Cochlear implantation is safe and effective in patients with MYH9-related disease

MYH9-related disease (MYH9-RD) is an autosomal-dominant syndromic disorder deriving from mutations in MYH9, the gene for the heavy chain of non-muscle myosin IIA (NMMHC-IIA) [1, 2]. MYH9-RD is characterized by a complex phenotype. All patients present at birth with thrombocytopenia, platelet macrocytosis, and pathognomonic cytoplasmic inclusions of the mutant protein in leukocytes; most of them subsequently develop sensorineural hearing loss, proteinuric nephropathy, presenile cataract, and/or alterations of liver enzymes [3, 4]. MYH9-RD encompasses four syndromes that have been considered for many years as distinct disorders, May-Hegglin Anomaly (MHA, MIM 155100), Sebastian syndrome (SBS, MIM 605249), Fechtner syndrome (FTNS, MIM 153640), and Epstein syndrome (EPTS, MIM 153650). The identification of MYH9 as the gene responsible for all of these syndromes led to demonstration that MHA, SBS, FTNS, and EPTS actually represented different clinical presentations of the same condition, for which the definition of MYH9-RD has been introduced [5‐7].

Sensorineural hearing loss is the most frequent non-congenital manifestation of the MYH9-RD. A recent analysis of 255 patients showed that deafness was present in about one-half of cases at a mean age at evaluation of 35 years, and was expected to develop over time in almost all cases [8]. The severity of hearing impairment is variable among different patients: while in some MYH9-RD subjects the hearing defect is mild or moderate even at advanced age, in other patients hearing loss presents during childhood and progresses to profound deafness within the first decades of life [8‐10]. Thus, in many MYH9-RD patients deafness greatly contributes to patients’ disability.

Non-muscle myosin-IIA is a cytoplasmic myosin expressed in most cell types and tissues, including the inner ear [11‐14]. As all conventional myosins, it has a hexameric structure formed by one heavy chain (NMMHC-IIA) dimer and two pairs of light chains. Each NMMHC-IIA molecule recognizes an N-terminal head domain (HD) responsible for enzymatic activity and a C-terminal tail domain (TD) mainly responsible for myosin assembly [15]. Genotype-phenotype studies showed that patients with mutations in the HD have a higher risk of early-onset and severe deafness than subjects with mutations in the TD [8‐10]. Of note, the MYH9 gene was identified as responsible also for a non-syndromic form of autosomal-dominant deafness, designated DFNA17 (MIM 603622), in two pedigrees carrying the same HD mutation, p.R705H [12, 16].

Cochlear implantation (CI) is a potential option for patients with MYH9-RD and severe to profound deafness, however, no consistent data are available about the risk to benefit ratio of CI in this condition. Reduced platelet counts of MYH9-RD patients obviously result in an increased risk of bleeding complications during of after surgery, and decision to perform CI and peri-operative management requires the co-operation of the hematologist with the ENT surgeon. To date, only one patient with a clinical diagnosis of EPTS who had received CI has been reported [17]. This patient benefited from CI, but surgery was complicated by delayed wound healing and subsequent severe chronic infection, which were attributed to the chronic thrombocytopenia and defective tissue repair due to the impaired NMMHC-IIA function [17]. The efficacy of CI in MYH9-related deafness is made even more uncertain in view of the discordant results obtained in non-syndromic deafness DFNA17: members of both the reported DNFA17 pedigrees received CIs, with good outcomes in one family and very poor results in the other one [16, 18].

This study includes 10 MYH9-RD patients who received CI between 1987 and 2009 at 8 different ENT centres in The Netherlands (1 centre/3 patients), Italy (3 centres/3 patients), France, Germany, Greece, and Argentina (1 patient each) [19‐22]. Mutational screening of the MYH9 gene was performed at 3 different institutions by previously reported methods [19, 20]. Immunofluorescence assay for the identification of NMMHC-IIA leukocyte inclusions was carried out as previously described [19, 23]. Severity of bleeding was graded according to the WHO bleeding score, except for severity of intraoperative bleeding during CI surgery, which was described according to Boezaart et al. [24]. Pure tone audiometric examinations were performed at the different ENT centers by standard methods. Pure tone average (PTA) was calculated using air conduction thresholds at 500, 1000, 2000 and 4000 Hz. Speech discrimination tests were administered at the different ENT centers too. Despite some differences in the utilized tests, all of them included the assessment of the percentages of discrimination of words and sentences at a conversation voice from an open list, which were therefore used to describe the CI outcomes. Unless otherwise specified, the speech perception scores are mentioned without the use of visual support. To describe short-term outcome of CI, evaluation at 6 or 12 months after switch-on of the implanted device was reported; for long-term outcome, evaluation at the last follow-up visit was used. The investigation was approved by the Institutional Review Board of the IRCCS Policlinico San Matteo Foundation, Pavia, Italy. All the patients or their legal guardians gave written informed consent for this retrospective study, which was conducted according to the declaration of Helsinki.

Patients affected by MYH9-RD with severe to profound deafness are potential candidates for CI. To date the only information about outcome of this procedure in MYH9-RD derives from a single case report. This subject benefited from CI, but surgery was complicated by delayed wound healing attributed to the chronic thrombocytopenia and/or NMMHC-IIA dysfunction. Moreover, the effectiveness of CI in MYH9-related deafness was questioned in view of the discordant results obtained in two families with the non-syndromic deafness DFNA17 [25]. In order to provide consistent information on the risk to benefit ratio of CI in MYH9-RD, we have gathered the data of 10 patients who received CI at 8 different institutions.

CI was effective in improving the hearing ability in 9 out of 10 MYH9-RD patients, while one subject did not take advantage from the procedure. In particular, 8 responders obtained excellent performances, with restoration of a practically regular hearing function since evaluation at 6–12 months after the switch-on of the implant. All of them referred restoration of the ability to engage in a normal conversation even in a noisy environment; they also reported good performance with phone conversations or listening to devices such as radio or television. These excellent responders were characterized by durations of severe deafness prior to CI ranging from 2 up to 11 years (mean, 7.0 years). Moreover, CI performance was similarly good in patients with different total durations of deafness before CI (7 to 55 years) and with different ages at implantation (childhood up to 72 years). Finally, similarly good performances have been obtained in 6 patients with mutations hitting the HD of NMMHC-IIA or in 2 patients with mutations in the TD, suggesting that CI outcome was independent on the specific MYH9 alteration. One subject (patient 9) benefited from CI, but performance was not as good as in the other responders. In fact, results of her speech perception tests were poor at the short-term evaluation after switch-on; however, her speech discrimination scores ameliorated over time until reaching fairly good levels that were maintained until 17 years after implantation. This patient differed from the other ones for the markedly longer duration of severe deafness before CI, i.e. 22 years. In CI responders affected by other forms of post lingual deafness, duration of severe deafness prior to implantation was a major predictor of CI performance [26, 27]. We therefore hypothesize that this feature affected the CI performance also in this subject and we suggest that CI should be offered to MYH9-RD patients shortly after they develop criteria for candidacy.

On the other hand, the reasons why the patient 6 did not benefit from CI remain undetermined. Among the different factors that could potentially affect CI outcome [26, 28, 29], we could not identify any feature of this patient explaining her poor response. The patient carried the p.R33W mutation of the HD of NMMHC-IIA, which represents a rare variant described in only one other MYH9-RD patient [30]. However, the clinical pictures of both patients with p.W33R do not suggest that this mutation induces a particularly severe NMMHC-IIA dysfunction with respect to other HD mutations.

CI surgery was carried out without major bleeding complications. In 4 cases with severe thrombocytopenia, bleeding risk was managed by prophylactic transfusion of 1–2 apheresis platelet concentrates. One center with experience in the care of MYH9-RD patients routinely uses tranexamic acid to prepare for surgery patients with moderate thrombocytopenia [31], and successfully used this drug for prophylaxis of patient 10. In 4 patients with automated platelet counts ranging from 70 to 129 × 10⁹/L prophylaxis for bleeding was not deemed necessary. On the whole, intraoperative bleeding was minimal in 8 patients and moderate in two cases; no postoperative bleedings occurred, with the exception of the formation, in one patient, of a hematoma at the site of surgical wound, which was drained by removing one stitch without any clinically relevant consequences. In clinical practice, two aspects should be considered for management of perioperative bleeding risk of MYH9-RD patients. First, routine automated cell counters usually underestimate platelet counts of these patients. In fact, electronic instruments identify platelets mainly based on their size and fail to recognize very large platelets typical of MYH9-RD [32]. Thus, microscopic counting should be used to assess the actual platelet counts of MYH9-RD patients for their proper management. Secondly, since in vitro platelet function in MYH9-RD is normal or only slightly reduced, the indication for prophylactic transfusions can be reasonably based on the general recommendations for thrombocytopenias. Recent guidelines recommend prophylaxis for patients with platelet counts below 100 × 10⁹/L before surgery at critical sites, and this threshold should be considered for CI [33]. On another line, none of the patients experienced complications related to delayed wound healing. It is therefore unlike that the complications observed in the previously reported MYH9-RD patient who received CI were dependent on factors specific to the disease [17]. Finally, three of our patients had conditions leading to increased risk of infection that are rather frequent among MYH9-RD patients [3]: two patients had been splenectomized because of a previous misdiagnosis with immune thrombocytopenia and one patient was on immunosuppressive treatment after kidney transplantation. None of them experienced infectious complications after administration of standard antimicrobial prophylaxis. We therefore conclude that CI is a safe procedure in MYH9-RD patients whenever adequate prophylactic interventions are carried out.

Pathogenesis of MYH9-related deafness is still unclear. Studies on mouse inner ear showed that NMMHC-IIA is extensively localized in the hair cells of the organ of Corti, the spiral ligament and the spiral limbus, with only minimal expression within the spiral ganglion [12, 13]. In hair cells, NMMHC-IIA is abundantly expressed in stereocilia [14]. Given that CI bypasses hair cells by directly stimulating the spiral ganglion, the finding that most MYH9-RD patients have excellent CI performances is consistent with the NMMHC-IIA expression pattern observed in animals and with the conclusion that MYH9 mutations primarily damage the hair cells. Altogether, these observations strengthen the notion that CI outcome is better in patients with deafness caused by defects of genes primarily expressed in the hair cells/membranous labyrinth as opposed to mutations causing spiral ganglion pathology [25, 34], and further point out the importance of genetic testing in CI candidates. The introduction of massively parallel sequencing technology led to recent development of approaches for an efficient screening of all known deafness genes simultaneously [35‐37]. The identification of definite correlations between genotype and CI outcome will pave the road to a tailored patients’ management and reduce the likelihood of ineffective CIs and unnecessary costs in healthcare.

AP designed the research, acquired data, analyzed and interpreted data, and drafted the manuscript. EJJV, NS, PC, CMB, HP, EK, MB, and VT acquired data, analyzed and interpreted data, and critically revised the manuscript. AG designed research, analyzed and interpreted data, and drafted the manuscript. All the authors revised and accepted the final version of the manuscript.