TESTING TOLERABILITY IN CHILDREN

I n t e r n a t i o n a l J o u r n a l o f P e d i a t r i c O t o r h i n o l a r y n g o l o g y 1 1 3 ( 2 0 1 8 ) 2 9 – 3 3 Contents lists available at ScienceDirect International Journal of Pediatric Otorhinolaryngology journal homepage: www.elsevier.com/locate/ijporl T Pediatric vestibular testing: Tolerability of test components in children Peter J. Cioleka, Elise Kangb, Julie A. Honakera, Erika A. Woodsona, Brandon S. Hopkinsa, Samantha Annea,∗ aHead and Neck Institute, Cleveland Clinic, Cleveland, OH, USA bLerner College of Medicine, Cleveland, OH, USA A R T I C L E I N F O A B S T R A C T Introduction: Objective of the study is to define rates of successful completion of components of pediatric vestibular testing (VT). Methods: Retrospective review of VT performed on patients less than 18 years of age from 2004 to 2015. Results: 188 pediatric patients (mean age: 13.9 ± 3.56 years old, range 2–17 years) underwent testing. Thirty- five (18.6%) had abnormal test results. Pediatric patients unable to complete all aspects of VT could still complete an average of 7.9 ± 4.0 of 12 test components. The optokinetic tracking test was the most commonly omitted component of the vestibular tests. In a multivariate analysis, failure to perform Nylen-Barany positional testing (χ227.5, p < 0.0001) or Dix-Hallpike (5.66, p= 0.0174) testing was associated with inability to obtain final diagnosis on VT. Conclusions: Interpretable VT may be obtained in most children, even in those that do not tolerate the full testing protocol. Spontaneous and gaze-evoked nystagmus testing maybe considered as part of initial testing protocol before attempting less well-tolerated components such as bithermal calorics or components that require VNG goggles. Keywords: Vestibular testing Pediatric Vertigo Dizziness 1. Introduction findings may not be compared to adult normative data. A skilled clinician must establish a rapport with the child and caregiver, anticipate potential problems, and conduct the VT accordingly, making modifications of the clinical examination based on age and overall tolerance to testing. Currently, there is a paucity of literature on the topic of pediatric vestibular testing and the ability of children to complete the components necessary to attain meaningful diagnostic data. The primary objective of our study was to define rates of successful completion of components of VT and identify factors predictive of cooperation in pediatric patients. Our secondary objective was to define rates of specific pathologies identified in children referred for vestibular testing in a tertiary care center. Pediatric vestibular disorders may present with a wide variety of symptoms including motor delay, torticollis, clumsiness, recurrent vo- miting, episodic spontaneous spells of dizziness, and vegetative in- activity [1,2]. The differential is diverse and differs markedly from adult vertigo [3]. Children may lack the language skills to adequately convey their symptoms to the clinician, adding a layer of complexity to the diagnosis. Workup of pediatric vestibular disorders may include a careful history, standardized questionnaires, routine physical examination, imaging, and more sophisticated testing included in a vestibular evaluation [4]. Vestibular testing (VT) is a valuable tool in evaluation of pediatric patients with dizziness. Furthermore, a thorough vestibular assessment is often indicated in cases of congenital hearing loss, such as prior to cochlear implantation. Vestibular testing in children presents sig- nificant challenges for the clinician. Testing may invoke unpleasant feelings of vertigo and nausea. Equipment such as the rotatory chair and video oculography goggles can be intimidating or not tailored to pediatric size. Additionally, based on maturation of the vestibular, visual, and proprioception systems, clinical decisions regarding abnormal 2. Methods This study is a retrospective review of VT performed on patients less than 18 years old at a tertiary care center from 2004 to 2015. The study was approved by the Institutional Review Board. The retrospective review of de-identified data included subtests from videonystagmography (VNG) and rotary chair assessment. VNG included standard subtests of ∗Corresponding author. Pediatric Otolaryngology, Head & Neck Institute, Cleveland Clinic Foundation, 9500 Euclid Ave / A-71, Cleveland, OH, 44195, USA. E-mail address: annes@ccf.org (S. Anne). https://doi.org/10.1016/j.ijporl.2018.07.009 Received 19 March 2018; Received in revised form 5 July 2018; Accepted 7 July 2018 0 1 6 5 - 5 8 7 6 / © 2 0 1 8 E l s e v i e r B . V . A l l r i g h t s r e s e r v e d . A v a i l a b l e o n l i n e 1 0 J u l y 2 0 1 8

I n t e r n a t i o n a l J o u r n a l o f P e d i a t r i c O t o r h i n o l a r y n g o l o g y 1 1 3 ( 2 0 1 8 ) 2 9 – 3 3 P.J. Ciolek et al. technicians had over 15 years of experience performing vestibular assessment and received annual training on the procedures. Modifications to the above battery were made based on age of patient. In general, in children between ages 2–3, rotary chair was attempted, often with the child positioned on a parent/caregivers lap. In instances where VNG goggles could not be fit properly or due to patient preference not to wear goggles, qualitative evaluation of VOR function via rotary chair was collected (i.e., present or absent nystagmus response). The VNG procedure was reserved for older (typically 5 years of age and older) children. Not all components of VNG or Rotary Chair were attempted on every patient. In general, the technicians always started with oculomotor testing, then rotational chair (SHA) followed by VNG (positional testing) and finally caloric irrigations. Results of the VT were reported as 12 components: central vestibulo-ocular pathway and peripheral vestibular system function, using Micromedical Technologies, System 2000 ™ (Chatham, IL). 2.1. VNG Patients wore 2D video eye goggles throughout testing, and cali- bration was conducted prior to eye movement recordings. Patients sat within 4 feet of an illuminating visual target. VNG ocularmotor examination consisted of saccade and smooth pursuit function and spontaneous nystagmus and gaze-evoked nystagmus testing. Saccade testing was conducted with a pseudorandom horizontal target presentation (5–50° steps). During smooth pursuit testing patients were instructed to follow a sinusoidal target moving from 0.1 to 0.4Hzat 10–60°/second. Spontaneous nystagmus searching was explored with eyes in primary position, with and without visual fixation. Gaze-evoked nystagmus was evaluated with eccentric gaze positions for both horizontal (25° target positioning to the right and left) and vertical (15° target positioning up and down). Evaluation of the presence of provocative nystagmus (i.e., Benign Paroxysmal Positional Vertigo, neck torsion and static positional nystagmus) was performed with positional testing (sitting head turn right and head turn left, supine head center, supine head right or right lateral and supine head left or left lateral) and standard Dix-Hallpike (with the additional of Nylen Barany) maneuvers. Dix-Hallpike maneuvers were performed by having the patient's head turned 45° to the right or left and brought back from a sitting to supine position. Whereas Nylen Barany maneuvers were performed with the patient sitting upright with head in primary position, then quickly moving to a supine position with head then turned 45° to the right or left. Finally, open loop bithermal (30 °C and 44°C) caloric irrigations were collected to determine presence of unilateral peripheral vestibular system paresis or directional preponderance of nystagmus responses. 1. Saccade Performance 2. Pursuit Tracking 3. Spontaneous & Gaze Evoke Nystagmus 4. Optokinetic tracking 5. Fixation Vestibular Interaction Rotational Chair 6. Sinusoidal Harmonic Vertical Axis Chair 7. Optokinetic-Vestibular Interaction Rotation Chair 8. Neck Torsion 9. Dix-Hallpike Positioning 10. Nylen-Barany Positioning 11. Positional Nystagmus 12. Caloric Testing Regardless of age, all testing began with a behavioral assessment including evaluation of gait and postural control. Following this objective testing was attempted; however, again not all aspects of VT were completed on every patient. Patients were referred for further sub- specialty evaluation and further testing including bloodwork and imaging as needed. All individual procedural reports were reviewed, and data was collected including patient demographics, testing compliance/ cooperation, testing results, and final medical diagnosis. Testing success was defined as completion of all attempted components. Testing was considered partially completed if patient did not cooperate with 1 or more attempted components. Final diagnosis was made after evaluation with a physician, with or without use of imaging studies, in conjunction with review of vestibular testing findings. Migraine disorder refers to patients that presented with migraine (headache, sensitivity to light/sound) either preceded, accompanied, or followed by symptoms of dizziness/vertigo, who were referred for vestibular testing. CNS Structural lesions include midline cerebellar defect, CPA tumor, Cerebellar lesion, Chiari malformation, and Symptomatic Rathke's cleft cysts. Cervicalgia refers to perceived symptoms of dizziness/imbalance accompanied by musculoskeletal pain of the neck. Motor delay, in itself, does not cause dizziness. However, these patients were initially referred for evaluation of dizziness/imbalance. Central unspecified refers to abnormal vestibular testing indicating a potential central pathology however no definitive etiology was identified. Summary statistics were calculated, and descriptive statistics of the sample are reported. Multivariate analysis was used to determine which components of the VT were essential to obtaining a final diagnosis. 2.2. Rotary chair Testing was conducted with Micromedical System 2000™ rotary chair using VNG 2D video eye goggle technique and included examination of optokinetic nystagmus and vestibular-ocular reflex (VOR) function via the Sinusoidal Harmonic Acceleration (SHA) test. Optokinetic testing was performed using full-field cylindrical surround presentation of visual targets (strips) presented at 50° per second. SHA testing was performed at 0.01, 0.02, 0.04, 0.08, 0.16, and 0.32 Hz. All frequencies were attempted on the children, with a mid-frequency response as a starting point (e.g., 0.08Hz), then low frequency responses if the child could tolerate the length of testing, alternating with high frequency responses. Suppression of the VOR response was confirmed with the patients visualizing a fixation target (red illuminating light) while rotating at 0.05 Hz, 50°/second and enhancement of the VOR response was confirmed with patients visualizing optokinetic strips while rotating at 0.05 Hz, at 50°/second. The very young children were recorded in the dark using video recordings and qualitatively evalu- ating the VOR response. The rotational chair results were interpreted based on laboratory norms for SHA testing. Normal VOR gain (mean values) based on frequency were as follows: 0.01 Hz (0.52), 0.02Hz (0.58), 0.04Hz (0.60), 0.08Hz (0.60), 0.16Hz (0.67), 0.32Hz (70), and 0.64 Hz (0.72). Normal responses fell within two standard deviations of normative data means. Abnormal responses could include single frequency (e.g., 0.01), reduced gain, phase and/or symmetry values that were clinically correlated with the patient's state and mental alertness and tasking at the time of testing. Step velocity testing was not performed. Time constants were not calculated for SHA testing. All VT was performed following a standard approach by trained technicians and interpreted by single reviewer. Due to the location of testing, most often one technician would perform the testing, but an- other technician may have helped in cases of young children. The 3. Results A total of 188 patients (Mean age: 13.9 ± 3.56 years old, range 2–17) underwent VT. Nineteen patients completed testing in the setting of cochlear implantation pre and/or post-evaluation. The remainder were tested after presenting for potential vestibular issues. Thirty-five (18.6%) children had abnormal test results. After combining clinical history and physical, VT, and other ancillary testing (e.g., imaging, labs), the most common final diagnosis was non-vestibular in origin 3 0

I n t e r n a t i o n a l J o u r n a l o f P e d i a t r i c O t o r h i n o l a r y n g o l o g y 1 1 3 ( 2 0 1 8 ) 2 9 – 3 3 P.J. Ciolek et al. optokinetic tracking test was not obtained in 37 encounters and was the most commonly omitted component within the vestibular tests (Fig. 1). Failure to cooperate during the vestibular testing limited full inter- pretation of results in 8 of 27 (29.6%) encounters. In a multivariate analysis, failure to perform Nylen-Barany positional testing (χ227.5, p < 0.0001) or Dix-Hallpike (5.66, p =0.0174) testing was associated with inability to obtain final diagnosis on VT. Table 1 Diagnosis in 169 Children Referred for Vestibular Testing for Symptoms of Dizziness and/or Vertigo. Diagnosis n % Central/Systemic Pathology Migraine Disorder Orthostatic Hypotension Psychogenic Concussion Cervicalgia CNS Structural Motor Delay Central Unspecified Total 45 17 9 7 5 5 3 6 97 26.6% 10.1% 5.3% 4.1% 3.0% 3.0% 1.8% 3.6% 57.4% 4. Discussion The prevalence of pediatric vestibular and balance disorders is re- latively low, however, likely to be underreported and underdiagnosed [5]. These disorders encompass a spectrum of pathologies. Common diagnoses in children include benign positional paroxysmal vertigo (BPPV and Meniere's [1,6]). Our data demonstrates migraine and Me- niere's to be common etiologies of peripheral disorder [1]. In our series, nearly a ¼ of the patients were diagnosed with migraine disorder. Migraine is the most common type of primary headache occurring in children [7]. About one quarter of migraine suffers report balance or vestibular symptoms that either precede, accompany, or remain following their headache [8–11]. The diagnosis of Meniere's disease as defined by rotary vertigo, hearing loss, tinnitus, and aural fullness in the pediatric population is rare Only 0.4–4% of patients that present with vestibular complaints are ultimately diagnosed with Meniere's [12]. This is similar to the rate in our series of 5%. It was not within the scope of this study to review details on how the final diagnosis was attained for migraine or meniere's disorders; further studies would need to clarify these findings. However, familiarity with the prevalence of specific pediatric vestibular and balance disorders can help guide the clinical evaluation and testing of the dizzy child. In prior studies, BPPV has been shown to a common etiology for peripheral vestibular disorder in children (6). In our own series, only Peripheral Vestibular Pathology Meniere's Disease BPVC Peripheral Vestibulopathy NOS Otitis Media Labyrinthitis SSCD BPPV Total 9 8 7 5 5 3 3 40 5.3% 4.7% 4.1% 3.0% 3.0% 1.8% 1.8% 23.7% No Final Diagnosis 32 17.8% Abbreviations: CNS, central nervous system; BPVC, Benign Paroxysmal Vertigo of Childhood; NOS, not otherwise specified; SSCD, Superior Semicircular Canal Dehiscence; BPPV, Benign Paroxysmal Positional Vertigo. (Table 1). A diagnosis was reached in 139 of 169 patients presenting with vestibular complaints. Patients unable to complete the all portions of the VT could still complete an average of 7.9 ± 4.0 of 12 components. Patients that were unable to tolerate were significantly younger than those that were able to tolerate (10.7 yrs vs. 14.5, p < 0.001). The Fig. 1. Completion of Vestibular Testing by Component that children were‘able to tolerate’ versus ‘unable to tolerate’. Order of components is by increasing tolerability. 3 1

I n t e r n a t i o n a l J o u r n a l o f P e d i a t r i c O t o r h i n o l a r y n g o l o g y 1 1 3 ( 2 0 1 8 ) 2 9 – 3 3 P.J. Ciolek et al. alternative to standard rotary chair (low-mid frequency response) and caloric irrigations (low frequency response). Patients who are unable tolerate the video goggles can participate in the bedside head thrust test with the use of an interesting fixation target. Furthermore, other ele- ments of the above objective examination can be performed at bedside as a starting point to collect meaningful information in the pediatric setting. For example, oculomotor assessment such as tracking an object can be assessed by qualitatively reporting age expected performance with the task. All portions of the testing can be modified and performed as part of direct office examination until the child can tolerate formal objective clinical examination. Additional considerations of VEMP testing, utilizing both cervical and ocular VEMP protocols can provide examination of both otolith organs as well as integrity of the superior and inferior vestibular nerves [4]. Similar to pediatric audiometric evaluation, utilizing assistants to collect the data and to distract or hold the interest in the patients can potentially increase overall compliance. While components of VT can be utilized in children as young as infants, it is important to note that pediatric norms should be used as comparisons when incorporating objective measures. The study is limited by the retrospective review of medical records with limited data extraction on diagnostic information used to desig- nate final diagnoses. It should be noted that the Micromedical 2000 unit is an enclosed rotational chair system. While enclosed rotation chair systems have advantages in terms of increased sensitivity and reliability by ensuring a light tight room, the enclosure may have limited some testing with the young pediatric population or adolescents. Newer rotational chair systems offer open (non-enclosure) systems and may be more favorable for the pediatric population by mitigating any fears associated with the enclosure room. about 2% of children were diagnosed with BPPV. Given the retrospective data review, it is uncertain why the patients had trouble with positional testing. It could be related to patient age, but may have been depended on a case by case basis due to medical co-morbidities. In general, most children are able to complete positional testing and our practice is to make modifications if necessary for any patient if we are unable to perform the DH. This might have resulted in some cases of BPPV to have been missed and would have otherwise potentially resulted in even a higher rate of diagnosis of BPPV. In our series, peripheral vestibulopathy was diagnosed in less than 18% of children referred for evaluation of dizziness, vertigo, or imbalance. The incidence of pediatric peripheral vestibulpathy is ill-defined in the literature [1]. In their large single center study, O'Reilly et al. found that 27% of patients referred for vestibular testing had a peripheral vestibulopathy. Meniere's disease, otitis media, and labyrinthitis were the most common pathologies identified in our series [13]. In addition, 3 children were found to have superior semicircular canal dehiscence on imaging. Two patients were referred for vestibular testing because of vertigo and were noted to have radiographic evi- dence of SSC. One patient was diagnosed based on radiographic evi- dence with symptoms including vertigo, aural fullness, autophony. None of the patients had vestibular evoked myogenic potentials. In the largest systematic review on the topic of vestibular disorders in children, SSCD was not reported in any of the 10 included articles [1]. Radiographic SSCD is found in less than 2% of children, and may de- crease with age [14]. Symptomatic SSCD is even rarer, but must remain on the differential, even in the pediatric population [15,16]. VT remains a useful adjunct for testing pediatric balance disorders. Vestibular tests developed for adults have been adapted to the pediatric population and normative data is now well established [4]. Clinicians may be wary of referring pediatric patients for VT for a variety of reasons including 1) the invasive nature of some portions of the battery, 2) expected poor patient cooperation, 3) lack of available pediatric testing centers, and 4) anticipation of poor quality or uninterpretable results. VT can be an anxiety provoking experience for the pediatric patient. Testing often involves dark rooms, intimidating equipment, and by its nature often invokes unpleasant feelings of vertigo and nausea. Testing centers should develop strategies and protocols to achieve interpretable results, even in the patients that are unable to tolerate all components of testing. The current study demonstrates that interpretable VT may be obtained in most children, even in those that do not tolerate the full testing protocol. Based on our findings, testing laboratories should choose to perform better tolerated components such as spontaneous and gaze-evoked nystagmus testing before attempting bithermal calorics or components that require VNG goggles. It comes as no surprise that with increasing age, comes improved cooperation with testing. Children 10 years or younger, seem to tolerate testing about half the time. Despite this, poor cooperation only limited our ability to interpret VT results in 30% of encounters. The work presented highlights the compliance rate when utilizing standardized VT in pediatric patients, but the results are limited to traditional computerized assessment measures that were previously recommended for all patients seeking this clinical evaluation in our facility. Modern vestibular assessment techniques, not previously collected in this retrospective review, but now included may be more suitable for pediatric population and overall compliance. Emerging technologies such as video head impulse testing (VHIT) achieve meaningful results in children that do not require darkened-rooms or the provocation of dizziness although pediatric patients do need to wear video googles that may still be ill-fitting. VHIT evaluates all six semicircular canals and is a viable complement to traditional testing (caloric irrigations and rotational chair testing) [17]. VHIT evaluates high frequency angular vestibulo-ocular reflex (VOR) responses and positive findings are indicators of impaired vestibular function [18]. Given the frequency response of the vestibular system, they are not a complete 5. Conclusions Interpretable VT may be obtained in most children, even in those that do not tolerate the full testing protocol. Spontaneous and gaze- evoked nystagmus testing maybe considered as part of initial testing protocol before attempting less well-tolerated components such as bithermal calorics or components that require VNG goggles. ‘ Conflicts of interest None. Funding sources None. References [1] F.M. Gioacchini, M. Alicandri-ciufelli, S. Kaleci, G. Magliulo, M. Re, Prevalence and diagnosis of vestibular disorders in children: a review, Int. J. Pediatr. Otorhinolaryngol. 78 (5) (2014) 718–724. [2] C.M. Bower, R.T. Cotton, The spectrum of vertigo in children, Arch. Otolaryngol. Head Neck Surg. 121 (8) (1995) 911–915. [3] S.H. Erbek, S.S. Erbek, I. Yilmaz, et al., Vertigo in childhood: a clinical experience, Int. J. Pediatr. Otorhinolaryngol. 70 (9) (2006) 1547–1554. [4] L.M. Valente, Assessment techniques for vestibular evaluation in pediatric patients, Otolaryngol. Clin. 44 (2) (2011) 273–290 (vii). [5] R.C. O'reilly, T. Morlet, B.D. Nicholas, et al., Prevalence of vestibular and balance disorders in children, Otol. Neurotol. 31 (9) (2010) 1441–1444. [6] Brodsky JR, LipsonS, R Wilber, G Zhou, Benign paroxysmal positional vertigo (BPPV) in children and adolescents: clinical features and response to therapy in 110 pediatric patients, Otol. Neurotol. 39(3):1. [7] M.A. Arruda, V. Guidetti, F. Galli, R.C. Albuquerque, M.E. Bigal, Primary headaches in childhood–a population-based study, Cephalalgia 30 (9) (2010) 1056–1064. [8] H. Neuhauser, M. Leopold, M. Von brevern, G. Arnold, T. Lempert, The interrela- tions of migraine, vertigo, and migrainous vertigo, Neurology 56 (4) (2001) 436–441. [9] J.U. Toglia, D. Thomas, A. Kuritzky, Common migraine and vestibular function. Electronystagmographic study and pathogenesis, Ann. Otol. Rhinol. Laryngol. 90 (3 Pt 1) (1981) 267–271. 3 2

I n t e r n a t i o n a l J o u r n a l o f P e d i a t r i c O t o r h i n o l a r y n g o l o g y 1 1 3 ( 2 0 1 8 ) 2 9 – 3 3 P.J. Ciolek et al. [15] C. Lagman, V. Ong, L.K. Chung, et al., Pediatric superior semicircular canal dehiscence: illustrative case and systematic review, J. Neurosurg. Pediatr. 20 (2) (2017) 196–203. [16] G.S. Lee, G. Zhou, D. Poe, et al., Clinical experience in diagnosis and management of superior semicircular canal dehiscence in children, Laryngoscope 121 (10) (2011) 2256–2261. [17] S.S. Hamilton, G. Zhou, J.R. Brodsky, Video head impulse testing (VHIT) in the pediatric population, Int. J. Pediatr. Otorhinolaryngol. 79 (8) (2015) 1283–1287. [18] H.G. Macdougall, L.A. Mcgarvie, G.M. Halmagyi, I.S. Curthoys, K.P. Weber, Application of the video head impulse test to detect vertical semicircular canal dysfunction, Otol. Neurotol. 34 (6) (2013) 974–979. [10] A. Szirmai, Vestibular disorders in patients with migraine, Eur. Arch. Oto-Rhino- Laryngol. 254 (Suppl 1) (1997) S55–S57. [11] P. Weisleder, T.D. Fife, Dizziness and headache: a common association in children and adolescents, J. Child Neurol. 16 (10) (2001) 727–730. [12] H. Akagi, K. Yuen, Y. Maeda, et al., Ménière's disease in childhood, Int. J. Pediatr. Otorhinolaryngol. 61 (3) (2001) 259–264. [13] R.C. O'reilly, J. Greywoode, T. Morlet, et al., Comprehensive vestibular and balance testing in the dizzy pediatric population, Otolaryngol. Head Neck Surg. 144 (2) (2011) 142–148. [14] N.M. Jackson, L.M. Allen, B. Morell, et al., The relationship of age and radiographic incidence of superior semicircular canal dehiscence in pediatric patients, Otol. Neurotol. 36 (1) (2015) 99–105. 3 3

TESTING TOLERABILITY IN CHILDREN