Tympanic membrane changesfollowing paediatric cochlear implantation

YVONNE CHAN, PAOLO CAMPISI, ADRIAN L JAMES, BLAKE C PAPSIN,Cochlear Implant Program, Hospital for Sick Children, Department ofOtolaryngology – Head & Neck Surgery, University of Toronto, Toronto,Canada

ABSTRACT Objective To investigate the effects of cochlear implantation on thetympanic membrane in children.Design Cross-sectional study.Methods Eighty-four cochlear implant recipients were recruited at a vaccination clinic.The implanted ear was examined and the appearance of the external auditory canal andtympanic membrane documented. Demographic and clinical data such as age at implan-tation, duration of implantation, preoperative tympanic membrane appearance andpreoperative middle-ear function were obtained retrospectively from hospital charts.Results Thirty-eight patients (46.9%) were found to have abnormal tympanicmembrane findings. The most common abnormality was scarring of the tympanicmembrane in the vicinity of the electrode array. No statistically significant association wasfound between the tympanic membrane findings and demographic and clinical variables.Conclusions Cochlear implantation may cause tympanic membrane scarring. The signif-icance of the scarring is unclear. Dehiscence of the posterior wall of the external auditorycanal should be avoided to prevent retraction pocket formation of the tympanicmembrane.

Keywords: cochlear implantation, tympanic membrane, cholesteatoma, tympanosclerosis, paediatric

Introduction

Paediatric cochlear implantation is a safe and effective method of auditory rehabil-itation. The benefits of paediatric cochlear implantation are well established(Papsin et al., 2000). These include improvement in the quality of life (Pulsifer etal., 2003), academic performance (Daya et al., 2000) and receptive as well as

10 Cochlear Implants International, 6(1), 10–15, 2005 © Whurr Publishers Ltd

CII 6.1_CRC 3/8/05 11:33 AM Page 10

expressive language skills (El-Hakim et al., 2001). In addition, recent advances inimplant design and surgical technique have rendered paediatric cochlear implan-tation a safe procedure with a low complication rate (Gysin et al., 2000).

The impact of cochlear implantation on the tympanic membrane is unknown.The purpose of this cross-sectional study was to characterize the appearance of thetympanic membrane in a series of 84 children who have received a cochlearimplant.

Materials and methods

Eight-four patients were recruited from a pneumococcal vaccination clinic forchildren with cochlear implants. During the patient interaction, the tympanicmembrane of the implanted ear was examined by one of three paediatric otolaryn-gologists. When required, cerumen was carefully removed from the externalauditory canal (EAC) by microdebridement. The appearance of the tympanicmembrane and EAC was documented. Digital photographs of abnormal tympanicmembranes were acquired with aural endoscopes for subsequent interpretation andclassification by all three paediatric otolaryngologists.

Demographic and clinical data were collected from patient charts. Thevariables of interest included patient age, gender, age at implantation, duration ofimplantation, preoperative tympanic membrane appearance and preoperativetympanometry. These variables were correlated with tympanic membrane andexternal auditory canal findings. Statistical analyses were performed using Chi-square and Student-t tests.

Results

Of the 84 patients recruited from the pneumococcal vaccination clinic, three wereexcluded because they were uncooperative and refused an otoscopic examination.As such, 81 patients were included in the final analysis. The study group had amean age of 9.1 years (range 1.4–18.5 years). Pertinent demographic and clinicaldata are summarized in Table 1. Retrospective review of hospital charts revealedthat prior to cochlear implantation, nine patients had tympanostomy tubes, fourhad at least one documented episode of middle-ear effusion and a further threepatients had abnormal tympanometric findings (but normal otoscopy).

The time interval between cochlear implantation and examination at thevaccination clinic ranged from 2 months to 8 years. A tympanic membrane abnor-mality was identified during otoscopy in 38 patients (46.9%). The vast majority ofthe abnormalities (92.1%) involved tympanic membrane scarring of unknownsignificance. Two patients had tympanic membrane retraction and one patient hadacute otitis media. Retraction pockets and cholesteatomas were not observed.These findings are summarized in Table 2. In addition, a dehiscent annulus wasidentified in two patients with a normal tympanic membrane and in two patientswith scarring.

Tympanic membrane following cochlear implantation 11

CII 6.1_CRC 3/8/05 11:33 AM Page 11

It was further noted that the tympanic membrane scarring developed inconsistent patterns in the postero-superior quadrant of the pars tensa. The patternof scarring was either linear, wedge-shaped or diffuse (Figure 1). The area ofscarring was positioned over the expected location of the electrode array inpatients with linear or wedge-shaped scarring. The base of the wedge-shaped scarswas typically located at the margin of the tympanic membrane. Diffusetympanosclerosis, encompassing more than 20% of the tympanic membranesurface area, was observed in five patients.

Abnormalities of the posterior external canal wall were observed in fourpatients. This resulted in retraction of the canal wall skin and annulus. In one ofthe cases, the retracted skin was draped over the implant electrode array (Figure2).

No statistical correlation was identified between the postoperative appearanceof the tympanic membrane and demographic and clinical variables.

12 Yvonne Chan et al.

Table 1: Patient demographic information

n = 81 Range

GenderMale 44 (54.3 %)Female 37 (45.7 %)

Age (years) 9.1 1.4–18.5 Age at implantation (years) 5.6 0.8–17.6Duration of implant (years) 3.3 0.2–9.4Side of implant

Left 14 (16.7%)Right 67 (83.3%)

Preoperative middle-ear statusTympanostomy tubes 9 (11.1%)Middle-ear effusion 4 (4.9%)Abnormal tympanometry (normal otoscopy) 3 (3.7%)

Table 2: Summary of post-operative tympanic membrane (TM) findings

n = 81 (%)

Normal TM 43 (53.1%)Abnormal TM 38 (46.9%)

Linear scar 18Wedge scar 12Diffuse scar 5Retraction 2Acute otitis 1

CII 6.1_CRC 3/8/05 11:33 AM Page 12

Discussion

The patients presented in this cross-sectional study received their cochlearimplants by means of a consistent surgical technique performed by a singlesurgeon. Patients were implanted with either a Nucleus 22 or 24 cochlear implant(Cochlear Corporation, Australia).

The effect of cochlear implant surgery on tympanic membrane physiology hasnot been previously studied. The electrode array travels a short distance across themesotympanum from the facial recess to the cochleostomy. This position does notobstruct the normal routes of communication between the meso- and epitym-panum. Thus, the array should not interfere with normal middle-ear physiology.Rather, opening of the mastoid and facial recess is expected to improve ventilationof the tympano-mastoid system, especially in poorly pneumatized mastoid cavities.

The postoperative appearance of the tympanic membrane was unremarkable in53.1% of patients. Of the 38 patients found to have an abnormal tympanicmembrane, the most common finding was a small area of tympanosclerosis in thevicinity of the electrode array (in an otherwise normal tympanic membrane). Thesignificance of the tympanosclerosis is not readily apparent. However, the

Tympanic membrane following cochlear implantation 13

Figure 1: Postoperative appearance of the tympanic membrane demonstrating (A) a linear, (B) awedge-shaped and (C) a diffuse tympanic membrane scar.

Figure 2: Retraction of the posterior canal wall skin overlying the cochlear implant electrode array.

CII 6.1_CRC 3/8/05 11:33 AM Page 13

positioning of the scarring in the vicinity of the electrode array suggests that thecochlear implant may be responsible for the observed tympanic membrane injury.

The linear pattern of scarring may be the result of adherence of the electrodearray to the overlying tympanic membrane. The wedge-shaped scars, whichparallel the radial vascular supply of the tympanic membrane, may be the result ofvascular disruption to a segment of the tympanic membrane at the time of surgery.Although less likely, the electrode array may have caused the scarring directlythrough electromagnetic radiation or heat. The potential for this form of injuryhas not been previously considered.

A breach in the posterior wall of the external auditory canal during the drillingof the facial recess resulted in retraction of meatal skin over the facial recess in fourpatients. Retraction pockets in this area represent a serious potential for in-growthof epithelium that may result in cholesteatoma formation. Althoughcholesteatoma was not identified in any of the patients in this study, there havebeen two cases of cholesteatoma in patients in our cochlear implant programmeexperience. In both instances, the epicentre of the cholesteatoma was the facialrecess. The occurrence of cholesteatoma following cochlear implantation has beendescribed (Kempf et al., 1999; Kumar et al., 1999). However, there is no evidencein the literature to suggest that the incidence of non-iatrogenic cholesteatoma isincreased by the presence of a cochlear implant.

Preservation of the posterior wall of the external auditory canal is paramountwhen performing a mastoidectomy and facial recess approach for cochlear implan-tation. A bony dehiscence may be avoided with careful and meticulous drilling ofthe facial recess with copious irrigation to prevent thermal injury and subsequentnecrosis of the posterior canal wall. The authors further suggest the preservation ofa ‘buttress’ of bone between the antrum and facial recess to offer protection againstretraction of the postero-superior tympanic membrane. All canal wall defectsshould be immediately repaired with bone paté and/or fragments of calvarial boneobtained from the receiver-stimulator well site. A breach of the meatal skin mayfurther increase the risk of cholesteatoma formation by epithelial implantation.Repair of the skin defect with a fascial graft should be considered in theseinstances.

Summary

The pneumococcal vaccination clinic provided a unique opportunity to examine across-section of children with cochlear implants in one setting. Consistentpatterns of tympanic membrane scarring were identified in 46.9% of patients inthe vicinity of the cochlear implant electrode array. The significance of thescarring is unclear. Furthermore, the presence of abnormal tympanic membranefindings could not be statistically correlated with demographic or clinicalvariables. However, the authors suggest that compromise of the posterior wall ofthe external auditory canal may predispose patients to retraction pockets that maylead to cholesteatoma formation.

14 Yvonne Chan et al.

CII 6.1_CRC 3/8/05 11:33 AM Page 14

References

Daya H, Ashley A, Gysin C, Papsin BC (2000) Changes in educational placement and speechperception ability after cochlear implantation in children. Journal of Otolaryngology 29(4):224–228.

El-Hakim H, Levasseur J, Papsin BC, Panesar J, Mount RJ, Stevens D, Harrison RV (2001)Assessment of vocabulary development in children after cochlear implantation. Archives ofOtolaryngology – Head and Neck Surgery 127(9): 1053–1059.

Gysin C, Papsin BC, Daya H, Nedzelski J (2000) Surgical outcome after paediatric cochlear implan-tation: diminution of complications with the evolution of new surgical techniques. Journal ofOtolaryngology 29(5): 285–289.

Kempf HG, Johann K, Lenarz T (1999) Complications in pediatric cochlear implant surgery.European Archives of Otorhinolaryngology 256(3): 128–132.

Kumar A, Mugge R, Lipner M4 (1999) Surgical complications of cochlear implantation: a report ofthree cases and their clinical features. Ear Nose Throat Journal 78(12): 913–919.

Papsin BC, Gysin C, Picton N, Nedzelski J, Harrison R (2000) Speech perception outcome measuresin prelingually deaf children up to four years after cochlear implantation. Annals of OtologyRhinology Laryngology Supplement 185: 38–42.

Pulsifer MB, Salorio CF, Niparko JK (2003) Developmental, audiological, and speech perceptionfunctioning in children after cochlear implant surgery. Archives of Pediatric and AdolescentMedicine 157(6): 552–558.

Address correspondence to: Paolo Campisi MSc, MD, FRCSC, FAA, Department of Otolaryngology– Head & Neck Surgery, Hospital for Sick Children, 6th Floor, Elm Wing, 555 University Avenue,Toronto, Ontario, M5G 1X8, Canada. Tel: +1 416 813 2190. Fax: +1 416 813 5036. Email:paolo.campisi@sickkids.ca

Accepted 13 December 2004

Tympanic membrane following cochlear implantation 15

CII 6.1_CRC 3/8/05 11:33 AM Page 15