Subtotal Petrosectomy and Cochlear Implantation: A Systematic Review and Meta-analysis

Key Points

Question  What are the indications for and complications of subtotal petrosectomy (STP) for cochlear implantation?

Findings  In this systematic review and meta-analysis with 27 unique studies and 397 unique STP procedures performed on 377 patients for cochlear implantation, the global complication rate was 12.4% and the cholesteatoma recidivism rate was 9.3%. The most common indications for STP included chronic otitis media, preexisting mastoid cavity, and cholesteatoma, and complication rates were not significantly different across single-stage vs multistage procedures, or in pediatric vs adult populations.

Meaning  This study’s findings suggest that STP is a safe and effective method in preparing the ear for cochlear implantation.

Importance  Subtotal petrosectomy (STP) has been more frequently performed to prepare ears with unfavorable conditions for cochlear implantation.

Objectives  To provide an overview of indications for and complications of STP and cochlear implantation and to compare outcomes between single vs multistage procedures and between pediatric vs adult populations.

Data Sources  A search of PubMed, Scopus, Ovid, and the Cochrane Library was performed from the databases’ inception to January 23, 2020, for studies evaluating STP for cochlear implantation.

Study Selection  Studies with a minimum follow-up of 3 months and no missing data regarding postoperative outcomes were included. Of the initial 570 studies identified, 27 (4.7%) met selection criteria.

Data Extraction and Synthesis  Two reviewers independently assessed study eligibility according to Preferred Reporting Items for Systematic Reviews and Meta-analyses guidelines; discrepancies were resolved by a third reviewer. Extracted data included patient demographics, indications for STP, rates of complications, and cholesteatoma recidivism when applicable. Data were pooled using a random- or a fixed-effects model when appropriate.

Main Outcomes and Measures  The primary study outcome was rate of global complications stratified by patient- and surgery-level characteristics.

Results  Twenty-seven unique studies with 377 unique patients (54.2% male; mean age, 50.6 [range, 1-99] years) undergoing 397 STP procedures and cochlear implantation were included. Of these procedures, 299 of 394 cases with the information reported (75.9%) were single procedures and 95 (24.1%) were multistage procedures. Of the total 397 STP procedures, most common indications included chronic otitis media (220 cases [55.4%]), previous open mastoid cavity (141 [35.5%]), cholesteatoma (74 [18.6%]), and cochlear ossification (29 [7.3%]). The overall complication rate was 12.4% (95% CI, 9.4%-15.9%); overall cholesteatoma recidivism rate was 9.3% (95% CI, 4.3%-17.1%). Complication rates did not significantly differ based on stage or age of patients. Cases with cholesteatoma more often underwent multistage vs single-stage procedures (23 of 54 [42.6%] vs 35 of 174 [20.1%]).

Conclusions and Relevance  Across all age groups, STP has been shown to be an effective surgical operation in preparing an ear with unfavorable conditions for cochlear implantation. The potential indications for which cochlear implantation can be performed have expanded with the use of STP. Presence of cholesteatoma might indicate that a multistage procedure should be performed. Lastly, with complication rates comparable to those in adult patients, STP can be considered in children requiring cochlear implantation to minimize ear-related issues and allow benefit from cochlear implantation.

The role of subtotal petrosectomy (STP) in preparing a diseased ear for cochlear implantation has expanded. Because STP involves complete exenteration of all air cells of the temporal bone, obliteration of the mastoid cavity, and closure of the external auditory canal (EAC) and eustachian tube, the resulting sterile field is optimal for cochlear implantation.¹ Subtotal petrosectomy is indicated for extensive infectious middle ear disease, temporal bone or skull base tumors, cerebrospinal fluid leaks, and potential infectious sequelae after temporal bone fractures.^2-5 Subtotal petrosectomy results in maximal conductive hearing loss in addition to any cochlear or sensorineural loss. Additional surgery, such as placement of an osseointegrated implant, can generally overcome the conductive loss if the cochlear reserve is reasonable. For patients with poor underlying sensorineural hearing, an osseointegrated device can provide stimulation to the contralateral ear, or a cochlear implant can be used to stimulate the ear in question.

Multiple ear disorders and conditions, such as noise exposure, ototoxic medications, congenital malformations, and infectious ear disease, lead to severe sensorineural hearing loss and consideration of cochlear implantation.^3,6,7 Many of these problems pose concerns for cochlear implantation, because placement of a foreign object into an unfavorable environment increases the potential for infection. The late 1990s saw the introduction of STP for cochlear implantation as a viable option for ears with chronic disease or unfavorable conditions.^8-11 Since then, STP has grown in popularity as an effective surgical option to seal off the middle ear and mastoid cavity from the external environment.

Multiple retrospective studies have described STP with cochlear implantation, either performed simultaneously in a single operation or in separate stages. The consideration and timing of a second procedure depend on the condition of the ear, because active inflammation and/or presence of cholesteatoma weigh into the decisions involving a staged operation. Varying graft materials have proven effective for mastoid obliteration, including abdominal fat, musculoperiosteal flaps, and bone pate.³ To date, a review of the literature in evaluating STP for cochlear implantation has not been conducted. Therefore, we have performed a systematic review to provide an overview of the indications for and complications of STP with cochlear implantation. Secondarily, we compared complication rates between single-stage vs multistage procedures, with differing graft materials, and in pediatric vs adult populations.

This systematic review was performed according to the Preferred Reporting Items for Systematic Reviews and Meta-analyses (PRISMA) guidelines. The PubMed, Scopus, Cochrane Library, and Ovid databases were queried from inception to January 23, 2020, for studies assessing STP performed for cochlear implantation. The search strategy included a combination of the following search terms: subtotal petrosectomy, mastoid obliteration, cochlear implant, and cochlear implantation. In addition, a manual review of included references was performed from which 1 additional study was included.

Inclusion criteria consisted of (1) STP or mastoid obliteration with overclosure of the EAC and eustachian tube, (2) cochlear implantation, (3) data regarding indications for STP and complications, and (4) minimum follow-up of 3 months. Given the relative scarcity of STP performed with cochlear implantation, case reports and series were included. Exclusion criteria consisted of (1) letters or reviews, (2) overlapping data from another study, (3) non-English language, and (4) nonhuman subjects. Two authors (F.Y. and P.D.R.) independently assessed study eligibility; any disputes were resolved by a third author (S.A.N.).

Extracted data included country of publication, study design, and patient and surgery characteristics. Specific patient characteristics included sex, age, follow-up time, indications for STP, and rates of cholesteatoma recidivism if applicable. Specific procedure-based characteristics included single-stage vs multistage procedure, obliteration graft material, and rates of complications, reoperations, and device dysfunction or failure.

The level of evidence of each study was evaluated using the Oxford Center for Evidence-Based Medicine criteria.¹² Next, risk of bias was assessed according to the Cochrane Handbook for Systematic Reviews of Interventions, version 5.1.0.¹³ Specifically, the ROBINS-I (Risk of Bias in Nonrandomised Studies) tool was used because this systematic review evaluated nonrandomized studies.¹⁴ Two authors (F.Y. and P.D.R.) performed a pilot assessment on 3 studies to check for consistency of assessment. Both then performed independent risk assessments on the remaining studies. All disagreements were resolved by a third author (S.A.N.). Risk of bias items included bias due to confounding, in selection of participants into the study, in classification of interventions, due to deviations from intended interventions, due to missing data, in measurement of outcomes, and in selection of reported results. The risk of bias for each aspect was graded as low, unclear, or high.

All statistical analyses were performed with MedCalc, version 18.10.2 (MedCalc Software). Meta-analysis of proportions was performed to pool rates of complications, device dysfunction or failure, cholesteatoma recidivism, and reoperation. The program MedCalc lists the proportions (expressed as a percentages) with their 95% CIs found in the individual studies included in the analysis. The weighted summary proportion is calculated by the Freeman-Tukey transformation.¹⁵ This pooled proportion is reported with 95% CIs for both fixed-effects and random-effects models.¹⁶ Both fixed-effects and random-effects models were used in this study. If there was high heterogeneity (I² > 50%), then a random-effects model was used; if low heterogeneity, then a fixed-effects model was allowable.

Finally, the Sterne and Egger tests were performed for further assessment of risk of publication bias.^17,18 Potential publication bias was evaluated by visual inspection of the funnel plot. In a funnel plot, treatment effect is plotted on the horizontal axis and the standard error is on the vertical axis.¹⁹ The vertical line represents the summary estimate derived using fixed-effect meta-analysis. Two diagonal lines represent (pseudo) 95% confidence limits (effect ±1.96 standard error) around the summary effect for each standard error on the vertical axis. These show the expected distribution of studies in the absence of heterogeneity or selection bias. In the absence of heterogeneity, 95% of the studies should lie within the funnel defined by these diagonal lines. Publication bias results in asymmetry of the funnel plot.

Of 570 initially identified articles, 403 underwent title and abstract screening. This process eliminated 317 articles, leaving 86 for full-text review. Twenty-seven unique studies were included for final analysis (Figure 1).^5,9-11,20-42

Table 1 provides an overview of all included studies. Each study’s level of evidence was assessed according to the 2011 Oxford Center for Evidence-Based Medicine criteria.¹² Evidence in 14 studies was level 4 and in 13 studies was level 2b. The risk of bias and heterogeneity were assessed for each included study (eFigure 1 in the Supplement). A funnel plot demonstrated all plots lying within the funnel, indicating little to no publication bias (eFigure 2 in the Supplement).

The 27 included studies^5,9-11,20-42 consisted of a total of 397 STP and cochlear implantation procedures performed in 377 unique patients. The overall mean age was 50.6 (range, 1-99) years, with 202 male (54.2%) and 171 female (45.8%) patients, and sex not reported for 4. The mean follow-up time was 36.7 (range, 3-252) months. Of the 25 studies^{5,9-11,20-30,32,34-42} reporting individual patient data, 33 of 261 (12.6%) were pediatric patients and 228 (87.4%) were adult patients. Of the 26 studies^{5,9-11,20-39,41,42} reporting single-stage vs multistage procedures, 299 of 394 cases (75.9%) were conducted in a single stage and 95 (24.1%) were conducted in a 2-stage procedure, with 92 of these (96.8%) having STP before cochlear implantation and 3 (3.2%) having STP after cochlear implantation. The time between stages ranged from 3 to 72 months. In terms of obliteration material, 308 cases (77.8%) used abdominal fat graft, 74 (18.7%) used a vascularized (musculoperiosteal or temporoparietal fascial) flap, 10 (2.5%) used both abdominal fat and a vascularized flap, and 2 (0.5%) used hydroxyapatite or bone graft. For 2 cases (0.5%), the obliteration material was unknown.

Table 2 provides the indications for STP. These included chronic suppurative or nonsuppurative otitis media in 220 cases (55.4%), a preexisting open mastoid cavity in 141 (35.5%), cholesteatoma in 74 (18.6%), other indications in 44 (11.1%), cochlear ossification in 29 (7.3%), inner ear malformations in 16 (4.0%), temporal bone fracture in 15 (3.8%), other unfavorable anatomy in 17 (4.3%), and previous STP requiring revision in 7 (1.8%).

Of the 17 cases with unfavorable anatomy, 4 included a sclerotic mastoid with an anterior sigmoid sinus; 3, temporal bone malformations; and 10, unspecified. Of the 16 cases of inner ear malformation, 4 resulted from Mondini dysplasia or incomplete partition type 2; 2, an aplastic or hypoplastic cochlea; and 10, unspecified. Of the 44 other indications for STP, 9 were recurrent acute otitis media; 2, atelectasis or retraction; 2, cerebrospinal fluid leak; 2, papillary adenomas of the temporal bone; 2, meningitis; 2, unknown; 1, granulomatosis with polyangiitis causing otitis media; 1, otosclerosis with unknown cochlear ossification; 2, tumefactive inflammatory pseudotumor of the temporal bone; 3, acute eosinophilic otitis media; 8, previous surgery not specified; 1, head trauma without fracture; 1, jugular foramen paraganglioma; 1, osteoradionecrosis of the temporal bone after cobalt irradiation; 1, previous cochlear implant electrode array extruded into an open cavity; 3, temporal bone abnormalities from CHARGE (coloboma, heart defects, atresia choanae, growth retardation, genital abnormalities, and ear abnormalities) syndrome; 2, failed simultaneous mastoidectomy and cochlear implantation procedures; and 1, device explantation from infection.

Table 3 provides an overview of complications, cholesteatoma recidivism, reoperations, and device dysfunction or failures. The pooled complication rate was 12.4% (95% CI, 9.4%-15.9%) (Figure 2); the pooled cholesteatoma recidivism rate was 9.3% (95% CI, 4.3%-17.1%); the pooled rate of reoperation was 10.2% (95% CI, 7.5%-13.4%); and the pooled rate of device dysfunction/failure was 7.3% (95% CI, 5.0%-10.2%). Of the 5 total cases of cholesteatoma recidivism, 4 resulted from recurrent disease and 1 resulted from residual disease. Three of these were removed during a second-stage operation simultaneously with cochlear implantation. Two of these were found at points past STP and cochlear implantation and required revision procedures.

Of the 48 reported complications (12.4% [95% CI, 9.4%-15.9%] pooled complication rate), the most common types of complications included 11 (22.9%) wound infections, 13 (27.1%) surgical site wound breakdowns, 4 (8.3%) abdominal hematomas, 4 (8.3%) cases of noninfectious inflammation, 5 (10.4%) facial palsies, 4 (8.3%) recurrent and 1 residual (2.1%) cholesteatomas, 2 (4.2%) cases of postoperative vertigo, 2 (4.2%) seromas, 2 (4.2%) cases of subcutaneous air accumulation, and 1 (2.1%) case of subcutaneous cerebrospinal fluid. Of the 13 surgical site wound breakdowns, 9 were related to the EAC blind sac closure, 2 were retroauricular fistulas, and 2 were unspecified wound breakdowns. One patient had both a wound infection causing explantation of the cochlear implant as well as recurrent cholesteatoma. Of the 4 cases of noninfectious inflammation, 1 included canal granulation, 1 included tumefactive inflammation, and 2 resulted from a foreign body reaction. Of the 5 facial palsies, 1 was permanent and 4 were transient.

Of the 39 cases requiring reoperation, 21 (53.8%) were related to cochlear implant repositioning, implantation, or explantation, and 18 (46.2%) were related to revision of the STP procedure. These 18 cases included 6 for revision of EAC closure, 4 for unspecified wound closure, 4 for recurrent cholesteatoma, 2 for reablation of the eustachian tube, 1 for wound debridement, and 1 for reversal of the mastoid obliteration.

Of the 26 cases with dysfunctional, failed, or explanted cochlear implants, 14 (53.8%) explantations were related to infection. Eight dysfunctional or failed cochlear implants (30.8%) were of unknown etiology, 2 (7.7%) were due to trauma, 1 (3.8%) was related to open exposure, and 1 (3.8%) was due to fibrotic resorption of the entire cochlea in a patient with a jugular foramen paraganglioma. These cases of cochlear implants that were found to be dysfunctional or that experienced failure occurred after STP and cochlear implantation were performed.

When STP and cochlear implantations performed in pediatric vs adult populations were compared, there were no significant differences in complication rates, proportion of single-stage vs multistage procedures, sex of patients, or number of cases with cholesteatoma. When the use of abdominal fat or temporalis graft as obliteration material was compared, STP performed in adults had a significantly higher proportion of cases using abdominal fat vs those performed in children (171 of 182 [94.0%] vs 10 of 22 [45.5%]) (eTable 1 in the Supplement).

When single-stage vs multistage procedures and the use of temporalis vs fat as obliteration material were compared, there were no significant differences in complication rates (eTables 2 and 3 in the Supplement). Multistage operations were significantly more often performed for cases with cholesteatoma than were single-stage operations (23 of 54 [42.6%] vs 35 of 174 [20.1%]).

Our results indicate that STP provides a safe environment for cochlear implantation in patients with challenging ears. Use of STP has allowed cochlear implantation to be performed in patients with chronic ear conditions, such as otitis media with cholesteatoma; these have previously been viewed as relative contraindications to cochlear implantation.⁴³ In fact, our findings reveal that 55.4% of patients had a history of otitis media and 35.5% had a preexisting mastoid bowl after canal wall–down (CWD) mastoidectomy for presumed infectious ear disease. These findings have shown a variety of indications for STP beyond infectious processes. Patients with unfavorable anatomy, most commonly resulting from sclerotic mastoids and inner ear malformations, can receive cochlear implantation after the excellent visual exposure that STP provides.

Our study demonstrates no significant differences in rates of complications, reoperation, or device failures between pediatric and adult populations. Because STP and cochlear implantation can be safely performed in pediatric populations, STP for cochlear implantation can be more frequently considered for children with ear disease resulting in complete sensorineural hearing loss, because earlier cochlear implantation has demonstrated improved speech and language development.⁴⁴ In addition, STP can eliminate middle ear disease at an earlier interval as well as the necessity of continuous microscopic reevaluation or additional procedures such as tympanostomy tube placement, tympanoplasty, or mastoidectomy.^21,45 In addition to congenital anatomical abnormalities, rarer ear disease such as acute eosinophilic otitis media, other autoinflammatory conditions (tumefactive inflammatory pseudotumor), and tumors (papillary adenoma) have been effectively managed by STP. Patients with these less frequently encountered ear diseases can be considered candidates for STP and cochlear implantation if they have significant hearing loss.

In the setting of cochlear implantation, STP is advantageous over alternatives such as a CWD mastoidectomy. First, a mastoid bowl after CWD mastoidectomy might require lifelong cleaning and subject the patient to continuous ear drainage, restrictions on activities of leisure (eg, swimming), and episodes of vertigo from caloric effects after water exposure or pressure changes.⁴⁶ Furthermore, open mastoid cavities have higher rates of electrode array extrusion. Alternatively, STP eliminates the need for routine open cavity monitoring as well as reduces the risks of infection and electrode array extrusion. In addition, without any residual hearing necessitating hearing restoration via cochlear implantation, the maintenance of a mastoid bowl becomes obsolete. However, STP is not without its risks; complications resulting from STP include infection of the obliteration graft, breakdown of the EAC blind sac closure, wound at the site of graft retrieval (eg, abdominal wound hematoma), and entrapped cholesteatoma. Our study revealed a global complication rate of 12.4%, which is lower than rates reported after CWD mastoidectomy with cochlear implantation (30% in a systematic review by Hunter et al⁴⁷). Meticulous surgical overclosure of both the EAC and eustachian tube can reduce this.^5,21 Specifically, reinforcement of the EAC overclosure by using a vascularized tissue flap might greatly reduce EAC overclosure breakdown. The comparable complication rate demonstrates STP to be an effective surgery with little morbidity for patients undergoing cochlear implantation.

Subtotal petrosectomy allows for a wide operative exposure conducive to meticulous removal of cholesteatoma. This is demonstrated by this study’s pooled cholesteatoma recidivism rate of 9.3%, which is comparatively less than that reported in the literature (15% after intact canal wall and 16% after CWD procedures).⁴⁸ Second-look operations to detect cholesteatoma might serve as an important safeguard to prevent residual or recurrent cholesteatoma from eroding temporal bone and potentially causing device failure.³⁰ In fact, our study revealed that 42.6% of cases with cholesteatoma underwent multistage operations while only 20.1% of cases underwent single-stage operations. This demonstrates appropriate surgical planning based on ear pathology, because during the second stage when cochlear implantation is usually performed, surgeons can simultaneously examine the middle ear and mastoid cavity for residual cholesteatoma.⁴¹ In addition, the inability to examine the middle ear after STP necessitates radiological monitoring for cholesteatoma recidivism. Although newer cochlear implant devices have had better magnetic resonance imaging compatibility, shadowing artifact still makes visualization of cholesteatoma recidivism poor.⁴⁹ Thus, clinicians might find high-resolution computed tomography to be the most effective imaging modality at present in visualizing cholesteatoma recidivism in patients following cochlear implantation.³ Overall, these results indicate that (1) middle ear disease with cholesteatoma can be safely managed by STP with cochlear implantation and (2) a multistage procedure is preferred in extensive cases with cholesteatoma to provide a second-look opportunity for recurrence simultaneously with cochlear implantation.

The choice of obliteration material often depends on surgeon preference and patient-specific factors. In this study, most included cases used abdominal fat, followed by a vascularized regional flap and bone pate. Abdominal fat is easily accessible, has a low metabolic rate, and can provide improved radiological visual contrast of cholesteatoma reformation on high-resolution computed tomography.^3,47 On the other hand, some studies have shown that a vascularized regional flap might be preferred in cases of infection because of its rich vascularization.⁵⁰ Our study demonstrated that temporalis musculofascial flap was used more frequently in pediatric populations than adult populations. Perhaps this is owing to surgeon preference or to the nature of ear abnormalities between pediatric and adult populations in this series, because STP might be indicated for cholesteatoma more frequently in adults than children.

This study has numerous limitations. First, we elected to include low-level evidence such as case series and case reports to capture a wide variety of indications for STP and cochlear implantation. Second, based on available and extractable data, we were not able to comment on surgical wound closure technique,⁵¹ rates of active vs stable ear disease, or surgeon experience. Third, we were unable to perform any direct comparison between STP and other surgical procedures in preparing the ear for cochlear implantation. These analyses might have provided useful insight into technical variations of STP and might be a subject for future prospective studies.

Subtotal petrosectomy has been shown to be effective in creating a sterile and closed-off environment of the middle ear and can allow for excellent visualization during cochlear implantation. In addition, STP can allow for cochlear implantation in ears with conditions previously believed to be contraindications, such as chronic ear infection. Presence of active inflammation or cholesteatoma might indicate that a multistage procedure should be performed. Last, STP can be considered in children requiring cochlear implantation to minimize ear-related issues and allow early benefit from cochlear implantation.

Accepted for Publication: August 5, 2020.

Corresponding Author: Flora Yan, BA, Department of Otolaryngology–Head and Neck Surgery, Medical University of South Carolina, 135 Rutledge Ave, MSC Room 550, Charleston, SC 29425 (yanf@musc.edu).

Published Online: October 15, 2020. doi:10.1001/jamaoto.2020.3380

Author Contributions: Ms Yan and Dr Nguyen had full access to all the data in the study and take responsibility for the integrity of the data and the accuracy of the data analysis.

Concept and design: Yan, Meyer.

Acquisition, analysis, or interpretation of data: All authors.

Drafting of the manuscript: Yan, Reddy.

Critical revision of the manuscript for important intellectual content: Isaac, Nguyen, McRackan, Meyer.

Statistical analysis: Yan, Nguyen.

Administrative, technical, or material support: Reddy.

Supervision: Isaac, McRackan, Meyer.

Conflict of Interest Disclosures: Dr McRackan reported receiving personal fees from Envoy Medical as well as funding from the American Cochlear Implant Alliance and National Institutes of Health. No other disclosures were reported.