REVIEW Open Access

Subjective and objective vestibular changesthat occur following paediatric cochlearimplantation: systematic review and meta-analysisMichael Yong, Emily Young, Jane Lea, Hannah Foggin, Erica Zaia, Frederick K. Kozak and Brian D. Westerberg*

Abstract

Objective: Cochlear implantation can result in post-operative vestibular dysfunction of unknown clinical significance.The objective of this study was to characterize the presence, magnitude, and clinical significance of vestibulardysfunction that occurs after pediatric cochlear implantation.

Data sources: The databases Embase, Medline (OvidSP), and PubMed were used. Only articles published in Englishwere included. Grey literature and unpublished sources were also reviewed.

Study selection: Articles published from 1980 until the present which documented pre-operative and post-operativevestibular testing on children under the age of 18 were used.

Data extraction: Parameters that were assessed included number of patients, pre- and post-operative vestibular-evoked myogenic potentials (VEMPs), head impulse testing (HIT), calorics, and posturography, timing of pre- and post-operative testing, symptomatology, and other demographic data such as etiology of the hearing loss.

Data synthesis: Ten articles were included. Relative risk values evaluating the effect of cochlear implantation onvestibular function were calculated for VEMPs and caloric testing due to the availability of published data. I2 valueswere calculated and 95% confidence intervals were reported. Separate analyses were conducted for each individualstudy and a pooled analysis was conducted to yield an overall relative risk. Assessment on risk of bias in individualstudies and overall was performed.

Conclusion: Pediatric cochlear implantation is associated with a statistically significant decrease in VEMP responsespost-operatively (RR 1.8, p < 0.001, I2 91.86, 95%CI 1.57–2.02). Similar results are not seen in caloric testing. Insufficientdata is available for analysis of HIT and posturography. Further studies are necessary to determine the effect of cochlearimplantation on objective vestibular measures post-operatively and whether any changes seen are clinically relevant inthis population.

IntroductionCochlear implantation in the pediatric population hasbecome a routine procedure for children with bilateral se-vere to profound sensorineural hearing loss [1]. The pastdecade has seen a rapid increase in the number of implantsbeing performed due to cost-effectiveness studies and theimproved quality of life enjoyed by implant recipients [2, 3].However, there have been recent efforts aimed at further

characterizing the potential side effects of cochlear implant-ation in children, specifically with respect to resultantvestibular dysfunction. There has been a growing trend to-wards bilateral simultaneous implantation in children due torecent studies showing an increased ability to preservenormal cortical preference and development, improvedsound localization contributing to enhanced preverbal com-munication and language development, and greatercost-effectiveness [1, 4–6]. The impact that bilateralimplantation has on vestibular and balance function inchildren has been studied but methodologies and results are

© The Author(s). 2019 Open Access This article is distributed under the terms of the Creative Commons Attribution 4.0International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, andreproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link tothe Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver(http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated.

* Correspondence: BWesterberg@providencehealth.bc.caBC Rotary Hearing and Balance Centre, St. Paul’s Hospital, 1081 Burrard St,Vancouver, BC V6Z 1Y6, Canada

Yong et al. Journal of Otolaryngology - Head and Neck Surgery (2019) 48:22 https://doi.org/10.1186/s40463-019-0341-z

variable, making it difficult to draw general conclusions fromthis research.The vestibular ramifications of implantation in the adult

population have been documented [7–9]. Cochlear im-plantation has the potential to cause a decrease in vestibu-lar function in adults measured by objective testing such asvestibular evoked myogenic potentials (VEMPs), calorics,and head impulse testing (HIT) [8, 9]. In addition, a recentmeta-analysis of vestibular function after cochlear implantsurgery reported a significant negative effect on caloric andVEMP testing in adults, although the overall patient-re-ported symptomatic manifestations of these objective find-ings were found to be insignificant [10].Documentation of pre- and post-operative vestibular

function surrounding pediatric cochlear implantation arelimited and perhaps reflective of the fact that many centresdo not routinely perform such testing. This may be due toseveral reasons such as cost, unclear benefit, difficulty intesting very young children, and lack of availableequipment. As such, no prior systematic reviews or meta-analyses are available. While pediatric cochlear implant-ation has proven to be a relatively safe procedure with fewsevere long-term complications, those related to the ves-tibular system have been reported [11, 12]. Data from theliterature to date has shown that vestibular end-organdysfunction can be a risk factor for implant failure inthe pediatric population and that there may be valuein identifying children with vestibular impairmentafter surgery to avoid this complication [13]. Inaddition, recent research has shown that cochlearimplantation has the potential to affect vestibularfunction through direct current spread from the coch-lea to the vestibular system [14].The now commonplace nature of this procedure, in-

cluding bilateral implantation, and the potential long-termramifications of undetected vestibular damage highlights aneed for systematic analysis of the current literature inorder to develop evidence-informed decisions regard-ing implantation. In turn, this may further elucidatepotential contraindications to surgery or highlight theneed for earlier vestibular rehabilitation interventionsfollowing implantation in young patients.

ObjectivesThe aim of this systematic review was to evaluate thevestibular side-effects of cochlear implantation in thepediatric population through examination of objectiveand subjective vestibular parameters. The reviewsought to answer the following questions:

(1) Is there a significant decrease in objectivevestibular testing responses in the implantedear after cochlear implantation?

(2) Are there any changes in subjective vestibularsymptoms post-operatively that suggest persistentlong-term clinical impairment?

According to the PICOS approach, the population ofinterest included pediatric patients under the age of 18undergoing cochlear implantation who have had pre-opera-tive and post-operative vestibular testing. Cochlear implant-ation represented the intervention, and the comparison orcontrol was the pre-operative vestibular status of the pa-tient prior to surgery. The outcome under investigation wasthe objective and subjective vestibular status of patientsafter surgery. The review was not limited to any specificstudy design criterium.

MethodsEthics exemption was obtained from the University ofBritish Columbia Research Ethics Board. The PreferredReporting Items for Systematic Reviews and Meta-analysis(PRISMA) statement was followed during the review.

ProtocolMethods of analysis and inclusion/exclusion criteriawere specified in advance and documented in a separateprotocol.

Study inclusion criteriaProspective or retrospective cohort studies reporting bothpre- and post-operative vestibular testing of pediatric pa-tients under the age of 18 at the time of cochlear implant-ation were included. Studies including children whounderwent either unilateral or bilateral cochlear implant-ation (simultaneous or sequential) were included. Studiesthat included adults as well as children were also included,providing individual relevant data was available for thepediatric patients. Children with additional comorbiditieswere included, even those with conditions that predisposeto vestibular dysfunction, as these form an importantcohort of patients currently undergoing cochlearimplantation.

Study exclusion criteriaStudies including only either pre- or post-operative test-ing were excluded. Non-English language papers wereexcluded as their exclusion has been shown to have lim-ited impact on review results [15]. Publications prior to1980 were intentionally excluded.

Types of outcome measures

1) Primary outcomes measures: pre and post operativevestibular function testing including at least one of:VEMPs, calorics, HIT, dynamic posturography.

Yong et al. Journal of Otolaryngology - Head and Neck Surgery (2019) 48:22 Page 2 of 10

2) Secondary outcome measures: subjective symptomsof vestibular function pre- and post-cochlearimplantation.

Information sourcesA structured search was performed of the followingthree databases between 1980 and August 2017: Medline(Ovid), EmBase (Ovid), and PubMed. The reference listof each eligible study was also examined for extractionof relevant published articles not identified during theoriginal search. A search of the grey literature, includingunpublished data and PhD theses was also performed.

Search strategyThe keywords, MeSH terms, and phrases searched includedthe following: “cochlear implant”, “pediatric”, “child”“VEMP”, “dizziness”, “posturography”, “vertigo” and “ves-tibular”. The search strategy was devised collectively andexecuted independently by two of the study authors (EY;MY). No study design or data limits were imposed uponthe search. The full search strategy can be found inAppendix.

Study selection and data collection processThe search results were collated on a Microsoft Excel(Microsoft, 2013) spreadsheet. The data collection sheetwas piloted on two randomly-selected included studiesand refined accordingly. Duplicate articles were elimi-nated. All abstracts were reviewed independently by au-thors EY and MY for relevance to the study topic. Fulltext studies of articles deemed to be relevant were fur-ther reviewed. Discrepancies regarding the inclusion orexclusion of studies were resolved through discussionbetween the two aforementioned authors. Neither of thereview authors was blinded to the journal titles or thestudy authors or institutions.

Data items/outcomesThe selected articles were searched for vestibular out-comes before and following cochlear implantation. VEMPtesting, calorics, HIT, and posturography were the primaryparameters. Demographic data on individual participantsin each of the selected studies were gathered. Theseincluded, whenever available, etiology of hearing loss,imaging status, age at implantation, unilateral or bilateralimplantation, method of implantation, simultaneous orsequential implantation, and timing of pre- and post-op-erative vestibular testing. Documentation of subjectivepost-operative vestibular symptoms was also extractedwhen possible. The two aforementioned authors inde-pendently extracted data from the selected articles accord-ing to these predetermined parameters. Any discrepanciesin extracted data were resolved by discussion between thereview authors.

Risk of bias in individual studiesA checklist including ten key questions adapted from theCritical Appraisal Skills Programme by the Institute ofHealth at Oxford University was used for the risk of biasassessment [16]. Studies were categorized into one of threecategories; (A) Low risk of bias; (B) Medium risk of bias;(C) High risk of bias. Both authors independently examinedeach study for risk of bias and assigned a categorization.Any discrepancies were resolved through discussion.An assessment of the risk of meta-bias was also per-

formed. This involved an independent review by bothauthors focusing on possible selective reporting withinstudies and a consideration of publication bias.

Summary measuresThe primary targeted measure calculated from the data wasthe relative risk of cochlear implantation on vestibular dys-function. This was performed by comparing each child’sobjective vestibular function pre-operatively to his or hervestibular function post-operatively. Abnormal test resultswere determined using each study authors’ definition.

Planned methods of analysisSeparate analyses were performed on each individualstudy, and then a pooled analysis was performed on allstudies to yield an overall relative risk. This involved acalculation of relative risk values, as well as I2 calcula-tions and 95% confidence intervals. Statistical signifi-cance was defined as p < 0.05.

Strength of evidenceThe strength of evidence was determined by the signifi-cance of the meta-analysis, relevance of objective testingin the context of clinical impact, and the risk of bias as-sessment performed on all studies.

ResultsStudy selectionApplying the predefined criteria, 410 studies were identifiedbased on initial search of the databases (Fig. 1). An add-itional three studies were identified based on a review ofstudy references and the grey literature. After duplicateswere removed, a review of 408 abstracts resulted in exclu-sion of 374 studies that did not meet one or more of theinclusion criteria. Of the remaining 34 studies that under-went full text review, 23 were excluded based on partici-pants over the age of 18 or not reporting both pre-operative and post-operative vestibular testing. Theremaining 11 studies were included for full analysis.

Study characteristicsDemographic data were extracted from all 11 studies whichincluded study design, number of patients (study size),mean age, timing of pre-operative and post-operative

Yong et al. Journal of Otolaryngology - Head and Neck Surgery (2019) 48:22 Page 3 of 10

vestibular testing, unilateral or bilateral implantation, se-quential or simultaneous implantation, etiology of hearingloss, relevant imaging findings, cochleostomy site, and im-plant brand/model (Table 1) [18–28].Data pertaining to pre- and post-operative VEMPs, ca-

lorics, HIT, posturography, and subjective symptomswere collected (Table 2). Of the 11 studies, four reportedsubjective dizziness data and only one reported HITdata. No studies reported posturography data.

Risk of bias within studiesA formal assessment of the risk of bias was performedon each study (Table 3). After review, 5 studies wereconsidered to have a low risk of bias, 4 had a mediumrisk of bias, and 2 had a high risk of bias. All studies thatwere examined were either prospective or retrospectivecohort studies.

Results of individual studiesData extracted from individual studies were analyzedand relative risk values were calculated for each study.The results for the VEMP analyses are displayed in theForest plot in Fig. 2. Five of the nine studies [18, 19, 21, 24,

28] that measured cervical VEMP (cVEMP) responsesshowed a statistically significant increase in abnormalVEMP responses following cochlear implantation. Thepooled relative risk of 1.800 was supported by a 95% confi-dence interval that did not span 1 (95%CI 1.442–2.281) anda significant p value of less than 0.001. On review of all con-tributing studies, a VEMP change was defined as any sig-nificant change in threshold, amplitude, or latencymeasured post-operatively when that ear had baseline nor-mal VEMP testing. A change was also defined as acomplete loss of VEMP responses in an ear that had base-line abnormal VEMP testing (threshold, amplitude, or la-tency). These results represented Level 3 evidenceaccording to OCEBM [17].The results for the calorics analysis are displayed in

the Forest plot in Fig. 3. One of the three studies [27]that measured calorics showed a statistically significantreduction in caloric responses following implantation.However, the pooled relative risk of 1.012 indicated nooverall significant change and was based on only threedifferent studies with the 95% confidence interval limitswidely spanning below and above 1 (95%CI 0.712–1.439). Additionally, each of the three studies had a

Fig. 1 Study article inclusion/exclusion algorithm

Yong et al. Journal of Otolaryngology - Head and Neck Surgery (2019) 48:22 Page 4 of 10

Table

1Stud

yde

mog

raph

ics

Stud

yDesign

Sample

size

Mean

age

Follow

upSide

Order

Etiology

Imagingmalform

ation

Surgicalsite

Brand

Ajalloueyan

Prospe

ctive

272

6–8weeks

Unilateral

N/A

Not

specified

Not

specified

Roun

dWindo

w“AdvancedBion

ics

Cochlear”and“

CochlearNucleus”

JinNot

specified

123.8

Not

specified

Not

specified

N/A

Not

specified

Mon

dini

(2),on

ebranch

ofvestibulocochlearne

rve

(1),EVA(1)

Not

specified

Not

specified

DeKege

lProspe

ctive

237m

o2years

Both

Both

Gen

eticno

Synd

rome(4),

Gen

eticsynd

rome(6),Acquired

CMV/Men

ingitis/premature(8),

ANSD

(3),Unkno

wn(2)

Not

specified

Not

specified

Not

specified

Licameli

Prospe

ctive

198

4–6weeks

Unilateral

N/A

Not

specified

Not

specified

Not

specified

Not

specified

Devroed

eRetrospe

ctive

246.75

90days

post-secon

dim

plant

Bilateral

Sequ

entialSynd

rome(7),Gen

eticmutation

(7),Men

ingitis

(1),CMV(1),ANSD

(2),Unkno

wn(8)

Vestibular

(3),cochlear

(1),

cochleo-vestibular

(3)

Antero-inferio

r“Cochlear”

Psillas

Prospe

ctive

102.85

6mon

ths

Unilateral

N/A

Not

specified

Not

specified

Antero-inferio

r“Freed

om”(6)and

“Med

elSonata”(4)

XuProspe

ctive

315.5

4weeks

Unilateral

N/A

Not

specified

Not

specified

Not

specified

Not

specified

Thierry

Retrospe

ctive

121.9

1.4years

Unilateral

N/A

Gen

etic(8),Synd

rome(7),CMV/men

ingitis

(5),Inne

rear/PD

Smutation(7),Kallm

an(1),Unkno

wn(15)

Bilateralm

alform

ation(7)

Antero-inferio

r“Cochlear”and

“AdvancedBion

ics”

Bogle

Retrospe

ctive

57.73

Not

specified

Unilateral

N/A

Con

genital(4),A

cquired(1)

Noabno

rmalities

Not

specified

Not

specified

Gup

taProspe

ctive

235.48

6weeks

Unilateral

N/A

Not

specified

Noabno

rmalities

Anterior-inferio

r“M

ed-ElP

ulsar”

Hazzaa

Prospe

ctive

405.7

1mon

thand

6mon

ths

Unilateral

N/A

Gen

etic(21),Syndrom

e(2),Ototoxicity

(2),Perin

atal(1),Unkow

n(13)

Not

specified

Not

specified

Not

specified

Yong et al. Journal of Otolaryngology - Head and Neck Surgery (2019) 48:22 Page 5 of 10

different definition of what constituted a significantcaloric response change. Ajalloueyan et al. did notreport a cut-off value and performed calorics usingcold air, Gupta et al. defined a difference of > 15% inmonothermal testing as significant and performedcalorics using warm air, and Devroede et al. defineda difference of > 20% in bithermal testing and > 27%for monthermal testing as significant and performedmonothermal calorics using cold water. This wasdeemed to represent Level 3 evidence.Only four studies [18, 19, 23, 30] reported on subject-

ive symptoms, and there were no studies that reportedoutcomes in a quantitative manner. Based on thequalitative reporting of data, no studies indicated thatthere were any post-operative symptoms such as imbal-ance or dizziness either immediately or long-termpost-operatively.

Synthesis of resultsData were combined for the purpose of a pooledmeta-analysis. Relative risk, I2 value, and the 95% con-fidence interval were yielded and displayed together inthe Forest plot (Fig. 2). The results showed a statisti-cally significant increase in abnormal VEMP responsesafter either unilateral or bilateral cochlear implantation(RR 1.8, p < 0.001, 95%CI 1.57–2.02). The heterogen-eity in the studies included for analysis was significant(I2 91.86, Q = 98.304). A statistically significantincrease in abnormal caloric testing was not similarlyseen (p = 0.60).

Risk of bias across studiesThe risk of bias when comparing between studies wasdeemed to be low. There were no concerns identified

Table 2 Study vestibular findings

Study VEMP Calorics vHIT Non-video HIT Symptoms

Ajalloueyan Pre: 7% (2) no cVEMP bilaterally.19% (5) no cVEMP unilaterally.

Pre: 8% (1)dysfunction.

Not performed Pre: 0% had dysfunction. N/A

Post: No change. Post: No change.Post: 20% (8) abnormal cVEMP.

Jin Pre: 8% (1) reduced cVEMP,42% (5) no cVEMP.

Not performed Not performed Not performed N/A

Post: 100% (12) abnormalcVEMP.

De Kegel Pre: 26% (5) had abnormalcVEMP unilaterally or bilaterally.

Not performed Not performed Not performed Significant drop in gross motorperformance, recovering towardage 2.

Post: No change.

Licameli Pre: 12% (2) no cVEMP. Not performed Not performed Not performed Data not reliable collected, butno significant difference found.

Post: 84% (16) abnormal cVEMP.

Devroede Pre: 21% (5) abnormal cVEMP. Pre: 71% (17)abnormal calorics.

Not performed Not performed Post-op dizziness in 33% (3) whohad post-op vestiublar dysfunction,which subsequently recovered.

Post: 38% (9) abnormal cVEMP. Post: 67% (16)abnormal calorics.

Psillas Pre: 30% (3) no cVEMP, 30%(3) reduced cVEMP.

Not performed Not performed Not performed No dizziness, vertigo, instability,or nystagmus noted post-op.

Post: 100% (10) abnormal VEMP.

Xu Pre: 26% (6) abnormal cVEMP. Not performed Not performed Not performed N/A

Post: 65% (15) abnormal VEMP.

Bogle Pre: 0%abnormal cVEMP. Not performed Not performed Not performed N/A

Post: 20% (1) abnormal cVEMP.

Gupta Not performed Pre: 22% (5) abnormalcalorics.

Not performed Not performed N/A

Post: 35% (8) abnormalcalorics.

Hazzaa Pre: 55% (22) abnormal orabsent cVEMP.

Not performed Not performed Not performed N/A

Post: 80% (32) abnormal orabsent cVEMP at 6 months.

Yong et al. Journal of Otolaryngology - Head and Neck Surgery (2019) 48:22 Page 6 of 10

Table

3Risk

ofbias

assessmen

t

Stud

yFocused

issue

Recruitm

ent

Expo

sure

accurately

measured

Outcome

accurately

measured

Con

foun

ding

factors

Follow

upResults

Precision

ofresults

Believability

ofresults

App

licability

Fitwith

eviden

ceIm

plications

Risk

ofbias

Ajalloueyan

Yes

Goo

dYes

Yes

Step

stakento

mitigate,

limitatio

nsno

tform

ally

addressed

Welld

ocum

ented

Welld

ocum

ented

Poor

Fair

Goo

dUnsure

Fair

B

JinYes

Poor

Yes

Yes

Few

step

stakento

mitigate,

limitatio

nsweaklyaddressed

Poorlyde

scrib

edWelld

ocum

ented

Fair

Fair

Goo

dUnsure

Unsure

C

DeKege

lYes

Goo

dYes

Yes

Step

stakento

mitigate,

limitatio

nsform

ally

addressed

Welld

ocum

ented

Welld

ocum

ented

Fair

Goo

dGoo

dUnsure

Fair

A

Licameli

Yes

Goo

dYes

Yes

Somestep

stakento

mitigate,lim

itatio

nsform

allyaddressed

Welld

ocum

ented

Welld

ocum

ented

Fair

Goo

dGoo

dGoo

dFair

A

Devroed

eYes

Fair

Yes

Yes

Step

stakento

mitigate,

limitatio

nsform

ally

addressed

Welld

ocum

ented

Welld

ocum

ented

Fair

Fair

Goo

dUnsure

Fair

B

Psillas

Yes

Fair

Yes

Yes

Few

step

stakento

mitigate,

limitatio

nssomew

hat

addressed

Welld

ocum

ented

Welld

ocum

ented

Fair

Fair

Goo

dGoo

dUnsure

B

XuYes

Goo

dYes

Yes

Somestep

stakento

mitigate,

limitatio

nssomew

hat

addressed

Welld

ocum

ented

Welld

ocum

ented

Fair

Goo

dGoo

dGoo

dFair

A

Thierry

Yes

Goo

dYes

Yes

Step

stakento

mitigate,

limitatio

nssomew

hat

addressed

Welld

ocum

ented

Welld

ocum

ented

Fair

Goo

dGoo

dGoo

dFair

A

Bogle

Yes

Poorly

describ

edYes

Poorly

describ

edNot

mitigated,

limitatio

nsform

allyaddressed

Poorlyde

scrib

edPo

orlyde

scrib

ed,

smallsam

plesize

Poor

Poorly

describ

edPo

orUnsure

Poor

C

Gup

taYes

Goo

dYes

Yes

Somestep

stakento

mitigate,

limitatio

nsno

tform

ally

addressed

Welld

ocum

ented

Welld

ocum

ented

Fair

Fair

Goo

dGoo

dFair

B

Hazzaa

Yes

Goo

dYes

Yes

Somestep

stakento

mitigate,

limitatio

nsno

tform

ally

addressed

Welld

ocum

ented

Welld

escribed

Goo

dGoo

dGoo

dGoo

dFair

A

Yong et al. Journal of Otolaryngology - Head and Neck Surgery (2019) 48:22 Page 7 of 10

over selective reporting of data, as patients in the studiesthat were reviewed were generally all accounted for inthe results.

DiscussionSummary of evidenceCochlear implantation carries a risk of post-operative ves-tibular dysfunction. Although its clinical significance cur-rently appears to be minimal, long-term findings in thetimeframe of decades have yet to be reported. Data fromadult studies suggests that following implant surgery, oto-lithic organs and canal function can be compromised, lead-ing to potential clinical manifestations such as post-operativeimbalance, vertigo, and falls [1, 4, 7]. This is especially con-cerning for children who are more commonly undergoingbilateral implantation, raising the theoretical concern of bilat-eral vestibular dysfunction in the long term following surgicalintervention. However, recent work on perceptual visual tiltusing the static subjective visual vertical test has also shownthat electrical pulses from the implant itself after initial coch-lear implantation may have a role in actually improving

vestibular function by helping to correct abnormal percep-tion of vertical [29].Cochlear implantation can potentially lead to disruption of

saccular function in pediatric patients. Review of the litera-ture would suggest there is a significant risk for increase inabnormal VEMP responses in pediatric patients undergoingcochlear implantation, with a relative risk of 1.8 (p < 0.001)despite several individual studies that did not display signifi-cant results [19, 20, 22, 23, 26]. These findings support simi-lar findings published in a recent systematic review studyingvestibular outcomes after adult cochlear implantation, show-ing a log relative risk of 0.5099 (p < 0.0001) after an analysisof 12 separate studies published since 2008 [9]. Proposedmechanisms for this finding have been previously discussedin the literature and include direct trauma related to inser-tion of the electrode, intraoperative perilymphatic fluid leak-age, electrical stimulation of the otolithic organs, and foreignbody labyrinthitis [17, 18].There was no significant change in caloric responses in

pediatric patients undergoing cochlear implantation. How-ever, this is based on a small number of available studies

Fig. 2 Meta-analysis Forest plots for VEMPs and caloric testing. 95% confidence intervals are includedin parentheses. *Denotes a statistically significant result at the p < 0.05 level. ** Denotes a statistically significant result at the p < 0.001 level

Yong et al. Journal of Otolaryngology - Head and Neck Surgery (2019) 48:22 Page 8 of 10

reporting on this parameter and poor precision of availabledata. In addition, the caloric testing method and definition ofcanal dysfunction was widely variable, contributing to a sig-nificant source of bias.There was a lack of available data concerning a number

of the parameters of interest, specifically rotary chair, HITand posturography. Challenges associated with collectingpediatric video-HIT data in particular have been reportedin the past, citing difficulties such as lack of interest, invol-untary or voluntary tensing of cervical muscles duringtesting, and problems with keeping eyes open during test-ing, especially in children under six years of age [30]. Asignificant proportion of pediatric implantations in this re-view occurred during the first five to six years of life.Regardless of the objective results, there was no evi-

dence for subjective symptoms of dizziness occurring inthe pediatric population after cochlear implantation.Whether this is a limitation in available data or in thesubjective reporting of symptoms, or reflective of a lackof clinically significant injury to the vestibular system inthis population is uncertain.

LimitationsThere is a paucity of data on vestibular function in thepediatric population undergoing cochlear implantation; sev-eral studies were excluded because of a lack of pre-operativeand post-operative objective measurements of vestibularfunction. As a result, there may be insufficient power to de-tect statistically significant differences in objective vestibularmeasures other than VEMPs. In addition, there was a gen-eral lack of data on motor function. Posturography can bedifficult to perform on children and cannot be performed onyoung children and infants, which explains the lack of datain this category. Future studies examining vestibular functionin implanted children may benefit from examining patientswith scales such as the Ghent modification of the Alberta in-fant motor scale in order to provide age-standardized mea-sures of motor function [30].The amount of variance or heterogeneity observed among

studies analyzing VEMP responses was substantial (I2 91.86,Q= 98.304). Reasons could include inherent differences inthe testing methods employed by different studies used todetermine whether children displayed objective vestibularfindings. Some studies reported detailed protocols of whatconstituted an abnormal VEMP response, including abnor-malities in latency, amplitude, or threshold [20, 23, 24],whereas one study simply classified abnormal VEMP re-sponses as hyporeflexic or areflexic [19]. Additionally, themean time to follow-up was widely variable and ranged fromas short as 4weeks post-implant surgery to as long as twoyears post-operatively; vestibular dysfunction likely has theability to recover or compensate over time.Six of the eleven studies that were included in this review

were graded as having a medium or high risk of bias due to

poor documentation of follow-up, insufficient measurestaken to reduce confounding factors in sample populations,and poor to fair precision of results.The lack of reporting etiology of hearing loss and imaging

findings in many of the studies may be relevant; certain con-ditions such as an enlarged vestibular aqueduct and otheranomalous cochleovestibular anatomy have been shown tobe responsible for higher rates of intraoperative CSF leaks,which could potentially put the patient at a higher risk forpost-operative vestibular dysfunction [31].We did not feel that there were significant limitations con-

cerning reporting bias or incomplete retrieval of publishedresearch. The number of studies that were included in thisanalysis approximated the expected number based on similarstudies on vestibular outcomes after cochlear implantation inadults [10].Finally, cochlear implant electrode technology con-

tinues to improve and smaller diameter electrodes arefacilitating atraumatic cochlear implantation. What rolethis may have in reducing the potential for vestibular in-jury in the future is uncertain.

ConclusionsPediatric patients experience a statistically significant changein vestibular function post-implantation as measured byVEMPs. This important finding supports previous workshowing vestibular changes post-implantation in the adultpopulation. However, data on post-operative subjectivesymptoms in the pediatric population is currently lacking.Additional longitudinal studies examining the correlation be-tween VEMP changes and clinical symptoms may be war-ranted to determine the clinical significance of the findingsin this review.

Appendix AFull search strategyThe search on PubMed included:((((cochlear implant [Title/Abstract]) OR cochlear implant-

ation [Title/Abstract])) AND ((((paediatric [Title/Abstract])OR pediatric [Title/Abstract]) OR child [Title/Abstract]) ORchildren [Title/Abstract])) AND ((((((((((vestibular [Title/Ab-stract]) OR VEMPs [Title/Abstract]) OR posturography[Title/Abstract]) OR dizziness handicap inventory [Title/Ab-stract]) OR DHI [Title/Abstract]) OR vertigo [Title/Ab-stract]) OR balance [Title/Abstract]) OR caloric [Title/Abstract]) OR vestibule [Title/Abstract]) OR labyrinth [Title/Abstract])The search on Medline (Ovid) and Embase (Ovid)

included:((cochlear implant.mp. or exp. Cochlear Implants/) and

(pediatric.mp. or exp. Pediatrics/) and (exp Dizziness/ or exp.Vestibule, Labyrinth/ or exp. Vertigo/ or exp. Vestibular Dis-eases/ or vestibular.mp.) and (exp Child/ or child.mp.))

Yong et al. Journal of Otolaryngology - Head and Neck Surgery (2019) 48:22 Page 9 of 10

AcknowledgementsNot applicable.

FundingNo funding was provided for this study.

Sources of support and disclosureNothing to disclose.

Authors’ contributionsMY drafted the protocol, refined the search algorithm, screened papers,conducted statistical analyses, and drafted the manuscript. EY assisted indrafting the protocol, screening papers, conducting statistical analyses, andrevising the manuscript. JL formulated the research question and revised themanuscript. HF conducted statistical analyses. EZ revised the manuscript. FKKrevised the manuscript. BDW formulated the research question and revisedthe manuscript. All authors read and approved the final manuscript.

Ethics approval and consent to participateEthics exemption was obtained from the University of British ColumbiaResearch Ethics Board. The Preferred Reporting Items for Systematic Reviewsand Meta-analysis (PRISMA) statement was followed during the review.

Consent for publicationNot applicable.

Competing interestsThe authors declare that they have no competing interests.

Publisher’s NoteSpringer Nature remains neutral with regard to jurisdictional claims inpublished maps and institutional affiliations.

Received: 1 January 2019 Accepted: 23 April 2019

References1. Lammers MJW, van der Heijden GJMG, Pourier VEC, Grolman W. Bilateral

cochlear implantation in children: a systematic review and best-evidencesynthesis. Laryngoscope. 2014;124(7):1694–9.

2. Chen JM, Amoodi H, Mittmann N. Cost-utility analysis of bilateral cochlearimplantation in adults: a health economic assessment from the perspectiveof a publicly funded program. Laryngoscope. 2014;124(6):1452–8.

3. McKinnon BJ. Cost effectiveness of cochlear implants. Curr Opin OtolaryngolHead Neck Surg. 2014;22(5):344–8.

4. Perez-Martin J, Artaso MA, Diez FJ. Cost-effectiveness of pediatric bilateralcochlear implantation in Spain. Laryngoscope. 2017;127(12):2866–72.

5. Gordon KA, Wong DD, Papsin BC. Bilateral input protects the cortex fromunilaterally-driven reorganization in children who are deaf. Brain. 2013;136:1609–25.

6. Papsin BC, Gordon KA. Cochlear implants for children with severe-to-profound hearing loss. N Engl J Med. 2007;357:2380–7.

7. Buchman CA, Joy J, Hodges A, Telischi FF, Balkany TJ. Vestibular effects ofcochlear implantation on the vestibular function. Laryngoscope. 2004;114(10Pt 2 Suppl 103):1–22.

8. Karsiari E, Balatsouras DG, Sengas J, Riga M, Korres GS, Xenelis J. Influence ofcochlear implantation on the vestibular function. Eur Arch Otorhinolaryngol.2013;270(2):489–95.

9. Melvin TA, Della Santina CC, Carey JP, Migliaccio AA. The effects of cochlearimplantation on vestibular function. Otol Neurotol. 2009;30(1):87–94.

10. Ibrahim I, da Silva SD, Segal B, Zeitouni A. Effect of cochlear implant surgeryon vestibular function: meta-analysis study. J Otolaryngol Head Neck Surg.2017;46(1):44.

11. Sivam SK, Syms CA 3rd, King SM, Perry BP. Consideration for routine outpatientpediatric cochlear implantation: a retrospective chart review of immediatepost-operative complications. Int J Pediatr Otorhinolaryngol. 2017;94:95–9.

12. Terry B, Kelt RE, Jeyakumar A. Delayed complications after Cochlearimplantation. JAMA Otolaryngol Head Neck Surg. 2015;141(11):1012–7.

13. Wolter NE, Gordon KA, Papsin BC, Cushing SL. Vestibular and balanceimpairment contributes to Cochlear implant failure in children. OtolNeurotol. 2015;36(6):1029–34.

14. Parkes WJ, Gnanasegaram JJ, Cushing SL, McKnight CL, Papsin BC, Gordon KA.Vestibular evoked myogenic potential testing as an objective measure ofvestibular stimulation with cochlear implants. Laryngoscope. 2017;127(2):E75–81.

15. Moher D, Pham B, Klassen TP, Schulz KF, Berlin JA, Jadad AR, et al. Whatcontributions do languages other than English make on the results ofmeta-analyses? J Clin Epidemiol. 2000;53(9):964–72.

16. Critical Appraisal Skills Programme (2017). CASP Checklist. [online] Availableat: http://www.casp-uk.net/casp-tools-checklists. Accessed 1 Jan 2018.

17. OCEBM 2011 Levels of Evidence. The Oxford 2011 Levels of Evidence. 2011.http://www.cebm.net/index.aspx?o=5653. Accessed 31 Jan 2018.

18. Licameli G, Zhou G, Kenna MA. Disturbance of vestibular function Attribuatableto Cochlear implantation in children. Laryngoscope. 2009;119:740–5.

19. Devroede B, Pauwels I, Le Bon SD, Monstrey J, Mansbach AL. Interest ofvestibular evaluation in sequentially implanted children: preliminary results.Eur Ann Otorhinolaryngol Head Neck Dis. 2016;133S:S7–S11.

20. Ajalloueyan M, Saeedi M, Sadeghi M, Abdollahi FZ. Int J PedOtorhinolaryngol. 2017;94:100–3.

21. Jin Y, Nakamura M, Shinjo Y, Kaga K. Vestibular-evoked myogenic potentialsin cochlear implant children. Acta Otolaryngol. 2006;126(2):164–9.

22. De Kegel A, Maes L, Van Waelvelde H, Dhooge I. Examining the impact ofCochlear implantation on the early gross motor development of childrenwith a hearing loss. Ear Hear. 2014;36(3):113–21.

23. Psillas G, Pavlidou A, Lefkidis N, Vital I, Markou K, Triaridis S, TsalighopoulosM. Vestibular evoked myogenic potentials in children after cochlearimplantation. Auris Nasus Larynx. 2014;41:432–5.

24. Xu X, Zhang X, Zhang Q, Hu J, Chen Y, Xu M. Ocular and cervical vestibular-evoked myogenic potentials in children with cochlear implant. ClinNeurophysiol. 2015;126:1624–31.

25. Thierry B, Blanchard M, Leboulanger N, Parodi M, Wiener-Vacher SR,Garabedian E, Loundon N. Cochlear implantation and vestibular function inchildren. Int J Ped Otorhinolaryngol. 2015;79:101–4.

26. Bogle, J. The effect of Cochlear implantation on the vestibular evokedmyogenic potential response in children and adults. Speech, Language, andHearing Services Graduate Theses and Dissertations. 2010. Paper 4.

27. Gupta A, Raj P. Compensated vestibular dysfunction post Cochlearimplantation in children with sensorineural hearing loss: a prospectivestudy. Indian J Otolaryngol Head Neck Surg. 2017. https://doi.org/10.1007/s12070-017-1054-0.

28. Hazzaa NM, Maksoud AIA, Khalil LH, Wahba HA, Galal IM. Effects of Cochlearimplantation on vestibular pathway: follow up study in Cochlear implantchildren. Glob J Otolaryngol. 2017;8(4):555743. https://doi.org/10.19080/GJO.2017.08.555743.

29. Gnanasegaram JJ, Parkes WJ, Cushing SL, McKnight CL, Papsin BC, GordonKA. Stimulation from Cochlear implant electrodes assists with recovery fromasymmetric perceptual tilt: evidence from the subjective visual vertical test.Front Integr Neurosci. 2016;10:32.

30. De Kegel A, Baetens T, Peersman W, Maes L, Dhooge I, Waelvelde H. Ghentdevelopmental balance test: a new tool to evaluate balance performance intoddlers and preschool children. Phys Ther. 2012;92:841–52.

31. Hulse R, Hormann K, Servais JJ, Hulse M, Wenzel A. Clinical experience withvideo head impulse test in children. Int J Pediatr Otoaryinolaryngol. 2015;79(8):1288–93.

Yong et al. Journal of Otolaryngology - Head and Neck Surgery (2019) 48:22 Page 10 of 10