decline during locomotion. This cerebral over-activation is effectively alleviated by visual restoration.
Introduction
Visual decline severely impacts motor ability and increases the risk of falling in the elderly. Through sensorimotor integration, the brain processes visual inputs and makes decisions for locomotor navigation. Investigating the impacts of visual impairment and its rehabilitation on cerebral function during walking could provide valuable insights for developing strategies to protect motor ability in the elderly through brain-level interventions. However, cerebral function related to sensorimotor integration during walking, both before and after visual restoration, has not yet been explored in elderly individuals. Age-related cataract is a major cause of chronic vision deterioration and plays a critical role in mobility deficits in the elderly. The satisfactory clinical outcomes of cataract surgery offer an ideal model for studying the effects of visual restoration on locomotion. Previous studies have demonstrated significant improvements in walking patterns and motor symmetry in elderly individuals following cataract surgery. However, the changes in cerebral activity during walking associated with optical restoration require further analysis to better understand the cerebral mechanisms underlying the effects of visual restoration on mobility.
Apart from the effects of visual impairments, the decline in sensorimotor integration has been recognized as a key contributor to age-related mobility deficits. To maintain normal motor performance, older adults recruit additional cortical or subcortical areas during motor control to compensate for the deterioration of cerebral function. This cerebral over-activation results in age-associated attentional movement control, which increasingly depends on somatosensory information and visual feedback. Thus, visual impairment could exacerbate the functional load on sensorimotor integration. In contrast, visual restoration could effectively alleviate this excessive activation. Based on functional magnetic resonance imaging (fMRI) before and after cataract surgery, previous studies have shown that visual restoration induces morphological and functional changes in regions related to sensorimotor integration in the elderly, indicating improved functioning of brain networks associated with motor control. By using a mobile brain/body imaging system, the study of electrocortical activity synchronous with walking provides further insight into the role of visual rehabilitation in protecting cerebral function related to sensorimotor integration in the elderly.
Observing patients with age-related cataract, a clinical model of chronic visual decline, the present study aimed to assess the impacts of visual impairment and its restoration on cerebral function associated with locomotion in elderly individuals. We hypothesized that (1) to compensate for the decline in visual inputs, there would be excessive activation of cerebral functions related to sensorimotor integration during level walking, and (2) optical reconstruction through cataract surgery would mitigate this functional over-activation. In line with previous investigations, theta band activity in EEG has been shown to be significant in sensorimotor integration during walking, while alpha band power is related to cerebral involvement in motor control. In addition, alterations in cerebral activity synchronous with locomotion resulting from visual manipulations have been found to be more prominent in the frontal and occipital regions. Consequently, in this study, we specifically analyzed the power spectrum of the theta and alpha bands recorded from the frontal and occipital regions.
Methods
The study design adhered to the principles of the Declaration of Helsinki for medical research involving human subjects, and ethics approval was obtained from the Institutional Review Board of Peking University Third Hospital. Informed consent was obtained from all participants after explaining the nature and potential consequences of the study.
This prospective case series recruited consecutive patients referred for standard minimal-incision phacoemulsification combined with intraocular lens implantation under topical anesthesia, performed by experienced surgeons from the same team in the Department of Ophthalmology at Peking University Third Hospital. Inclusion criteria were: diagnosis of binocular age-related cataract; eligibility for consecutive cataract surgery; scheduled for implantation of monofocal intraocular lenses; female patients aged 60–75 years; uncorrected distant visual acuity (UCDVA) equal to corrected distant visual acuity (CDVA) preoperatively, not worse than 20/200 on the Snellen chart; neurologically healthy, as confirmed by self-report and a preoperative medical examination by a physician; and right-foot dominance. Exclusion criteria included: previous ocular surgery or laser treatment; an interval longer than 1 week between the first and contralateral eye surgeries; postoperative visual acuity worse than 20/25 on the Snellen chart 4 weeks after cataract surgery; need for refractive correction postoperatively; intraoperative or postoperative complications; detection of macular disease; neuro-ophthalmic disease; uncontrolled hypertension; history of diabetes; cardiovascular or cerebrovascular disease; vestibular or cerebellar dysfunction; mental illness; neurological motor disorders; use of psychotropic drugs; musculoskeletal injuries within 6 months before the study; symptoms of neurological disease detected during physician consultation; or locomotor deficits identified by a physiotherapist in the gait laboratory.
All participants underwent a standardized baseline ophthalmological assessment preoperatively. The examinations included visual acuity, refraction, slit-lamp biomicroscopy, fundoscopy, and non-contact tonometry. Lens opacity was classified by two ophthalmologists according to the Lens Opacities Classification System III (LOCS III) criteria under slit-lamp examination. Postoperative ophthalmological assessments were performed at 1 day, 1 week, and 4 weeks, with the same items as the preoperative assessment.
EEG signals synchronous with level walking were recorded for each participant in the week before the first eye surgery and 4 weeks after the contralateral eye surgery, once refraction and visual restoration had reached a stable status. Participants were required to complete five barefoot walking trials at their usual pace, each covering 10 meters. Before the formal tests in both the pre- and postoperative phases, participants were provided detailed instructions on the testing process. Practice trials were conducted to familiarize them with the procedures and minimize the influence of learning effects on the results. Data from the five successful walking trials were collected, and the average was analyzed.
The tests were performed in a professional interdisciplinary laboratory at our hospital. The gait assessment device, coupled with a wearable wireless EEG device, formed the mobile brain/body imaging system. Temporal-spatial gait parameters were recorded by a 2-meter Footscan plate system. The data of gait cycle, stride length, and walking base were extracted. The walking speed was calculated.
EEG data synchronous with each walking trial were recorded by a wireless 32-channel dry EEG system, with a sampling rate of 1000 Hz. The EEG signals were bandpass-filtered from 0 to 100 Hz. Silver/silver chloride (Ag/AgCl) electrodes were mounted on a custom-made cap according to the international 10–20 system, with electrode impedance maintained below 5 kΩ. All recorded signals were referenced to the left mastoid and re-referenced by bio-mastoid average. The ground electrode was placed on the frontal head. During EEG recording while walking, participants were instructed to fix their gaze on a wall at eye level in front of them and to remain relaxed throughout the experiment, minimizing head movement such as turning or bobbing while walking. The EEG signal at rest was recorded for 3 minutes while the subject was seated on a chair with eyes closed and awake. The rest-phase EEG was recorded before the walking trials. Offline analysis was performed using WinEEG software, with a low-pass filter at 30 Hz. Independent component analysis (ICA) was used to eliminate blink artifacts and those related to the contraction of frontal, temporal, or neck muscles. EEG epochs were segmented and baseline corrected. A 1000 ms epoch with a 100 ms time window overlap was extracted from the continuous data. Epochs exceeding 80 µV were excluded as artifacts. The spectral power of theta (5–8 Hz) and alpha (9–13 Hz) bands was calculated for the frontal sites (F3, Fz, and F4) and occipital sites (O1 and O2), respectively.
Statistical analyses were conducted using SPSS version 22.0. The normality of the data distribution was assessed using the Kolmogorov–Smirnov test. Descriptive statistics for normally distributed continuous variables, including preoperative visual acuity, gait cycle, stride length, walking base and walking speed, were expressed as the mean ± standard deviation. Non-normally distributed variables, such as postoperative visual acuity, were presented as the median (P25 and P75). A two-tailed paired t-test was used to compare gait parameters before and after cataract surgery. The Wilcoxon rank-sum test was applied for comparisons of visual acuity. EEG data were analyzed using repeated-measures analysis of variance (ANOVA), with “Assignment” (rest and walking), “Phase” (preoperative and postoperative), and “Electrode sites” (F3, Fz, F4, O1, and O2) as within-subject factors. The EEG results are presented as the mean ± standard error. A significance level of P <0.05 (two-sided) was used, with the Greenhouse–Geisser correction applied to the P-values. Spearman correlation analysis was performed to investigate associations between postoperative changes in visual acuity, changes in walking speed, and the changes in the spectral power of the theta band at each electrode site. For the correlation analysis, visual data were averaged for both eyes. A sample size calculation, conducted using PASS 11.0 software based on data in pilot study, indicated that 11 participants would be required to detect a significant difference in the spectral power of the theta band during walking between the preoperative and postoperative phases, with 80% statistical power and an α level of 0.05. Considering a potential dropout rate of 50%, a total of 16 participants were recruited.
Results
Sixteen female patients, referred for cataract surgery at our hospital, were enrolled in the study. Two patients were lost to follow-up at the postoperative phase due to their refusal to undergo the EEG test. Ultimately, 14 patients were included in the analysis. For these 14 patients, uncorrected distant visual acuity (UCDVA, logMAR) for the right and left eye both significantly improved 4 weeks after cataract surgery. Walking speed increased from 1.09 ± 0.26 m/s at the preoperative phase to 1.22 ± 0.26 m/s at the postoperative phase. The change in walking speed was reflected as a decrease in gait cycle. The stride length was comparable between the preoperative and postoperative phases. No statistically significant difference was detected in comparison of walking base.
The three-way ANOVA for EEG band power in the theta frequency revealed a significant main effect of Assignment, indicating that the power of the theta band was significantly higher during level walking compared to rest status. The main effect of Phase was also significant, with significantly lower theta activity at the postoperative phase compared to the preoperative phase. The main effect of Electrode sites was not statistically significant. The interaction of Phase × Assignment was significant. Subsequent multiple comparison tests revealed that, compared to the rest status, theta activity increased significantly during walking, both in the preoperative phase and the postoperative phase. Compared to the preoperative phase, the power of the theta band significantly decreased during walking in the postoperative phase. However, theta power at rest did not differ significantly between the two phases. These findings suggest that elderly individuals recruit cerebral functions related to theta activity during level walking to compensate for visual impairment. After visual restoration, the overactivation of theta activity is alleviated, as evidenced by the reduced theta activity during walking.
No other interactions reached statistical significance. The spectral power of the theta band under different conditions at the relevant electrode sites is presented in Supplementary Table 2.
A potential correlation between changes of walking speed and theta power at electrode sites of O1 and O2 was observed. To explore the potential association among functional parameters, we conducted a correlation analysis between postoperative changes in visual acuity, walking speed, and the spectral power of the theta band at the five electrode sites. The correlation analysis between changes in visual acuity and changes in spectral power of the theta band at the five electrode sites revealed no significant results, either at rest or during walking. Statistically significant correlations were observed between changes in walking speed and changes in theta band power at the electrode sites of O1 and O2 during the rest status. This potential relationship suggests that cerebral activity in the occipital regions may play an important role in the changes in sensorimotor integration mediating the effects of visual restoration on walking performance.
The three-way ANOVA detected no significant main effects of Phase, Assignment, or Electrode sites for the alpha band power. The interactions of Phase × Assignment, Phase × Electrode sites, Assignment × Electrode sites, or Phase × Assignment × Electrode sites were not statistically significant. These results indicate that the spectral power of the alpha band remained stable during level walking and was not affected by visual restoration. The spectral power of the alpha band under different conditions at the relevant electrode sites is presented in Supplementary Table 3.
Discussion
In the field of brain science, cerebral activity synchronized with locomotion has been a focal point of current research. However, the changes in cerebral function induced by visual restoration during walking remain largely unclear. By observing patients with age-related cataracts before and after cataract surgery, this study aimed to assess how chronic visual impairment and its restoration affect electrocortical activity during level walking in elderly individuals. The results showed that the brain recruited additional cerebral functions related to sensorimotor integration to compensate for blurred visual inputs and maintain locomotor navigation, as evidenced by enhanced theta activity. After visual restoration, this overactivation of cerebral function was effectively alleviated. These findings highlight the importance of visual restoration, such as cataract surgery, in preserving motor function in the elderly by reducing cerebral load during walking. Electrocortical activities related to the theta band should be further explored as potential indicators of decompensated cerebral function in response to visual impairment, which may serve as an early warning for risk events.
In the present study, a significant increase in theta band activity was observed in patients with age-related cataracts during level walking. Previous studies in sighted individuals have shown changes in cerebral activity related to the theta band during walking with visual perturbation. For example, EEG analysis during treadmill walking revealed that walking with eyes closed led to significant theta desynchronization during the stance phase. In addition, studies have reported that subjects training in virtual reality with rotating visual views during treadmill walking exhibited increased theta power compared to those with unaltered visual input. We previously demonstrated that blurred visual inputs, simulating cortical cataract with foggy goggles targeted at a Snellen visual acuity of 20/60, elicited a generalization of theta activity from the occipital to the frontal regions in healthy young volunteers. Increased theta activity has also been observed during the decision-making phase of visual navigation when walking toward specific targets. In immersive virtual hallway locomotor tasks, with or without conflicts between self-motion and landmarks, theta activity in frontal-midline regions was considered a crucial index for the early processing of spatial representations during navigation. Theta band activity is known to be associated with sensorimotor integration in challenging locomotor tasks, such as responses to abrupt balance perturbations during treadmill walking, adaptation on perturbed stepping tasks, or mobility in indoor real-world environments. As a model of chronic progressive visual impairment, age-related cataracts affect cerebral functions related to walking navigation. Thus, the higher spectral power of theta band observed at the preoperative phase suggests that the brain compensates for visual decline by recruiting additional sensorimotor integration functions to maintain locomotor navigation during walking.
Based on our results, the spectral power of theta activity decreased after visual restoration. Several studies have examined neural changes following visual restoration in congenitally blind individuals. These studies found that the visual cortex’s response to non-visual stimuli was significantly reduced, indicating a decrease in compensatory neurofunctional remodeling. Age-related cataracts, which cause visual impairment, can be effectively treated, making it an ideal model to observe the impact of visual restoration on cerebral activity related to motor control in the elderly. A previous fMRI study in patients with age-related cataracts found that visual improvement after cataract surgery was accompanied by increased gray matter volume in areas important for vision processing, cognition, somatic movement, and somatosensory processing. Further research on global and local graph metrics in fMRI showed that, following cataract surgery, eigenvector centrality values in regions such as the inferior occipital gyrus, superior parietal gyrus, and several cerebellar areas significantly decreased. This was interpreted as attenuation of the over-activation in cerebral functions associated with motor control. In accordance with these findings, the postoperative decrease in theta band spectral power in our study provides direct electrocortical evidence of the alleviation of sensorimotor integration recruitment during walking, following visual restoration.
The correlation between changes in walking speed and the spectral power of the theta band in the occipital regions suggests that cortical activity related to visual processing may play a significant role in modulating sensorimotor integration in response to visual decline in the elderly. A study examining the relationship between gait speed and EEG signals in older adults found a significant increase in theta band power in 68% of participants with walking speeds below 0.8 m/s, indicating an early functional change associated with mobility decline. A recent study also identified theta band activation during treadmill walking as a neural signature of mobility-related brain function in elderly individuals at risk for cognitive impairment. The authors attributed the increase in theta activity to compensatory responses for age-related declines in sensorimotor integration. Functional near-infrared spectroscopy studies have shown that aging leads to excessive cerebral activation during walking. The results of that study indicated that this compensatory over-activation reached its resource limit in the 70s, contributing to gait impairments. Consistent with these findings, our results suggest that EEG signals, particularly in the theta band recorded from the occipital regions, may serve as a predictive marker for vision-related mobility decline in the elderly. Further investigation is needed to explore the correlation between electrocortical activities in the occipital region and areas directly involved in motor control, such as the cerebellum and basal ganglia.
According to our data, the alpha-band power remained stable during walking and was unaffected by visual restoration. Numerous studies have established that changes in alpha activity are associated with sensorimotor integration and active brain involvement in motor control. For example, suppression of the alpha band in the parieto-occipital cortex has been observed during visually guided precision stepping, highlighting its role in planning and executing leg movements with visual navigation. In response to visual inputs that conflict with somatosensory and vestibular inputs, alpha band power increases over the occipital cortex during postural control on a balance board, indicating an adjustment in visual weighting within sensorimotor integration. During treadmill walking on a balance beam, a prolonged increase in alpha power was detected across the occipital, temporal, sensorimotor, and posterior parietal cortices after intermittent visual occlusion. In our study, the lack of significant changes in alpha band power during walking can likely be attributed to the simplicity of the level walking task, which may not have been demanding enough to trigger alpha activity fluctuations. More challenging tasks, such as obstacle recognition or route planning, should be incorporated in future studies to observe the potential effects of visual rehabilitation on alpha activity. In addition, since the current study focused on cerebral function changes before and after visual restoration, the data collection sequence could not be randomized. The consistent alpha activity across both phases suggests that the impact of learning effects on the data was minimal.
Several limitations should be considered before drawing conclusions. To eliminate the influence of gender, only female patients were included in this study. Due to strict inclusion and exclusion criteria and the low compliance with EEG testing, the sample size was relatively small. Future studies should include a larger and more diverse participant group, including male patients, to enhance the generalizability of the results. The current study did not stratify participants based on the severity of preoperative visual impairment. A future study should investigate how different levels of visual impairment affect electrocortical activity during walking. In addition, the health status of the nervous system and cognitive function are crucial factors influencing electrophysiological responses to visual rehabilitation. For this reason, we used a self-controlled study design and focused on short-term changes in cerebral dynamics post-cataract surgery. However, fluctuations in physiological and psychological factors, such as fatigue and anxiety, may have acted as confounding variables. Therefore, including fMRI analysis and cognitive assessments via questionnaires in long-term studies would provide a more comprehensive understanding of changes in the nervous system and cognitive function. Finally, EEG signals contain complex dynamic information. Future studies should incorporate nonlinear dynamical parameters, such as approximate entropy, to capture the characteristic changes in cerebral function related to visual impairment and its restoration.
In summary, using age-related cataract as a clinical model, the present study examined the impact of chronic visual impairment and its rehabilitation on electrocortical activity during level walking in elderly individuals. The results indicated that blurred visual inputs led to over-activation of cerebral functions related to sensorimotor integration during walking in aged females. Visual restoration effectively alleviated the compensatory upregulation of functional load associated with locomotion. The theta band activity should be prioritized in investigations of predictive factors for declines in walking performance due to visual impairment, particularly in elderly females. Brain function interventions targeting theta frequency may serve as a promising method to enhance tolerance to age-related visual decline and protect motor ability in older adults.
doi.org/10.1097/CM9.0000000000003549
Was this helpful?
0 / 0