Diabetes is a significant public health concern (International Diabetes Federation [IDF], 2021). In South Africa, an estimate of about 4.2 million adults (10.8%) between the ages of 20 and 79 years live with diabetes, with a projected proportion of undiagnosed cases of approximately 1922.2 (45.4%) (IDF, 2021). Without proactive intervention to prevent this disease, this trend is predicted to double by 2045 (IDF, 2021). With the increase in new diabetes cases, it is critical to explore the link between diabetes and hearing impairment.

Hearing impairment in diabetes may occur because of multisystem complications caused by the disease (Kibirige et al., 2019). Despite extensive literature confirming the association between diabetes and hearing loss, there is still some controversy surrounding this relationship (Akinpelu et al., 2017; Hlayisi et al., 2018; Sameli et al., 2017). This controversy is seen in the reported variation in diabetes-related hearing impairment prevalence worldwide. A study conducted in the United States (US) reported a prevalence rate of 13.1% (Kakarlapudi et al., 2003), while in Iran, a rate of 45% was reported (Mozaffari et al., 2010). The highest prevalence rates were observed in India and South Africa at 78.2% and 55.0%, respectively (Bhaskar et al., 2014; Hlayisi et al., 2018). There are various factors that could contribute to the aforementioned variation in hearing loss prevalence. Such factors may include diverse study designs, sample size composition, audiologic assessment methods used, hearing loss classification methods and presence of comorbidities (Hlayisi et al., 2018; Pillay et al., 2021; Sameli et al., 2017). It was noted that the majority of the studies used different threshold cutoff to determine the presence or absence of hearing loss. The use of pure-tone hearing threshold average (PTA) with different normative values of either 15 dB HL or 25 dB HL was evident in studies conducted by Thimmasettaiah and Shankar (2012) and Agarwal et al. (2013), whereas Hlayisi et al. (2018) used conventional pure-tone average (CPTA) and high-frequency pure-tone average (HPTA), thus contributing to the variation in hearing loss prevalence. While this difference in the hearing loss classification may result in variation in prevalence, co-morbidities such as hypertension may also be a contributing factor. It is important to note that diabetes and hypertension are inextricably linked diseases, and the prevalence of hypertension is reported to double in the presence of diabetes (Venugopal & Mohammed, 2014). Hypertension is known to cause damage in the microvasculature of the inner ear, similar to the damage seen in diabetes (Chao, 2004; Kakarlapudi et al., 2003). Therefore, both these conditions heighten the risk of acquiring hearing loss. This variation in the reporting of hearing loss prevalence is a potential reason why hearing impairment has not been considered a definite complication of diabetes despite the abundance of scientific evidence available.

Hearing impairment is a significant global public health issue, affecting approximately 430 million people (WHO, 2021). Low- and middle-income countries, such as South Africa, face the most substantial impact of hearing loss (WHO, 2021). Regardless of its characteristics, hearing impairment has been associated with adverse effects such as social isolation and depression, which ultimately impact quality of life (Nordvik et al., 2019). In type 2 diabetes, the commonly identified hearing loss is comparable to presbycusis and involves a more pronounced loss at higher frequencies (Akinpelu et al., 2014; Hlayisi et al., 2018). In some cases, the lower and mid-frequencies are also affected, and hearing loss has been reported to be bilateral and sensorineural (Bhaskar et al., 2014; Mozaffari et al., 2010; Ren et al., 2017). Although authors such as Agarwal et al. (2009), Krishnappa and Naseeruddin (2014) and Bhaskar et al. (2014) believe that the causes of hearing loss present in type 2 diabetes could be multifactorial, there appears to be more evidence pointing to microangiopathy of the inner ear as a potential contributor to auditory abnormalities in this patient population (Helzner & Contrera, 2016). Therefore, the permanent nature of hearing loss and its consequent effect on patient quality of life highlight the necessity of including audiological services in diabetes care.

Recent research reports on diabetes-related hearing impairment emphasise the importance of various audiological modalities for assessing and monitoring hearing in individuals with type 2 diabetes mellitus (DM). High frequency audiometry has been favoured as a method for early identification of hearing loss in patients with diabetes (Akinpelu et al., 2014; Kakarlapudi et al., 2003; Ozkurt et al., 2016) because of diabetes being associated with high-frequency hearing loss. This form of audiometry is valuable in detecting hearing loss at a subclinical phase (Valiente et al., 2014). Objective testing methods such as the Distortion Product Otoacoustic Emissions test (DPOAE) and the auditory brainstem response test (ABR) (Meena et al., 2016) have also shown value in detection of hearing impairment for patients with diabetes. The complexity of this metabolic disease, resulting in various forms of neuropathies in the body, including neural hearing loss (Feldman et al., 2019) further supports the need for a comprehensive test battery approach when assessing hearing loss in persons with diabetes. Also, the lack of universally acceptable protocol suggested for this population provides an impetus to profile hearing loss for persons living with type 2 diabetes.

Based on the available literature, it is evident that there is a high prevalence of type 2 DM in South Africa, which is known to be associated with hearing loss resulting from cochlear damage. Current studies in South Africa have been limited to peripheral auditory assessment (Hlayisi et al., 2018; Pillay et al., 2021). However, literature suggests a neurological impact associated with diabetes (Farmaki et al., 2020). Therefore, the study aimed to investigate the audiological profile of patients with type 2 diabetes in South Africa using a comprehensive audiological assessment that reflects the integrity of the entire auditory pathway.

The study found that 25.7% (n = 9) of our participants reported bilateral self-perceived hearing difficulties, which is higher than Soares et al. (2018), who indicated a prevalence of self-perceived hearing impairment of 6.69% in their diabetic population. This variation in self-perceived hearing impairment findings could be attributed to the fact that Soares et al. (2018) found that age, race, education, occupational exposure, smoking and alcohol consumption were positively associated with self-reported hearing impairment. In contrast, the current study could not account for the impact of such variables in the findings. There was a lower prevalence of tinnitus (11.4%) among participants with confirmed hearing loss compared to previous reports of 68.7% (Berner et al., 2017) and 45.1% (Pillay et al., 2021). A possible reason for the lower prevalence of tinnitus in the current study is that 51.4% (n = 18) of the participants had diabetes for less than 5 years. This could have possibly indicated reduced microvascular complications such as those associated with tinnitus.

In light of self-perceived audiological and otologic symptoms, there was a lower prevalence of dizziness (17.1%) compared to previous reports by Nemati et al. (2018) and Pillay et al. (2021), who reported higher percentages of 26.0% and 42.4%, respectively. However, caution should be exercised when interpreting these findings, as these studies could not distinguish dizziness resulting from inner ear defects or glucose alterations. A comprehensive vestibular assessment is necessary to make this distinction. Regrettably, this was beyond the scope of the current study.

The prevalence of hearing loss among individuals with type 2 diabetes varies across regions of the world (Bhaskar et al., 2014; Hlayisi et al., 2018; Kakarlapudi et al., 2003). The present study found the prevalence of hearing loss in adults with type 2 diabetes to be 31.4%, which is lower than that reported in the literature (Dhasmana et al., 2021; Hlayisi et al., 2018; Idugboe et al., 2018; Mozaffari et al., 2010; Pillay et al., 2021). However, the current findings are consistent with the prevalence rate of 21.6% reported in Nigeria by Adebola et al. (2016). It should be noted that the comparison of prevalence rates is complex, as it is influenced by various factors, such as the type of assessment protocols employed (diagnostics vs. screening), study sample compositions and normative data utilised. For example, Pillay et al. (2021) and Bashkar et al. (2014) used the pure tone average (PTA) to determine the prevalence of hearing loss, and Hlayisi et al. (2018) utilised low-frequency PTA and high-frequency PTA cutoffs, while the current study took all frequencies into account (125 Hz-8 kHz) when determining the degree of hearing loss. Regardless of this variation in the methods of establishing the prevalence of hearing loss across the literature, the present study still confirms the existence of hearing loss in type 2 diabetes and emphasises the importance of incorporating audiological assessments in the routine care of these patients.

Hearing loss can be categorised into three types: conductive, sensorineural or mixed (Alshuaib et al., 2015). Of 22 ears with hearing loss, 81.1% had sensorineural hearing loss. This finding is consistent with previous literature reports indicating that sensorineural hearing loss is the most common type of hearing loss observed in persons with type 2 diabetes (Hlayisi et al., 2018; Idugboe et al., 2018; Pillay et al., 2021; Samocha-Bonet et al., 2021). The pathophysiological basis of sensorineural hearing loss specific to DM may be because of microangiopathy of the inner ear, neuropathy of the cochlear nerve or a combination of both factors, as suggested in the literature (Kakarlapudi et al., 2003). Further support for this theory comes from the study conducted by Fukushima et al. (2005), who noted anatomic changes in the cochlear structures of insulin and non-insulin diabetic patients. Evidence from Makishima and Tanaka (1971), though dated centuries ago, still provides insightful knowledge regarding the auditory anatomic defects occurring in persons with diabetes. The authors identified atrophy of spiral ganglia in the basal to middle turns of the cochlea and demyelination of the VIII Cranial nerve myelin sheaths (Makishima & Tanaka, 1971), thus ultimately affecting the physiological aspect, which is hearing, potentially resulting in neural hearing loss. These findings highlight the multifactorial nature of hearing impairment in diabetes and the need for early detection and management of hearing loss in people with diabetes.

The current study observed slight hearing loss (16 dB – 25 dB) in 36.4% of the participants, consistent with the findings of Hlayisi et al. (2018), who reported a similar rate of 35.0% in their diabetic cohort (Hlayisi et al., 2018). Minimal or slight hearing loss is used to describe hearing thresholds within 16 dB – 25 dB (Kaderavek & Pakulski, 2002; Martin & Clark, 2003). Despite its subtle nature, slight hearing loss can negatively impact speech reception, discrimination and comprehension, as previously documented in the literature (Arlinger, 2003). Moreover, hearing loss has been associated with various adverse outcomes, including social isolation, depression and cognitive decline (Croll et al., 2021; Kim et al., 2023). Recently, there has been growing evidence linking hearing loss with dementia (Griffiths et al., 2020), highlighting the importance of early detection and management of hearing loss to minimise the negative consequences accompanying hearing loss.

In this study, the most common hearing loss configuration observed among individuals with diabetes was rising (36.0%) and sloping (31.8%). To our knowledge, there is a lack of studies that have specifically investigated and reported the configuration of hearing loss in diabetes (Al-Rubeaan et al., 2021). Understanding the specific configuration of hearing loss in diabetes may improve intervention outcomes, such as those requiring amplification.

In the literature, high-frequency involvement has been reported as a hallmark feature of diabetes-related hearing impairment (Meena et al., 2016; Ozkurt et al., 2016). The current study also found high-frequency hearing loss in 22.7% of cases. However, this prevalence rate is lower than those reported by Hlayisi et al. (2018) and Ren et al. (2017), who found high-frequency hearing loss in 55% to 72.2% of cases. While the previous studies of Hlayisi et al. (2018) and Ren et al. (2017) determined high-frequency loss by averaging frequencies of 4 kHz, 6 kHz and 8 kHz, the current study did not utilise frequency averages when determining the prevalence. The analysis took the frequency range of 125 Hz to 8 kHz into account when determining the degree of hearing loss, which aligns with the hearing loss classification scheme from the ASHA (1981), which may explain why the prevalence is perceived as low.

To further peruse hearing loss beyond the frequency range routinely tested in clinical practice, we employed EHF testing (> 8 kHz to 20 kHz), which is more sensitive to subclinical hearing loss than conventional audiometry (Valiente et al., 2014). The results showed that as the frequency increased, the hearing loss also increased, with most participants showing ‘no responses’ at frequencies of 16 kHz, 18 kHz and 20 kHz. These findings are consistent with those of Ozkurt et al. (2016) and Nemata et al. (2018), who observed increased hearing loss as the frequency increased. As seen in diabetes, the higher hearing frequencies are reported to be affected first before the standard hearing frequency range is affected (Das et al., 2018). Hearing loss in the higher frequencies (> 4000 Hz) can have a negative impact on the perception of certain speech sounds, like ‘f’, ‘s’ and ‘th’, including sounds perceived in nature and music (Paken et al., 2023). However, the detection of sounds between 5000 Hz and 9000 Hz is important for humans as it ensures proper acoustic perception as they ‘guarantee a good part of speech intelligibility by favouring consonant discrimination and speech recognition’ (Anastasio et al., 2012, p. 39). Therefore, this finding further favours EHF audiometry – a valuable measure in this patient population. Additionally, the results provide tangible evidence that EHF hearing thresholds could be considered a potential sensitive marker of hearing loss in type 2 diabetics.

This study utilised the SRT test to determine the validity of pure tone audiometry. The majority of the ears with hearing loss showed a good correlation between SRT and pure-tone average (PTA), which indicates reliable pure-tone results (Table 5). A correlation of less than 5 dB was considered good, a correlation between 6 dB and 9 dB was considered fair and a correlation greater than 10 dB was considered poor (Gelfand, 2001). Though this study found a good SRT-PTA correlation, Kiakojouri et al. (2014) reported worse SRT in their diabetic group compared to the control group. There seems to be no current evidence reporting SRT outcomes in diabetic patients.

Previous studies have investigated various patient characteristics associated with diabetes-related hearing impairment, including diabetes control status, duration and associated comorbidities (Al-Rubeaan et al., 2021; Hlayisi et al., 2018; Pillay et al., 2021). The present study examined the impact of hypertension on hearing loss in persons with type 2 diabetes. In hypertension, adverse synergistic effects with inner ear microvascular damage resembling damage caused by diabetes have been observed (Chang et al., 2011; Chao, 2004; Kakarlapudi et al., 2003). From a physiological point of view, the combined presence of hypertension and diabetes is believed to lead to damage in the cochlea, thus resulting in high-frequency sensorineural hearing loss (Duck et al., 1997). Recognising the risk profile can enhance the understanding of how these risk factors interplay in type 2 diabetes. Simultaneous exposure to multiple risk factors such as co-morbidities (i.e. hypertension) can exacerbate auditory dysfunction more than if each factor was considered separately, thus implying synergistic effect (Chao, 2004; Paken et al., 2021). This may be plausible in this study, as a statistically significant association between hypertension and hearing loss was observed in both the right ear (p = 0.006) and the left ear (p = 0.02).

In Nigeria, Babarinde et al. (2021) found a statistically significant association between hypertension and hearing loss (p < 0.010) in their study of 500 individuals, which included a control group. Hearing loss in hypertension was further reported to be mild in degree, sensorineural in nature, as well as increased with age, severity and duration of the disease. This study’s findings and previous research suggest that comorbidities such as hypertension, when present with type 2 DM, may increase the risk of acquiring hearing impairment, thus implying a synergistic effect.

Previous research has consistently reported that the duration of diabetes impacts the development of sensorineural hearing loss (Samocha-Bonet et al., 2021). However, the current study did not find a statistically significant association between diabetes duration and hearing impairment. The researchers postulate that the lack of statistical significance may be attributed to the observed relatively lower prevalence rate of hearing loss because of the small sample size. Further studies with a larger sample size and a wider range of diabetes durations are needed to confirm this hypothesis. The researchers speculate that this lack of association in the present study could also be attributed to a large proportion (88.9%) of the participants having diabetes for less than 5 years. It is also possible that a longer duration of diabetes may be necessary for the development of hearing impairment, which emphasises the need for glucose monitoring.

The auditory system depends on glucose for energy and functioning (Elibol & Baran, 2020). Monitoring glucose levels is crucial for persons with diabetes to prevent diabetes-related complications. Previous studies have established a link between poor glucose control and hearing loss in persons with diabetes (Al-Rubeaan et al., 2021; Elibol & Baran, 2020). However, the current study did not find a significant statistical association between glycaemic control status and hearing impairment, even though most participants (85.7%) had poor glycaemic values, indicating poor control of the disease. In contrast, Al-Rubeaan et al. (2021) reported that poor glycaemic control of > 8% was associated with hearing loss in their study. Although this study could not establish a statistically significant relationship, there is still abundant research showing the influence of poor glycaemic control on the auditory acuity of diabetes patients (Ebol & Baran, 2020; Nagahama et al., 2018; Srinivas et al., 2016). Audiologists are urged to emphasise the importance of glucose monitoring in their counselling or education with diabetes patients to prevent the possible occurrence of hearing impairment or worsening of hearing impairment when present.

There is scientific evidence suggesting that patients with diabetes may experience damage in the outer hair cells (OHCs) of the cochlea (Fukushima et al., 2005; Hlayisi et al., 2018). It is postulated that the dysfunction of OHCs in diabetic patients could be attributed to disruptions in glucose metabolism (Kumar et al., 2021). Glucose is recognised as a potential energy source for the cochlea (Kumar et al., 2021). Therefore, any disturbances in glucose metabolism could potentially impact the proper functioning of OHCs. The present study revealed that most participants had absent DPOAE in the frequency range of 6 kHz – 8 kHz bilaterally, suggesting that diabetes impacts the higher frequencies more. The loss of OHCs within the frequencies mentioned above could imply that the impact is more on the region of basal turns of the cochlea (Fukushima et al., 2005). Consistent with the current findings of OHCs dysfunction are Ferreira et al. (2016) and Hlayisi et al. (2018), who reported a high prevalence of absent OAEs in diabetic patients (Hlayisi et al., 2018; Selvarajah et al., 2011). These findings emphasise the importance of incorporating cochlear monitoring using measures such as DPOAE testing and that DPOAEs are sensitive in detecting high-frequency hearing loss in this patient population.

One common complication in individuals affected by diabetes is peripheral neuropathy (Feldman et al., 2019). Other studies have reported the presence of central neuropathy in diabetic patients (Selvarajah et al., 2011). The current study utilised ABR testing to assess the integrity of the auditory neural system. The findings reveal that most participants exhibited delayed absolute latency of wave III in both the right ear (n = 24, 69%) and the left ear (n = 18, 51%), which is consistent with the findings reported by Siddiqi et al. (2013) who utilised a neurological protocol in their testing. The high proportion of delayed absolute latency of wave III observed in this study suggests a possible lesion in the superior olivary complex (SOC). The SOC is an essential site of auditory stimuli convergence from both ears and plays a significant role in sound localisation (Grothe et al., 2010). Therefore, the observed abnormalities in wave III latency may affect the perception and processing of sound localisation cues in individuals with diabetes-related hearing impairment. This finding supports the assertion that ABR is sensitive to subtle auditory neural changes in diabetes. To the researcher’s knowledge, currently in the South African published literature regarding diabetes-related hearing impairment, the current study has pioneered the comprehensive audiological evaluation achieved by incorporating ABR in the assessment of diabetic patients. It has provided insight into the auditory neural integrity of diabetic patients. Although the findings advocate for the inclusion of ABR in audiologic test batteries for diabetes, this measure may not be economical for low- and middle-income countries because of the costs associated with procuring and maintaining such equipment.

The intricate relationship between type 2 DM and its associated complications, including hearing loss, emphasises the urgency for a holistic, multidisciplinary approach in managing this complex condition. Collaboration among healthcare professionals such as audiologists, primary healthcare nurses, endocrinologists, physicians and health promoters is essential to provide comprehensive care addressing both metabolic and audiological concerns. Audiologists play a crucial role in mitigating the adverse effects of hearing impairment in diabetes. Their involvement spans prevention, screening, diagnosis and management of auditory disorders. Employing an audiologic test battery approach guided by the cross-check principle enables early identification of auditory alterations, crucial for timely intervention. Additionally, proactive engagement in diabetes health promotion campaigns increases awareness among individuals with diabetes, influencing self-help seeking behaviour.

Diabetes and hearing loss are significant public health concerns, which necessitates attention from policymakers and health professionals. Knowledge generated from this study should inform policymakers in crafting service delivery models tailored for South Africa’s diabetic population. Strong advocacy for the inclusion of audiological services and adequate resources becomes imperative, promoting health equity and accessibility. By acknowledging the complex connection between diabetes and hearing impairment across policy and clinical realms, individuals with diabetes in South Africa stand to benefit immensely. Recognition and integration of audiological care within diabetes management promise an enhanced quality of life, ensuring timely identification, diagnosis and intervention of auditory disorders.