Hearing impairment in X-linked a review

Authors: Seongeun Oh, Jessica L Sandy, Craig F Munns, Peter J Simm, Aris Siafarikas, Lucy Collins, Christie-Lee Wall, Maria E Craig, Christine P Rodda, Andrew Biggin

Abstract

Although hearing impairment is often listed as a nonskeletal complication of X-linked hypophosphatemia (XLH), the prevalence, etiology, pathology, and natural history are poorly described. This review aims to summarize existing literature with a view to guide the clinical management of hearing impairment in XLH. The review was conducted by 2 researchers independently. Four databases (PubMed/Medline, EMBASE, Web of Science, and Cochrane Library) were searched between January 1, 2000 and July 31, 2024, with keywords related to “X-linked hypophosphataemic rickets” and “hearing loss” including synonyms. Identified records were screened for inclusion and exclusion criteria. Human and animal studies were included. Out of 82 records found excluding duplicates, 12 studies met the final criteria and were reviewed. Studies described both conductive and sensorineural hearing loss in 13%-76% of adults with XLH, with sensorineural hearing loss more commonly reported, with impairment developing in adulthood, affecting high and low frequencies, and may be fluctuating. Evidence suggests that endolymphatic hydrops (ELH) may be a major underlying cause of hearing loss in XLH. Individuals with XLH have generalized osteosclerosis with petrous bone thickening and narrowing of the auditory meatus. No studies have looked at burosumab, a monoclonal antibody that inhibits FGF23, and its effect on the development of hearing loss in individuals with XLH. Animal studies of XLH mouse models (Hyp and Gy) describe both conductive and sensorineural hearing impairment. Mouse models demonstrate high Auditory Brainstem Response (ABR) thresholds and signs of osteomalacia of auditory ossicles and ELH. In conclusion, there is an association between hearing loss in XLH and, most commonly, adult-onset sensorineural hearing loss. Pathogenesis of hearing loss in XLH is incompletely understood, but possible contributing factors include thickening of the temporal bones, osteomalacia of the auditory ossicles, and development of ELH. There is currently no evidence that treatment with conventional therapy or burosumab reduces the risk or severity of hearing impairment.

Introduction

Hypophosphataemic rickets refers to a group of inherited or acquired disorders characterized by chronic hypophosphatemia due to reduced renal tubular reabsorption of phosphate.^1^ The majority of hypophosphataemic rickets is hereditary, with X-linked hypophosphataemic rickets (XLH) being the most common form, accounting for 80% of all types of rickets, with an estimated prevalence of approximately 1 in 20 000.^2^ XLH has an X-linked dominant pattern of inheritance, where loss-of-function mutations of the Phosphate Regulating Endopeptidase Homolog X-Linked (PHEX) gene cause inappropriate elevation of serum fibroblast growth factor 23 (FGF23) levels.^1^ In addition to skeletal manifestations, hearing impairment in XLH patients has also been reported in numerous studies since the 1980s.

FGF23 is a bone-derived phosphaturic hormone involved in the downregulation of renal sodium-phosphate co-transporters in the proximal tubules as well as in the inhibition of renal \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{upgreek} \usepackage{mathrsfs} \setlength{\oddsidemargin}{-69pt} \begin{document} $1\alpha$\end{document}-hydroxylase and stimulation of \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{upgreek} \usepackage{mathrsfs} \setlength{\oddsidemargin}{-69pt} \begin{document} $24\alpha$\end{document}-hydroxylase, causing a reduction in calcitriol levels.^3^ Excess FGF23 level in XLH subsequently leads to a decrease in both renal reabsorption and intestinal absorption of phosphate, resulting in chronic hypophosphatemia. Defective phosphate metabolism contributes to a broad range of local and systemic consequences in affected children and adults.^2^

Historically, the conventional treatment for XLH has been with a combination of phosphate and activated vitamin D supplementation (calcitriol or where this is unavailable 1 alpha hydroxyvitamin D3).^4^ In 2018, the development of burosumab, a fully human monoclonal antibody against FGF23, resulted in a ground-breaking transition in the management of XLH. With its specific FGF23 antagonism, more direct normalization of biochemical parameters can be achieved requiring less frequent monitoring for dosing, with considerably less side effects such as secondary/tertiary hyperparathyroidism and nephrocalcinosis.^5^

With regard to hearing impairment, evidence from early key studies largely suggests a predominant sensorineural hearing loss in XLH adults of cochlear origin (similar to that which occurs in Ménière’s disease), rather than neural lesions.^6–9^ These findings are summarized in Table 1. Key cochlear pathology noted across hearing-impaired XLH patients was Endolymphatic Hydrops (ELH), potentially precipitated by generalized structural changes in the petrous bone as noted by O’Malley et al.,^8^ which may alter the endolymph turnover between the stria vascularis and the endolymphatic sac, or cause obstruction of the endolymphatic duct/sac. Multiple potential molecular mechanisms for hearing loss in XLH have been proposed.^1^ These may involve PHEX directly, be mediated by increased FGF23, or via other inflammatory processes. Despite these early findings, however, details on the prevalence, etiology, and natural history of hearing loss in XLH remained poorly understood, limited by a small sample size, especially of pediatric patients, and limited technologies. The impact of treatment, either with conventional therapy or FGF-23 inhibition, on hearing impairment is also unclear.

Since the 2000s, more reliable studies have been published with larger adult and pediatric populations, along with the use of more advanced otological methodologies. Screening in older children and adults is most often performed via pure-tone audiometry, which enables the classification of hearing impairment into conductive, sensorineural, or mixed patterns. Other modes of assessment are also often performed, including Auditory Brainstem Response (ABR) thresholds, which reflect the electrical signal of the central auditory pathways and the auditory nerve, as well as Distortion Product Otoacoustic Emission (DPOAE) thresholds, which reflect the outer hair cell function.^10^

Animal studies including numerous mice models of different PHEX mutation alleles have been identified, enabling modeling of hearing impairment in human XLH pathogenesis. Widely recognized strains include male mice hemizygous for Hyp, Hyp-Duk, Hyp-2J, and Gy alleles, each denoted as Hyp/Y, Phex^Hyp-Duk^/Y, Phex^Hyp-2J^/Y, and Gy/Y, respectively.^11^^,^^12^

This review aims to clarify the association between XLH and hearing impairment and to establish the pattern, age of onset, and pathophysiology of hearing impairment in XLH. It is anticipated that these results will guide future clinical management and investigate whether early routine hearing assessment and treatment can effectively delay or improve the degree of hearing impairment in XLH.

Materials and methods

Databases PubMed, EMBASE, Web of Science, Cochrane Library

Search and screening were conducted between November 2023 and July 2024

Keywords: X-linked hypophosphataemic rickets (synonyms X-linked hypophosphataemic rickets, familial hypophosphataemic rickets); hearing loss (synonyms deafness, hearing impairment, deaf, hard of hearing)

Search Query: (((familial hypophosphat^^) OR (X linked hypophosphat^^) OR (hereditary hypophosphat^^) OR (congenital hypophosphat^^) OR (genetic hypophosphat^^)) AND ((hearing loss) OR (deafness) OR (hearing impair^^) OR (deaf) OR (hard of hearing) OR (hearing disorder^*^)))

^*^Filters English language only, published date January 1, 2000–Current.

Inclusion

• – Full-text articles
• – English only
• – Published date January 1, 2000 to July 31, 2024
• – Case series/reports, observational studies (cohort, case–control, cross-sectional studies)
• – Human & animal studies

Exclusion

• – Reviews, conference abstracts, notes, comments, book chapter, surveys
• – Non-X-linked familial hypophosphataemic rickets, Vitamin D-deficient rickets

Results

The search process is outlined as a PRISMA diagram in Figure 1. Twelve studies were selected for the review. The summary of key study characteristics for the selected records is presented in Table 2 for human studies and Table 3 for animal studies. Note that the study by Delsmann et al.^13^ consisted of both human and animal studies hence has been included in both tables.

Nine human studies,^13–21^ including 3 case series^18–20^ and one case report,^21^ and 4 animal studies^13^^,^^22–24^ were included. Within the 9 human studies, a total sample size of 229 XLH patients was represented, consisting of 157 adults and 72 children. Five studies included both adult and pediatric XLH patients, while the other 4 studied adults exclusively.

All 4 animal studies utilized male mutant Phex mice models with their respective age-matched male wildtype (WT) control mice.

Hearing in XLH adults

Various methodologies were used to assess hearing in XLH patients across the studies. Five studies^13–17^ performed formal audiometry on all their samples. Tympanometry was performed by Ivanovic-Zuvic et al.^14^ and Fishman et al.,^17^ with the latter also testing stapedial reflex thresholds. Hearing impairment was noted in a total of 30 XLH patients, with 28 adults and 2 children. The prevalence of hearing impairment was 71%^13^, 23%,^14^ 32%,^15^ 14%^16^, and 15%^17^ of XLH adults across the 5 studies. Of the 28 adults with hearing impairment, 20 had sensorineural hearing loss, 3 had conductive hearing loss, and 5 had mixed hearing loss. Each individual study also consistently reported predominant sensorineural hearing loss in their XLH adult cohorts.

Delsmann et al.^13^ performed cranial CT scans of the auditory ossicles on 2 adults with severe bilateral conductive hearing loss and compared these findings to those of healthy controls. Scans of the XLH adults showed impaired mineralization and severe osteomalacia of the auditory ossicles. Transiliac crest biopsies also showed lower mean calcium content and higher fraction of low mineralized bone areas compared to the control group, indicating severe osteomalacia.

Of note, Ivanovic-Zuvic et al.^14^ reported that subjective hearing loss was reported in 12 adults, while only 4 had confirmed hearing loss. Episodic tinnitus and vertigo were commonly reported in both patients with and without confirmed hearing loss. All 4 adults with confirmed hearing loss reported episodic tinnitus; however, no significant association was observed between otologic manifestations and the severity of other XLH manifestations. There was also no symmetry of severity and pattern of hearing loss between either ear in a single patient.

Of the 20 adults included in this study, 8 (40%) were on conventional treatment with activated vitamin D and phosphate. Analysis of biochemical parameters showed that serum phosphate was lower in patients with hearing loss compared to those with no hearing loss, with no differences seen with other biochemical markers, treatment exposure rate, dosage regime, demographics, or imaging. Studies done on those on treatment, conventional or burosumab, showed similarly high prevalences of hearing impairment, including Kato et al.^15^ reporting a prevalence of 32% of hearing impairment despite all their patients being treated with either conventional therapy or burosumab. Another study by Ivanovic-Zuvic et al.,^14^ where 40% of adults were on conventional therapy demonstrated a prevalence of 20% of hearing impairment.

Hearing in XLH children

Hearing impairment was found only in 2 children with XLH, reported in the study by Ivanovic-Zuvic et al.,^14^ which recruited only 6 children in total. They were 5 and 6 yr old and both were found to have conductive hearing loss and both showed signs of otitis media with effusion, a common condition in the general pediatric population. Neither of these children reported subjective hearing loss. All children were on conventional therapy with activated vitamin D, such as calcitriol and phosphate supplementation. Fishman et al.^17^ reported no hearing loss in 16 of its pediatric patients, which could be attributable to XLH. One child had a profound bilateral hearing loss, but this was due to a coincidental Mondini congenital inner ear malformation on temporal bone CT.

Auditory phenotypes in Phex mutant mice

A wide range of methods was utilized across the studies, including ABRs, DPOAEs, imaging and histology of the cochleae, temporal bones and auditory ossicles. ABR threshold analysis was performed in all 4 animal studies in the review.

Megerian et al.^23^ observed first onset of mild-moderate unilateral hearing loss in Phex^Hyp-Duk^/Y mice of BALB/cUrd background at Postnatal-Day-21 (P21). By P25 and beyond, most mice developed severe bilateral and asymmetric hearing loss, that was progressive with age (~P90). Vestibular dysfunction was also observed around P15 with balance dysfunction becoming prominent in most mutant mice by P30.

Studies however showed variable expression of auditory phenotypes between different background genetic strains of mutant mice. When the same Hyp-Duk mice from a different C57BL/6 (B6) background strain were studied, however, ABR threshold analysis showed no significant hearing impairment in mutant mice compared to WT from P21 to P131, with only some mutants with mild unilateral hearing loss at P61. This was also seen in the study by Lorenz-Depiereux et al.,^24^ which studied 4 groups of mutant mice, each with 4 different PHEX genetic alleles. Hyp and Hyp-2J mice were from B6 strain, Hyp-Duk from BALB/cUrd strain, and Gy from B6EiC3SnF1-a/A strain. All 4 groups clinically presented very similar phenotypes of shortened tail, trunk, and hind legs, with hypophosphatemia, hypocalcaemia, and rachitic bone disease. All 4 mutations have been found to cause a complete loss of PHEX function. On ABR analysis, however, despite all groups except Hyp-2J exhibiting significant hearing impairment compared to their respective WT, the hearing of Hyp-Duk mice from the BALB/cUrd strain was significantly more impaired compared to mutants from different a background strain. The study further analyzed the effects of background strain effects with inter-strain breeding of Hyp-Duk mice, which resulted in an attenuated hearing impairment that was then comparable to Hyp and Gy mutant mice. This indicated the presence of recessive genetic modifiers in the BALB/cUrd strain that exacerbate the auditory phenotypes in Phex mutant mice, supporting the findings by Megerian et al.^23^

Pathological changes in the auditory ossicles and the cochleae in Phex mutant mice

Histological analysis of the ossicles and/or the inner ears was conducted in all 4 studies.

Normal postnatal development of the ossicles in the WT mice (B6 strain) was observed by Delsmann et al.,^13^ which showed voids in the malleus with blood vessels at 3 wk, which then rapidly mineralized by 6 wk as blood vessels regressed and were replaced with osteoblast-laid bone matrix. By 24 wk, a significant decrease in areas of osteoid and increase in mean calcium content was noted. This contrasted with the prominent osteomalacia of the ossicles in 24-wk-old mutant Hyp mice (B6 strain) with a markedly high osteoid volume per bone volume. A mild degree of ELH was detected in the cochleae of Hyp mice. High-resolution CT and electron imaging of the ossicles supported this with significantly higher porosity and lower mean calcium content demonstrating severe hypomineralisation. Histology of the temporal bones and the cochleae in Hyp-Duk mice at different ages by Megerian et al.^23^ provided a detailed picture of the progressive pathological changes in the ears of Phex mutants.

BALB/cUrd Hyp-Duk mice showed no cochlear abnormalities or signs of ELH at birth. At P25, thickening and poor mineralization of the otic capsule was observed with mild ELH, closely following the onset of unilateral hearing loss at P21 noted on ABR analysis. Beyond P40, severe ELH was noted with displacement of the cochlear duct and by P90, degenerative changes were seen in the spiral ganglia and the Organ of Conti. Interestingly, despite severe ELH, no endolymphatic duct obstruction was observed throughout the course in both mutants and WT.

B6 Hyp-Duk mice, which had no significant hearing impairment on ABR compared to WT, showed marked thickening and impaired mineralization of the otic capsule at 4 mo. However, no apparent ELH or other degenerative changes in the cochleae were observed, in contrast to the BALB/cUrd Hyp-Duk mice, which developed progressive hearing loss. A few B6 Hyp-Duk mice with mild unilateral hearing loss at P61 typically had otitis media. This suggests susceptibility to otitis media may aggravate hearing loss, although to a minimal degree.

Comparison of hearing-impaired Hyp-Duk variant and normal-hearing Hyp-2J variant of the same background strain at 5 mo by Lorenz-Depiereux et al.^24^ further supports the role of ELH in pathogenesis. Both groups had thickened temporal bones with areas of poor mineralization. However, the degenerative changes of the organ of Conti and the spiral ganglia in the hearing-impaired Hyp-Duk mice were absent in normal-hearing Hyp-2J mice. Enlarged endolymph volume was also noted in hearing-impaired Hyp-Duk mice, evident through displaced Reissner’s membrane, diminished scala vestibule, and enlarged scala media. The authors suggest this increase in endolymph volume as the likely cause of sensorineural degeneration and subsequent hearing impairment.

Therapeutic interventions in Phex mutant mice

Wick et al.^22^ added further insight into the findings described above by investigating the effects of supplemental phosphate and calcitriol on hearing in BALB/cUrd Phex^Hyp-Duk^/Y mice. Four cohorts were WT, Hyp-Duk control, Phex prevention diet, and Phex rescue diet. The Phex prevention cohort received supplementation between P7 and P40, while the rescue cohort received supplementation between P20 and P40. ABR analysis showed significantly better hearing in WT at all frequencies compared to all 3 Phex cohorts. There was no significant difference across the 3 Phex cohorts.

Histologically, all 3 Phex groups showed gross cochlear malformation compared to WT, but no significant differences were present within the 3 Phex groups. However, both rescue and prevention groups showed marked improvement in the mineralization of the otic capsule compared to the Hyp-Duk control group, although they did not reach the full organization level of the WT control. Of note, there was no evidence of improvement in the degree of ELH in rescue and prevention cohorts despite supplementation. In this study, serum phosphate levels were also significantly lower in the 3 Phex groups compared to WT, but no statistically significant deviation within the 3 Phex cohorts. This may reflect inadequate dosing or timing of treatment, which requires further detailed investigation.

Discussion

This review summarizes the limited literature exploring hearing loss in XLH. The variable methodologies and study designs, while a limitation of this review, were also a strength, as each study presented unique insights into pathogenesis and implications of hearing loss in this population. The paucity of adequate patient studies is another limitation of this review. The few studies meeting the inclusion criteria were also limited in that most were cross-sectional with no longitudinal follow-up. Therefore, the natural pathogenesis of hearing impairment could not be determined. Incomplete assessment and lack of formal audiometry were also a limiting factor and likely under-reports the true prevalence of hearing impairment in XLH. There were also various inconsistencies across the studies, which meant drawing definitive conclusions was not possible. However, these findings highlight several areas for future research. One unanswered question is whether the introduction of burosumab will impact the development of hearing loss in individuals with hearing loss. Therefore, longitudinal studies looking at the impact of burosumab should include an assessment of hearing loss if possible.

Hearing loss in XLH in adults and children

While methodologies were variable across only a small number of studies, the reported studies suggest that hearing impairment is more prevalent in adults with XLH than in the general population, with prevalences ranging from 14%-71%. This is generally higher than the global prevalence of hearing impairment of 6.8%-14.1% in adults from the general population.^14^ The majority of hearing loss in adults with XLH was noted to be sensorineural.

The prevalence of younger adult patients with hearing loss was notably higher in adults with XLH compared to the general population.

Hearing loss is a common chronic condition that increases with age and is influenced by both genetic and environmental factors. Prevalence varies across studies due to genetic and socioeconomic differences, as well as nonstandardized methods of epidemiologic data collection.^25^ However, cross-sectional national data from the USA (1999-2018) indicate that hearing loss prevalence is less than 10% among individuals aged 20-29 and 30-39, but rises sharply to at least 30% in those aged 60-69.^25^^,^^26^ A similar trend is observed globally, as reported in the Global Burden of Disease 2019 study.^27^ The studies included in our review examined XLH patients across a broad age range (17-79 yr). Four^13–15^^,^^17^ of the 5 studies provided age ranges for a total of 20 hearing-impaired XLH patients, all of whom were under 64 yr old. Additionally, 2 studies^13^^,^^15^ provided raw demographic data, showing that 4 out of 6 (67%) and 4 out of 10 (40%) XLH adults between 20 and 39 yr old were hearing impaired. This reported prevalence is substantially higher than those observed in the general population for this age group (less than 10%), suggesting that hearing impairment may develop earlier in XLH patients compared to the general population.

While the pediatric studies looking specifically at hearing loss in XLH were small,^14^^,^^17^ and therefore do not allow any definitive conclusions to be drawn, they did not suggest that hearing impairment in children with XLH was more common than in the general population. When it does occur, it appears to be unrelated to the underlying condition and more likely due to common childhood conditions such as otitis media with effusion.

Looking beyond these cross-sectional studies, there are a number of case series describing single-family pedigrees who did not all undergo systematic formal audiological testing, reporting hearing loss or tinnitus in over 40% of the 22 affected family members^18^ or mixed/sensorineural hearing loss in at least 2 out of 4 affected family members.^19^ A single center case series of 59 people from 36 unrelated families reported sensorineural hearing loss in at least 3 children^20^ but, again, not all participants underwent audiological testing. These findings highlight the scarcity of data in this area but suggest that clinicians should consider formal audiological testing for all adults with XLH. For children, there is not enough evidence to suggest the need for routine audiological testing. Clinical guidelines are variable in what they recommend; Haffner et al.^4^ suggest an initial assessment once the audiological testing is feasible (at around 5 yr of age), with hearing evaluation sought if there are symptoms from 8 yr of age into adulthood.

Hearing loss in Phex mutant mouse models

Despite the variable expression of auditory phenotypes and cochlear pathologies between different background genetic strains, the findings from the 4 animal studies collectively suggest ELH as the main cochlear pathology responsible for hearing impairment in Phex mutant mice. A wide range of methods were utilized in Phex mutant mice studies, with each providing unique insights into the pathogenesis of hearing loss in XLH. The findings demonstrate that hearing is affected from a very young stage of life in Hyp-Duk mice but that it only develops hearing loss postnatally.^23^ The background genetic strain may have a significant effect on the auditory phenotypes in Phex mutant mice, with different strains of Phex mutant mice demonstrating different hearing phenotypes.^14^^,^^23^

Histological analysis demonstrated prominent osteomalacia of auditory ossicles across Phex mutant mice regardless of background genetic strains. There was also evidence that functional degeneration may precede structural degeneration.^23^ Presence of ELH in hearing-impaired Hyp-Duk (BALB/cUrd) mice but absence in hearing-preserved Hyp-Duk (B6) mice highlights the crucial role of ELH in the pathogenesis of hearing loss in XLH and that the bone abnormality alone cannot cause hearing loss. Interestingly, in the analysis by Megerian et al.,^23^ ELH was seen without endolymphatic duct obstruction, suggesting that endolymphatic duct obstruction is not a prerequisite for the pathogenesis of ELH. The presence of otitis media in a small number of B6 Hyp-Duk mice also suggests that a susceptibility to otitis media may also aggravate hearing loss, something that may evidently also impact humans with XLH.

Effect of XLH treatments on hearing loss

Lower serum phosphate levels were associated with hearing loss in 2 1 human^14^ and 1 animal study.^22^ In the clinical study by Ivanovic-Zuvic et al.,^14^ there was no association between prevalence of hearing loss and other biochemical markers, including those that better indicate longer-term phosphate levels (such as alkaline phosphatase). Furthermore, the treatment exposure rate/dosage regimen, demographics, and imaging between patients with and without hearing loss also showed no significant differences. This suggests either limited benefits of conventional therapy in hearing loss or may be a reflection of the small sample size. The other reviewed studies presented no evidence that treatment of XLH impacted incidence or severity of hearing impairment. Studies looking at treated populations did not show significantly different prevalences of hearing loss than those with only a fraction of their patients on treatment.^14^^,^^15^

This is consistent with findings from Wick et al.,^22^ who looked at the effect of supplemental phosphate and calcitriol on Phex mutant mouse models. This study demonstrated that conventional XLH therapy may alleviate hypomineralisation, hence the conductive component of hearing loss, but does not necessarily improve the severity of ELH or the overall hearing. Together, these findings, while based on small studies with limited subjects, highlight the need to consider hearing assessments in those with XLH, regardless of therapy or severity of disease.

Wick et al.^22^ also demonstrated that rescue and prevention cohorts had significantly higher serum FGF-23 levels compared to Hyp-Duk control cohort, reflecting the expected limitation of conventional therapy in XLH. Direct FGF23 inhibition with burosumab may induce greater improvement in bone mineralization, potentially to the extent that affects the degree of ELH and overall hearing. Hence, this may be a potential target for future studies using Phex mutant mice.^22^ There were no clinical studies looking at the impact of burosumab on hearing loss. In addition, a recent article summarizing the opinion of a group of 8 expert clinicians on anticipated impacts of long-term burosumab reported a low level of agreement on the effect of burosumab on hearing loss in XLH.^28^ Future studies aiming to identify whether burosumab will reduce the risk of hearing loss in XLH should include regular monitoring of hearing with formal audiology in their long-term follow-up of patients.

Clinical implications for individuals with XLH and future directions

As summarized in Table 1, historical studies prior to 2000 have noted low-frequency hearing impairment in XLH patients with audiometry and tympanic electrocochleogram suggestive of cochlear hearing loss with ELH as the main pathology, similar to that in Ménière’s disease.^6^^,^^7^ Petrous bones showed generalized osteosclerosis and thickening with narrowing of internal auditory meatus at its mid-point, compared to control groups.^8^ Despite our 5 human studies not performing targeted studies such as electrocochleography necessary to assess for the presence of ELH, results show predominant sensorineural hearing loss with vertigo and tinnitus frequently co-reported in hearing-impaired XLH patients, resembling Ménière’s disease.

The exact pathophysiology of ELH in XLH is unclear. Previously proposed mechanisms involved morphological changes through petrous bone osteomalacia, disrupting physiologic endolymphatic turnover via affecting endolymph production in the stria vascularis and/or causing obstruction of endolymphatic duct or sac.^8^ Evidence suggests that bone abnormality alone is not the sole factor responsible for marked hearing impairment in XLH nor endolymphatic duct obstruction. Early treatment may be crucial to alleviate the hypomineralisation of the auditory ossicles observed in both our animal and human studies, but there is limited evidence regarding the efficacy of conventional therapy in altering the pathogenesis of ELH and hearing impairment in XLH. Currently, management of hearing impairment in XLH is limited to supportive measures such as hearing aids or cochlear implants, prevention of noise exposure and avoidance of ototoxic drugs.

In Ménière’s disease, which also presents with ELH and sensorineural hearing loss as in XLH, a defective periductal interstitial connective tissue network has been proposed as the key factor that leads to disrupted homeostasis of the endolymphatic system contributing to the pathogenesis of ELH.^29^^,^^30^ Management of Meniere’s disease is complex and multimodal, from dietary to pharmacological and surgical interventions. Several studies have investigated the role of corticosteroids in treating Meniere’s disease and ELH. The evidence for the efficacy of oral corticosteroids in improving hearing impairment remains weak, with some improvements noted in vertigo but not in hearing.^31^^,^^32^ Pantel et al.^21^ reported a 3-yr audiological follow-up of a 55-yr-old man with untreated XLH, who initially presented with fluctuating bilateral hearing loss. Audiometry confirmed symmetrical moderate to severe sensorineural hearing loss in both low and high-frequency ranges. After a 2-wk course of oral corticosteroids, the patient showed marked but temporary improvement in low to medium frequencies, lasting for 4 wk. Hearing thresholds continued to fluctuate until stabilizing spontaneously at 1 yr. Intratympanic corticosteroids, another modality used in Meniere’s disease after failed first-line noninvasive therapy, have shown mixed results regarding hearing improvement in ELH.^33–35^ The effectiveness of management options for Ménière’s disease in treating hearing impairment in XLH remains to be elucidated and warrants further research.

Lower serum phosphate levels were noted in both hearing-impaired XLH patients^14^ and hearing-impaired Phex^Hyp-Duk^/Y mice,^22^ however, whether phosphate levels can be a reliable biochemical marker to monitor or predict hearing impairment in XLH requires further research. In addition, the utility of serum phosphate is limited due to the fluctuation of levels throughout the day. In these studies, markers suggestive of chronic hypophosphatemia, such as elevated alkaline phosphatase (ALP) or radiological rickets, did not show any correlation with the rate of hearing loss. Hearing impairment is likely part of the natural history of XLH in adults and given the significant impact on quality of life, close follow-up and hearing monitoring are recommended.

The existing literature does not suggest that hearing impairment occurs more frequently in children with XLH than in the general population, and there is wide variability regarding the prevalence, pattern, and onset of hearing impairment in children with XLH. It is unclear whether long-term treatment with conventional therapy or burosumab from early childhood can prevent or delay the onset of hearing loss in XLH. Despite the onset of hearing impairment seen at very early stages of life in Phex mutant mice, hearing loss from cross-sectional cohort studies was reported in only 2 children of ages 5 and 6 yr across the included human studies, both with asymptomatic conductive hearing loss. Otitis media with effusion was also observed on physical examination, which is very common in otherwise healthy children of that age and known to cause conductive hearing loss. Hence, whether this is an independent disease from XLH or a progression to ELH due to XLH is unclear, and longitudinal follow-up of larger numbers of affected children would be beneficial for future studies to monitor the trajectories of otitis media in XLH. Consequently, routine hearing assessment may not be justified in asymptomatic children with XLH below the age of 5 yr, and for any symptomatic children with XLH, other otological causes should be thoroughly investigated and excluded in the first instance.

Conclusions

There is a clear association between sensorineural hearing loss and XLH in adulthood; however, the pattern of hearing loss in children with XLH varies widely across studies. Findings from Phex mice models are inconclusive with many confounding factors, but largely suggest ELH as the key pathology that contributes to hearing loss in XLH. Bone abnormality alone likely does not contribute to hearing impairment in XLH. There is limited evidence on the efficacy of conventional therapy and burosumab in the management of hearing impairment in XLH despite clear benefits in preserving the mineralization of the auditory ossicles and the optic capsule and remains to be determined in future studies.