Diabetes mellitus has become one of the most pressing non-communicable disease challenges of the present century, and India, frequently described as the diabetes capital of the world, carries a disproportionate share of the global burden. [1,2] Successive Indian population surveys have documented the scale of the problem. The National Urban Diabetes Survey reported a prevalence of roughly 12% among urban adults, together with a high prevalence of impaired glucose tolerance, [3] and the Chennai Urban Rural Epidemiology Study showed a steady secular rise in prevalence in urban South India. [4] The more recent ICMR–INDIAB study demonstrated marked inter-state variation, with prevalence ranging from roughly 5% to 14% across the first regions surveyed, and a large reservoir of prediabetes that will feed future T2DM cases. [5] Because Asian Indians tend to develop T2DM at a younger age and at lower body mass index than Western populations, patients are exposed to hyperglycemia for longer periods and are therefore at risk of chronic complications for many years. [1,2]

The classical complications of diabetes (retinopathy, nephropathy, neuropathy and macrovascular disease) are the products of chronic hyperglycemia acting on small and large blood vessels and on peripheral nerves. The inner ear is rich in capillary networks and in specialised sensory and neural elements, which makes it theoretically vulnerable to the same pathological processes. An association between diabetes and sensorineural hearing impairment was reported as early as the middle of the last century, when Jorgensen and Buch described impaired inner-ear function in diabetic subjects, [6] and subsequent clinical series documented hearing thresholds that were poorer than in non-diabetic individuals, most often at high frequencies. [7,8] An early Indian study from Vellore likewise found significantly poorer hearing thresholds in patients with diabetes. [9]

Over the last two decades, larger epidemiological and pooled analyses have strengthened this evidence. A meta-analysis of observational studies found that diabetes was associated with roughly a twofold higher risk of hearing impairment in adults, [10] and a systematic review restricted to T2DM reached a similar conclusion. [11] Audiometric data from the United States National Health and Nutrition Examination Survey showed that hearing impairment was about twice as common among adults with diabetes as among those without, even after accounting for age and other confounders. [12] Histopathological studies of human temporal bones have shown thickening of the capillary walls of the stria vascularis and loss of outer hair cells in T2DM, [13] findings that parallel the microangiopathic changes described in earlier work. [14] A neuropathic component involving the cochlear nerve has also been proposed. [15]

Hearing loss is a major but under-recognised public health problem in India, where the burden of disabling deafness is high and screening services are limited. [16] In an ageing, diabetic population, unaddressed hearing loss can lead to communication difficulty, social withdrawal, reduced adherence to counselling and a poorer quality of life. Indian audiological studies in T2DM remain relatively few, [9,17] and the relationship between the degree of hearing loss and indices of glycemic status such as HbA1c, which reflects average glycemia over the preceding two to three months, has not been consistently demonstrated. Detailed pure tone audiometric characterisation, including the frequency pattern, severity, type and configuration of the loss, may help clinicians recognise at-risk patients early.

The present study was therefore undertaken to (i) describe the pure tone audiometric profile of patients with T2DM in comparison with non-diabetic controls, and (ii) examine the association of hearing thresholds with glycemic status, duration of diabetes and diabetic complications.

Study design and setting

This was a hospital-based, cross-sectional, comparative study conducted in the Departments of Otorhinolaryngology of a tertiary care teaching hospital in India over a period of 12 months (January 2015 to December 2015). Approval of the Institutional Ethics Committee was obtained before the commencement of the study, and written informed consent was taken from all participants. The study was conducted in accordance with the Declaration of Helsinki.

Study population

Patients with T2DM attending the diabetes clinic and medicine outpatient department were recruited consecutively. T2DM was diagnosed according to the American Diabetes Association criteria (fasting plasma glucose ≥ 126 mg/dL, 2-hour post-load glucose ≥ 200 mg/dL, HbA1c ≥ 6.5%, or a documented history of treatment with antidiabetic drugs). [18] Healthy, non-diabetic volunteers (hospital staff, attendants of patients and persons attending the general health check-up clinic), frequency-matched for age and sex, formed the control group, and normal glycemia was confirmed by FBS < 100 mg/dL and HbA1c < 5.7%.

Inclusion criteria

Age 30–70 years; established T2DM of at least one year’s duration (cases); normal otoscopic findings with an intact tympanic membrane; and willingness to participate.

Exclusion criteria

Type 1 diabetes or secondary diabetes; history of occupational or recreational noise exposure; use of ototoxic drugs (aminoglycosides, loop diuretics, platinum-based agents, high-dose salicylates); previous ear surgery, head injury or chronic suppurative otitis media; family history of hearing loss; known Meniere’s disease, otosclerosis or retrocochlear pathology; impacted cerumen not resolved after removal; estimated glomerular filtration rate below 30 mL/min/1.73 m²; and inability to cooperate with audiometric testing.

Sample size

The minimum sample size was calculated for comparing two proportions, assuming a hearing-loss prevalence of 60% in T2DM and 30% in controls, a two-sided significance level of 5% and a power of 80%; this gave about 42 participants per group. To permit stratified analyses by age, HbA1c category and duration of diabetes, 100 participants were enrolled in each group (total n = 200).

Clinical and biochemical evaluation

A structured proforma recorded demographic data, duration and treatment of diabetes, history of hypertension and otological symptoms. Height and weight were measured and the body mass index (BMI) was calculated. Blood pressure was recorded after five minutes of rest; hypertension was defined as a systolic pressure ≥ 140 mmHg, diastolic pressure ≥ 90 mmHg, or the use of antihypertensive medication. Peripheral neuropathy was assessed with a 10-g monofilament and a 128-Hz tuning fork, retinopathy by dilated fundus examination, and nephropathy by a spot urine albumin–creatinine ratio (≥ 30 mg/g). Venous blood was collected after an overnight fast. Fasting blood sugar was estimated by the glucose oxidase–peroxidase method and HbA1c by high-performance liquid chromatography. Patients with T2DM were grouped by HbA1c into good control (< 7.0%), fair control (7.0–8.9%) and poor control (≥ 9.0%).

Audiological evaluation

Every participant underwent otoscopic examination and tympanometry (226-Hz probe tone) to exclude middle-ear pathology. Pure tone audiometry was performed by the same audiologist, who was blinded to the glycemic status of the participants, using a calibrated diagnostic audiometer with supra-aural headphones and a bone vibrator in a sound-treated room with ambient noise within permissible limits. Air-conduction thresholds were measured at 250, 500, 1000, 2000, 4000 and 8000 Hz and bone-conduction thresholds at 500, 1000, 2000 and 4000 Hz, using the modified Hughson–Westlake ascending technique, [19] with testing procedures and masking according to ISO 8253-1. [20] The better-hearing ear was tested first.

Definitions and classification

The pure tone average (PTA) was the mean air-conduction threshold at 0.5, 1, 2 and 4 kHz. Hearing loss was defined as a PTA greater than 25 dB HL in one or both ears. The degree of loss in the worse ear was graded as mild (26–40 dB HL), moderate (41–55), moderately severe (56–70), severe (71–90) or profound (> 90) according to Clark. [21] Loss was classified as sensorineural (air–bone gap ≤ 10 dB), conductive or mixed. Audiogram configuration was described as sloping (high-frequency), flat or other (rising, notched or irregular).

Statistical analysis

Data were entered in Microsoft Excel and analysed with SPSS software (version 26). Continuous variables are presented as mean ± standard deviation (SD) and categorical variables as number (percentage). Group comparisons used the independent-samples t-test and the chi-square test (Fisher’s exact test where expected counts were below five). Differences among the three HbA1c categories were assessed by one-way ANOVA with Bonferroni post hoc comparison. Pearson’s correlation coefficient (r) was used to study associations of PTA with continuous variables. Multivariable logistic regression, restricted to patients with T2DM, identified independent predictors of hearing loss and is reported as adjusted odds ratios (aOR) with 95% confidence intervals (CI). A two-sided p value < 0.05 was considered statistically significant.

Baseline characteristics

A total of 200 participants (100 with T2DM and 100 controls) completed the evaluation. The two groups were comparable with respect to age, sex distribution and BMI (Table 1). Hypertension was more frequent among patients with T2DM (38.0% vs 19.0%, p = 0.003). As expected, FBS and HbA1c were significantly higher in the T2DM group. The mean duration of diabetes was 8.6 ± 5.2 years (range 1–24 years); 64 patients were on oral hypoglycemic agents alone and 36 on insulin-containing regimens. By HbA1c, 24 patients (24.0%) had good, 38 (38.0%) fair and 38 (38.0%) poor glycemic control.

Table 1. Baseline characteristics of the study participants

SD, standard deviation; BMI, body mass index; OHA, oral hypoglycemic agents; FBS, fasting blood sugar; HbA1c, glycated haemoglobin.

Prevalence of hearing loss and pure tone average

Hearing loss (PTA > 25 dB HL in either ear) was detected in 62 of 100 patients with T2DM (62.0%) compared with 28 of 100 controls (28.0%) (Table 2). The odds of hearing loss were more than four times higher in patients with T2DM (OR 4.20; 95% CI 2.31–7.61). Bilateral loss was about three times as common among patients with T2DM (49.0% vs 17.0%), whereas unilateral loss did not differ significantly. The mean PTA was significantly poorer in the T2DM group in both ears: 26.8 ± 11.9 dB HL versus 18.0 ± 7.6 dB HL in the right ear, and 27.8 ± 12.3 dB HL versus 18.6 ± 7.8 dB HL in the left ear (both p < 0.001).

Table 2. Prevalence of hearing loss and pure tone average in patients with T2DM and controls

PTA, pure tone average at 0.5, 1, 2 and 4 kHz; values are mean ± SD unless stated; CI, confidence interval.

Frequency-wise hearing thresholds

Mean air-conduction thresholds were higher in patients with T2DM at every test frequency in both ears (Table 3 and Figure 1). The difference widened progressively with frequency, from about 6 dB at 250 Hz to about 12 dB at 4000 and 8000 Hz in the right ear, producing a gently sloping, high-frequency-predominant audiometric profile. The steepest rise in thresholds occurred between 2000 and 4000 Hz in both groups, but the slope was greater in T2DM patients.

Table 3. Mean air-conduction thresholds (dB HL, mean ± SD) by frequency

Independent-samples t-test between T2DM and control groups at each frequency.

Degree of hearing loss

Of the 62 patients with T2DM who had hearing loss, 38 had mild, 17 moderate, 6 moderately severe and 1 severe loss in the worse ear; no participant had profound loss (Table 4, Figure 2). Among controls, the 28 affected individuals were almost all in the mild category (24 mild, 4 moderate). The distribution of severity differed significantly between the groups (p < 0.001), with a shift towards greater severity in T2DM.

Table 4. Degree of hearing loss in the worse ear (Clark’s classification)

Chi-square test for the distribution across categories (Fisher’s exact test for sparse cells).

Type, laterality and configuration of hearing loss

Sensorineural loss predominated in both groups but accounted for a larger share in T2DM (87.1%) than in controls (71.4%) (Table 5). Conductive loss, which was due to middle-ear effusion and cerumen-related causes in controls, was uncommon in T2DM. Most T2DM patients with hearing loss had bilateral, symmetrical involvement (79.0%), and a gently sloping high-frequency configuration was the commonest audiogram pattern (66.1%), followed by a flat pattern (21.0%).

Table 5. Type, laterality and audiogram configuration among participants with hearing loss

Values are n (%); p values by Fisher’s exact test (type) and chi-square test (laterality).

Age-stratified prevalence

Hearing loss became more frequent with advancing age in both groups, but at every age band the prevalence was higher in patients with T2DM (Table 6). The relative difference was notable in the 40–49 and 50–59 year groups, and nine in ten T2DM patients aged 60 years or more had hearing loss. The age-adjusted (Mantel–Haenszel) common odds ratio for hearing loss in T2DM was 5.2 (p < 0.001), indicating that the excess burden was not attributable to age differences.

Table 6. Prevalence of hearing loss by age group

Mantel–Haenszel common odds ratio = 5.2, p < 0.001.

Association with glycemic status

The prevalence and severity of hearing loss increased stepwise with worsening glycemic control (Table 7, Figure 3). Hearing loss was present in 33.3% of patients with HbA1c below 7.0%, 60.5% of those with HbA1c between 7.0% and 8.9%, and 81.6% of those with HbA1c of 9.0% or more (χ² = 14.6, p = 0.001). Mean PTA also differed significantly across the three groups (F = 11.3, p < 0.001). On Bonferroni post hoc testing, the poorly controlled group had significantly higher PTA than both the good-control group (p < 0.001) and the fair-control group (p = 0.01), whereas the difference between good and fair control did not reach significance (p = 0.14).

Table 7. Hearing loss and pure tone average according to HbA1c category in T2DM

Bonferroni post hoc: good vs fair p = 0.14; good vs poor p < 0.001; fair vs poor p = 0.01.

Duration of diabetes

The prevalence of hearing loss increased with the duration of diabetes (Table 8). Fewer than four in ten patients with diabetes of less than five years’ duration had hearing loss, compared with more than eight in ten of those with duration exceeding ten years (χ² = 15.9, p < 0.001).

Table 8. Hearing loss according to duration of diabetes

Correlation analysis

Among patients with T2DM, HbA1c showed the strongest positive correlation with mean PTA (r = 0.58) and with the 4000 Hz threshold (r = 0.54), followed by age and duration of diabetes (Table 9). FBS correlated moderately with PTA, whereas BMI showed no significant correlation.

Table 9. Correlation of hearing thresholds with clinical and biochemical variables in T2DM (n = 100)

Pearson correlation coefficient; PTA = mean of both ears.

Diabetic complications and hearing loss

Hearing loss was significantly more frequent in patients with peripheral neuropathy, retinopathy and nephropathy (Table 10). More than 80% of patients with any of these microvascular complications had hearing loss. Hypertension showed a similar trend that did not reach statistical significance.

Table 10. Association of hearing loss with diabetic complications and hypertension

n = number of patients with the complication; percentages are of patients with (present) and without (absent) the complication who had hearing loss.

Multivariable analysis

In the multivariable logistic regression model including age, sex, BMI, duration of diabetes, HbA1c and hypertension, three variables were independent predictors of hearing loss in patients with T2DM (Table 11): HbA1c (aOR 1.74 for each 1% increase; 95% CI 1.28–2.37), age (aOR 1.92 for every 10-year increase) and duration of diabetes (aOR 1.58 for every 5-year increase). Sex, BMI and hypertension were not independently associated.

Table 11. Multivariable logistic regression of predictors of hearing loss in T2DM

OR, odds ratio; CI, confidence interval.

In this cross-sectional comparative study, nearly two-thirds of patients with T2DM had audiometrically confirmed hearing loss, against slightly more than one-quarter of age- and sex-matched non-diabetic controls. Patients with diabetes had poorer thresholds at all frequencies, a gently sloping high-frequency pattern, a graded relationship between hearing loss and HbA1c, and an independent contribution of long-term glycemic exposure after adjustment for age. These observations are in keeping with pooled evidence that diabetes is associated with roughly double the risk of hearing impairment, [10,11] and with population-based audiometric data from the NHANES surveys. [12]

The prevalence we observed lies within the wide range reported in the studies, which varies with the age profile of the sample, the threshold used to define hearing loss, the frequencies analysed and whether one or both ears are considered. [22,23] Indian investigators have also reported poorer hearing thresholds among diabetic patients than among non-diabetic subjects, [9,17] and our findings extend this work by describing the severity, type and configuration of hearing loss and by relating them to HbA1c, a long-term marker of glycemia. The strong association with age in our data is expected, since age is the principal determinant of hearing loss in any population. [24,25] Importantly, however, the excess risk in T2DM persisted in every age band and the age-adjusted odds ratio was higher than the crude estimate, suggesting that diabetes adds to, rather than merely coincides with, age-related cochlear change. This is in line with the longitudinal observations of Mitchell and co-workers from the Blue Mountains cohort. [26]

The predominance of bilateral, symmetrical, high-frequency sensorineural loss in our T2DM patients agrees with the audiometric profile described in earlier clinical series [8,15,27] and in more recent controlled studies. [28–30] The basal turn of the cochlea, which encodes high frequencies, has the highest metabolic demand and is the region most susceptible to ischaemic and oxidative injury, which probably explains why thresholds were most elevated at 4000 and 8000 Hz. Human temporal bone histology in T2DM has shown atrophy of the stria vascularis, thickening of strial capillary walls and loss of outer hair cells, [13] changes that echo the microangiopathy described in insulin-dependent diabetes. [14] The predominance of the sensorineural type in our cases, with only 3.2% having a purely conductive component, supports a cochlear or retrocochlear rather than a middle-ear mechanism. Early cochlear involvement has also been demonstrated in younger patients with type 1 diabetes, who often have normal conventional thresholds but abnormal findings on more sensitive tests, [31,32] indicating that cochlear changes may begin well before overt hearing handicap.

A distinctive feature of the present study is the graded association of hearing loss with HbA1c. Prevalence rose from one-third in the well-controlled group to more than four-fifths in those with HbA1c of 9% or above, mean PTA was approximately 13 dB HL worse in the poorly controlled group, and each 1% increase in HbA1c carried a 1.7-fold increase in the odds of hearing loss after adjustment. The correlation was stronger with HbA1c than with FBS, and the duration of diabetes also contributed independently. Taken together, these results suggest that cumulative exposure to hyperglycemia, rather than momentary glucose levels, drives cochlear injury, which is consistent with a report that cochlear dysfunction in T2DM may develop independently of acute hyperglycemia and of peripheral neuropathy. [33]

Plausible biochemical mechanisms include non-enzymatic glycation of basement-membrane proteins, accumulation of advanced glycation end products, oxidative stress and endothelial dysfunction in the strial and spiral modiolar vessels. [34] Studies that examined mediators of diabetes-related hearing impairment have likewise pointed to a role for the vascular and neuropathic complications of the disease. [35] Not all earlier reports, however, found a clear relationship with glycemic control; differences in study design, single-point HbA1c measurements, age profiles and the audiometric end-points used may explain this inconsistency. [36]

Hearing loss was significantly more frequent among patients with neuropathy, retinopathy and nephropathy. The co-occurrence of these complications supports the concept that the inner ear is another end-organ of diabetic microangiopathy, a view that has long been advanced on the basis of clinical and histopathological observations. [14,15] The cochlear nerve may be affected by the same metabolic neuropathy that damages peripheral nerves. [15] Hypertension was more common in patients with diabetes and tended to coexist with hearing loss, but it lost significance in the multivariable model, indicating that the association was largely mediated by age and glycemic burden. Experimental and clinical data suggest that hypertension may act synergistically with diabetes in producing cochlear damage, [37] and larger studies are needed to settle this question.

These findings have particular relevance for India. The country has one of the largest populations of people with diabetes in the world, and the disease appears at a younger age than in Western countries, so that patients live longer with hyperglycemia and accumulate complications. [1,5] At the same time, hearing loss is a major cause of disability, and most of it remains undetected and unmanaged because of limited access to audiological services. [16] Routine diabetes care in India currently focuses on glycemic, retinal, renal, neurological and cardiovascular surveillance and rarely includes hearing assessment. Our data suggest that simple pure tone audiometry, ideally including 4000 and 8000 Hz, could be added to the annual complication screen of patients with long-standing or poorly controlled diabetes, or those with other microvascular complications. Early detection can allow counselling, use of hearing aids where appropriate, avoidance of additional ototoxic exposures and reinforcement of the need for strict glycemic control. Whether improved glycemic control can slow or prevent the progression of hearing loss remains to be established by prospective interventional studies.

Strengths and limitations

Strengths of this study include the use of an age- and sex-matched control group, the exclusion of the major confounders of sensorineural hearing loss (noise exposure, ototoxic drugs, otological disease and family history), the use of standardised audiometric procedures by a blinded audiologist, the use of HbA1c as an index of chronic glycemia, and the multivariable analysis. Limitations include the single-centre, hospital-based design, which limits generalisability, and the cross-sectional approach, which cannot establish temporality or causation.

HbA1c was measured on a single occasion, and glycemic variability, lipid profile and other metabolic factors were not studied. More sensitive audiological tests such as high-frequency audiometry (above 8 kHz), otoacoustic emissions and auditory brainstem responses were not performed and could identify subclinical dysfunction in patients with normal conventional audiograms. Longitudinal studies with serial audiometry would be useful to establish whether the rate of hearing decline differs according to the degree of glycemic control.

Patients with type 2 diabetes mellitus have a significantly higher prevalence and greater severity of hearing loss than non-diabetic persons of similar age and sex. The loss is predominantly bilateral, symmetrical, sensorineural and high-frequency in nature. Poorer glycemic status, as reflected by HbA1c, and longer duration of diabetes are independently associated with worse hearing, and hearing loss is more common in patients with other microvascular complications. Pure tone audiometry is a simple and inexpensive test that should be considered as part of routine complication screening in T2DM, and maintaining optimal glycemic control may help to preserve hearing. Prospective multicentre studies are warranted to confirm these findings and to determine whether tighter glycemic control alters the course of diabetic hearing loss.

1. Mohan V, Sandeep S, Deepa R, Shah B, Varghese C. Epidemiology of type 2 diabetes: Indian scenario. Indian J Med Res. 2007;125(3):217-30.
2. Ramachandran A, Ma RC, Snehalatha C. Diabetes in Asia. Lancet. 2010;375(9712):408-18.
3. Ramachandran A, Snehalatha C, Kapur A, Vijay V, Mohan V, Das AK, et al. High prevalence of diabetes and impaired glucose tolerance in India: National Urban Diabetes Survey. Diabetologia. 2001;44(9):1094-101.
4. Mohan V, Deepa M, Deepa R, Shanthirani CS, Farooq S, Ganesan A, et al. Secular trends in the prevalence of diabetes and impaired glucose tolerance in urban South India - the Chennai Urban Rural Epidemiology Study (CURES-17). Diabetologia. 2006;49(6):1175-8.
5. Anjana RM, Pradeepa R, Deepa M, Datta M, Sudha V, Unnikrishnan R, et al. Prevalence of diabetes and prediabetes (impaired fasting glucose and/or impaired glucose tolerance) in urban and rural India: phase I results of the Indian Council of Medical Research-INdia DIABetes (ICMR-INDIAB) study. Diabetologia. 2011;54(12):3022-7.
6. Jorgensen MB, Buch NH. Studies on inner-ear function and cranial nerves in diabetics. Acta Otolaryngol. 1961;53:350-64.
7. Axelsson A, Fagerberg SE. Auditory function in diabetics. Acta Otolaryngol. 1968;66(1):49-64.
8. Taylor IG, Irwin J. Some audiological aspects of diabetes mellitus. J Laryngol Otol. 1978;92(2):99-113.
9. Kurien M, Thomas K, Bhanu TS. Hearing threshold in patients with diabetes mellitus. J Laryngol Otol. 1989;103(2):164-8.
10. Horikawa C, Kodama S, Tanaka S, Fujihara K, Hirasawa R, Yachi Y, et al. Diabetes and risk of hearing impairment in adults: a meta-analysis. J Clin Endocrinol Metab. 2013;98(1):51-8.
11. Akinpelu OV, Mujica-Mota M, Daniel SJ. Is type 2 diabetes mellitus associated with alterations in hearing? A systematic review and meta-analysis. Laryngoscope. 2014;124(3):767-76.
12. Bainbridge KE, Hoffman HJ, Cowie CC. Diabetes and hearing impairment in the United States: audiometric evidence from the National Health and Nutrition Examination Survey, 1999 to 2004. Ann Intern Med. 2008;149(1):1-10.
13. Fukushima H, Cureoglu S, Schachern PA, Paparella MM, Harada T, Oktay MF. Effects of type 2 diabetes mellitus on cochlear structure in humans. Arch Otolaryngol Head Neck Surg. 2006;132(9):934-8.
14. Smith TL, Raynor E, Prazma J, Buenting JE, Pillsbury HC. Insulin-dependent diabetic microangiopathy as a cause of sensorineural hearing loss. Laryngoscope. 1995;105(2):236-40.
15. Friedman SA, Schulman RH, Weiss S. Hearing and diabetic neuropathy. Arch Intern Med. 1975;135(4):573-6.
16. Garg S, Chadha S, Malhotra S, Agarwal AK. Deafness: burden, prevention and control in India. Natl Med J India. 2009;22(2):79-82.
17. Panchu P. Auditory acuity in type 2 diabetes mellitus. Int J Diabetes Dev Ctries. 2008;28(4):114-20.
18. American Diabetes Association. Classification and diagnosis of diabetes. Diabetes Care. 2015;39(Suppl 1):S13-22.
19. Carhart R, Jerger JF. Preferred method for clinical determination of pure-tone thresholds. J Speech Hear Disord. 1959;24(4):330-45.
20. International Organization for Standardization. ISO 8253-1:2010. Acoustics - Audiometric test methods - Part 1: Pure-tone air and bone conduction audiometry. Geneva: ISO; 2010.
21. Clark JG. Uses and abuses of hearing loss classification. ASHA. 1981;23(7):493-500.
22. Maia CA, Campos CA. Diabetes mellitus as etiological factor of hearing loss. Braz J Otorhinolaryngol. 2005;71(2):208-14.
23. Bamanie AH, Al-Noury KI. Prevalence of hearing loss among Saudi type 2 diabetic patients. Saudi Med J. 2011;32(3):271-4.
24. Agrawal Y, Platz EA, Niparko JK. Prevalence of hearing loss and differences by demographic characteristics among US adults: data from the National Health and Nutrition Examination Survey, 1999-2004. Arch Intern Med. 2008;168(14):1522-30.
25. Gates GA, Mills JH. Presbycusis. Lancet. 2005;366(9491):1111-20.
26. Mitchell P, Gopinath B, McMahon CM, Rochtchina E, Wang JJ, Boyages SC, et al. Relationship of type 2 diabetes to the prevalence, incidence and progression of age-related hearing loss. Diabet Med. 2009;26(5):483-8.
27. Tay HL, Ray N, Ohri R, Frootko NJ. Diabetes mellitus and hearing loss. Clin Otolaryngol Allied Sci. 1995;20(2):130-4.
28. Austin DF, Konrad-Martin D, Griest S, McMillan GP, McDermott D, Fausti S. Diabetes-related changes in hearing. Laryngoscope. 2009;119(9):1788-96.
29. Frisina ST, Mapes F, Kim S, Frisina DR, Frisina RD. Characterization of hearing loss in aged type II diabetics. Hear Res. 2006;211(1-2):103-13.
30. Sakuta H, Suzuki T, Yasuda H, Ito T. Type 2 diabetes and hearing loss in personnel of the Self-Defense Forces. Diabetes Res Clin Pract. 2007;75(2):229-34.
31. Lisowska G, Namyslowski G, Morawski K, Strojek K. Early identification of hearing impairment in patients with type 1 diabetes mellitus. Otol Neurotol. 2001;22(3):316-20.
32. Pessin AB, Martins RH, Pimenta WP, Simoes AC, Marsiglia A, Amaral AV. Auditory evaluation in patients with type 1 diabetes mellitus. Ann Otol Rhinol Laryngol. 2008;117(5):366-70.
33. Sasso FC, Salvatore T, Tranchino G, Cozzolino D, Caruso AA, Persico M, et al. Cochlear dysfunction in type 2 diabetes: a complication independent of neuropathy and acute hyperglycemia. Metabolism. 1999;48(7):885-92.
34. Hong O, Buss J, Thomas E. Type 2 diabetes and hearing loss. Dis Mon. 2013;59(4):139-46.
35. Bainbridge KE, Cheng YJ, Cowie CC. Potential mediators of diabetes-related hearing impairment in the U.S. population: National Health and Nutrition Examination Survey 1999-2004. Diabetes Care. 2010;33(4):811-6.
36. Dalton DS, Cruickshanks KJ, Klein R, Klein BE, Wiley TL. Association of NIDDM and hearing loss. Diabetes Care. 1998;21(9):1540-4.
37. Duck SW, Prazma J, Bennett PS, Pillsbury HC. Interaction between hypertension and diabetes mellitus in the pathogenesis of sensorineural hearing loss. Laryngoscope. 1997;107(12):1596-605.