Outcomes of Cochlear implantation in early-deafened patients with Waardenburg syndrome: A systematic review and narrative synthesis

2.1 Eligibility criteria

Population: The participants included were children or adults with Waardenburg syndrome causing profound hearing loss.

Intervention: The intervention was cochlear implantation. No restrictions were placed on implant device or method of insertion.

Comparison: No formal comparison group was used as hearing function was not expected to change without implantation. However, any comparison to other indications for CI was noted.

Outcomes: The primary outcomes were pre- and post-implantation audiometric outcomes using audiometry and/or speech perception and/or speech production. Where pre-implantation outcomes were not available, post-implantation audiometric outcomes were analyzed only. The secondary outcomes considered were genetic analysis, pre-implantation radiological findings and intra- or post-operative complications.

2.2 Study inclusion criteria

Clinical studies of cochlear implantation in patients with Waardenburg Syndrome with hearing outcomes reported at a minimum of 3 months post implantation were included. Diagnosis of WS may be clinical or genetic and of any subtype. Human studies of any methodology other than case reports or case series <3 patients were included. Studies without report of postoperative audiometric outcomes or where the abstract or full text were unavailable were excluded.

2.3 Search strategy

Searches were initially performed by our information specialist librarian and subsequently repeated by ME, and abstracts were independently screened by two reviewers (AL/ME). The following databases were searched: Medline, Pubmed, Embase, Web of Science, Cochrane Collection, ClinicalTrials.gov (via Cochrane).

No limit was placed on language or year of publication.

2.4 Selection of studies

Two reviewers (AL/ME) independently screened all records identified from the database searches. Studies describing cochlear implantation in patients with WS were assessed against the inclusion and exclusion criteria, with any disagreement resolved by discussion with a third reviewer (CM). Studies without an accessible abstract or full text after the title/abstract screening were followed up by attempting to contact the study authors. If they remained unavailable the study was excluded. Studies were excluded if they did not report post intervention audiometric outcomes at a minimum of 3 months post procedure. Studies presenting overlapping populations were limited to the largest study sharing data. Potentially relevant studies highlighted from the initial searches and abstract screening underwent full-text screening by two independent reviewers (AL/ME) prior to data extraction. Conflicts on the selection were resolved by discussion between the reviewers.

2.5 Data extraction

Data was extracted by the first reviewer (AL) and then checked by a second reviewer (ME). Extracted data was collected in a spreadsheet (Excel, Microsoft Corp, WA, USA). The data of interest comprised of location, study design, participant characteristics (including age at implantation, WS subtype, genetic analysis, and anatomy on imaging), intervention characteristics (including operative technique, implant type and perioperative complications), and primary outcome data (including follow-up timeframe).

2.6 Risk of bias quality scoring

Two reviewers (AL/ME) independently assessed the risk of bias using the Brazzelli Risk of Bias Tool for NonRandomized Studies.²¹ Studies were also graded according to the Oxford Centre for Evidence Based Medicine Grading System.²² Discrepancies between the reviewers were resolved by discussion with a third reviewer (CM).

2.7 Synthesis of results

Study results have been presented by outcome measures. Additional study characteristics and findings of WS subtype, genetic analysis and radiological findings have been collated. No meta-analysis was undertaken due to the heterogeneity in methodology and outcome measures reported between and within studies.

3.1 Description of studies

Sixteen studies met the inclusion criteria with a total of 179 patients and at least 194 implants. There were nine case series, two case–control, and five cohort studies; all included between three and 30 WS patients. All studies were published between 2000 and 2022. Study characteristics are summarized in Table 1.^{15, 24-38}

3.2 Demographics

Fifteen studies included pediatric patients only, one study included both children and adults.²⁶ The average age at time of cochlear implantation ranged from 12 months to 6.5 years; the oldest patient included was 22 years old. One study did not report the age at implantation.³³ Thirteen studies reported on the type of implant used.^{15, 24-32, 37, 38} Reporting on WS type was variable, with 10 studies specifying the subtype of the condition based on clinical manifestation of the syndrome.^{24, 25, 27, 28, 30, 32, 34, 37, 38} Five studies reported on genetic analysis in 29 patients following identification of hearing loss or clinical diagnosis; range of mutations including SOX10/PAX3/SNA12/SW2B/SW2C/EDN3/EDNRB/MITF, though the method of identification of these mutations was not reported in any series.^{24, 25, 30, 34, 37} Radiological assessment with either preoperative computed tomography (CT) or magnetic resonance imaging (MRI) was reported in 100 patients across 10 studies (Table 1)^{15, 24, 25, 27, 28, 30-32, 34, 37}; findings varied from normal anatomy, to malformations of the cochlea, vestibule or semicircular canals. Only one study discussed pre-implantation radiological findings in relation to genetic analysis, highlighting one case of bilateral cochlear hypoplasia in a patient with SOX10 mutation.³⁷

3.3 Quality of studies

The methodological quality of included studies was modest, predominantly consisting of retrospective noncontrolled case series with small numbers of patients. All studies were OCEBM grade III-IV (Table 1). In addition to the heterogeneity of study types included, there were inconsistencies in reporting of pre-implantation assessment and use of variable audiological outcomes presented within and between studies which precluded formal meta-analysis. There were also limitations in reporting of WS subtype, genetic analysis, pre-implantation radiological findings, surgical technique and rehabilitation protocols (Table 2).

3.4 Audiological outcomes

Reporting of preoperative audiological assessment was highly variable. Seven studies provided pre-implantation pure tone audiometry (PTA) and five utilized speech perception scores, six reported vague description of hearing level with one study providing no details of baseline hearing (Table 3). Where provided pre-op hearing loss was in keeping with severe to profound hearing loss.

Post-implantation hearing outcomes demonstrated improvement across all studies; however, reporting of follow up duration and outcome measures was heterogenous across studies. The shortest average duration of follow up was 12 months and the longest follow up period was 15 years.

Post-implantation hearing outcomes are summarized in Table 4. A total of 22 different audiological outcome measures were used with heterogeneity in outcome measures reported between studies. There was also variability seen within studies particularly regarding pre- and post-implantation outcome measures used.

PTA was reported both pre- and post-implantation in four studies^{24, 26, 28, 34} all of which demonstrated improvement in average thresholds across their respective cohorts. Claros et al. demonstrated an improvement in average thresholds from 95.4 dB ± 4.2 to 23.2 ± 9.9 post-CI with no significant difference in post-implantation thresholds when compared to a reference group of 86 nonsyndromic children (p = .49).²⁴

Speech perception was assessed in all studies through varied tools. The Meaningful Auditory Integration Scale (MAIS) or the Infant-Toddler Meaningful Auditory Integration Scale (IT-MAIS) was used most frequently across six studies to assess postoperative speech perception.^{24, 26, 30, 32, 37, 38} Polanski et al., reported both pre and post intervention IT-MAIS/MAIS demonstrating improvement from average scores of 3%–90%. Categories of Auditory Performance (CAP) were also commonly used to assess postoperative performance in six studies,^{24, 27, 29-32} with two also utilizing CAP preoperatively.^{27, 29} Amilsari et al. demonstrated statistically significant improvement in CAP scores post-CI in 6 WS children (0.33 ± 0.5 to 4.00 ± 1.26 (p < .05)) with no significant difference in outcome when compared to a reference group of 75 implanted nonsyndromic children.²⁹ Other assessment tools included speech recognition scores (SRS), phoneme scores, Open-Set Words (OSW), Closed-Set Words (CSW), CI-2004 test and the Melbourne Speech Perception Score.

Speech intelligibility was assessed in 10 studies through a variety of measures. The Speech Intelligibility Rating (SIR) was utilized in five studies^{24, 27, 29, 30, 32} and the Meaningful Use of Speech Scale (MUSS) was implemented in six to assess post-implantation speech intelligibility.^{24, 26, 30, 34, 37, 38} Claros et al., demonstrated significantly better speech intelligibility outcomes (SIR/MUSS) following CI compared to a reference group of 86 implanted non syndromic hearing loss cases (p < .05).²⁴ No study used a formal framework to evaluate quality of life following CI, although four studies provided descriptive analysis through main method of communication and attendance of mainstream education.^{25, 27, 30, 35}

Overall, outcomes were favorable following cochlear implantation regardless of assessment method used. However, follow-up post implantation was variable, as were the assessment tools used to report audiological outcomes. Pre-implantation hearing status audiology was not routinely reported and in many cases measures did not correlate with those used post-CI limiting comparison. A number of studies also provided comparison with reference groups however, these groups were often much larger, limited baseline data and not matched to the study population.

3.5 Surgical outcomes

Five studies reported on intra- or post-operative complications.^{15, 24, 28, 31, 38} A total of 12 complications were reported in as many patients. The most frequently reported were implant failure and CSF gusher accounting for four cases each. In patients with CSF gusher inner ear abnormalities were identified including two cases of enlarged vestibular aqueduct (EVA), one common cavity deformity (CCD) and one incomplete partition type II (IP-2) deformity. The EVA and IP-II were related to WS type 2, and CCD to WS type 4. Other less frequent complications included displaced electrode, seroma formation and wound infection (Table 2).

4.1 Limitations and recommendations

This is an uncommon condition and as such studies present small sample sizes and varied outcome measures limits pooled analysis. These issues could be addressed by large-scale mandatory registries of implantation recipients and results, particularly useful for such types of rare diseases. Such registries are not yet in use but the proliferation of electronic patient records and increased interest in outcome measures is likely to drive adoption. While a number of challenges exist in the implementation of national registries, including oversight, funding and legal implications, there are trends toward the development of such a database for cochlear implantation.^{40, 41}

Additionally, the inclusion of studies presenting a diverse spectrum of audiological outcomes, many without reported pre-implantation audiological assessment, resulted in heterogeneity. Moreover, the categorical scales used for measurement of performance outcome following CI are subjective, being dependent on the assessing clinician's judgment. The audiological tools have been developed for assessment of individuals with hearing loss and therefore comparison to normative population outcomes is not possible. The use of standardized auditory and spoken language assessment tools, in addition to audiology, would allow clinicians the ability to critically assess outcomes following cochlear implantation as well as facilitate the synthesis of a wider pool of data and provide opportunities to more accurately assess outcomes on a larger scale.

Furthermore, assessment of appropriate diagnosis, either by genetic analysis or clinical classification, was difficult to ascertain in the included studies. For the five studies including genetic analysis to aid or support diagnosis of WS, it was evident that not all genetic mutations had been tested for in each patient. In the future, it would be useful if genetic testing and appropriate counseling could be undertaken whenever possible, with a variety of gene loci mutations being assessed regardless of the phenotypical clinical classification of the disease.