Vestibular and Oculomotor Function in Children with Cerebral Palsy: A Scoping Review

Abstract

Cerebral palsy (CP) is a nonprogressive permanent brain injury that causes an impairment of movement and posture. This scoping review aimed to answer the following questions: (1) “What is the status of oculomotor function in children with CP?” (2) “What is the status of vestibular function (i.e., gaze stability, perception of vertical, vestibular-related balance abilities) in children with CP?” Using Arksey's and O'Malley's five-stage framework, we searched six online databases for relevant articles. The inclusion criteria were: (1) participants of the studies included individuals with CP; (2) a primary outcome in the studies was measurement of oculomotor, vestibular, and/or balance; (3) studies were published within the past 20 years; and (4) the participants in the studies were between 0 and 21 years of age. Twenty-one articles were found that described impairments in oculomotor function ( n  = 9), vestibular function ( n  = 1), and oculomotor and vestibular integration ( n  = 11) in children with CP. The evidence suggests that children with CP may have altered saccadic and smooth pursuit eye movements, abnormal saccular function, poor eye–hand coordination, and abnormal use of vestibular information for balance. Future studies should explore peripheral and central vestibular function using reliable and valid methods for this population. This scoping review demonstrated a paucity of rigorous and objective research to describe the status of oculomotor and vestibular function in children with CP. However, preliminary studies suggest that more research is warranted.

Learning Outcomes: As a result of this activity, the participant will be able to (1) describe the methodology of a scoping review and (2) discuss the evidence available on oculomotor and vestibular function in children with cerebral palsy.

Cerebral palsy (CP) is a group of nonprogressive permanent disorders caused by disturbances to the fetal or infant's brain. ¹ This group of disorders affects the development of movement and posture causing activity and participation limitations. In addition to impairments of movement and posture, visual deficits are well documented in this population and include restricted visual fields, strabismus, reduced visual acuity, and abnormal refractive errors. ^{2 3 4 5 6 7} Recent studies also have documented oculomotor impairments to include abnormal saccades (i.e., quick eye movements to a randomly moving target), ^{6 8 9 10 11} volitional smooth pursuit (i.e., ability to follow a moving target), ^{6 8 11 12 13} and optokinetic nystagmus (OKN; i.e., involuntary smooth pursuit) ^{6 11 14 15} in children with CP. Visual and oculomotor impairments can affect important activities that include reading, driving (e.g., vehicles and power chairs), and responding to environmental stimuli. ^{16 17}

To better describe this population, researchers developed the Gross Motor Function Classification System (GMFCS), ¹⁸ which consists of five levels of mobility with level one (I) being the least impaired and level five (V) being the most impaired. Across the spectrum of these motor impairments in individuals with CP, limitations in vision and oculomotor abilities may affect the quality and quantity of participation in the community. Some examples include difficulty using technology such as eye gaze communication devices, problems with reading speed, longer fixation duration (e.g., fixating on in-vehicle objects), and less flexible horizontal search strategies when driving. ¹⁷

Gaze stability during head movement is made possible using the smooth pursuit system if the head movement velocity is less than 100 degree/second (e.g., during activities of daily living such as reading or watching TV). ¹⁹ However, activities involving head movements faster than 100 degree/second or 2 Hz (e.g., walking, running, crawling, jumping, playing sports) gaze stability can only be achieved with a functional vestibulo-ocular reflex (VOR). The VOR maintains a stable image on the retina by producing a slow-phase eye movement in the opposite direction but same velocity and amplitude as the head movement. ^{20 21} The vestibular system is not only responsible for gaze stability but also for perception of one's orientation in space, and balancing in challenging sensory environments (e.g., unstable surface and visual cues). ^{21 22} As researchers consider potential sensory and motor impairments in children with CP, they must also consider whether the peripheral vestibular end organs are properly functioning, and whether or not the corresponding central vestibular pathways are producing the proper motor output and sensory perception.

A scoping review is a process of mapping available evidence on a certain topic that has not been thoroughly reviewed in previous studies. A scoping review does not aim to evaluate the evidence; instead, it aims to present the breadth and/or depth of the evidence in the literature. ²³ One reason to conduct a scoping review is to examine the range, extent, and nature of research activity on a given topic. ^{23 24} Therefore, the purpose of this study is to review evidence on oculomotor and vestibular function in children with CP. The research questions are (1) “What is the status of oculomotor function in children with CP?” and (2) “What is the status of vestibular function (i.e., gaze stability, perception of upright orientation, and vestibular-related balance abilities) in children with CP?”

Methods

Arksey and O'Malley proposed a five-stage framework to guide the process of conducting a scoping review. ²³ The framework stages are (1) identify the research question; (2) identify relevant studies; (3) study selection; (4) chart the data (i.e., extract pertinent information from each study); and (5) compile, summarize, and report the results. Levac et al proposed recommendations to advance the utilization of Arksey's and O'Malley's framework to include acute focus on the research questions and purpose throughout the study, and constant communication between researchers in the stages of study selection. This will enable researchers to better refine inclusion criteria based on the extent of search results at each stage (i.e., beginning, midpoint, and final stages of the search process). ²⁴ We employed the recommendations of Levac et al in this scoping review.

Identifying the Research Question

We used the Arksey's and O'Malley's framework ^{23 24} to identify the research questions for this scoping review and hypothesized that oculomotor and vestibular function might contribute to the deficits in movement and postural control seen in children with CP. We kept our search broad and comprehensive, given the complexity of the topic. Therefore, our research questions are (1) “What is the status of oculomotor function in children with CP?” and (2) “What is the status of vestibular function (i.e., gaze stability, perception of upright orientation, vestibular-related balance abilities) in children with CP?” For this review, vestibular system function is operationally defined as any one of the following: (1) gaze stability (i.e., ability to see clearly with head movements greater than 100 degree/second or 2 Hz, mediated primarily by the semicircular canals), (2) perception of vertical and horizontal orientation (i.e., proper perception of upright mediated primarily by the utricle, or (3) vestibular-related balance abilities (i.e., ability to maintain balance when vision is either absent or in conflict at the same time that the support surface is compliant or moving, mediated primarily by the otoliths). The definition of oculomotor system function is as follows: (1) saccades (i.e., quick eye movement from one target to another, (2) smooth pursuit movement (i.e., ability to track a moving target), and (3) OKN (defined as the involuntary pursuit system driven by a full field moving target such as vertical stripes or dots).

Identifying Relevant Studies

We searched six online databases including PubMed, Embase, Scopus, CINAHL, Psych Info, and the Cochrane Library, to find relevant peer-reviewed articles. The keywords were a combination of “cerebral palsy,” “vestibular,” “inner ear,” “labyrinth,” “oculomotor,” “ocular motility,” and “visual abilities.” The combinations used as search terms always included “cerebral palsy” and one or more of the keywords (e.g., “cerebral palsy” AND “vestibular system”). Articles were included if they were published within the past 20 years and written in English. Published abstracts were also searched for relevant content.

Study Selection

Three reviewers (L.V., J.B.C., and A.A.) developed the inclusion criteria that were stated previously. We excluded studies that focused on interventions; until we truly understand the phenomenon of oculomotor and vestibular function in this population, it is difficult to know if interventions were appropriately addressing these impairments. Two reviewers (L.V. and A.A.) reviewed the abstracts independently and then discussed the selected abstracts with a third reviewer (J.B.C.) to resolve any conflicts. After a full article review (L.V. and A.A.), all three researchers (L.V., J.B.C., and A.A.) met to finalize article selection.

Charting the Data

The three reviewers (L.V., J.B.C., and A.A.) determined pertinent information to include in the data-charting form. After reviewing several articles independently, the reviewers met again to discuss the charting process and refine the data-charting form ( Appendix A ). ²⁴ Because of the breadth of the research question, a thematic criterion was used to choose and chart the selected articles. A theme is operationally defined as the primary focus of each article related to the scoping review research questions. The themes of interest included oculomotor function, vestibular function, and integration, defined as a primary measure that included functions that indirectly assessed the integration of vestibular and/or oculomotor functions such as eye–hand coordination or balance with eyes closed. The following information was included in the final version of the data-charting form: first author, year of publication, article focus (i.e., oculomotor, vestibular, integration), country of research, sample size, age range, description of population, inclusion/exclusion criteria, instruments used, data points, data analysis procedures, findings, and conclusion (see Appendix A ).

Appendix A: Data-Charting Variables used in the Scoping Review.

First author	The name of the first author of the study
Year of publication	The year of publication of the study
Article focus	The focus of the article based on the question of interest of the scoping review (i.e., oculomotor, vestibular, integration)
Country of research	The country where the research took place
Sample size	The number of the participants, specified for each group
Age range	The age range of the participants, specified for each group if different
Description of population	Any classification used to further describe the participant (e.g., GMFCS and/or topographic distribution of brain damage)
Inclusion/Exclusion criteria	Specific criteria for inclusion and exclusion of participants
Instruments used	The instruments that were used to implement methodology
Data points	The number of points in time where the methodology was implemented
Data analysis procedures	Statistical procedures in the study
Findings	Brief summary of the results of the study
Conclusion	The conclusion of the study

Abbreviation: GMFCS, Gross Motor Function Classification System.

Compiling, Summarizing, and Reporting the Results

After charting each study, the researchers compiled and summarized the data. This stage of the scoping review was described by Arksey and O'Malley ²³ as “thematic analysis.” The thematic analysis enabled the researchers to explore the studies in more depth and included a comprehensive summary of the geographical spread, years of publication, characteristics of individuals with CP, and outcome measures used to describe oculomotor/vestibular function and integration. To assist in summarizing data, studies were assigned to one of the three themes of interest (oculomotor function, vestibular function, and integration).

Results

The initial search produced 192 articles: 41 in PubMed, 20 in CINAHL, 47 in Embase, 32 in Scopus, 47 in Psych Info, and 5 in Cochrane. After reviewing the study abstracts, we excluded duplicates and irrelevant studies based on the inclusion criteria, resulting in 59 articles. Further review of abstracts, along with consulting with the third reviewer (J.B.C) to reach consensus, led to the exclusion of 38 articles. The exclusion of articles was primarily due to the study not meeting the inclusion criteria of the age range and/or the oculomotor variables measured. The final number of articles included in the scoping review was 21 articles; one of these was a published abstract. ²⁵

Study Characteristics

The 21 reviewed studies were published between 1999 and 2016. We did not find any published in 2017 or later. Four of these studies were from the United States, ^{6 26 27 28} and three articles were from Brazil. ^{13 14 29} Greece, ^{9 10} Sweden, ^{17 30} United Kingdom, ^{31 32} and Belgium ^{33 34} each produced two articles, while Italy, ¹¹ Turkey, ²⁵ Iran, ³⁵ Canada, ⁸ Israel, ¹⁵ and China ³⁶ each produced one article. Nine articles examined oculomotor function in children with CP. Eleven articles examined integration of vestibular and oculomotor systems in children with CP. Only one article examined the vestibular system in children with CP. ³⁵

Oculomotor Function in Children with CP

Nine articles discussed oculomotor function in children with CP. ^{6 8 9 10 11 13 14 15 33} In some studies, oculomotor function was included with other visual abilities ^{6 8 10 11 13 14 15} ( Tables 1 and 2 ). Of these, six studies examined OKN in children with CP. OKN is a reflexive eye movement induced by large-field motion of images, to reset them on the retina. ³⁷ The areas of the brain, cortical and subcortical, responsible for this reflex are the primary visual cortex (Brodmann area 17, V1), posterior parietal cortex, frontal eye field (FEF), supplementary eye field (SEF), temporal visual area, presupplementary motor area, basal ganglia, superior colliculus, vestibular nuclei, and cerebellum. ³⁸ Five studies assessed OKN subjectively by the examiner using a piece of cardboard with vertical stripes ^{11 13 14 15 39} and one study used digital infrared videography which measured the eyes in response to an OKN stimulus. ⁶ These studies showed that abnormal OKN (i.e., absent OKN response, or asymmetrical response to the right or the left) is evident in children with CP, ranging from 50 to 80% of those assessed in the research. Abnormal OKN affects the child's ability to fixate large-field images on the retina while moving. We hypothesize that this could contribute to the poor balance and difficulty with visual focus seen in this population. ^{11 37}

Table 1. Descriptive Summary of Studies Examining Oculomotor Function.

Author (year)	Sample	Age	GMFCS	Topographic distribution	Instrument	Variable tested
Guzzetta et al (2001)	CP ( n  = 47)	8 mo to 13 y	–	Hemiplegia ( n  = 47)	Clinical testing	OKN
Jan et al (2001)	CP ( n  = 14)	4 mo to 13 y	–	Dyskinesia ( n  = 14)	Clinical testing	Smooth pursuit Saccades
Kozeis et al (2006)	CP ( n  = 105) TD (no. not specified)	6–15 y	–	Spastic diplegia ( n  = 185) Other types ( n  = 5)	Clinical testing	OKN
Ghasia et al (2008)	CP ( n  = 50)	2–19.5 y	I ( n  = 10) II ( n  = 10) III ( n  = 10) IV ( n  = 10) V ( n  = 10)	–	Videography (digital infrared recordings)	OKN Smooth pursuit Saccades
da Cunha Matta et al (2008)	CP ( n  = 123)	4–12 y	I ( n  = 0) II ( n  = 80) III ( n  = 33) IV ( n  = 10) V ( n  = 0)	Diplegia ( n  = 44) Hemiplegia ( n  = 50) Triplegia ( n  = 14) Mixed ( n  = 15)	Clinical testing	OKN Smooth pursuit
Ferziger et al (2011)	CP ( n  = 77)	3–20 y	V ( n  = 77)	Spastic quadriplegia ( n  = 61) Athetoid ( n  = 5) Mixed ( n  = 8) Hemiplegia ( n  = 3)	Clinical testing	OKN
Fazzi et al (2012)	CP ( n  = 129)	3 mo to 15 y	–	Diplegia ( n  = 51) Hemiplegia ( n  = 17) Tetraplegia ( n  = 61)	Clinical testing	OKN Smooth pursuit Saccades
Kozeis et al (2015)	CP ( n  = 100) Pre-term ( n  = 70) Full term ( n  = 30)	6–15 y	–	Diplegia ( n  = 32) Hemiplegia ( n  = 28) Quadriplegia ( n  = 40)	DEM test	Saccades
Ego et al (2015)	CP ( n  = 51) TD ( n  = 78)	5–16 y	I ( n  = 33) II ( n  = 14) III ( n  = 3) IV ( n  = 1)	Diplegia ( n  = 22) Hemiplegia ( n  = 29)	Infrared eye tracker	Asked to follow a target [change direction (1st condition) > change direction and velocity (2nd condition]

Abbreviations: CP, cerebral palsy; DEM, developmental eye movement; GMFCS, Gross Motor Function Classification System; OKN, optokinetic nystagmus; TD, typically developed.

Table 2. Qualitative Summary of Studies Examining Oculomotor Function.

Author (year)	Research question(s)	Conclusion(s)
Guzzetta et al (2001)	Are visual abnormalities in children with congenital, or early acquired, hemiplegia related to type of lesion, timing of insult, site and size of the lesion?	Visual abnormalities are frequent in children with hemiplegia, and that the presence and the severity of these defects cannot be always predicted by MRI findings
Jan et al (2001)	Do dyskinetic eye movement disorders that are associated with CP mask the presence of good visual acuity and result in poor visual function?	Children with dyskinetic CP suffer from severe difficulties with voluntary smooth pursuit, saccades, and fixation, which is usually misdiagnosed with this population
Kozeis et al (2006)	Is microsaccadic abilities in school-aged children with CP related to reading skills?	Children with CP suffer severely affected microsaccade compared with TD children, which might affect their reading skills
Ghasia et al (2008)	Do children with CP with different severities, defined by GMFCS, have different degrees or types of visual dysfunction?	Visual deficits are evident in children with CP, and they differ in children who have mild vs. severe CP
da Cunha Matta et al (2012)	Is there a simple, practical protocol that can be used by nonspecialist, to assess visual function in children with CP that can be used in an outpatient sitting?	Clinical ophthalmological evaluation is a feasible tool to describe the visual abilities of children with CP
Ferziger et al (2011)	What is the integrity of the visual abilities in children with CP, level V on GMFCS?	Variability in visual performance along with limited communication abilities among children with profound neurological impairments presents a challenge for accurate and meaningful measurement of visual function
Fazzi et al(2012)	1. What is the overall clinical picture of visual dysfunction in children with CP? 2. Does it differ with different types of CP?	Children with CP suffer from a wide spectrum of visual deficits. Characterization of the clinical picture of CP can be based not only on the topography and characteristics of the motor deficit but also on neuro-ophthalmological aspects
Kozeis et al (2015)	Does being born preterm impose the risk of developing microsaccades deficits in children with CP?	Preterm birth in children with CP is not an additional risk factor for the development of microsaccades disorders
Ego et al (2015)	1. Do children with CP suffer from oculomotor deficits and how do these deficits evolve during development? 2. Is there any link between the oculomotor deficit of children with CP and their motor impairment (side of hemiplegia)?	Oculomotor function spontaneously improves with age in children with CP, even faster than typically developing children for some of the parameters. This improvement indicates that the lesioned brain of children with CP retains the ability to reorganize

Abbreviations: CP, cerebral palsy; GMFCS, Gross Motor Function Classification System; TD, typically developed.

Five articles reported abnormal saccadic eye movements in children with CP. ^{6 8 9 10 11} Saccadic eye movement is controlled by many cortical and subcortical brain areas including FEF, SEF, V1, dorsolateral prefrontal cortex, cingulate cortex, pontine reticular formation, superior colliculus, and cerebellum. ³⁸ Two of these studies ^{8 11} used clinical methods where the examiner asks the child to look from one target to another (e.g., holding up two fingers). These studies found that 80 to 100% of participants with CP had altered saccadic eye movement, which was defined differently across studies (e.g., dysmetric, absent, or inaccurate). ^{8 11} Two studies ^{9 10} used the developmental eye movement (DEM) test, a verbal-based outcome measure where the child is asked to read numbers from a page and the examiners count the errors that the child makes. These studies found that 73 to 80% of children with CP tested had abnormal results, with more than 50% of these results attributed to problems with saccadic eye movements and/or perceptual disabilities (i.e., type 2: oculomotor dysfunction, and type 4: automaticity and oculomotor dysfunction, of DEM test). ^{9 10} Another study used infrared ocular motor recordings to assess saccadic eye movements. ⁶ In this study, participants were asked to follow a target, while it moved horizontally. This study found that 98% of children with CP had abnormally long saccadic latencies and/or subnormal saccadic amplitudes. ⁶ Despite the different methodologies used in these studies, the evidence shows that saccadic eye movements are impaired in children with CP, regardless of GMFCS level or motor impairment distribution. Altered saccadic eye movement affects a child's ability to accurately switch between two or more targets. This could contribute to participation restrictions such as difficulty reading, playing sports, or simply shifting focus from one place to another while walking.

Four studies examined volitional smooth pursuit abilities, ^{6 8 13 33} defined as the ability to smoothly follow a target that is moving side to side or up and down. As with saccades, smooth pursuit involves many cortical and subcortical areas to include FEF, SEF, V1, cingulate cortex, temporal visual area, pre-supplementary motor area, basal ganglia, superior colliculus, brainstem reticular formation, and cerebellum. ³⁸ Two of the studies ^{8 13} used a clinical method to examine smooth pursuit where the child was asked to follow a single target that was moved horizontally or vertically in room light. The examiners scored the test by using categories (i.e., continuous/interrupted or absent/saccadic/normal). These investigators found that children with CP had abnormal smooth pursuit abilities (100 and 26% ³ ). ^{8 13} Another study ⁶ examined smooth pursuit using infrared eye tracking technology and a laser target. Abnormal smooth pursuit abilities were operationally defined as eye-lagging and/or using catch-up saccades. They found that 48% of children with CP had abnormal smooth pursuit abilities. ⁶ A study by Ego et al focused on examining smooth pursuit in children with CP using eye trackers. ³³ In this cross-sectional study of 51 children with CP between the ages of 5 and 16 years, GMFCS levels I–IV, the authors described smooth pursuit function in children with CP compared to preterm children without brain lesions and an age-matched control group. The participants were asked to follow a laser target, which, at certain time points, changed velocity, direction, or both. The analysis of initial eye acceleration and pursuit gain through different velocities showed that children with CP, aged between 14 and 16 years, had smooth pursuit abilities that were similar to the control group. These results led the authors to conclude that smooth pursuit function in children with CP improves with age. ³³ However, the study was cross-sectional rather than longitudinal and, therefore, the participants' smooth pursuit abilities as young children were unknown. Smooth pursuit is related to many activities (e.g., reading, school work, and plying sports); any dysfunction would compromise the child's participation in their community and with their peers.

Vestibular Function in Children with CP

The only article that examined function of the vestibular system was by Akbarfahimi and colleagues ³⁵ ( Tables 3 and 4 ). These investigators examined cervical vestibular-evoked myogenic potential (cVEMP) in children with CP between the ages of 7 and 12 years, GMFCS levels I–II (i.e., the most functional levels), compared with an age-matched group of children who were typically developing. The results indicated that cVEMP was a feasible test to use for children with CP, GMFCS levels I and II. The amplitude asymmetry ratio (AAR), although not statistically significant, was higher in children with CP when compared with the control group. ³⁵ In the group of children with CP that had a normal AAR, the P ₁₃ –N ₂₃ amplitude of both ears was significantly smaller when compared with control group ( p  < 0.001). ³⁵ The authors indicated that not only was there an asymmetry but an overall decrease in saccular function, evident by the smaller P ₁₃ –N ₂₃ amplitudes children with CP exhibited. ³⁵ No study could be found that measured VOR function or utricular function (i.e., ocular VEMP [oVEMP]) in children with CP. Also, we did not find any studies that examined perception of upright (i.e., subjective visual vertical [SVV] and horizontal) in this population. We hypothesize that if children with CP have abnormal function of the peripheral vestibular systems or its central pathways, it might contribute to the poor postural control and visual focus problems seen in this population, especially when combined with an abnormal oculomotor system.

Table 3. Descriptive Summary of Studies Examining Vestibular Function.

Author (year)	Sample	Age	GMFCS	Topographic distribution	Instruments	Procedure
Akbarfahimi et al (2016)	CP ( n  = 31) TD ( n  = 31)	7–12 y	I ( n  = 9) II ( n  = 21)	Diplegia ( n  = 5) Hemiplegia ( n  = 15) Quadriplegia ( n  = 11)	Two-channel cVEMP test	Both groups went through cVEMP testing protocol

Abbreviations: CP, cerebral palsy; cVEMP, cervical vestibular-evoked myogenic potential; GMFCS, Gross Motor Function Classification System; TD, typically developed.

Table 4. Qualitative Summary of Studies Examining Vestibular Function.

Author (year)	Research question(s)	Conclusion(s)
Akbarfahimi et al (2016)	What is the integrity of the saccular function of the vestibular in children with CP (7–12 y) when using compared with healthy age-matched control participants?	cVEMP is feasible in children with CP in assessing the saccular function. Significant differences in cVEMP parameters were noted between children with spastic CP when compared with TD children

Abbreviations: CP, cerebral palsy; cVEMP, cervical vestibular-evoked myogenic potential; TD, typically developed.

Integration of Vestibular and Oculomotor Systems in Children with CP

Eleven articles ^{17 25 26 27 28 29 30 31 32 34 36} described vestibular and oculomotor integration through the following functions: static balance, eye–hand coordination, and head stability ( Tables 5 and 6 ). Six articles examined static standing balance in children with CP compared to a control group of children who were typically developing. ^{25 26 29 30 31 36} Four of them ^{25 26 31 36} measured static standing balance using force plates, either custom or commercially available systems, while two studies used a custom moving room ²⁹ and a three-dimensional motion analysis system. ³⁰ The common variable measured in these studies was the sway score, operationally defined as the mean center of pressure (COP) and the sway path length. These studies reported that children with CP demonstrated higher sway frequency and amplitude, especially when the vestibular system was required to maintain balance (i.e., unstable surface with eyes closed). ^{25 26 29 30 31 36} One article ³⁰ also showed that these results are significantly different in subgroups of CP, for example between children who required support and those who do not. Without extensive testing, clinicians would not know if the balance impairments experienced by children with CP, typically attributed to a poor neural drive, would be compounded by an aberrant vestibular and/or oculomotor system.

Table 5. Descriptive Summary of Studies Examining Oculomotor and Vestibular Integration.

Author (year)	Sample	Age	GMFCS	Topographic distribution	Instruments	Procedure
Cherng et al (1999)	CP ( n  = 7) TD ( n  = 14)	–	–	Diplegia ( n  = 7)	A force platform system	CTSIB): (1) full vision and fixed foot support; (2) occluded vision and fixed foot support; (3) sway referenced vision and fixed foot support; (4) full vision and compliant foot support; (5) occluded vision and compliant foot support; (6) sway referenced vision and compliant foot support
Dan et al (2000)	CP ( n  = 12) TD ( n  = 12)	3–12 yo	–	–	Optoelectric ELITE system (motion analysis system)	Reflective markers were placed on the lateral aspect of the nose at the height of the infraorbital edge and the ear tragus, during rapid squatting from standing position with arms extended and rapid standing up from stable squatting
Falkmer and Gregersen	CP ( n  = 15) TD ( n  = 20) experienced drivers ( n  = 20)	–	–	–	Head-mounted eye tracker, driving Chrysler Voyager equipped with an NTSC video recording	Visual strategies were tested during driver's education sessions
Donker et al (2008)	CP ( n  = 10) TD ( n  = 9)	5–11 y		Hemiplegia ( n  = 10)	Custom-made strain gauge force plate and vertical screen for COP feedback	Three conditions (EO, EC, COP feedback)
Saavedra et al (2010)	CP ( n  = 10) TD ( n  = 10)	CP (6–16 y) TD (4–15 y)	I ( n  = 6) II ( n  = 2) III ( n  = 2)	Hemiplegia ( n  = 3) Diplegia ( n  = 4) Ataxia ( n  = 2) Mixed ( n  = 1)	An ASL remote eye tracker/computer screen and magnetic sensors	4 tasks × 3 support conditions Tasks: Control task —maintaining central fixation when a peripheral target appeared Eye-only task —quickly look at the target Eye–hand task —concurrently looked at and touched the target Hand-only task —quickly touched the target while maintaining central fixation. Support conditions: No support Upper torso support —providing an external brace at the level of the xiphoid process Hip suppor t—the pelvis was stabilized in vertical alignment with straps
Saavedra et al (2000)	CP ( n  = 15) TD ( n  = 26) young healthy adults ( n  = 11)	CP (6–16 y) TD (4–14 y)	I ( n  = 10), II ( n  = 2), III ( n  = 3)	Spastic hemiplegia ( n  = 3) Spastic diplegia ( n  = 5) Dystonia ( n  = 1) Ataxic ( n  = 4) Mixed ( n  = 2)	Magnetic tracking	Sensor placed on the participants' center of the forehead and sitting quietly. [2 visual conditions (EO/EC) × 3 support conditions (no support/pelvic support/torso support)]
Barela et al (2011)	CP ( n  = 15) TD( n  = 15)	8–16 y		Spastic hemiplegia ( n  = 15) (L = 6/R = 9)	A moving room through a custom software (3 white walls and 4th wall is covered by vertical black stripes) An infrared emitter	Participants were asked to stand as still as possible, 1 m in front of the frontal wall, barefoot, and hands beside their body. 9 trials [2 visual condition (EO/EC), 2 frequencies (0.2 Hz/0.5 Hz), 2 amplitudes (1.2 cm/3.2 cm), 2 peak velocity (0.6 cm/s/3.5 cm/s)]
Bustamante Valles et al (2011)	CP ( n  = 29) TD ( n  = 20)	5–18 y		Diplegia ( n  = 17), Hemiplegia ( n  = 12)	Two six-axis force plates/A monitor provided visual feedback to the test subjects	Participants were asked to stand quietly during 3 different conditions (EC, EO [stare at the dot on screen], and EOF [keeping COP as stable as possible]
van Kampen et al (2012)	CP ( n  = 10)	11–19 y		Hemiplegia ( n  = 10)	A DV camera recorded the grasping action of the participants. MobileEyeTM was used to record the visual point of gaze	Walk toward the belt, grasp the ball when it moved through the interception [2 velocities conditions × 2 occlusion visions (no occlusion/occlusion) × 2 approach conditions (ipsilateral/contralateral)]
Lidbeck et al (2016)	CP ( n  = 36) TD ( n  = 27)	6–17 y	I ( n  = 5) II ( n  = 12) III ( n  = 2)		3D eight-camera motion analysis system/Surface electromyography	Neuro-ophthalmological (saccadic movements, smooth pursuit movements, fixation 3 tasks (no task, blindfolded, attention task—watching a video)
Ozal et al (2016)	CP ( n  = 20) TD ( n  = 20)	5–17 yo	I and II (no. not specified)	Hemiplegia ( n  = 10) Diplegia ( n  = 10)	Computer dynamic posturography (sensory organization test)/timed up and go and pediatric balance scale	Both groups were tested using all instruments

Abbreviations: 3D, three-dimensional; COP, center of pressure; CP, cerebral palsy; CTSIB, clinical test of sensory interaction in balance; EC, eye closed; EO, eye opened; EOF, eyes opened with feedback; GMFCS, Gross Motor Function Classification System; TD, typically developed.

Table 6. Qualitative Summary of Studies Examining Oculomotor and Vestibular Integration.

Author (year)	Research question(s)	Conclusion(s)
Cherng et al (1999)	What is the difference in static standing balance between children with diplegic CP and nondisabled children under altered sensory conditions?	The children with spastic diplegic CP had more difficulties in maintaining standing balance than gender- and age-matched nondisabled children under sensory conflict conditions
Dan et al (2000)	Does head stability differ during different visual conditions in children with CP when compared with age-matched control?	Children with CP exhibit high head movement during activities when compared with TD children
Falkmer and Gregersen (2001)	What is the fixation pattern of young adults with CP during driver's education sessions compared with typically developed young adults and experienced drivers?	Driver's education programs should be tailor when serving learners with CP, and they should include a program teaching visual search strategies to this population
Donker et al (2008)	What is the effect of visual information on postural sway in children with CP when compared with typically developed children?	Visual deprivation (EC condition) influenced sway characteristics differently in CP and TD children
Saavedra et al (2009)	What is the integrity of functional coupling of eye, head, and hand during reaching in children with CP in various diagnostic groups?	Children with CP, like TD 4–6 y olds, have rapid accurate saccadic responses but initiate and complete hand movements more slowly than their peers; they use more concurrent head movement and have significantly greater difficulty isolating eye, head, and hand movements
Saavedra et al (2010)	Does trunk control and vision contribute to head stability during quit sitting in children with CP?	1. Children with CP have deficits in postural mechanisms of head stability in the sagittal plane and trunk control that affect head stability in the frontal plane 2. The head stability in the frontal plane is impacted by the type of movement disorder in children with CP
Barela et al (2011)	1. What is the integrity of coupling abilities between visual information and body sway in children with CP? 2. Are children with CP able to adapt with changes in coupling abilities between visual information and body sway?	Children with CP have similar sensory–motor coupling skills but more variable compared with TD children. Children with CP also showed adaptive properties, but it was less calibrated and did not persist, when compared with TD children
Bustamante Valles et al (2011)	Does postural sway behavior differ under different sensory conditions between typically developing children and children with CP? 2. Do these differences extend to be evident between children with spastic diplegia and spastic hemiplegia?	1. Time domain parameters were significantly increased in both CP groups with eyes open when compared with controls. Significant differences were not found in the frequency domain parameters 2. Under the EC condition, only mediolateral metrics were sensitive enough to discriminate among the CP and control groups 3. Body sway deficits in diplegic CP and hemiplegic CP are significantly different
van Kampen et al (2012)	1. What is the integrity of gaze pattern during interceptive movement in children with spastic unilateral CP? 2. Do these gaze patterns relate to lesion side?	1. The gaze behavior of children with spastic unilateral CP during interceptive movement is anticipatory 2. Differences in gaze behavior, especially those patterns which were involved in planning, were dependent on lesioned side (right hemiplegia vs. left hemiplegia)
Lidbeck et al (2016)	What is the influence of visual stimuli on standing posture in children with CP that might, or might not, require support when compared with typically developing children?	Visual stimuli influenced standing posture in children with CP. The removal of the visual stimulus would cause for more deviation in the standing posture in children who require support, when compared with those who don't require support
Ozal et al (2016)	What is the effect of visual function on balance and functional movement skills in children with CP?	Visual function is an important aspect in children with CP, especially in terms of balance and functional skills

Abbreviations: COP, center of pressure; CP, cerebral palsy; EC, eye closed; EO, eye opened; TD, typically developed.

Three studies measured eye–hand coordination in children with CP in different environments; ^{17 27 32} all used computerized eye trackers to examine eye movement patterns. Falkmer and Gregersen ¹⁷ examined visual strategies during driver education sessions in young adults with CP compared with young adults with typical development and experienced drivers. ¹⁷ These investigators found that young adults with CP lack flexibility in visual search abilities and tend to keep their focus and attention close to the vehicle. Saavedra et al ²⁷ examined the functional employment of eye, hand, and head coupling of children with CP of different functional (e.g., GMFCS) and physiological levels (e.g., spastic hemiplegia). The investigators examined postural demand as the children performed three tasks—(1) control task: maintain central fixation of the eyes when a peripheral target appeared; (2) eyes only task: looking at the peripheral target; (3) eye–hand task: touching the target while maintaining central fixation; and during three different support conditions—(1) no support, (2) upper torso support, and (3) hip support. Their results showed that the saccadic responses of children with CP were rapid and accurate. However, it was evident that children with CP had difficulties isolating eye, hand, and head movements, when compared with their peers with typical development. ²⁷ van Kampen et al measured the gaze pattern during anticipatory grasp tasks in children with spastic hemiplegic CP. ³² A tennis ball was placed on a moving conveyor belt, and children were asked to walk to the belt and grasp the tennis ball. Children completed the task during various conditions including two different belt velocities (50 and 70% of the child's maximum walking speed), tunnel occlusion of the ball on the belt versus no occlusion, and grasping the ball with the more involved side versus the less involved side. ³² This study showed that the longer the child spent viewing the target, the more accurate the performance. They also found that children with right hemiplegia spent longer time viewing the object before starting the task, when compared with children with left hemiplegia.

Only two articles examined head stability in children with CP. ^{28 34} Dan et al used a motion analysis system to examine the head stability of children with spastic diplegic CP, compared with children who were typically developing, aged 3 to 12 years, during two tasks: squatting from standing position with hands extended forward and standing from squatting position with hands extended forward. Children with spastic diplegic CP had significantly worse head stability when compared with typically developing children who maintained their heads relatively stable throughout the task. ³⁴ To examine the effect of postural support and visual deprivation on head stability, Saavedra et al assessed head movement in children with CP (GMFCS levels I–III) while they were sitting on a backless bench, during two visual conditions (eyes open [EO] and eyes closed [EC]), and three postural support conditions (no support, pelvic support, and torso support). ²⁸ During EO and EC conditions, children with CP had significantly greater head displacement in the frontal plane, compare with children who were typically developing during no support and pelvic support conditions. In the sagittal plane, children with CP had greater head displacement and velocity during all support conditions, compared with children who were typically developing. ²⁸ Therefore, if the VOR is not functioning in these children, they may experience disruptions in gaze stability with head movement. The integration between vestibular and oculomotor function is related to exploring the environment while moving and during different challenges. Impairments affecting this integration would affect a child's ability to discover the world around them.

Discussion

This scoping review aimed to answer the following questions: “What is the status of oculomotor function in children with CP?” and “What is the status of vestibular function (i.e. gaze stability, perception of upright, vestibular-related balance abilities) in children with CP?” The review was needed to examine the range, extent, and nature of research activity on oculomotor and vestibular function in children with CP. ^{23 24} In many of the reviewed studies, the research questions were not focused on specific exploration of the oculomotor and vestibular systems, or the integration of oculomotor and vestibular function in children with CP. Disparity between the instrumentation used within each theme was also notable. Some studies used clinical methods, which tends to be subjective and less specific. ^{8 13 14 15} The few studies that used objective testing methods, such as eye-tracking systems, did not fully describe the oculomotor abilities in children with CP. ^{6 33} None of the articles discussed the complexity and difficulty of testing oculomotor or vestibular function in this heterogeneous population. For example, some of the articles combined GMFCS levels as one group for analysis when reporting their results when in reality the performance across GMFCS levels may vary widely. ^{10 11} Other studies included a wide age range of children with CP and combined their results, although age and developmental status could affect oculomotor and vestibular function. ^{6 8 11 14 15 25 34} The one study that measured vestibular function only measured the saccule, which includes only one-fifth of the entire vestibular system. ³⁵ Therefore, the depth and extent of oculomotor, vestibular, and integration problems that this population faces was not addressed and should be the focus of future studies.

Oculomotor Function in Children with CP

Although visual deficits are well established in children with CP, ^{2 5 11 40 41 42} only two articles used eye tracking to assess oculomotor abilities (i.e., saccades, smooth pursuit, OKN). ^{6 33} The other articles used clinical measures, which are more subjective. ^{8 11 13 14 15} Another problem with clinical testing is the lack of consistent administration procedures which hinders the repeatability of these tests and the ability to compare results across studies. Future studies should focus on validating reliable and valid oculomotor function testing tools and methods that can be easily implemented in clinical settings and with children of different functional and cognitive abilities. Valid and reliable measures of oculomotor function will provide valuable information regarding the child's ability to view the world around them. Impairment of oculomotor function may contribute to limited participation in school activities, play, or social interactions and could potentially be addressed with targeted interventions.

Vestibular Function in Children with CP

Only one article examined vestibular function in children with CP. ³⁵ This article included the highest functioning levels of children with CP (i.e., GMFCS levels I and II), and examined only one part of the peripheral vestibular system, the saccule. No studies examined the integrity of the semicircular canals or utricle in children with CP .

Currently accepted tests of semicircular canal function include the caloric, rotary chair, and video head impulse tests. ⁴³ The caloric test primarily determines horizontal canal function by measuring the VOR while using a water or air stimulus. This nonphysiologic stimulus mimics very low frequencies of head movement estimated from 0.003 to 0.008 Hz. ⁴⁴ The rotary chair test determines horizontal canal function by measuring the VOR in response to low to middle frequency sinusoidal head movements (e.g., 0.01–1.28 Hz) or a step test where the chair rotates at a constant velocity (e.g., 100 or 240 degree/s) for 1 minute. ⁴⁴ The Video Head Impulse Test examines the response of the horizontal, anterior, and posterior semicircular canals to very fast head movements (i.e., >150 degree/s). ⁴⁵ Peripheral utricular function can be assessed with the oVEMP test. ^{43 46} All of these tests have been included in studies involving child participants, but not in children with CP. ^{43 44 46 47 48}

The perception of upright (i.e., true vertical and horizontal) is thought to be mediated by the utricle, central utricular pathways, and vestibular cortex. ⁴⁹ This function is measured using the test of SVV. For this test, the individual must look at a laser-generated straight line that is tipped off center in the absence of other visual cues, and then must align it to vertical (i.e., neither tipped left nor right). This is accomplished, typically, by pushing buttons on a remote control. Accepted values for adults without vestibular impairment is within 2 degrees of true vertical. ⁵⁰ Recent studies have determined normative values and age/gender effects of SVV for children. ^{47 48} Brodsky et al ⁴⁸ tested 33 children aged 7 to 18 years, n  = 12 typically developing and n  = 21 with dizziness. Typically developing children (mean age = 13.9, SD = 2.8) had a mean SVV of 0.7 (SD = 0.5) and a score of more than 2 degrees identified children with peripheral vestibular impairment (100% sensitivity and 75% specificity). ⁴⁸ Toupet et al ⁴⁷ reported SVV tilt and results of vestibular diagnostic tests in 492 children aged 4 to 19 years with various vestibular disorders ( n  = 73 with normal vestibular function). Although the mean SVV tilt was within accepted ranges for adults, −0.02 ± 1.64, all children perceived vertical as tilted toward the side of initial presentation, which was noted significantly more in children aged 4 to 7 years than in children 15 to 17 years. In addition, the amount of tilt in the older group was more in girls than in boys. This suggests a potential effect of age and gender on SVV that should be explored. Christy et al ⁵⁰ tested SVV in children aged 6 to 12 (mean age = 9.8, SD = 1.8) with severe to profound sensorineural hearing loss. The children without vestibular hypofunction as measured with rotary chair and cVEMP ( n  = 11) had a mean SVV of 2.15 (SD = 1.5), which would have been considered abnormal if using the cutoff score suggested by Brodsky et al. ⁴⁸ More studies are needed to determine reliability, normative values, cutoff scores, and the age at which children of all ages are able to reliably complete the test. No study has examined SVV in children with CP. This test could prove to be a challenge for this population, especially if the child has cognitive, behavioral, or attentional issues, or poor head or upper extremity control. Creative researchers, then, will need to develop and test alternative methodology to test SVV in children with CP.

We do not know whether the balance impairments experienced by children with CP are due in part to aberrant vestibular function. The cognitive issues, poor head and trunk control, and behavior and attention issues may affect performance in a vestibular testing protocol. Protocols for rotary chair, VEMP, video head impulse test, and SVV may need to be adapted for younger children, and children with functional impairments.

Integration of Oculomotor and Vestibular Function in Children with CP

Static standing balance, eye–hand coordination, and head stability with visual focus require not only that the sensory and motor systems work but that they work together to produce the desired function. We reviewed two articles that investigated head stability in children with CP compared with typically developing children, both of which showed that children with CP exhibited more head movement during activities, regardless of the visual or support condition, than their typically developing peers. ^{28 34} Children with CP also had poor eye–hand coordination while sitting, ²⁷ driving, ¹⁷ and walking. ³² Given the findings from these studies, we hypothesize that the poor eye–hand coordination could potentially be due in part to poor gaze stability with head movement. However, no study has examined whether children with CP have gaze instability (i.e., poor visual focus) during head movement. This function can be measured in children using the clinical DVA test, which measures the difference in visual acuity with the head static, versus moving in the yaw plane at 2 Hz. ⁵⁰ This test has not been validated for children with CP, although studies are underway. ⁵¹ If children with CP have gaze instability with head movement, there needs to be understanding as to why. Is it because of aberrant oculomotor, visual, and/or peripheral vestibular function? Is it because of poor central integration? Is it a combination of these? The answers to these important questions might help clinicians to better guide interventions to improve gaze stability and balance. For example, gaze stabilization exercises (i.e., maintaining visual focus during head movement) is known to improve DVA in adults ⁵² as well as children with vestibular hypofunction. ⁵³ Could these exercises also help improve DVA in children with CP? If so, would this carry over into the improvement of eye–hand coordination, functional activities, and participation?

It is not surprising that standing balance, especially during visually challenging conditions, was worse in children with CP compared with typically developing children; ^{25 26 29 30 31 36} children with CP had higher sway frequency and amplitude, and were prone to falls, especially when unsupported. Of note, however, is that the standing balance of children with CP was worse during visual and support surface challenges, conditions that require the vestibular system to resolve conflict and maintain balance. ^{29 36} Balance and motor challenges experienced by children with CP are attributed, at its core, to brain damage at or before birth leading to an aberrant sensory and motor system and causing weakness, spasticity, and poor skeletal alignment. ⁵⁴ This scoping review led to many unanswered questions: Could the poor balance ability, in part, be the result of an abnormal peripheral vestibular system as suggested by Akbarfahimi et al ³⁵ ? No study has examined utricular function, or SVV in this population which could explain postural control deficits. Could the poorly functioning oculomotor and visual systems ^{6 8 14 15 33} affect the child's ability to use vision during balance? Finally, could the lack of integration of these systems, combined with an abnormal somatomotor system, lead to poor development of the postural control system? The answers to these questions could guide clinicians to targeted interventions that might improve postural control in children with CP. For example, practice of balancing during conditions that normally required the vestibular system (e.g., unstable surface with eyes closed) improved somatosensory and visual effectiveness in children with vestibular hypofunction. The mechanism behind this improvement is unknown, but thought to be due to sensory substitution, ²¹ that is substituting other central nervous system pathways for an abnormal vestibular system.

These findings support the need for more studies that objectively measure the oculomotor, peripheral, and central vestibular systems in children with CP to explain the underlying causes of functional impairments. This will be challenging, given that children with CP often have cognitive, behavioral, and physical impairments that might preclude accurate testing. Future research should utilize existing technology such as mobile eye trackers and develop new technology to modify vestibular function tests and validate the modified methods. Clinical tests should also be developed so that clinicians can easily screen for potential vestibular and oculomotor pathology and then refer to a specialist such as a pediatric audiologist or developmental optometrist for more comprehensive testing in the laboratory. This line of research could provide important information regarding impaired postural control and function in children with CP. It could reveal the underlying causes of functional impairments not previously addressed in this population such as vestibular-mediated gaze instability or poor perception of vertical. Finally, this information could lead to the development of interventions that target the use of the vestibular and oculomotor systems, thus improving function and community participation in children with CP.

Conclusion

This scoping review revealed a paucity of rigorous and objective research to describe the status of oculomotor and vestibular function in children with CP. However, results of the few studies examining these functions warrant further exploration. This line of research is important, considering that children with CP have poor static balance and eye–hand coordination, poor head stability, and difficulty with visual focus in function. Therefore, future research should focus on developing valid and reliable oculomotor and vestibular tests in children with CP. Researchers doing this work will need to consider the challenges (e.g., behavior, attention, motor skills, head control), GMFCS levels, and type of CP, as well as the potential effect of age and development.