546605
research-article2014
AORXXX10.1177/0003489414546605Annals of Otology, Rhinology & LaryngologyLevi et al
Article
Magnetic Resonance Imaging Findings in Children With Tinnitus
Annals of Otology, Rhinology & Laryngology 2015, Vol. 124(2) 126–131 © The Author(s) 2014 Reprints and permissions: sagepub.com/journalsPermissions.nav DOI: 10.1177/0003489414546605 aor.sagepub.com
Eric Levi, MBBS, MPHTM1, Elhamy K. Bekhit, MBBS, FRANZCR2,3, and Robert G. Berkowitz, MD, FRACS1,4,5
Abstract Objective: Tinnitus in adults is generally investigated by contrast-enhanced magnetic resonance imaging (MRI) to rule out the diagnosis of acoustic neuroma. Acoustic neuroma is rare in children and, therefore, the role of MRI in children with tinnitus is unclear. This study was undertaken to determine the value of MRI in the investigation of tinnitus in children. Methods: Retrospective study of children younger than 18 years who underwent MRI for the investigation of tinnitus over a 10-year period. Results: Sixty-five patients were identified, but there were only 34 who had also undergone audiologic assessment. Among the 25 patients with normal audiology, MRI abnormalities were present in 9, but these were all thought to be nonspecific. Nine patients had abnormal audiograms and the MRI was abnormal in 4 of these cases, which included 3 children who were found to have multiple sclerosis. Conclusion: Magnetic resonance imaging would appear to be mandatory in the investigation of tinnitus in children who are found to have sensorineural hearing loss, particularly to rule out the diagnosis of multiple sclerosis. Although our study does not support the routine use of MRI in children with normal audiology, the numbers in our series are too small for a conclusive recommendation. Keywords children, MRI, tinnitus
Introduction Tinnitus is a subjective experience of noise, such as buzzing, ringing, or whistling, in the absence of any external auditory stimuli. The prevalence of tinnitus increases with age, peaking at 14.3% between 60 and 69 years of age. In a large study of more than 14 000 people, the prevalence of tinnitus is found to be highest among older adults, former smokers, and adults with hypertension, hearing impairment, loud noise exposure, or generalized anxiety disorder.1 Savastano2 reported that tinnitus occurs at some stage in 34% of children, but only 6.5% were found to complain spontaneously. Of these, 76% demonstrated normal hearing. Using a more rigorous questioning to avoid bias in children, Stouffer et al3 placed the prevalence of tinnitus in normal hearing children at 6% to 13% and at 24% to 29% in those children with hearing impairment. In children, there are a few risk factors associated with tinnitus that have been explored by Coelho et al,4 including age, sex, hearing loss, history of noise exposure, motion sickness, and hyperacusis. In adults, tinnitus is generally classified as being
pulsatile or nonpulsatile, to help direct the appropriateness of investigation.5 However, in children, qualitative description of tinnitus is hard to ascertain, as children are not often able to characterize their tinnitus.3 Tyler and Smith6 have suggested the use of the terms congenital versus acquired and middle ear versus sensorineural tinnitus as useful etiological classifications. The latter differentiates in particular those children in whom tinnitus may be related to otitis 1
Department of Otolaryngology, Royal Children’s Hospital Melbourne, Victoria, Australia 2 Department of Medical Imaging, Royal Children’s Hospital Melbourne, Victoria, Australia 3 Department of Radiology, The University of Melbourne, Melbourne, Australia 4 Murdoch Children’s Research Institute, The University of Melbourne, Melbourne, Australia 5 Australian School of Advanced Medicine, Macquarie University, Sydney, Australia Corresponding Author: Eric Levi, MBBS, MPHTM, 28 Malabar Road, Blackburn, Victoria 3130, Australia. Email: [email protected]
Downloaded from aor.sagepub.com at MCMASTER UNIV LIBRARY on March 24, 2015
127
Levi et al media from those in whom tinnitus may be related to significant inner ear or brainstem pathology. In adults, tinnitus, especially unilateral nonpulsatile tinnitus, may be due to acoustic neuroma. Acoustic neuromas are benign tumors of the eighth cranial nerve occurring at the cerebellopontine angle, with only about 10 acoustic neuromas newly diagnosed each year per million persons,7 corresponding to between 2000 and 3000 new cases each year in the United States. Acoustic neuromas occur largely in adults and are very rare in children. Only 39 cases in children had been reported in the literature as of 2001.8 The standard investigation for tinnitus includes an audiogram and magnetic resonance imaging (MRI).5 Gadoliniumenhanced MRI still represents the gold standard investigation in adults reporting nonpulsatile tinnitus.5 Whereas MRI has been shown to be of benefit in the investigation of tinnitus in adults to identify acoustic neuroma,9,10 this condition is rare in the pediatric age group, and therefore the value of performing an MRI in children with tinnitus is unclear. Magnetic resonance imaging in children may require the use of sedation or anesthesia, and therefore it is costly and may involve some risk. Magnetic resonance imaging findings in the investigation of tinnitus in adults are established but have not yet been confirmed in children. With the aim of determining the value of MRI in the investigation of children presenting with tinnitus, we carried out a 10-year institutional retrospective review of MRI findings in the investigation of tinnitus in children.
Methods After approval from our institutional ethics committee (HREC 31258A), a keyword search of the database of the Department of Medical Imaging at the Royal Children’s Hospital Melbourne, Australia, was performed. All MRI studies conducted in the 10-year period prior to January 1, 2012, in children 18 years of age or younger where the term tinnitus was recorded on the request or report were extracted. These MRI findings were then reviewed by an attending radiologist (E.K.B.) to ensure uniformity of reporting. Clinical data were gathered from retrospective chart review of patient histories. Patient characteristics including age, sex, tinnitus laterality, quality, duration, relevant otological past history, examination findings, and audiology reports were recorded. Patients without formal audiograms were excluded from the study. In this study, normal hearing was regarded as an audiometrically confirmed pure-tone average hearing sensitivity at 500 Hz, 1000 Hz, and 2000 Hz of 25 dB or better, and a type A tympanogram. Any deviation from this definition was regarded as an abnormal audiogram or abnormal hearing. We then evaluated the MRI findings to determine effect on the diagnosis and management of these children. An abnormal MRI report was regarded as one where any abnormal
signal change arising from a pathological lesion or anything other than a normal anatomy or an acceptable normal variant of that anatomy was present.
Results Sixty-five MRI studies were identified with 29 males and 36 females, average age 11.5 years (range, 3-18 years). The MRI was requested by Otolaryngology in 40 cases, Pediatric Neurology in 20, and Neurosurgery in 5. However, audiometry was performed in only 34 patients, and the remainder were excluded from this study. The excluded group contained 1 patient with an incidental Chiari 1 malformation, 1 patient with an enlarged right vestibular aqueduct, and 1 case of posttraumatic right extradural hematoma. The distribution of patients is detailed in Figure 1. Of the 34 patients for analysis, there were 25 with normal hearing on audiometry and 9 with abnormal audiological findings. In the group with normal audiometry, there were 16 normal MRI studies and 9 abnormal MRI findings, detailed in Table 1. None of the abnormal MRI findings were related to pathology involving the cerebellopontine angle or the vestibulocochlear system. The most common finding was sinus disease, present in 4 cases. In the group of 9 patients with tinnitus and abnormal audiometry, the MRI was normal in 5 children and abnormal in 4. The details of these 9 children are listed in Table 2. This group included 3 children who were found on MRI to have multiple sclerosis (MS) (Figure 2 for representative image). All 3 of these children presented with unilateral sensorineural hearing loss, and this was the only symptom at presentation in 1 child. In the other abnormal MRI, postinfective demyelination was identified. The other 5 children with abnormal audiometry but normal MRI findings included 2 children who had clinical and audiological signs of middle ear disease.
Discussion Our study has looked at the role of MRI in the investigation of tinnitus in the pediatric population. To our knowledge, there has not been any published data in the literature specifically exploring MRI findings in children with tinnitus. In a group of 34 children, we identified 3 children with the diagnosis of MS. All 3 children in this study had asymmetrical sensorineural hearing loss and tinnitus. We have previously reported on 1 of these children, who presented with acute tinnitus and sudden hearing loss as the initial complaint leading to the diagnosis of MS.11 Multiple sclerosis is an inflammatory immune-mediated disease affecting myelinated axons in the central nervous system. The hallmark of MS is symptomatic episodes that occur months or years apart and affect different central nervous system anatomic sites. Therefore, MS is a diagnosis made on a combination of clinical and radiological grounds. The
Downloaded from aor.sagepub.com at MCMASTER UNIV LIBRARY on March 24, 2015
128
Annals of Otology, Rhinology & Laryngology 124(2)
Figure 1. Representation of result. Table 1. Patients With Tinnitus, Normal Audiometry, and Abnormal Findings on MRI. Patient 1 2 3 4
3F 17M 9F 11M
Tinnitus Laterality
Tinnitus Quality
Other Clinical Findings
Bilateral Bilateral Bilateral Bilateral
High-pitched Nonspecific High-pitched Nonspecific
Disequilibrium Disequilibrium Nil Nil
5 9F
Bilateral
Nonspecific
6 16M
Unilateral (left)
High-pitched
Disequilibrium, nausea, migraine, chronic fatigue syndrome Ehlers-Danlos syndrome, disequilibrium, headaches
7 16F
Bilateral
High-pitched
8 14M
Bilateral
High-pitched
Difficulty understanding speech, depression, chronic fatigue syndrome NF1
9 15F
Bilateral
High-pitched
Nil
MRI Findings Right sphenoid and maxillary disease Paranasal sinus disease Maxillary and ethmoid sinus disease Sinus disease and borderline cerebellar tonsillar ectopia Blake’s pouch cyst Nodular heterotopia subependymal position paralleling bilateral lateral ventricles Hyperintensities on corpus callosum; multiple sclerosis ruled out on repeat MRI Left corona radiate small focus of signal intensity Bilateral temporal lobes mild increased signal intensity
Abbreviations: MRI, magnetic resonance imaging; NF1, neurofibromatosis type 1.
Downloaded from aor.sagepub.com at MCMASTER UNIV LIBRARY on March 24, 2015
129
Levi et al Table 2. Patients With Tinnitus and Abnormal Audiometry. Patient
Tinnitus Laterality
Tinnitus Quality
Other Clinical Findings Sudden left hearing loss CN5&7 palsies, ataxia, dysarthria, nystagmus
1 13M
Unilateral (left)
High-pitched
2 15M
Unilateral (left)
Nonspecific
3 13F
Unilateral (right)
Buzzing
4 15F
Unilateral (left)
High-pitched
5 11M
Bilateral
Nonspecific
6 13F
Unilateral (right)
Nonspecific
7 10F
Bilateral
High-pitched
8 18F
Unilateral (right)
Humming
9 15M
Bilateral
Nonspecific
Audiology
Details
Unilateral SNHL Left moderate to severe SNHL Unilateral SNHL Left mild to moderate SNHL
MRI Findings Normal
Multifocal white matter hyperintensities consistent with demyelination (MS) (Figure 2) Unilateral right hearing Unilateral SNHL Right mild to Multifocal loss over 6 weeks moderate white matter SNHL hyperintensities consistent with demyelination (MS) Left hearing loss, ataxia Unilateral SNHL Left mild to Multifocal moderate white matter SNHL hyperintensities consistent with demyelination (MS) Developmental delay; Bilateral SNHL Asymmetrical Asymmetrical right middle ear SNHL: right hyperintensities abnormality. MS mild to in frontal, ruled out on repeat moderate parietal, and MRI SNHL with posterior additional mild temporal lobes conductive loss; consistent with left mild SNHL postinfective demyelination change Hearing loss Bilateral SNHL Mild symmetrical Normal SNHL Hearing loss Bilateral SNHL Mild symmetrical Normal SNHL Previous MEVTs Bilateral CHL Mild symmetrical Normal CHL Previous MEVTs Bilateral CHL Mild symmetrical Normal CHL
Abbreviations: CHL, conductive hearing loss; CN, cranial nerve; MEVT, middle ear ventilation tube; MRI, magnetic resonance imaging; MS, multiple sclerosis; SNHL, sensorineural hearing loss.
American Academy of Neurology Practice Guidelines12 advocates for the use of MRI in making an early diagnosis of MS that would guide early intervention. Tinnitus is an uncommon complaint in patients with MS. Less than 2% of patients with MS report tinnitus and hearing loss.13 In a series by Fischer et al13 on patients with MS, only 12 out of 705 patients experienced acute hearing loss. In our series, MS was diagnosed in 3 out of 9 children with tinnitus and hearing loss, and it is clear that performing an MRI is essential in this group of children. In children with tinnitus but otherwise normal hearing, we found MRI abnormalities in 9 of 25 children, but none of these were related to any otologic or significant central
nervous system pathology. The most common finding was that of incidental paranasal sinus disease. This finding and other abnormalities listed in Table 1 are of questionable relevance to the presenting complaint of tinnitus in our study population. von Kalle et al14 studied paranasal sinus magnetic resonance images and found that mucosal swelling in paranasal sinuses and in mastoid cells is a frequent incidental finding in children and is not necessarily a sign of disease. The significance of the associated sinus disease and minor central nervous system abnormalities detected is therefore unclear. Coupled with the relatively small patient numbers in the study, we cannot determine whether or not MRI should be recommended as part of the routine
Downloaded from aor.sagepub.com at MCMASTER UNIV LIBRARY on March 24, 2015
130
Annals of Otology, Rhinology & Laryngology 124(2)
Figure 2. Coronal Fluid Attenuation Inversion Recovery (FLAIR) image demonstrating multifocal white matter hyperintensities (arrows), consistent with demyelination and multiple sclerosis (MS).
evaluation of a child presenting with tinnitus who is found to have normal audiology. Although the MRI study itself is relatively safe, performing an MRI in some children may require sedation and therefore may be associated with some risk and additional cost in comparison with MRI in adults. Tinnitus is a common complaint in adults. Sixteen percent to 19% of people older than 17 years have reported experiencing spontaneous tinnitus at some stage. Of these individuals, 8% experienced tinnitus as a moderate to severe disturbance. Tinnitus in children, however, may be a clinically under-recognized problem. Its prevalence has been reported in various studies as being in the order of 6% to 34%.2,15,16 Some authors believe that these percentages may potentially be either an underestimation or an overestimation due to difficulties with language in children or due to the children’s desire to satisfy the questioner.17 Tinnitus can occur in children with normal hearing and those with hearing deficits. Mills et al16,18 suggested that up to 29% of normally hearing children reported having had tinnitus,16 compared to 43.9% of children with conductive hearing loss reporting tinnitus and 29.5% of those with sensorineural hearing loss.18 In our series of children with tinnitus, 73.5% had normal hearing and 26.5% had abnormal hearing confirmed on audiometry. Tinnitus in children may potentially be due to past or present middle ear disease. This was carefully looked at by
Savastano,2 who found that in children who complained of tinnitus, 14% had clinical features of middle ear inflammation and 14.6% had abnormal negative pressure tympanometry. However, these rates are not statistically significant compared to the group who did not complain of tinnitus. A history of middle ear disease was reported in 31.6%, but it was not thought to be a significant factor in the genesis of tinnitus, as there was no statistical difference in the incidence of tinnitus between the group of children who had middle ear disease and those without middle ear disease.2 In our study population, there were 2 children who had clinical and audiological features of middle ear disease with mild conductive hearing loss and type B tympanograms. They both had normal MRI reports. This is consistent with Tyler and Smith’s6 proposed etiology of middle ear tinnitus. It would appear that performing an MRI for tinnitus in the context of a child with middle ear disease is less likely to identify an associated abnormality, compared with a child who has sensorineural hearing loss. Our study underscores the importance of considering an MRI to investigate tinnitus in a child with sensorineural hearing loss because of the potential to make an early diagnosis of MS. In the adult population, imaging investigation, usually with a gadolinium-enhanced MRI, is directed at confirming pathology in the vestibulocochlear nerve or cerebellopontine angle.5 Weissman and Hirsch19 found that most adult patients with subjective tinnitus and normal examination have no imaging abnormalities. However, Sonmez and colleagues20 reviewed 74 adult patients and described radiographic abnormalities in 61.9% of patients with subjective tinnitus and in all patients with pulsatile tinnitus. On closer inspection, the radiographic abnormalities, namely, high jugular bulb, atherosclerosis, dehiscent jugular bulb, aneurysm of internal carotid artery, dural arteriovenous fistula, aberrant internal carotid artery, jugular diverticulum, and glomus tumor, may not necessarily be the true cause of tinnitus, as they can be found incidentally in patients who are asymptomatic.21 It is unclear what percentage of these imaging abnormalities was the true cause of tinnitus in these patients. Other studies have investigated various temporal bone anomalies that have shown no significant clinical correlation with tinnitus.21 A large study would be required to have enough statistical power to fully determine the relationship between common or uncommon normal variants and their associated clinical symptoms.22 The prevalence of cerebellopontine tumor in children is rare.8 We have not found any children in our retrospective 10-year series with an intracranial or cerebellopontine tumor presenting with tinnitus. Hence, the investigation of a child with tinnitus by means of a gadolinium-enhanced MRI is of limited benefit. Based on our experience, the role of MRI in the evaluation of tinnitus appears to be primarily to identify or rule out the diagnosis of MS.
Downloaded from aor.sagepub.com at MCMASTER UNIV LIBRARY on March 24, 2015
131
Levi et al
Conclusion Magnetic resonance imaging would appear to be mandatory in the investigation of tinnitus in children who are found to have sensorineural hearing loss, particularly to rule out the diagnosis of MS. Although our study does not support the routine use of MRI in children with normal audiology, the numbers in our series are too small for a conclusive recommendation. Authors’ Note E.L., E.K.B., and R.G.B. were involved in all stages of the preparation of this article. E.K.B. is a staff radiologist who reviewed all MRI images and reports examined in this article. R.G.B. is a professor of pediatric otolaryngology and senior author of this article.
Declaration of Conflicting Interests The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding The author(s) received no financial support for the research, authorship, and/or publication of this article.
References 1. Shargorodsky J, Curhan GC, Farwell WR. Prevalence and characteristics of tinnitus among US adults. Am J Med. 2010;123(8):711-718. 2. Savastano M. Characteristics of tinnitus in childhood. Eur J Pediatr. 2007;166:797-801. 3. Stouffer JL, Tyler RS, Booth JC, Buckrell B. Tinnitus in normal-hearing and hearing-impaired children. In: Aran JM, Dauman R, eds. Tinnitus 91—Proceedings of the Fourth International Tinnitus Seminar. Amsterdam, The Netherlands: Kugler Publications; 1992:255-258. 4. Coelho C, Sanchez TG, Tyler RS. Tinnitus in children and associated risk factors. Prog Brain Res. 2007;166:179-194. 5. Sajisevi M, Weissman JL, Kaylie DM. What is the role of imaging in tinnitus? Laryngoscope. 2014;124:583-584. 6. Tyler RS, Smith RJ. Management of tinnitus in children. In: Newton VE, ed. Paediatric Audiological Medicine. Philadelphia, PA: Whurr Publishers; 2002:397-404. 7. Evans DG, Moran A, King A, Saeed S, Gurusinghe N, Ramsden R. Incidence of vestibular schwannoma and neurofibromatosis
2 in the North West of England over a 10-year period: higher incidence than previously thought. Otol Neurotol. 2005;26(1): 93-97. 8. Pothula VB, Lesser T, Mallucci C, May P, Foy P. Vestibular schwannomas in children. Otol Neurotol. 2001;22(6):903-907. 9. Sidman J, Carrasco V, Whaley R. Gadolinium the new gold standard for diagnosing cerebellopontine angle tumours. Arch Otolaryngol Head Neck Surg. 1989;115:1244-1277. 10. Murphy MR, Selesnick SH. Cost-effective diagnosis of acoustic neuromas: a philosophical, macroeconomic, and technological decision. Otolaryngol Head Neck Surg. 2002;127:253-259. 11. Rodriguez-Casero MV, Mandelstam S, Kornberg AJ, Berkowitz RG. Acute tinnitus and hearing loss as the initial symptom of multiple sclerosis in a child. Int J Ped Oto. 2005;69:123-126. 12. American Academy of Neurology. Practice guidelines: multiple sclerosis. https://www.aan.com/Guidelines/Home/ ByTopic?topicId=18. Accessed August 15, 2014. 13. Fischer C, Mauguiere F, Ibanez V, Confavreux C, Chazot G. The acute deafness of definite multiple sclerosis. Clin Neurophysiol. 1985;61(1):7-15. 14. von Kalle T, Fabig-Moritz C, Heumann H, Winkler P. Incidental findings in paranasal sinuses and mastoid cells: a cross-sectional magnetic resonance imaging (MRI) study in a pediatric radiology department. Rofo: Fortschritte auf dem Gebiete der Rontgenstrahlen und der Nuklearmedizin. 2012;184(7):629-634. 15. Nodar RH. Tinnitus aurium in school age children. J Aud Res. 1972;12:133-135. 16. Mills RP, Albert DM, Brain CE. Tinnitus in childhood. Clin Otolaryngol Allied Sci. 1986;11:431-434. 17. Baguley DM, McFerran DJ. Tinnitus in childhood. Int J Pediatr Otorhinolaryngol. 1999;49:99-105. 18. Mills RP, Cherry JR. Subjective tinnitus in children with otological disorders. Int J Paediatr Otorhinolaryngol. 1984;7:21-27. 19. Weissman JL, Hirsch BE. Imaging of tinnitus: a review. Radiology. 2000;216(2):342-349. 20. Sonmez G, Basekim CC, Ozturk E, Gungor A, Kizilkaya E. Imaging of pulsatile tinnitus: a review of 74 patients. Clin Imaging. 2007;31(2):102-108. 21. Koesling S, Kunkel P, Schul T. Vascular anomalies, sutures and small canals of the temporal bone on axial CT. Eur J Radiol. 2005;54(3):335-343. 22. Kang M, Escott E. Imaging of tinnitus. Otolaryngol Clin N Am. 2008;41:179-193.
Downloaded from aor.sagepub.com at MCMASTER UNIV LIBRARY on March 24, 2015
research-article2014
AORXXX10.1177/0003489414546605Annals of Otology, Rhinology & LaryngologyLevi et al
Article
Magnetic Resonance Imaging Findings in Children With Tinnitus
Annals of Otology, Rhinology & Laryngology 2015, Vol. 124(2) 126–131 © The Author(s) 2014 Reprints and permissions: sagepub.com/journalsPermissions.nav DOI: 10.1177/0003489414546605 aor.sagepub.com
Eric Levi, MBBS, MPHTM1, Elhamy K. Bekhit, MBBS, FRANZCR2,3, and Robert G. Berkowitz, MD, FRACS1,4,5
Abstract Objective: Tinnitus in adults is generally investigated by contrast-enhanced magnetic resonance imaging (MRI) to rule out the diagnosis of acoustic neuroma. Acoustic neuroma is rare in children and, therefore, the role of MRI in children with tinnitus is unclear. This study was undertaken to determine the value of MRI in the investigation of tinnitus in children. Methods: Retrospective study of children younger than 18 years who underwent MRI for the investigation of tinnitus over a 10-year period. Results: Sixty-five patients were identified, but there were only 34 who had also undergone audiologic assessment. Among the 25 patients with normal audiology, MRI abnormalities were present in 9, but these were all thought to be nonspecific. Nine patients had abnormal audiograms and the MRI was abnormal in 4 of these cases, which included 3 children who were found to have multiple sclerosis. Conclusion: Magnetic resonance imaging would appear to be mandatory in the investigation of tinnitus in children who are found to have sensorineural hearing loss, particularly to rule out the diagnosis of multiple sclerosis. Although our study does not support the routine use of MRI in children with normal audiology, the numbers in our series are too small for a conclusive recommendation. Keywords children, MRI, tinnitus
Introduction Tinnitus is a subjective experience of noise, such as buzzing, ringing, or whistling, in the absence of any external auditory stimuli. The prevalence of tinnitus increases with age, peaking at 14.3% between 60 and 69 years of age. In a large study of more than 14 000 people, the prevalence of tinnitus is found to be highest among older adults, former smokers, and adults with hypertension, hearing impairment, loud noise exposure, or generalized anxiety disorder.1 Savastano2 reported that tinnitus occurs at some stage in 34% of children, but only 6.5% were found to complain spontaneously. Of these, 76% demonstrated normal hearing. Using a more rigorous questioning to avoid bias in children, Stouffer et al3 placed the prevalence of tinnitus in normal hearing children at 6% to 13% and at 24% to 29% in those children with hearing impairment. In children, there are a few risk factors associated with tinnitus that have been explored by Coelho et al,4 including age, sex, hearing loss, history of noise exposure, motion sickness, and hyperacusis. In adults, tinnitus is generally classified as being
pulsatile or nonpulsatile, to help direct the appropriateness of investigation.5 However, in children, qualitative description of tinnitus is hard to ascertain, as children are not often able to characterize their tinnitus.3 Tyler and Smith6 have suggested the use of the terms congenital versus acquired and middle ear versus sensorineural tinnitus as useful etiological classifications. The latter differentiates in particular those children in whom tinnitus may be related to otitis 1
Department of Otolaryngology, Royal Children’s Hospital Melbourne, Victoria, Australia 2 Department of Medical Imaging, Royal Children’s Hospital Melbourne, Victoria, Australia 3 Department of Radiology, The University of Melbourne, Melbourne, Australia 4 Murdoch Children’s Research Institute, The University of Melbourne, Melbourne, Australia 5 Australian School of Advanced Medicine, Macquarie University, Sydney, Australia Corresponding Author: Eric Levi, MBBS, MPHTM, 28 Malabar Road, Blackburn, Victoria 3130, Australia. Email: [email protected]
Downloaded from aor.sagepub.com at MCMASTER UNIV LIBRARY on March 24, 2015
127
Levi et al media from those in whom tinnitus may be related to significant inner ear or brainstem pathology. In adults, tinnitus, especially unilateral nonpulsatile tinnitus, may be due to acoustic neuroma. Acoustic neuromas are benign tumors of the eighth cranial nerve occurring at the cerebellopontine angle, with only about 10 acoustic neuromas newly diagnosed each year per million persons,7 corresponding to between 2000 and 3000 new cases each year in the United States. Acoustic neuromas occur largely in adults and are very rare in children. Only 39 cases in children had been reported in the literature as of 2001.8 The standard investigation for tinnitus includes an audiogram and magnetic resonance imaging (MRI).5 Gadoliniumenhanced MRI still represents the gold standard investigation in adults reporting nonpulsatile tinnitus.5 Whereas MRI has been shown to be of benefit in the investigation of tinnitus in adults to identify acoustic neuroma,9,10 this condition is rare in the pediatric age group, and therefore the value of performing an MRI in children with tinnitus is unclear. Magnetic resonance imaging in children may require the use of sedation or anesthesia, and therefore it is costly and may involve some risk. Magnetic resonance imaging findings in the investigation of tinnitus in adults are established but have not yet been confirmed in children. With the aim of determining the value of MRI in the investigation of children presenting with tinnitus, we carried out a 10-year institutional retrospective review of MRI findings in the investigation of tinnitus in children.
Methods After approval from our institutional ethics committee (HREC 31258A), a keyword search of the database of the Department of Medical Imaging at the Royal Children’s Hospital Melbourne, Australia, was performed. All MRI studies conducted in the 10-year period prior to January 1, 2012, in children 18 years of age or younger where the term tinnitus was recorded on the request or report were extracted. These MRI findings were then reviewed by an attending radiologist (E.K.B.) to ensure uniformity of reporting. Clinical data were gathered from retrospective chart review of patient histories. Patient characteristics including age, sex, tinnitus laterality, quality, duration, relevant otological past history, examination findings, and audiology reports were recorded. Patients without formal audiograms were excluded from the study. In this study, normal hearing was regarded as an audiometrically confirmed pure-tone average hearing sensitivity at 500 Hz, 1000 Hz, and 2000 Hz of 25 dB or better, and a type A tympanogram. Any deviation from this definition was regarded as an abnormal audiogram or abnormal hearing. We then evaluated the MRI findings to determine effect on the diagnosis and management of these children. An abnormal MRI report was regarded as one where any abnormal
signal change arising from a pathological lesion or anything other than a normal anatomy or an acceptable normal variant of that anatomy was present.
Results Sixty-five MRI studies were identified with 29 males and 36 females, average age 11.5 years (range, 3-18 years). The MRI was requested by Otolaryngology in 40 cases, Pediatric Neurology in 20, and Neurosurgery in 5. However, audiometry was performed in only 34 patients, and the remainder were excluded from this study. The excluded group contained 1 patient with an incidental Chiari 1 malformation, 1 patient with an enlarged right vestibular aqueduct, and 1 case of posttraumatic right extradural hematoma. The distribution of patients is detailed in Figure 1. Of the 34 patients for analysis, there were 25 with normal hearing on audiometry and 9 with abnormal audiological findings. In the group with normal audiometry, there were 16 normal MRI studies and 9 abnormal MRI findings, detailed in Table 1. None of the abnormal MRI findings were related to pathology involving the cerebellopontine angle or the vestibulocochlear system. The most common finding was sinus disease, present in 4 cases. In the group of 9 patients with tinnitus and abnormal audiometry, the MRI was normal in 5 children and abnormal in 4. The details of these 9 children are listed in Table 2. This group included 3 children who were found on MRI to have multiple sclerosis (MS) (Figure 2 for representative image). All 3 of these children presented with unilateral sensorineural hearing loss, and this was the only symptom at presentation in 1 child. In the other abnormal MRI, postinfective demyelination was identified. The other 5 children with abnormal audiometry but normal MRI findings included 2 children who had clinical and audiological signs of middle ear disease.
Discussion Our study has looked at the role of MRI in the investigation of tinnitus in the pediatric population. To our knowledge, there has not been any published data in the literature specifically exploring MRI findings in children with tinnitus. In a group of 34 children, we identified 3 children with the diagnosis of MS. All 3 children in this study had asymmetrical sensorineural hearing loss and tinnitus. We have previously reported on 1 of these children, who presented with acute tinnitus and sudden hearing loss as the initial complaint leading to the diagnosis of MS.11 Multiple sclerosis is an inflammatory immune-mediated disease affecting myelinated axons in the central nervous system. The hallmark of MS is symptomatic episodes that occur months or years apart and affect different central nervous system anatomic sites. Therefore, MS is a diagnosis made on a combination of clinical and radiological grounds. The
Downloaded from aor.sagepub.com at MCMASTER UNIV LIBRARY on March 24, 2015
128
Annals of Otology, Rhinology & Laryngology 124(2)
Figure 1. Representation of result. Table 1. Patients With Tinnitus, Normal Audiometry, and Abnormal Findings on MRI. Patient 1 2 3 4
3F 17M 9F 11M
Tinnitus Laterality
Tinnitus Quality
Other Clinical Findings
Bilateral Bilateral Bilateral Bilateral
High-pitched Nonspecific High-pitched Nonspecific
Disequilibrium Disequilibrium Nil Nil
5 9F
Bilateral
Nonspecific
6 16M
Unilateral (left)
High-pitched
Disequilibrium, nausea, migraine, chronic fatigue syndrome Ehlers-Danlos syndrome, disequilibrium, headaches
7 16F
Bilateral
High-pitched
8 14M
Bilateral
High-pitched
Difficulty understanding speech, depression, chronic fatigue syndrome NF1
9 15F
Bilateral
High-pitched
Nil
MRI Findings Right sphenoid and maxillary disease Paranasal sinus disease Maxillary and ethmoid sinus disease Sinus disease and borderline cerebellar tonsillar ectopia Blake’s pouch cyst Nodular heterotopia subependymal position paralleling bilateral lateral ventricles Hyperintensities on corpus callosum; multiple sclerosis ruled out on repeat MRI Left corona radiate small focus of signal intensity Bilateral temporal lobes mild increased signal intensity
Abbreviations: MRI, magnetic resonance imaging; NF1, neurofibromatosis type 1.
Downloaded from aor.sagepub.com at MCMASTER UNIV LIBRARY on March 24, 2015
129
Levi et al Table 2. Patients With Tinnitus and Abnormal Audiometry. Patient
Tinnitus Laterality
Tinnitus Quality
Other Clinical Findings Sudden left hearing loss CN5&7 palsies, ataxia, dysarthria, nystagmus
1 13M
Unilateral (left)
High-pitched
2 15M
Unilateral (left)
Nonspecific
3 13F
Unilateral (right)
Buzzing
4 15F
Unilateral (left)
High-pitched
5 11M
Bilateral
Nonspecific
6 13F
Unilateral (right)
Nonspecific
7 10F
Bilateral
High-pitched
8 18F
Unilateral (right)
Humming
9 15M
Bilateral
Nonspecific
Audiology
Details
Unilateral SNHL Left moderate to severe SNHL Unilateral SNHL Left mild to moderate SNHL
MRI Findings Normal
Multifocal white matter hyperintensities consistent with demyelination (MS) (Figure 2) Unilateral right hearing Unilateral SNHL Right mild to Multifocal loss over 6 weeks moderate white matter SNHL hyperintensities consistent with demyelination (MS) Left hearing loss, ataxia Unilateral SNHL Left mild to Multifocal moderate white matter SNHL hyperintensities consistent with demyelination (MS) Developmental delay; Bilateral SNHL Asymmetrical Asymmetrical right middle ear SNHL: right hyperintensities abnormality. MS mild to in frontal, ruled out on repeat moderate parietal, and MRI SNHL with posterior additional mild temporal lobes conductive loss; consistent with left mild SNHL postinfective demyelination change Hearing loss Bilateral SNHL Mild symmetrical Normal SNHL Hearing loss Bilateral SNHL Mild symmetrical Normal SNHL Previous MEVTs Bilateral CHL Mild symmetrical Normal CHL Previous MEVTs Bilateral CHL Mild symmetrical Normal CHL
Abbreviations: CHL, conductive hearing loss; CN, cranial nerve; MEVT, middle ear ventilation tube; MRI, magnetic resonance imaging; MS, multiple sclerosis; SNHL, sensorineural hearing loss.
American Academy of Neurology Practice Guidelines12 advocates for the use of MRI in making an early diagnosis of MS that would guide early intervention. Tinnitus is an uncommon complaint in patients with MS. Less than 2% of patients with MS report tinnitus and hearing loss.13 In a series by Fischer et al13 on patients with MS, only 12 out of 705 patients experienced acute hearing loss. In our series, MS was diagnosed in 3 out of 9 children with tinnitus and hearing loss, and it is clear that performing an MRI is essential in this group of children. In children with tinnitus but otherwise normal hearing, we found MRI abnormalities in 9 of 25 children, but none of these were related to any otologic or significant central
nervous system pathology. The most common finding was that of incidental paranasal sinus disease. This finding and other abnormalities listed in Table 1 are of questionable relevance to the presenting complaint of tinnitus in our study population. von Kalle et al14 studied paranasal sinus magnetic resonance images and found that mucosal swelling in paranasal sinuses and in mastoid cells is a frequent incidental finding in children and is not necessarily a sign of disease. The significance of the associated sinus disease and minor central nervous system abnormalities detected is therefore unclear. Coupled with the relatively small patient numbers in the study, we cannot determine whether or not MRI should be recommended as part of the routine
Downloaded from aor.sagepub.com at MCMASTER UNIV LIBRARY on March 24, 2015
130
Annals of Otology, Rhinology & Laryngology 124(2)
Figure 2. Coronal Fluid Attenuation Inversion Recovery (FLAIR) image demonstrating multifocal white matter hyperintensities (arrows), consistent with demyelination and multiple sclerosis (MS).
evaluation of a child presenting with tinnitus who is found to have normal audiology. Although the MRI study itself is relatively safe, performing an MRI in some children may require sedation and therefore may be associated with some risk and additional cost in comparison with MRI in adults. Tinnitus is a common complaint in adults. Sixteen percent to 19% of people older than 17 years have reported experiencing spontaneous tinnitus at some stage. Of these individuals, 8% experienced tinnitus as a moderate to severe disturbance. Tinnitus in children, however, may be a clinically under-recognized problem. Its prevalence has been reported in various studies as being in the order of 6% to 34%.2,15,16 Some authors believe that these percentages may potentially be either an underestimation or an overestimation due to difficulties with language in children or due to the children’s desire to satisfy the questioner.17 Tinnitus can occur in children with normal hearing and those with hearing deficits. Mills et al16,18 suggested that up to 29% of normally hearing children reported having had tinnitus,16 compared to 43.9% of children with conductive hearing loss reporting tinnitus and 29.5% of those with sensorineural hearing loss.18 In our series of children with tinnitus, 73.5% had normal hearing and 26.5% had abnormal hearing confirmed on audiometry. Tinnitus in children may potentially be due to past or present middle ear disease. This was carefully looked at by
Savastano,2 who found that in children who complained of tinnitus, 14% had clinical features of middle ear inflammation and 14.6% had abnormal negative pressure tympanometry. However, these rates are not statistically significant compared to the group who did not complain of tinnitus. A history of middle ear disease was reported in 31.6%, but it was not thought to be a significant factor in the genesis of tinnitus, as there was no statistical difference in the incidence of tinnitus between the group of children who had middle ear disease and those without middle ear disease.2 In our study population, there were 2 children who had clinical and audiological features of middle ear disease with mild conductive hearing loss and type B tympanograms. They both had normal MRI reports. This is consistent with Tyler and Smith’s6 proposed etiology of middle ear tinnitus. It would appear that performing an MRI for tinnitus in the context of a child with middle ear disease is less likely to identify an associated abnormality, compared with a child who has sensorineural hearing loss. Our study underscores the importance of considering an MRI to investigate tinnitus in a child with sensorineural hearing loss because of the potential to make an early diagnosis of MS. In the adult population, imaging investigation, usually with a gadolinium-enhanced MRI, is directed at confirming pathology in the vestibulocochlear nerve or cerebellopontine angle.5 Weissman and Hirsch19 found that most adult patients with subjective tinnitus and normal examination have no imaging abnormalities. However, Sonmez and colleagues20 reviewed 74 adult patients and described radiographic abnormalities in 61.9% of patients with subjective tinnitus and in all patients with pulsatile tinnitus. On closer inspection, the radiographic abnormalities, namely, high jugular bulb, atherosclerosis, dehiscent jugular bulb, aneurysm of internal carotid artery, dural arteriovenous fistula, aberrant internal carotid artery, jugular diverticulum, and glomus tumor, may not necessarily be the true cause of tinnitus, as they can be found incidentally in patients who are asymptomatic.21 It is unclear what percentage of these imaging abnormalities was the true cause of tinnitus in these patients. Other studies have investigated various temporal bone anomalies that have shown no significant clinical correlation with tinnitus.21 A large study would be required to have enough statistical power to fully determine the relationship between common or uncommon normal variants and their associated clinical symptoms.22 The prevalence of cerebellopontine tumor in children is rare.8 We have not found any children in our retrospective 10-year series with an intracranial or cerebellopontine tumor presenting with tinnitus. Hence, the investigation of a child with tinnitus by means of a gadolinium-enhanced MRI is of limited benefit. Based on our experience, the role of MRI in the evaluation of tinnitus appears to be primarily to identify or rule out the diagnosis of MS.
Downloaded from aor.sagepub.com at MCMASTER UNIV LIBRARY on March 24, 2015
131
Levi et al
Conclusion Magnetic resonance imaging would appear to be mandatory in the investigation of tinnitus in children who are found to have sensorineural hearing loss, particularly to rule out the diagnosis of MS. Although our study does not support the routine use of MRI in children with normal audiology, the numbers in our series are too small for a conclusive recommendation. Authors’ Note E.L., E.K.B., and R.G.B. were involved in all stages of the preparation of this article. E.K.B. is a staff radiologist who reviewed all MRI images and reports examined in this article. R.G.B. is a professor of pediatric otolaryngology and senior author of this article.
Declaration of Conflicting Interests The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding The author(s) received no financial support for the research, authorship, and/or publication of this article.
References 1. Shargorodsky J, Curhan GC, Farwell WR. Prevalence and characteristics of tinnitus among US adults. Am J Med. 2010;123(8):711-718. 2. Savastano M. Characteristics of tinnitus in childhood. Eur J Pediatr. 2007;166:797-801. 3. Stouffer JL, Tyler RS, Booth JC, Buckrell B. Tinnitus in normal-hearing and hearing-impaired children. In: Aran JM, Dauman R, eds. Tinnitus 91—Proceedings of the Fourth International Tinnitus Seminar. Amsterdam, The Netherlands: Kugler Publications; 1992:255-258. 4. Coelho C, Sanchez TG, Tyler RS. Tinnitus in children and associated risk factors. Prog Brain Res. 2007;166:179-194. 5. Sajisevi M, Weissman JL, Kaylie DM. What is the role of imaging in tinnitus? Laryngoscope. 2014;124:583-584. 6. Tyler RS, Smith RJ. Management of tinnitus in children. In: Newton VE, ed. Paediatric Audiological Medicine. Philadelphia, PA: Whurr Publishers; 2002:397-404. 7. Evans DG, Moran A, King A, Saeed S, Gurusinghe N, Ramsden R. Incidence of vestibular schwannoma and neurofibromatosis
2 in the North West of England over a 10-year period: higher incidence than previously thought. Otol Neurotol. 2005;26(1): 93-97. 8. Pothula VB, Lesser T, Mallucci C, May P, Foy P. Vestibular schwannomas in children. Otol Neurotol. 2001;22(6):903-907. 9. Sidman J, Carrasco V, Whaley R. Gadolinium the new gold standard for diagnosing cerebellopontine angle tumours. Arch Otolaryngol Head Neck Surg. 1989;115:1244-1277. 10. Murphy MR, Selesnick SH. Cost-effective diagnosis of acoustic neuromas: a philosophical, macroeconomic, and technological decision. Otolaryngol Head Neck Surg. 2002;127:253-259. 11. Rodriguez-Casero MV, Mandelstam S, Kornberg AJ, Berkowitz RG. Acute tinnitus and hearing loss as the initial symptom of multiple sclerosis in a child. Int J Ped Oto. 2005;69:123-126. 12. American Academy of Neurology. Practice guidelines: multiple sclerosis. https://www.aan.com/Guidelines/Home/ ByTopic?topicId=18. Accessed August 15, 2014. 13. Fischer C, Mauguiere F, Ibanez V, Confavreux C, Chazot G. The acute deafness of definite multiple sclerosis. Clin Neurophysiol. 1985;61(1):7-15. 14. von Kalle T, Fabig-Moritz C, Heumann H, Winkler P. Incidental findings in paranasal sinuses and mastoid cells: a cross-sectional magnetic resonance imaging (MRI) study in a pediatric radiology department. Rofo: Fortschritte auf dem Gebiete der Rontgenstrahlen und der Nuklearmedizin. 2012;184(7):629-634. 15. Nodar RH. Tinnitus aurium in school age children. J Aud Res. 1972;12:133-135. 16. Mills RP, Albert DM, Brain CE. Tinnitus in childhood. Clin Otolaryngol Allied Sci. 1986;11:431-434. 17. Baguley DM, McFerran DJ. Tinnitus in childhood. Int J Pediatr Otorhinolaryngol. 1999;49:99-105. 18. Mills RP, Cherry JR. Subjective tinnitus in children with otological disorders. Int J Paediatr Otorhinolaryngol. 1984;7:21-27. 19. Weissman JL, Hirsch BE. Imaging of tinnitus: a review. Radiology. 2000;216(2):342-349. 20. Sonmez G, Basekim CC, Ozturk E, Gungor A, Kizilkaya E. Imaging of pulsatile tinnitus: a review of 74 patients. Clin Imaging. 2007;31(2):102-108. 21. Koesling S, Kunkel P, Schul T. Vascular anomalies, sutures and small canals of the temporal bone on axial CT. Eur J Radiol. 2005;54(3):335-343. 22. Kang M, Escott E. Imaging of tinnitus. Otolaryngol Clin N Am. 2008;41:179-193.
Downloaded from aor.sagepub.com at MCMASTER UNIV LIBRARY on March 24, 2015
