Acta Oto-Laryngologica, 2013; 133: 475–480

ORIGINAL ARTICLE

Vestibular compensation after vestibular schwannoma surgery:normalization of the subjective visual vertical and disability

ANGEL BATUECAS-CALETRIO1, SANTIAGO SANTACRUZ-RUIZ1,ANGEL MUÑOZ-HERRERA1, PABLO SOUSA1, ALVARO OTERO1 &NICOLAS PEREZ-FERNANDEZ2

1Department of Otorhinolaryngology, Unidad de Base de Cráneo, Hospital Universitario de Salamanca,University of Salamanca and 2Clínica Universidad de Navarra, University Hospital and Medical School,University of Navarra, Spain

AbstractConclusions:The degree of caloric weakness before surgery influences faster or slower recovery of patients undergoing vestibularschwannoma surgery. The Dizziness Handicap Inventory (DHI) is a good index to show the recovery of patients as it relatesdirectly to an improvement or not of the subjective visual vertical (SVV). Objective: To evaluate the process of recovery ofpatients as measured by the SVV and the DHI after surgical removal of vestibular schwannoma. Methods: We studied24 consecutive patients of the University Hospital of Salamanca who underwent vestibular schwannoma surgery. We assessedage, tumour size, degree of canalicular weakness and preoperative SVV, and their relationship with DHI and SVV at dischargeand also at 1, 3 and 6 months postoperatively. Results: Patients with lesser degrees of caloric weakness took longer to normalizeSVV than those with a higher caloric weakness before surgery (p < 0.05). There was a significant correlation between DHI andimprovements in SVV with time. The differences disappeared in 6 months where all patients, with greater or lesser degree ofcaloric weakness, had the same results.

Keywords: Vertigo, caloric test, videonystagmography, vestibular areflexia

Introduction

The perception of the subjective visual vertical (SVV)is the ability of a subject to judge whether an object iscorrectly located on the vertical axis. The partici-pation of different vestibular receptors in this task isa matter of debate; however, vestibular lesions influ-ence gravity perception mostly via ocular torsionaleffects rather than by disrupting any internal repre-sentation of verticality, thus SVV inclines towards thelesioned side [1]. There is a very close relationshipbetween the torsional eye position measurements andthe SVV tilts in patients with vestibular disorders, andin those patients with large tilts in the SVV theability to correctly perceive verticality is maintainedby other means [2].

After vestibular schwannoma (VS) surgery there is atilt in the SVV towards the operated side in weakcorrelation with the preoperative amount of vestibulardeficiency as registered with the caloric test [3]. In thefollow-up the tilt returns to normal and may evendeviate to the normal side [4]. In the short term, itis more intense in patients relying more on vision tocontrol balance before surgery than in those who relyless; in both groups, the long-term results are similar[5]. They do not correlate with the amount of distres-sing unsteadiness or imbalance after surgery [6].The Dizziness Handicap Inventory (DHI) is a

standardized, validated 25-item questionnaire usedto determine if subjects perceive a handicap due todizziness, and its subscales reflect physical, emotionaland social limitations [7].

Correspondence: Dr Angel Batuecas Caletrío, Department of Otorhinolaryngology, Pso S. Vicente 58-182, 37007 Salamanca, Spain. Tel: +34 923291430.E-mail: abatuc@yahoo.es

(Received 11 October 2012; revised 20 November 2012; accepted 2 December 2012)

ISSN 0001-6489 print/ISSN 1651-2251 online � 2013 Informa HealthcareDOI: 10.3109/00016489.2012.757798

Act

a O

tola

ryng

ol D

ownl

oade

d fr

om in

form

ahea

lthca

re.c

om b

y M

rs S

hirl

ey D

emps

ter

on 0

5/20

/13

For

pers

onal

use

onl

y.

The purpose of this study was to assess the results inthe SVV test in patients with VS before and aftersurgery, and to analyze correlations with the DHI andcaloric stimulation.

Material and methods

Subjects

The study evaluated 24 patients diagnosed with aunilateral VS who were prospectively scheduled forsurgery between May 2008 and June 2010. All weresubjected to retrolabyrinthine or translabyrinthineVS surgery and were reviewed at 1, 3 and 6 monthspostoperatively. The study was performed in accor-dance with the ethical guidelines of the 1975 Declara-tion of Helsinki. Patients with other pathologies thatcould directly affect postural control were excluded.

Auditory and vestibular evaluation

Audiometric findings were reported in terms of thepure-tone averages (PTA), calculated by taking theaverage of the readings at four frequencies (0.5, 1,2 and 3 kHz).In all the patients the head-impulse test (HIT) was

performed before caloric testing. The physician stoodin front of the patient and grasped the patient’s headwith both hands at a distance of approximately 30 cm.The patient was told to maintain eyes fixed on theexaminer’s nose. The head of the patient was slowlyturned through 20� in the horizontal plane to theopposite side from that the physician wished toexplore (i.e. to the left if the right side was beingexplored). The patient’s head was then rapidly rotatedto the other side, maintaining it in the horizontalplane. In normal conditions (healthy patients), thephysician should observe that the patient is ableto maintain the eyes fixed on the stationary target(the physician’s nose) during the high-speed headrotation. When unilateral vestibular weakness exists,the eyes drift in the same direction as the head, andthen compensatory refixation saccades are used toreset the visual fixation on the target. The test wasrepeated three times on each side and was consideredas positive (pathological) when refixation saccadeswere observed in at least two of the three repetitions.The Fitzgerald and Hallpike bithermal caloric test

was used, in which eye movements were recorded bymeans of a video-based system (Ulmer VNG, v. 1.4,SYNAPSIS, Marseille, France). The maximum slowphase velocity (SPVmax) of nystagmus was calculatedafter each irrigation. The total response from each ear(TotE) was calculated, and caloric weakness (CW)and directional preponderance (DP) were determinedaccording to Jongkees’ formula.

Subjective visual vertical (SVV)

The SVV was examined using a custom-made faintlyilluminated bar visible in the darkness; no othervisual references were available during the test.The patient was sitting in a chair facing the barwith his/her head resting in such a manner thatthe Reid’s angle was horizontal. The bar was 1 meterin front of the patient and placed deviating fromthe vertical. The range of the baseline deviation ofthe SVV rod was displaced in all cases and variedbetween each test.The patient was then asked to set the bar precisely

in a vertical position with the use of a controlmodule. The different pretest tilts of the bar werepresented to the patient randomly in clockwise (CW)and counter-clockwise (CCW) directions. Once thepatient considered the procedure was complete,the amount of tilt was measured and the resultwas positive if CW deviated and negative if CCWdeviated. The SVV was determined from the averageof four consecutive measurements with the patientstarting from a random offset position. A tilt of morethan 2� with regards to the true vertical positionwas considered pathological [8]. This test wasperformed before surgery, again on day the patientwas discharged from the hospital, and 1, 3 and6 months later.

Disability and handicap assessment

The DHI questionnaire was translated and adaptedto the Spanish language following the methodof cross-translation. The reliability of the scale wasevaluated and the Cronbach alpha coefficientwas 0.9226 [9]. The aim of the questionnaire wasexplained by the doctor, and the patient filled out thequestionnaire. Where necessary, the technician incharge of the vestibular tests (who was well versedin the scope of the work and who had consider-able experience in the vestibular laboratory) helpedto clarify some items. The questionnaires wereanswered in a similar fashion to the original Englishversion. In the DHI, the patient had to answer‘yes’, ’sometimes’ or ‘no’ to each question, theresponses being given a value of 4, 2 and 0, respec-tively. The questionnaire has 25 items, so the totalscore (DHITS) ranged from 0 to 100. The ques-tionnaire was completed on the same day as the SVVevaluation.

Statistical analysis

The statistical software program SPSS 12.0.1 forWindows was used to analyze the data; a p value of

476 A. Batuecas-Caletrio et al.

Act

a O

tola

ryng

ol D

ownl

oade

d fr

om in

form

ahea

lthca

re.c

om b

y M

rs S

hirl

ey D

emps

ter

on 0

5/20

/13

For

pers

onal

use

onl

y.

<0.05 was considered to be statistically significant.Since the data were not normally distributed, non-parametric tests were used. We calculated frequencies,means and ranges for the variables studied. To recordand analyse the relation between variables, contingencytables were used and the measure of association wasdone with the c2 test. Spearman’s rank correlationcoefficient (rho) was the non-parametric measureused to assess statistical dependence between variables.Following radiological evaluation, patients wereclassified according to Koos et al. [10]: grade Iwhen the tumour was in the internal auditory canal(IAC); grade II when the tumour was in the ponto-cerebellar angle (PCA) and <2 cm; grade III when thetumour was in the IAC and 2–4 cm; and grade IV if

the tumour was in the IAC and >4 cm. According tocanal paresis the patients were classified into threegroups: group 1 or normal when canal paresis was<20%, group 2 or moderate if canal paresis was21–70% and group 3 or severe if it was >71% [11].

Results

Patients

During the period of this study (May 2008 to June2010), 24 patients were surgically treated: 15 womenand9men,withameanageof48±14years. In58%, theright side was affected. With regard to radiologicalevaluation 4 patients were classified as grade I,11 patients as grade II, 5 as grade III and 4 asgrade IV. Mean hearing level in the ear with the VSwas56±30dBHL.Meancanalparesiswas50±46%(inthe lesion side in all the patients) and the patientswere classified into three groups: 7 patients in group1 (0–20% canal paresis), 7 in group 2 (21–70% canalparesis) and10 ingroup3 (71–100%canalparesis).Thehead impulse test was correlated with the amount ofcanal paresis (c2 test, p = 0.002).Mean SVVwas 1.25 ±1o and itwas considerednormal in21/24patients.MeanDHI was 18 ± 10. Age, sex and tumour side did notcorrelate with the mean canal paresis or with the meanSVV(Figure1).Thesizeof the tumourandcanalparesiswere positively correlated (Spearman rho = 0.47,p = 0.022).

Post-treatment results

Figures 2 and 3 show the results for the SVV and DHIduring the post-surgical study period: at the time ofhospital discharge (mean 12 ± 4 days), and 1, 3 and6 months later. There was a good correlation betweenSVV and DHI results in each of the periods ofassessment (Pearson, p = 0.001).The amount of canal paresis before surgery was well

correlated (Pearson, p = 0.0001) with a better result inthe SVV test and DHI at post-surgery hospitaldischarge, and at 1- and 3-month follow-up. That

SVV before surgery

0

0.5

1

1.5

2

2.5

3

Grade I Grade II Grade III Grade IV

Figure 1. Subjective visual vertical (SVV) before surgery according to tumour size: 1 = grade I, 2 = grade II, 3 = grade III, 4 = grade IV. Each barrepresents an individual patient.

Evaluation time

SV

V

0

2.00

4.00

6.00

8.00

10.00

12.00

2

5983

66

6550

One week One month Three months Six months

VNG Group 1, 0−20%VNG Group 2, 21−70%VNG Group 3, 71−100%

Figure 2. Subjective visual vertical (SVV) follow-up score. Resultsare given for patients according to results in canal paresisgroups. Group 1, 0–20%; group 2, 21–70% and group 3,71–100%, at visit 1 (1 week after surgery), visit 2 (1 month aftersurgery), visit 3 (3 months after surgery) and visit 4 (6 months aftersurgery).

DHI and SVV in vestibular schwannoma 477

Act

a O

tola

ryng

ol D

ownl

oade

d fr

om in

form

ahea

lthca

re.c

om b

y M

rs S

hirl

ey D

emps

ter

on 0

5/20

/13

For

pers

onal

use

onl

y.

is, patients with a normal caloric test before surgeryhad worse results in SVV and DHI at 1 and 3 monthsafter surgery, and patients with great caloric weak-ness had better results in SVV and DHI at 1 and3 months after surgery. This correlation was notfound in the 6-month post-surgery results. Patientswith more caloric weakness had better results for SVVand DHI at the time of discharge and at 1- and3-month follow-up.

Discussion

The main findings in our study are the high correla-tion between the caloric weakness before surgery andthe SVV and DHI outcomes after surgery. The headimpulse test correlated with the amount of canalparesis, as did the size of the tumour (patients withlarge tumours had higher canal paresis).Balance is one of the most important aspects

regarding VS. Preoperative unsteadiness is commonand balance dysfunction is the most difficult aspect ofrecovery in 10% of patients. For this reason this aspectmust be taken into account in patients with VS. Thereare many variables that may influence the recovery ofpatients after VS surgery as far as balance is con-cerned. Most of them have been studied accuratelywithout reaching agreement on the main factorsaffecting recovery. As in other studies, we found norelationship between age or sex, or the affected sideand period of time for recovery of the patients [7]. In

this work we were interested in the follow-up of twomeasures of vestibular compensation: the SVV andthe DHI.Normal subjects are able to align an illuminated bar

within ±2� of accuracy in darkness, but patients whohave an acute unilateral peripheral vestibulopathysystematically tilt the bar towards the affected earabove that measure and, when compensation occurs,the results return to normal limits. This is because thesubjective visual perception of the verticality of a lineprojected ahead is affected by the head and bodyposition relative to gravity. In this case, the relativecontribution of utricular and vertical canal afferents tothe SVV is controversial.In the present study we have shown that, before

surgery, 87.5%of the patients displayed normal resultsfor the SVV independently of the amount of canalparesis or tumour size, as has been shown by otherswhen studying the subjective visual horizontal [12].Wehave seen that they systematically tilt the bar to the sideof thedisease close to2� before surgery.Webelieve thatthis finding reflects that compensation is proceeding toa normal extent in response to a deficient unilateralutricular nerve and function. This is expected, as ina recent study 75% of the patients with VS showedutricular dysfunction as measured with ocularvestibular evoked myogenic potentials [13].Long-term post-surgical evaluation of the SVV in

our patients has shown that the majority recovered tonormal values set at ±2� in accordance with what

Evaluation time

DH

I

0

20.00

40.00

60.00

80.00

100.002

26

8350

7490

One week One month Three months Six months

VNG Group 1, 0−20%VNG Group 2, 21−70%VNG Group 3, 71−100%

Figure 3. Dizziness Handicap Inventory (DHI) follow-up score. Results are given for patients according to results in canal paresis groups.Group 1, 0–20%; group 2, 21–70% and group 3, 71–100%, at visit 1 (1 week after surgery), visit 2 (1 month after surgery), visit 3 (3 monthsafter surgery) and visit 4 (6 months after surgery).

478 A. Batuecas-Caletrio et al.

Act

a O

tola

ryng

ol D

ownl

oade

d fr

om in

form

ahea

lthca

re.c

om b

y M

rs S

hirl

ey D

emps

ter

on 0

5/20

/13

For

pers

onal

use

onl

y.

others have obtained [3]. We consider that, for thefollow-up of vestibular compensation in the ocularcomponent of the disrupted vestibulo-ocular reflex,this is of great value as others have shown in the case ofvestibular neurectomy for Meniere’s disease [14] orVS [3,15].The importance of SVV relies on its sensitivity to

the behavioural component of the compensation pro-cess as shown by the visit-to-visit modification thatshows close similarity between SVV and DHI. Itwould be interesting to analyse in detail whether ornot visual dependence modifies this close similarity asexpected by findings in dynamic SVV [15] or dynamicvisual acuity [16].Regarding the DHI, there is a close similarity

between our results and those reported byUehara et al. [17]. In both studies the amount ofdisequilibrium at hospital discharge is correlated withthe degree of damage during surgery but also with theamount of canal paresis before treatment. In thatsense patients with an important vestibular deficiencybefore surgery display a lower amount of disability orhandicap in the immediate postoperative period. Thiswas not the case when only the size of the tumour wastaken into account, which confirms the importance ofpreoperative vestibular evaluation in these patients.This confirms the importance in the treatment ofthese patients of pretreatment before surgery withintratympanic gentamicin; in that way post-surgical rehabilitation will be facilitated [18].It is interesting to note that the results are not

significantly different at the 6-month postoperativefollow-up. This is probably related to the existenceof different variables that influence recovery through-out that period and the exposure of the patient to verydifferent environments in which all the sensory influ-ences for equilibrium and previous experience coop-erate to provide good visual stability [19].As noted before, both measures behave in a parallel

fashion throughout the period of follow-up [20]. Athospital discharge the perception of disability isvery high as well as the degree of misalignment inthe subjective vertical. With the passage of time, thepatient returns to normality in both measures. In thisway we can recommend the follow-up of the patientafter surgery with assistance of the SVV analysis asthe main objective measure, because it is easy toperform and the information is valuable. It wouldbe interesting to address the issue of the relevantimportance of utricles and vertical canals in thesepatients in a more detailed way, i.e. by placingthe patient in a deviated position with regard to theearth-vertical axis [21].Regarding the methodology, performing the study

under binocular condition is not a limitation as it has

been found that there were no significant differencesin the tilts of the SVV between binocular and mon-ocular measurements in VS patients [22].

Declaration of interest: The authors report noconflicts of interest. The authors alone are responsiblefor the content and writing of the paper.

References

[1] Curthoys IS, Dai MJ, Halmagyi GM. Human oculartorsional position before and after unilateral vestibularneurectomy. Exp Brain Res 1991;85:218–25.

[2] Bronstein AM, Pérennou DA, Guerraz M, Playford D,Rudge P. Dissociation of visual and haptic vertical in twopatients with vestibular nuclear lesions. Neurology 2003;61:1260–2.

[3] Hafström A, Fransson PA, Karlberg M, Magnusson M.Idiosyncratic compensation of the subjective visualhorizontal and vertical in 60 patients after unilateralvestibular deafferentation. Acta Otolaryngol 2004;124:165–71.

[4] Ogawa Y, Hayashi M, Otsuka K, Shimizu S, Inagaki T,Hagiwara A, et al. Subjective visual vertical in patientswith ear surgery. Acta Otolaryngol 2010;130:576–82.

[5] Parietti-Winkler C, Gauchard GC, Simon C, Perrin PP.Visual sensorial preference delays balance control compen-sation after vestibular schwannoma surgery. J Neurol Neu-rosurg Psychiatry 2008;79:1287–94.

[6] Hafstrom A, Fransson PA, Karlberg M, Magnusson M.Subjective visual tilt and lateral instability after vestibulardeafferentation. Acta Otolaryngol 2006;126:1176–81.

[7] Saman Y, Bamiou DE, Gleeson M. A contemporary reviewof balance dysfunction following vestibular schwannomasurgery. Laryngoscope 2009;119:2085–93.

[8] Friedman G. The judgement of the visual vertical and hor-izontal with peripheral and central vestibular lesions. Brain1970;93:313–28.

[9] Perez N, Garmendia I, García-Granero M, García-Tapia R.Factor analysis and correlation between Dizziness HandicapInventory and Dizziness Characteristics and impact onquality of life scales. Acta Otolaryngol Suppl 2001;545:145–54.

[10] Koos WT, Spetzler RF, Böck FW. 1976. Microsurgery ofcerebellopontine angle tumors. In Koos WT, Böck FW,Spetzler RF, editors. Clinical microneurosurgery. Stuttgart:Thieme. p 91–112.

[11] Tringali S, Charpiot A, Ould MB, Dubreuil C, Ferber-Viart C. Characteristics of 629 vestibular schwannomasaccording to preoperative caloric responses. Otol Neurotol2010;31:467–72.

[12] Ushio M, Iwasaki S, Chihara Y, Murofushi T. Abnormaldeviation of subjective visual horizontal in patients withvestibular schwannoma. Ann Otol Rhinol Laryngol 2008;117:641–4.

[13] Huang CH, Wang SJ, Young YH. Correlation betweencaloric and ocular vestibular evoked myogenic potentialtest results. Acta Otolaryngol 2011;131:123–5.

[14] Leveque M, Seidermann L, Tran H, Langagne T, Ulmer E,Chays A. Vestibular function outcomes after vestibular neur-ectomy in Meniere disease: can vestibular neurectomy pro-vide complete vestibular deafferentation? Auris Nasus Larynx2010;37:308–13.

DHI and SVV in vestibular schwannoma 479

Act

a O

tola

ryng

ol D

ownl

oade

d fr

om in

form

ahea

lthca

re.c

om b

y M

rs S

hirl

ey D

emps

ter

on 0

5/20

/13

For

pers

onal

use

onl

y.

[15] Goto F, Kobayashi H, Saito A, Higashino K, Kunihiro T,Kanzaki J. Compensatory changes in static and dynamicsubjective visual vertical in patients following vestibularschwannoma surgery. Auris Nasus Larynx 2003;30:29–33.

[16] Cohen HS, Bloomberg JJ. Modified dynamic visual acuitytests after acoustic neuroma resection. Acta Otolaryngol2007;127:825–8.

[17] Uehara N, Tanimoto H, Nishikawa T, Doi K, Katsunuma S,KimuraH, et al. Vestibular dysfunction and compensation afterremoval of acoustic neuroma. J Vestib Res 2011;21:289–95.

[18] Magnusson M, Kahlon B, Karlberg M, Lindberg S, Siesjo P.Preoperative vestibular ablation with gentamicin and vestib-ular ‘prehab’ enhance postoperative recovery after surgery forpontine angle tumors – first report. Acta Otolaryngol 2007;127:1236–40.

[19] Perez N, Martin E, Garcia-Tapia R. Dizziness: relating theseverity of vertigo to the degree of handicap by measuringvestibular impairment. Otolaryngol Head Neck Surg 2003;128:372–81.

[20] Humphriss RL, Baguley DM, Moffat DA. Change in dizzi-ness handicap after vestibular schwannoma excision. OtolNeurotol 2003;24:661–5.

[21] Tribukait A. Subjective visual horizontal in the uprightposture and asymmetry in roll-tilt perception: indepen-dent measures of vestibular function. J Vestib Res 2006;16:35–43.

[22] Sainoo Y, Terakado M, Fujiyama D, Kumagami H,Takahashi H. Binocular and monocular measurements ofsubjective visual vertical in vestibular loss. Eur Arch Otorhi-nolaryngol 2012;269:57–60.

480 A. Batuecas-Caletrio et al.

Act

a O

tola

ryng

ol D

ownl

oade

d fr

om in

form

ahea

lthca

re.c

om b

y M

rs S

hirl

ey D

emps

ter

on 0

5/20

/13

For

pers

onal

use

onl

y.