Frequency of Audiology Testing Among Individuals with Osteogenesis Imperfecta and Suggestions for Improving Audiogram Participation

Abstract

Background:

Osteogenesis imperfecta (OI), characterized by brittle bones and frequent fractures, often manifests with earlier onset hearing loss compared with the general population.

Purpose:

This study aims to assess the frequency of audiology testing in OI individuals and evaluate a portable audiometry device to enhance audiogram participation.

Research Design:

This is a prospective observational study.

Study Sample:

Ninety-seven participants were prospectively enrolled.

Data Collection and Analysis:

Participants underwent a one-time audiology test using SHOEBOX Audiometry Pro with RadioEar DD450 circumaural headphones (Clearwater Clinical Limited, Ottawa, ON, Canada). Hearing loss was defined as having a pure tone threshold (PTT) of ≥25 dB at one or more tested frequencies.

Results:

Most participants (54/97) reported undergoing professional audiology testing less often than once every 2 years. The most common reported reason for infrequent testing was because patients did not perceive issues with their hearing, even if hearing loss was subsequently found during screening. Seventy-one percent (69/97) of participants had hearing loss (PTT ≥ 25 dB) at one or more frequencies.

Conclusions:

Using a portable audiometry device in OI clinics could facilitate early hearing loss detection and improve follow-up care, enhancing quality of life.

INTRODUCTION

Osteogenesis imperfecta (OI) is a group of connective tissue disorders characterized by brittle bones and a tendency to fracture (Rauch and Glorieux, 2004). Although genetically heterogeneous, more than 90 percent of people with OI have a dominant variant in one of the type I procollagen genes, COL1α1 and COL1α2, resulting in either altered quality or decreased quantity of type I collagen (Van Dijk and Sillence, 2014). Clinically, OI is grouped into several types, with type I being the mildest, type IV a moderate form of OI, and type III the most severe, survivable form of OI (Rauch and Glorieux, 2004; Van Dijk and Sillence, 2014).

Although fractures are the hallmark of OI, hearing loss is a critical, less visible extraskeletal manifestation that can compromise safety and significantly impact quality of life (World Health Organization, 2021; Borre et al, 2023). Focus groups of individuals with OI indicated many difficulties attributed to hearing loss, including loss of conversation ease, impact to education or career, and loss of social participation (Goderie et al, 2023). Therefore, participation in audiology testing can potentially improve early detection and treatment of hearing loss in affected individuals, thus improving safety and well-being.

Hearing loss generally begins at an earlier age in individuals with OI than in the general population, first presenting in the second to third decade of life and progressing into adulthood (Riedner et al, 1980; Sillence, 1981; Stewart and O’Reilly, 1989; Carré et al, 2019; Machol et al, 2020). The prevalence of hearing loss in the OI population varies widely between reports, from 2 to 94.1 percent, depending on methodologies and how “hearing loss” is defined (Carré et al, 2019). The reported prevalence of hearing loss in the OI population can also differ based on participant selection criteria, because hearing loss is thought to be most common in patients with OI type I and least frequent in those with OI type IV (Paterson et al, 2001; Ugarteburu et al, 2022). For instance, a study by Garretsen et al (1997) found the incidence of hearing loss to be 95 percent when looking only at individuals with type I OI who were older than 30 years, after controlling for age as a significant factor in their analysis. In contrast, Machol et al (2020) included audiometry results from 312 individuals with OI and found the overall prevalence of hearing loss in OI to be 28 percent and increased with age in type I, but not in types III and IV. Current literature supports that hearing loss in OI develops bilaterally initially as a conductive loss, but often progresses to a mixed conductive and sensorineural loss (Riedner et al, 1980; Stewart and O’Reilly, 1989; Swinnen et al, 2011). Just as other manifestations of OI are expressed as a broad range of phenotypes, it can be concluded that OI-related hearing loss also tends to be diverse in terms of type, progression, and severity.

Because it is widely reported that hearing loss is a significant clinical feature of OI (Stewart and O’Reilly, 1989; Garretsen et al, 1997; Swinnen et al, 2011), many studies have indicated the importance of standardized screening measures in this population. This is particularly true for children, whose hearing loss may impact communicative abilities and language development (Imani et al, 2003; Goderie et al, 2023). A 2023 collaboration of OI experts recommended at least annual audiological testing for OI patients with confirmed hearing loss and every 5 years for those without (Goderie et al, 2023). However, to our knowledge, there have been no investigations into the proportion of the OI population that routinely undergoes audiology testing as part of their standard of care. In addition, awareness of hearing loss and perception of audiology testing among individuals with OI remain unknown.

This study evaluates the frequency of professional audiology testing among individuals with OI, as well as patient-reported factors that influence the regularity (or lack thereof) of such testing. Based on clinical experience, we hypothesize that the majority of individuals with OI undergo professional audiology examinations less frequently than once per year even if they have hearing loss. Therefore, we will evaluate the prevalence of hearing loss among study participants and review hearing loss screening measures employed within our clinic as potential means to increase detection and monitoring of hearing loss in the OI population.

PATIENTS AND METHODS

All work was performed at Hospital for Special Surgery (New York, NY). This study was performed in accordance with the ethical standards in the 1964 Declaration of Helsinki and regulations of the Health Insurance Portability and Accountability Act. Institutional review board approval was obtained at Hospital for Special Surgery, and this study was conducted with the understanding and full informed consent of the subjects.

All individuals 6 years and older with a clinical and/or genetic diagnosis of OI were eligible for inclusion. Participants were enrolled prospectively during routine visits to the Kathryn O. and Alan C. Greenberg Center for Skeletal Dysplasias at Hospital for Special Surgery from 2019 to 2023 or while attending the 2019 Osteogenesis Imperfecta Foundation (OIF) regional conferences in Wilmington, Delaware and Atlanta, Georgia. In addition to undergoing hearing loss screening using a portable audiometer, participants were interviewed regarding how often they obtain professional audiology testing at an audiological center and the reasons behind their visit patterns.

Audiograms were collected using the SHOEBOX Audiometry Pro with an iPad for data collection and RadioEar DD450 circumaural headphones (Clearwater Clinical Limited). The SHOEBOX Audiometry Pro Edition is a clinically validated tool for use as a diagnostic audiometer, even when used outside of an audiometric testing booth (Yeung et al, 2013; Saliba et al, 2017; Bastianelli et al, 2019; Thai-Van et al, 2023). In adults with at least moderate hearing loss (pure tone average > 40 dB), sensitivity and specificity levels of the SHOEBOX were 100 percent and 95.9 percent, respectively (Saliba et al, 2017). In a cohort of children, the SHOEBOX also demonstrated strong sensitivity (93.3 percent) and specificity (94.5 percent) (Yeung et al, 2013). No statistically significant difference between repeat assessments has been reported (Yeung et al, 2013; Saliba et al, 2017; Bastianelli et al, 2019).

The SHOEBOX portable audiometer has advanced clinical capabilities for pure tone audiometry, masking, bone conduction, speech reception threshold (SRT), and speech recognition. Sound generation ranges from −10 to 120 dB HL (decibels hearing level), with speech recognition conducted at a comfortable listening level of 55 dB. It also has built-in technology to test background noise level and alerts the test administrator if background noise is too high for appropriate testing. The audiometer is calibrated annually in compliance with ANSI S3.6 standards.

Hearing loss in this study was defined as having a pure tone threshold (PTT) of ≥25 dB HL at one or more tested frequencies: 250, 500, 1000, 2000, 3000, 4000, or 8000 Hz. Previous studies have used this method of defining hearing loss, although with cutoff levels ranging from 15 to 25 dB (Pedersen, 1984; Waissbluth et al, 2020). We used this screening criterion to pick up early signs of hearing loss and sensitively monitor OI patients longitudinally at annual follow-up visits, given that they may be predisposed to develop broader hearing loss in the future. Patients found to have a PTT ≥ 25 dB were referred to establish care with an ear, nose, and throat (ENT) specialist and/or audiologist for professional audiometric evaluation and treatment. The number of patients with a pure tone average (PTA) ≥ 25 dB, a more commonly used assessment to define mild hearing loss in the general population, is also reported (World Health Organization, 2020). Finally, the number of patients with SRT ≥ 25 dB was reported to evaluate the potential effect of hearing loss on daily social interactions (Olusanya et al, 2019).

Statistical analyses consisted of descriptive statistics used to describe the study population. All analyses were conducted using SPSS 29 for Windows (SPSS Statistics; IBM, Armonk, NY).

RESULTS

This study included 97 patients with OI (66 females, 30 males, 1 transgender man), aged 6–79 years, with an average age of 33.6 years. The majority of the cohort (71.1 percent) were adults (69 adults between 22 and 79 years old), and 28.9 percent of the cohort was pediatric (28 participants between 6 and 21 years old) (Figure 1). Type I OI was the most common diagnosis (n = 41, 42.2 percent), followed by type IV (n = 27, 27.8 percent), type III (n = 16, 16.5 percent), type V (n = 2), and type IX (n = 1) (Figure 2). Ten individuals (10.3 percent of the cohort) had a clinical diagnosis of OI, but because of the lack of genetic testing results, their type of OI was not molecularly confirmed.

Figure 1.

Distribution of audiograms across age groups (n = 97). yo = years old.

Figure 2.

Distribution of osteogenesis imperfecta (OI) types within the study population (n = 97).

Hearing loss (PTT ≥ 25 dB at one or more frequencies) was detected in 69 (71.1 percent) individuals, of whom 50 individuals had bilateral hearing loss (72.5 percent) and 19 individuals had unilateral hearing loss (27.5 percent). The average age of individuals with hearing loss at one or more frequencies was 38.4 ± 19.8 years, whereas the average age of those without hearing loss was 21.7 ± 11.7 years. Ninety-two of the 97 individuals underwent both speech recognition and SRT testing. Results showed that 40 (43.4 percent) individuals had SRT ≥ 25 dB in one or both ears, 15 of whom had a speech recognition score of less than 80 percent words correct in one or both ears. Of the 69 individuals with hearing loss at one or more frequencies, 18 (26.1 percent) were males and 51 (73.9 percent) were females. There was no significant association between sex and the prevalence of hearing loss in the OI population, as noted with a Pearson’s χ² test (α = 0.05, p = 0.097). Finally, 40/97 (41.2 percent) individuals had a left or right PTA ≥ 25 dB, the standard indication of mild hearing loss.

More than half of the total study cohort (55/97) reported undergoing hearing tests less often than once every 2 years (Figure 3). Of the 69 individuals with hearing loss, 47 (68.1 percent) indicated that they undergo follow-up audiological testing less often than the recommended regularity of once per year (10 reported biennially and 37 reported at intervals exceeding 2 years). Frequency of audiology testing did not appear to differ by sex, with 57 percent of females and 60 percent of males reporting that they received audiology testing less frequently than once every 2 years. In addition, type of OI did not appear to influence the frequency of audiology testing, because 53 percent of individuals with type I OI, 56 percent of those with type III OI, and 64 percent of those with type IV OI reported obtaining audiology testing less frequently than once every 2 years. When evaluated by age group, 11/25 (44 percent) children (age < 18 years) and 45/72 (62.5 percent) adults (age ≥18 years) had testing less frequently than once every 2 years.

Figure 3.

Frequency of audiology testing (n = 95).

If participants reported testing their hearing less often than once per year, they were asked the main reason for infrequent follow-up. Reasons included prioritizing management of other symptoms of OI, financial or insurance barriers to obtaining audiology examinations, and transportation issues among others, but the most common reason by far was patients did not perceive issues with their hearing (Table 1).

Table 1.

Patient-Reported Reasons for Infrequent Testing Among Participants Who Reported Undergoing Audiology Testing Less Frequently than Once per Year (n = 69)

Reason for Infrequent Testing	No. of Participants (Percent)
No issues with hearing	37 (54)
Not priority compared with other symptoms	13 (19)
Never been referred to an audiologist	6 (9)
Financial/insurance barriers	5 (7)
Other	3 (4)
No time	3 (4)
Transportation issues	1 (1.5)
Unknown	1 (1.5)
Total	69

Just before the hearing examination, participants were asked whether they thought they had hearing loss. Of the 97 individuals, 43 expressed having issues with hearing at the time of testing, whereas 50 did not recognize problems with hearing (Figure 4). Four participants noted that they were unsure about having hearing loss. Notably, of those 50 individuals who self-reported no hearing loss, 26 did in fact have PTT ≥ 25 dB at one or more frequencies (Figure 4). Of the 48 participants who underwent SRT testing and self-reported no hearing loss, six had SRT ≥ 25 dB (Figure 4). Furthermore, 18 of the 26 individuals found to have hearing loss received hearing examinations less frequently than once every 2 years, 13 of whom cited “no issues with hearing” as the reason for infrequent testing (Figure 4). We found that 10 of those 13 individuals had an SRT < 25 dB, which may explain their lack of perceived hearing loss.

Figure 4.

Flowchart of hearing test results with subgrouping by self-reported hearing loss status (n = 97). *Not every participant completed SRT testing.

DISCUSSION

Despite early-onset hearing loss being one of the known extraskeletal manifestations of OI, more than half of the participants in this study expressed that they perceived no issues with their hearing. Similarly, most participants reported going to a professional audiologist for hearing testing less frequently than once every 2 years. Although most of these patients had mild hearing loss with SRT < 25, several patients had greater degrees of hearing loss. Although mild hearing loss may not be recognized by a patient, early detection offers the opportunity to consider intervention and/or monitoring for progression. Ten of the 13 individuals who reported “no issues with hearing” as the reason for their infrequent testing had an SRT < 25 dB, indicating that their hearing loss was mild and may not have noticeably impacted their daily communication. However, many patients in our series had more severe levels of hearing loss despite the absence of a perceived impairment. Hearing loss has been recognized to have significant health consequences. Detection and monitoring of early signs of hearing loss are important to health maintenance, particularly in at-risk populations such as OI population.

Interestingly, this trend has similarly been reported in larger longitudinal studies of OI, including the Linked Clinical Research Center’s (LCRC’s) multicenter study of individuals with OI (Patel et al, 2015). Seventeen percent (5/29) of LCRC participants with documented hearing loss by audiometry did not perceive they had hearing loss, further demonstrating the importance of incorporating audiological testing into the standard treatment protocol for OI. Previous literature also supports the poor correlation between perceived hearing loss and measured hearing loss in other populations (Nondahl et al, 1998; Sindhusake et al, 2001; Kerr et al, 2003). Nondahl et al (1998) collected audiometric results and three different measures of perceived hearing loss in older adults and found that the question “do you feel you have a hearing loss?” had a sensitivity and specificity of only 71 percent when compared with pure tone audiometry. Kerr et al (2003) studied a group of laborers and farmers and noted that although a high percentage of both groups exhibited hearing loss, they found that perceived hearing loss based on a self-reported measure was not a strong indicator of actual hearing loss based on audiometry results. These researchers concluded that an audiometric screening tool should be incorporated into a complete hearing loss prevention program in industries with high levels of noise exposure. However, to the best of our knowledge, there have been no reports on whether individual perception of hearing loss aligns with actual hearing test results in patients with other genetic conditions involving early hearing loss, such as other skeletal dysplasias.

Care for patients with OI at our specialty orthopedic institution is coordinated through a dedicated, multidisciplinary center; a detailed description of our skeletal dysplasia center’s multidisciplinary process has been previously published (Carter et al, 2008). Portable, automated audiometers are a valuable tool in the OI clinic setting because they allow for the possibility of routine pure tone audiometry and speech testing, creating an easy, user-friendly way to screen for hearing loss.

Given the high prevalence (71.1 percent) of hearing loss in our OI cohort detected with the portable SHOEBOX Pro audiometer, we were able to make many referrals for professional audiological evaluation and thereby provide earlier detection and treatment for hearing loss in those individuals. Implementing this screening tool in OI clinics can not only aid in capturing hearing loss that might potentially go undiagnosed but also provide audiology referrals for patients who may otherwise not be motivated to see an audiologist regularly. From Table 1, we noted that the most common patient-reported reasons for following up with an audiologist less than once per year are that they have no issues with their hearing (37/69, 54 percent) or that it is not a priority compared with other symptoms (13/67, 19 percent). Providing all individuals with OI a standard-of-care SHOEBOX audiometry assessment, regardless of whether they perceive hearing loss, may alleviate some of these reported reasons for infrequent audiologist visits.

Altogether, this study is not without its limitations. First, our study population comprises people who attended OIF regional conferences and those followed by our multidisciplinary center. This cohort may represent a more health-conscious group along with those who may have experienced OI-related health problems significant enough for them to seek help or resources, thus creating a potential sampling bias. However, the low rate of follow-up with audiologists paired with the lack of realization regarding their own hearing loss in this already health-conscious group of individuals suggests that these numbers may be even more exaggerated when looking at the overall OI population, who may be less engaged in their care and therefore even less attentive to the possibility of developing hearing loss. Thus, the fact that our study cohort may consist of those people who are more interested in following up at the clinic only serves to increase the weight of these findings and emphasize the need for hearing screening tools to be incorporated into the OI standard of care. By offering hearing screening to people with OI who seek care at our orthopedic hospital, multidisciplinary care centers such as ours can help coordinate referrals to facilitate better care and management of the extraskeletal features that may impact function and quality of life for this unique patient population.

Another limitation of this study is that we did not determine how many patients who were referred to a professional audiologist after visiting our clinic actually pursued hearing testing. It is important to consider that even when individuals with OI are aware of their hearing loss, there may be other socioeconomic barriers that prevent these patients from receiving proper treatment, including stigma surrounding wearing hearing aids and challenges with insurance. Because insurance companies do not typically cover hearing aids for patients with OI once they reach adulthood, it becomes more difficult for low-income patients to afford hearing aids. In the future, it would be interesting to conduct a follow-up study assessing patient adherence to professional audiology referrals after a patient discovers that they have some level of hearing loss detected through the initial screening tests.

Overall, this work implies that people with OI may show signs indicative of early hearing loss even before developing SRT ≥ 25 dB. These findings highlight the value of using portable audiometry in multidisciplinary OI clinics. This tool not only facilitates early detection of hearing loss but also raises patient awareness about potential hearing problems. In addition, it ensures timely referrals to audiology and ENT specialists. Together, these measures can improve earlier detection of hearing loss, leading to earlier intervention and potentially enhanced quality of life for individuals with OI.

Abbreviations

dB HL

decibels hearing level

ENT

ear, nose, and throat

LCRC

linked clinical research center

OI

osteogenesis imperfecta

OIF

Osteogenesis Imperfecta Foundation

PTA

pure tone average

PTT

pure tone threshold

SRT

speech reception threshold