Association Between HbA1c Level and Hearing Impairment in a Nondiabetic Adult Population

Abstract

Background:

The aim of this study was to investigate whether glycosylated hemoglobin (HbA1c) level in nondiabetic patients is associated with hearing impairment in the general Korean population.

Patients and Methods:

Data from the Korean National Health and Nutrition Examination Survey 2011–2013 were used in the analyses. Participants were excluded from this study for the following reasons: they could not provide data regarding pure tone audiometry, they had ear disease, they had brain disorders, asymmetric sensory neural hearing loss (HL), or they were younger than 40 years or had diabetes mellitus. Finally, 7449 participants were included in this study.

Results:

The mean HbA1c levels in the low, middle, and high tertiles were 5.3% ± 0.2%, 5.7% ± 0.1%, and 6.1% ± 0.2%, respectively. The numbers of participants in the low, middle, and high tertiles were 2808, 2509, and 2132, respectively. The low-frequency, mid-frequency, high-frequency, and average hearing thresholds were significantly increased with increasing HbA1c tertile. Linear regression analyses showed that HbA1c level in the nondiabetic participants was associated with components of metabolic syndrome. The mean numbers of metabolic syndrome components in the low, middle, and high HbA1c tertiles were 1.22, 1.53, and 2.02, respectively. The participants in the middle and high HbA1c tertiles had a 1.239- and 1.253-fold increased risk of HL, respectively, compared with those in the low HbA1c tertile.

Conclusion:

HbA1c level was associated with hearing impairment in the nondiabetic participants of this study. Therefore, the participants with high HbA1c levels should be closely monitored for hearing impairment.

Introduction

Hearing loss (HL) affects communication and functional ability and is significantly associated with decreased quality of life.^1,2 The prevalence of HL in the adult population is ∼4.9%–17.0%.³ The prevalence of HL may gradually increase as the general population ages. HL is caused by many factors, such as age, genetic factors, alcohol intake, micronutrient intake, medication use, noise, and metabolic disturbances, including diabetes mellitus (DM) or dyslipidemia.^4

–8 Hyperglycemia leads to microangiopathy, which results in major complications, such as nephropathy, neuropathy, and retinopathy. These show the association between hyperglycemia and HL through microangiopathy in the cochlea.⁹

Glycosylated hemoglobin (HbA1c) level has been widely accepted as a diagnostic parameter of DM in nondiabetic patients and an indicator of long-term glycemic control in DM patients.¹⁰ HbA1c is an advanced glycosylation end product, and increased HbA1c level is correlated with increased advanced glycosylation end product.¹¹ Advanced glycosylation end products, including HbA1c, induce endothelial dysfunction and oxidative stress. These consequently result in micro- and macroangiopathies.^12,13

Recent studies have shown that HbA1c level is associated with increased risk of DM, cardiovascular disease, and mortality in nondiabetic patients.^14
–16 Park et al. showed the usefulness of HbA1c level as a diagnostic criterion for metabolic syndrome in nondiabetic patients.¹⁷ These studies suggest that high HbA1c in nondiabetic patients is associated with metabolic syndrome or insulin resistance. These metabolic disturbances consequently induce microangiopathies, in the cochlea, and may result in hearing impairment. Therefore, the aim of this study was to investigate whether HbA1c level in nondiabetic patients is associated with hearing impairment in the general Korean population.

Patients and Methods

Study population

Data from the Korean National Health and Nutrition Examination Survey (KNHANES) 2011–2013 were used in the analyses. The KNHANES is a nationwide, multistage, stratified survey of a representative sample of the South Korean population conducted by the Korea Centers for Disease Control and Prevention. The total number of participants in the KNHANES was 24,594. Participants were excluded from this study for the following reasons: they could not provide data regarding pure tone audiometry (n = 9032) or HbA1c (n = 505); they had external or middle ear disease (n = 484); they had brain disorders, such as stroke (n = 294); asymmetric sensory neural HL [defined as average hearing threshold (AHT) asymmetry ≥15 dB; n = 1005]; or they were younger than 40 years (n = 4348) or had DM (a self-reported history of a DM diagnosis, a fasting glucose level ≥126 mg/dL, or HbA1c level ≥6.5%; n = 1477). Finally, 7449 participants were included in this study. This study was approved by the Institutional Review Board of the Yeungnam University Hospital (2015-03-027). The board waived the need for informed consent as the subjects' records and information were anonymized and deidentified before the analysis.

Study variables

Clinical and laboratory data collected from the participants during the health examination included the following: age, sex, serum creatinine level (mg/dL), body mass index (BMI; kg/m²), waist circumference (cm), HbA1c level (%), fasting blood glucose level (mg/dL), total cholesterol level (mg/dL), high-density lipoprotein cholesterol (HDL-C) level (mg/dL), triglyceride level (mg/dL), systolic blood pressure (SBP; mmHg), diastolic blood pressure (DBP; mmHg), hearing thresholds, smoking behavior, and alcohol consumption.

HbA1c levels were measured using high-performance liquid chromatography system (HLC-723G7; Tosoh Co., Tokyo, Japan). The participants were divided into three groups according to their HbA1c tertiles. Hypertension (HTN) was defined as an SBP ≥140 mmHg, a DBP ≥90 mmHg, a self-reported history of HTN, or the use of anti-HTN drugs. Smoking behaviors were classified as current smoker, ex-smoker, or nonsmoker. Alcohol intake was defined by the Korean version of standard drinking, which was based on the WHO classification.^18,19 We classified alcohol intake into three categories as follows: abstinence (not having had an alcohol drink within the last year), moderate drinking (women: 0.1–19.99 grams pure alcohol/day; men: 0.1–39.99 grams pure alcohol/day), and heavy drinking (women: ≥20 grams pure alcohol/day; men: ≥40 grams pure alcohol/day). Metabolic syndrome was defined using the National Cholesterol Education Program Adult Treatment Panel III guidelines.^20,21

Otological physical examinations and pure tone audiometry were performed as previously described.⁷ None of these subjects was currently receiving medication that is associated with ototoxicity. Histories of exposure to explosive or occupational noise were classified as positive or negative, as previously described.⁷ Briefly, an explosive noise was defined as a sudden loud noise, such as an explosion or a gunshot. Exposure to occupational noise was determined according to whether the participants had worked in a location with loud machinery for ≥3 months. Loud noise was defined by whether the participants had needed to raise his or her voice to have a conversation. The hearing thresholds were measured using an automatic audiometer at 0.5, 1, 2, 3, 4, and 6 kHz. For both ears of each subject, the pure tone averages at 0.5 and 1 kHz were averaged to obtain the low-frequency (Low-Freq) value, those at 2 and 3 kHz to obtain the mid-frequency (Mid-Freq) value, and those at 4 and 6 kHz to obtain the high-frequency (High-Freq) value. In the present study, the AHT was calculated as pure tone average at four frequencies (0.5, 1, 2, and 3 kHz). HL was defined according to an AHT >25 dB in the better ear.

Statistical analyses

Data were analyzed using SPSS version 21 (SPSS, Chicago, IL). Categorical variables were expressed as numbers and percentages. The continuous variables were expressed as mean ± standard deviation or standard error. The Pearson chi-squared test or Fisher exact test was used to analyze categorical variables. For continuous variables, means were compared using the t-test or one-way analysis of variance, followed by a post-hoc Tukey comparison. Linear regression analysis was performed to assess the independent predictors of HbA1c level. Logistic regression analyses were used to estimate the odds ratios and 95% confidence intervals (CI), which were used to determine the relationship between HbA1c level and HL.

Multivariate analysis was adjusted for age, sex, alcohol intake, smoking habit, exposure to explosive noise, and exposure to occupational noise. Multivariate analyses using analyses of covariance, multiple linear regression, or multiple logistic regression were used to determine the independent predictors of metabolic disturbances or hearing. Discrimination, which is the ability of the model to differentiate between participants with and those without HL or metabolic syndrome, was examined using the area under the receiver operating characteristic curve (AUROC). AUROC analysis was also performed to calculate cutoff values, sensitivity, and specificity. The best cutoff risk point was defined from the maximum of the Youden index in the AUROC. The AUROC was calculated using MedCalc version 11.6.1.0 (MedCalc, Mariakerke, Belgium). The level of statistical significance was set at P < 0.05.

Results

Clinical characteristics of the participants

The mean HbA1c levels in the low, middle, and high tertiles were 5.3% ± 0.2% (4.0%–5.5%), 5.7% ± 0.1% (5.6%–5.8%), and 6.1% ± 0.2% (5.9%–6.4%), respectively (Table 1). The numbers of participants in the low, middle, and high tertiles were 2808, 2509, and 2132, respectively. The proportions of participants with HTN, exposure to explosive noise, and HL were increased as the HbA1c tertile increased. Age, BMI, waist circumference, fasting blood glucose level, total cholesterol level, triglyceride level, and SBP increased with increasing HbA1c tertile. Serum creatinine level, DBP, exposure to occupational noise, and smoking habit were not significantly associated with the HbA1c tertile.

Table 1.

Clinical Characteristics of the Participants According to HbA1c Tertile

	Low tertile	Middle tertile	High tertile	P
Age (years)	53.2 ± 10.7	56.7 ± 11.0	59.8 ± 10.8	<0.001
Body mass index (kg/m²)	23.3 ± 2.9	23.8 ± 2.9	24.5 ± 3.0	<0.001
Creatinine (mg/dL)	0.83 ± 0.20	0.83 ± 0.18	0.84 ± 0.27	0.169
Waist circumference (cm)	79.8 ± 8.7	81.6 ± 8.7	83.6 ± 8.8	<0.001
HbA1c (%)	5.3 ± 0.2	5.7 ± 0.1	6.1 ± 0.2	<0.001
Fasting blood glucose (mg/dL)	91.4 ± 8.0	94.5 ± 8.3	99.7 ± 10.0	<0.001
Total cholesterol (mg/dL)	188.7 ± 32.6	197.3 ± 33.9	201.2 ± 37.2	<0.001
Triglyceride (mg/dL)	122.1 ± 87.1	135.0 ± 89.7	147.7 ± 102.5	<0.001
High-density lipoprotein cholesterol (mg/dL)	53.7 ± 13.3	52.2 ± 12.3	51.0 ± 11.8	<0.001
Systolic blood pressure (mmHg)	118.9 ± 16.6	120.8 ± 16.7	123.9 ± 16.7	<0.001
Diastolic blood pressure (mmHg)	76.9 ± 10.4	76.8 ± 10.1	77.3 ± 10.4	0.198
Hypertension, n (%)	796 (28.3)	839 (33.5)	957 (44.9)	<0.001
Alcohol intake				<0.001
Abstinence, n (%)	709 (25.2)	771 (30.7)	742 (34.8)
Moderate drinking, n (%)	1906 (67.9)	1581 (63.0)	1278 (59.9)
Heavy drinking, n (%)	128 (4.6)	83 (3.3)	55 (2.6)
Unknown, n (%)	65 (2.3)	74 (2.9)	57 (2.7)
Smoking				0.524
Nonsmoker, n (%)	1683 (59.9)	1490 (59.4)	1268 (59.5)
Ex-smoker, n (%)	608 (21.7)	534 (21.3)	430 (20.2)
Current smoker, n (%)	457 (16.3)	416 (16.6)	380 (17.8)
Unknown, n (%)	60 (2.1)	69 (2.8)	54 (2.5)
Occupational noise exposure, n (%)	374 (13.3)	379 (15.1)	299 (14.0)	0.177
Explosive noise exposure, n (%)	618 (22.0)	611 (24.4)	549 (25.8)	0.008
Hearing loss, n (%)	335 (11.9)	431 (17.2)	435 (20.4)	<0.001

Data are expressed as numbers (percentages) for categorical variables and means ± standard deviations for continuous variables. The P values were tested using one-way analysis of variance for continuous variables and Pearson chi-squared test or Fisher exact test for categorical variables.

HbA1c, glycosylated hemoglobin.

Association between HbA1c tertile and hearing thresholds

For Low-Freq, the mean hearing thresholds were 12.9 (95% CI 12.5–13.3), 14.7 (95% CI 14.2–15.1), and 16.3 dB (95% CI 15.8–16.9), respectively (Fig. 1). For Mid-Freq, the mean hearing thresholds were 17.6 (95% CI 17.0–18.2), 20.5 (95% CI 19.8–21.1), and 22.7 dB (95% CI 22.0–23.4), respectively. For High-Freq, the mean hearing thresholds were 30.2 (95% CI 29.5–31.0), 33.8 (95% CI 32.9–34.6), and 36.6 dB (95% CI 35.7–37.5), respectively. AHTs were 13.3 (95% CI 12.8–13.7), 15.5 (95% CI 15.0–16.0), and 17.3 dB (95% CI 16.7–17.9), respectively. The Low-Freq, Mid-Freq, High-Freq, and AHT were significantly increased with increasing HbA1c tertile. The multivariate analysis revealed that the four hearing thresholds in the low HbA1c tertile were significantly less than those in the middle and high HbA1c tertiles.

FIG. 1.

Hearing thresholds according to glycosylated hemoglobin (HbA1c) tertile. The mean hearing thresholds in low, middle, and high HbA1c tertiles were as follows—for low-frequency (Low-Freq): 12.9 ± 0.2 (mean ± standard error), 14.7 ± 0.2, and 16.3 ± 0.3 dB, respectively; for mid-frequency (Mid-Freq): 17.6 ± 0.3, 20.5 ± 0.3, and 22.7 ± 0.4 dB, respectively; for high-frequency (High-Freq): 30.2 ± 0.4, 33.8 ± 0.4, and 36.6 ± 0.4 dB, respectively; and for average hearing threshold (AHT): 13.3 ± 0.2, 15.5 ± 0.3, and 17.3 ± 0.3 dB, respectively. In the multivariate analyses, the mean hearing thresholds in the low, middle, and high HbA1c tertiles were as follows—for Low-Freq: 13.6 ± 0.2, 14.5 ± 0.2, and 15.0 ± 0.2 dB, respectively; for Mid-Freq: 18.6 ± 0.3, 20.1 ± 0.3, and 20.7 ± 0.3 dB, respectively; for High-Freq: 31.8 ± 0.3, 33.2 ± 0.3, and 34.4 ± 0.4 dB, respectively; and for AHT: 14.2 ± 0.2, 15.3 ± 0.2, and 15.7 ± 0.2 dB, respectively. The multivariate analysis was adjusted for age, sex, alcohol intake, smoking habit, exposure to explosive noise, and exposure to occupational noise (P < 0.05 for trend in the univariate and multivariate analyses; *P < 0.05, compared with the participants with low HbA1c tertile; ^# P < 0.05, compared with the participants with middle HbA1c tertile). The data are expressed as mean and standard error values.

Association between HbA1c level and metabolic disturbances in the nondiabetic participants

Univariate linear regression analyses of HbA1c levels according to metabolic syndrome components revealed a standardized β ± standard error of 0.097 ± 0.000 for HDL-C level (P < 0.001), 0.188 ± 0.000 for waist circumference (P < 0.001), 0.121 ± 0.000 for triglyceride level (P < 0.001), 0.384 ± 0.000 for fasting blood glucose level (P < 0.001), 0.118 ± 0.000 for SBP (P < 0.001), and 0.010 ± 0.000 for DBP (P = 0.379). Multivariate analyses revealed a standardized β ± standard error of 0.081 ± 0.000 for HDL-C level (P < 0.001), 0.195 ± 0.000 for waist circumference (P < 0.001), 0.132 ± 0.000 for triglyceride level (P < 0.001), 0.396 ± 0.000 for fasting blood glucose level (P < 0.001), and 0.082 ± 0.000 for SBP (P < 0.001). Univariate and multivariate linear regression analyses showed that HbA1c level in the nondiabetic participants was associated with components of metabolic syndrome (except DBP).

The numbers of metabolic syndrome components in the low, middle, and high HbA1c tertiles were 1.20 (95% CI 1.15–1.24), 1.53 (95% CI 1.48–1.58), and 2.03 (95% CI 1.98–2.09), respectively (P < 0.001 for trend). Multivariate analyses revealed that the mean numbers of metabolic syndrome components in the low, middle, and high HbA1c tertiles were 1.22 (95% CI 1.18–1.27), 1.53 (95% CI 1.48–1.57), and 2.02 (95% CI 1.97–2.07), respectively (P < 0.001 for trend). The mean numbers of metabolic syndrome components increased as the HbA1c tertile increased. The AUROC of HbA1c level for detecting metabolic syndrome was analyzed and was 0.648 (95% CI 0.637–0.659; P < 0.001). The cutoff value was >5.7%. The sensitivity and specificity for predicting metabolic syndrome were 56.0% and 66.3%, respectively.

Association between HbA1c level and HL

Univariate logistic regression showed that the participants in the middle and high HbA1c tertiles had a 1.531-fold (95% CI 1.312–1.786; P < 0.001) and 1.892-fold (95% CI 1.620–2.210; P < 0.001) increased risk of HL, respectively, compared with those in the low HbA1c tertile. The multivariate analysis revealed that the participants in the middle and high HbA1c tertiles had a 1.239-fold (95% CI 1.034–1.485; P = 0.020) and 1.253-fold (95% CI 1.050–1.496; P = 0.012) increased risk of HL, respectively, compared with those in the low HbA1c tertile.

Association between metabolic syndrome and hearing

For Low-Freq, the mean hearing thresholds of the participants with and those without metabolic syndrome were 16.2 ± 12.0 dB (mean ± standard deviation) and 14.0 ± 11.2 dB, respectively (P < 0.001). For Mid-Freq, the mean hearing thresholds of the participants with and those without metabolic syndrome were 22.4 ± 16.4 dB and 19.3 ± 16.1 dB, respectively (P < 0.001). For High-Freq, the mean hearing thresholds of the participants with and those without metabolic syndrome were 36.8 ± 20.5 dB and 32.2 ± 20.6 dB, respectively (P < 0.001). For AHT, the mean hearing thresholds of the participants with and those without metabolic syndrome were 17.1 ± 13.0 dB and 14.6 ± 12.5 dB, respectively (P < 0.001). Multivariate analyses showed that for High-Freq, the mean hearing thresholds of the participants with and those without metabolic syndrome were 15.2 dB (95% CI 14.8–15.7) and 14.0 dB (95% CI 13.8–14.3), respectively (P < 0.001). For Mid-Freq, the mean hearing thresholds of the participants with and those without metabolic syndrome were 20.5 dB (95% CI 19.9–21.2) and 19.6 dB (95% CI 19.3–20.0), respectively (P = 0.020). For High-Freq, the mean hearing thresholds of the participants with and those without metabolic syndrome were 34.1 dB (95% CI 33.3–35.0) and 32.7 dB (95% CI 32.3–33.1), respectively (P = 0.003). For AHT, the mean hearing thresholds of the participants with and those without metabolic syndrome were 15.8 dB (95% CI 15.3–16.3) and 14.8 dB (95% CI 14.5–15.0), respectively (P < 0.001). The numbers of HL participants with and those without metabolic syndrome were 342 (20.4%) and 859 (14.9%), respectively (P < 0.001).

Discussion

Fasting blood glucose level is a diagnostic criterion for metabolic syndrome. However, some studies showed the limitation of using fasting blood glucose level and the benefit of using HbA1c level comparison with fasting blood glucose level in the diagnosis of metabolic syndrome.^{14,22

–26} In 2010, the American Diabetes Association recommended the use of HbA1c level as an alternative to fasting blood glucose level to define the category of increased DM risk and defined prediabetes as an HbA1c 5.7%–6.4% and DM as an HbA1c level ≥6.4%.²³ Furthermore, some investigators showed the usefulness of HbA1c level as a single predictor of metabolic syndrome.^14,24
–26 As previous studies suggested, the present study showed that HbA1c level was inversely associated with HDL-C level and positively associated with waist circumference, triglyceride level, fasting blood glucose level, and SBP among the metabolic syndrome components. The AUROC also showed a diagnostic usefulness of HbA1c level for predicting metabolic syndrome.

We focused on age-related sensorineural HL. Age-related sensorineural HL increases from the fifth decade, but HL in underlying diseases, such as diabetes or HTN, can develop earlier than the fifth decade of life.²⁷ Therefore, no definite age criteria have been established for evaluating HL. Previous studies included age as a criterion variable, such as age ≥35, ≥40, and ≥50 years.^7,28
–30 If young participants were mainly included, the prevalence of HL could be underestimated, and the young population may be less sensitive to the effects for metabolic disturbances on HL than the old population. Therefore, inclusion of a population with a wide age range may induce confusion in the interpretation of results. If mainly older participants are included, smaller study cohorts and the presence of co-morbidities can be a weakness. In the present study, age ≥40 years may be a suitable criterion for resolving these problems.

DM has been reported to be associated with increased incidence of HL.^31,32 The pathogenesis for DM-related hearing impairments has included cochlear microangiopathy, hyperglycemia of the perilymph, and variable neuropathy or encephalopathy.³² A study using autopsy results showed the association between DM and vessel wall thickening or loss of cochlear outer hair cell.³³ Metabolic syndrome as a prediabetic status may be associated with these pathological conditions. In addition, previous studies that used each metabolic syndrome components showed the association between metabolic syndrome or components and decrease in hearing thresholds.^30

–35 In the present study, all hearing thresholds, namely Low-Freq, Mid-Freq, High-Freq, and AHT, were higher in the participants with metabolic syndrome than in those without metabolic syndrome. Multivariate analyses showed similar trends. In addition, the number of HL participants was higher among the participants with metabolic syndrome than among those without metabolic syndrome.

Some studies evaluated the relationship between glycemic control and hearing impairment.^36,37 A study focused on the association between diabetic duration or glycemic control and HL but did not show any positive association.³⁶ A recent study evaluated the association between HbA1c level and hearing impairment in Asians, including both DM and nondiabetic patients. Through a cross-sectional analysis that enrolled both DM and nondiabetic patients, an association between HbA1c level and hearing impairment was demonstrated.³⁷ Longitudinal analyses using the same cohort also showed an association between HbA1c level and hearing impairment. However, significant association between HbA1c level and hearing impairment was not observed only in the nondiabetic patients, but the trend was similar with data from both DM and nondiabetic patients. The present study focused on the association between HbA1c level and hearing impairment in nondiabetic participants. HbA1c level in the nondiabetic participants was associated with increases in all the four hearing thresholds and HL. These associations may be correlated with the development of metabolic syndrome or increases in the metabolic syndrome components.

This study had a few limitations. First, the study was limited by its retrospective nature, and ethnic differences were not evaluated. This study could not establish causality. Second, this study did not evaluate sensitive components of hearing problems, such as speech discrimination. Third, the statistical significant difference in hearing threshold between groups was small. Fourth, the present study might also have included a selection bias due to the exclusion of 9032 participants without audiometric data. The portion excluded represents ∼36.7% of the population. Further prospective analysis, including follow-up data and speech discrimination, will be needed to evaluate a possible strong correlation between HbA1c level and hearing impairment.

In conclusion, HbA1c level was associated with hearing impairment in the nondiabetic participants of this study. Therefore, the participants with high HbA1c levels should be closely monitored for hearing impairment.

Footnotes

Acknowledgment

This work was supported by the 2014 Yeungnam University Research Grant.

Author Disclosure Statement

No conflicting financial interests exist.

References

Jennings

, Shaw

. Impact of hearing loss in the workplace: Raising questions about partnerships with professionals. Work, 2008; 30:289–295.

Walling

, Dickson

. Hearing loss in older adults. Am Fam Physician, 2012; 85:1150–1156.

Stevens

, Flaxman

, Brunskill

, et al. Global and regional hearing impairment prevalence: An analysis of 42 studies in 29 countries. Eur J Public Health, 2013; 23:146–152.

Lee

, Matthews

, Dubno

, et al. Longitudinal study of pure-tone thresholds in older person. Ear Hear, 2005; 26:1–11.

Van Eyken

, Van Camp

, Van Laer

. The complexity of age-related hearing impairment: Contributing environmental and genetic factors. Audiol Neurootol, 2007; 12:345–358.

Cruickshanks

, Nondahl

, Tweed

, et al. Education, occupation, noise exposure history and the 10-yr cumulative incidence of hearing impairment in older adults. Hear Res, 2010; 264:3–9.

Kang

, Choi

, Kim

, et al. Dietary vitamin intake correlates with hearing thresholds in the older population: The Korean National Health and Nutrition Examination Survey. Am J Clin Nutr, 2014; 99:1407–1413.

Simpson

, Matthews

, Dubno

. Lipid and C-reactive protein levels as risk factors for hearing loss in older adults. Otolaryngol Head Neck Surg, 2013; 148:664–670.

Akinpelu

, Mujica-Mota

, Daniel

. Is type 2 diabetes mellitus associated with alterations in hearing? A systematic review and meta-analysis. Laryngoscope, 2014; 124:767–776.

10.

American Diabetes Association. Standards of medical care in diabetes-2014. Diabetes Care, 2014; 37(Suppl 1):S14–S80.

11.

Cai

, Li

, Chen

, et al. Glycated hemoglobin level is significantly associated with the severity of coronary artery disease in non-diabetic adults. Lipids Health Dis, 2014; 13:181.

12.

Brownlee

, Cerami

, Vlassara

. Advanced products of nonenzymatic glycosylation and the pathogenesis of diabetic vascular disease. Diabetes Metab Rev, 1988; 4:437–451.

13.

Brownlee

. The pathobiology of diabetic complications: A unifying mechanism. Diabetes, 2005; 54:1615–1625.

14.

Selvin

, Steffes

, Zhu

, et al. Glycated hemoglobin, diabetes, and cardiovascular risk in nondiabetic adults. N Engl J Med, 2010; 362:800–811.

15.

Timmer

, Hoekstra

, Nijsten

, et al. Prognostic value of admission glycosylated hemoglobin and glucose in nondiabetic patients with ST-segment-elevation myocardial infarction treated with percutaneous coronary intervention. Circulation, 2011; 124:704–711.

16.

Silbernagel

, Grammer

, Winkelmann

, et al. Glycated hemoglobin predicts all-cause, cardiovascular, and cancer mortality in people without a history diabetes undergoing coronary angiography. Diabetes Care, 2011; 34:1355–1361.

17.

Park

, Yoon

, Won

, et al. Usefulness of glycated hemoglobin as diagnostic criteria for metabolic syndrome. J Korean Med Sci, 2012; 27:1057–1061.

18.

Rehm

, Room

, Monteiro

, et al. Alcohol use. In: Ezzati

, Lopez

, Rodgers

, Murray

CJL

(eds). Comparative Quantification of Health Risks: Global and Regional Burden of Disease Attributable to Selected Major Risk Factors. Geneva: World Health Organization; 2004: 959–1108.

19.

The Korea Centers for Disease Control and Prevention. Alcohol drinking. Accessed at http://health.mw.go.kr/HealthInfoArea/HealthInfo/View.do?idx=5800&subIdx=4&searchCate=&searchType=&searchKey=&pageNo=&category=1&category_code=301301306&dept=3&sortType=date&page=1&searchField=titleAndSummary&searchWord= on December 31, 2015.

20.

Grundy

, Cleeman

, Daniels

, et al. Diagnostic and management of the metabolic syndrome: An American Heart Association/National Heart, Lung, and Blood Institute Scientific Statement. Circulation, 2005; 112:2735–2752.

21.

Seo

, Rhee

, Park

, et al. Metabolic syndrome criteria as predictors of subclinical atherosclerosis based on the coronary calcium score. Korean J Intern Med, 2015; 30:73–81.

22.

Dilley

, Ganesan

, Deepa

, et al. Association of A1C with cardiovascular disease and metabolic syndrome in Asian Indians with normal glucose tolerance. Diabetes Care, 2007; 30:1527–1532.

23.

American Diabetes Association. Diagnosis and classification of diabetes mellitus. Diabetes Care, 2010; 33(Suppl 1):S62–S69.

24.

Osei

, Rhinesmith

, Gaillard

, et al. Is glycosylated hemoglobin A1c a surrogate for metabolic syndrome in nondiabetic, first-degree relatives of African-American patients with type 2 diabetes?. J Clin Endocrinol Metab, 2003; 88:4596–4601.

25.

Veeranna

, Ramesh

, Zalawadiya

, et al. Glycosylated hemoglobin and prevalent metabolic syndrome in nondiabetic multiethnic U.S. adults. Metab Syndr Relat Disord, 2011; 9:361–367.

26.

Janghorbani

, Amini

. Glycated hemoglobin as a predictor for metabolic syndrome in an Iranian population with normal glucose tolerance. Metab Syndr Relat Disord, 2012; 10:430–436.

27.

Pacala

, Yueh

. Hearing deficit in the older patients: “I didn't notice anything.”. JAMA, 2012; 307:1185–1194.

28.

Kim

, Park

, Kim do

, et al. Visceral adipose tissue is significantly associated with hearing thresholds in adult women. Clin Endocrinol (Oxf), 2014; 80:368–375.

29.

Shargorodsky

, Curhan

, Eavey

, et al. A prospective study of cardiovascular risk factors and incident hearing loss in men. Laryngoscope, 2010; 120:1887–1891.

30.

Hwang

, Wu

, Hsu

, et al. Association of central obesity with the severity and audiometric configurations of age-related hearing impairment. Obesity (Silver Spring), 2009; 17:1796–1801.

31.

Bainbridge

, Hoffman

, Cowie

. 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; 149:1–10.

32.

Austin

, Konrad-Martin

, Griest

, et al. Diabetes-related changes in hearing. Laryngoscope, 2009; 119:1788–1796.

33.

Fukushima

, Cureoglu

, Schachern

, et al. Effects of type 2 diabetes mellitus on cochlear structure in humans. Arch Otolaryngol Head Neck Surg, 2006; 132:934–938.

34.

Rosenhall

, Sundh

. Age-related hearing loss and blood pressure. Noise Health, 2006; 8:88–94.

35.

Evans

, Tonini

, Shope

, et al. Dyslipidemia and auditory function. Otol Neurotol, 2006; 27:609–614.

36.

Dalton

, Cruickshanks

, Klein

, et al. Association of NIDDM and hearing loss. Diabetes Care, 1998; 21:1540–1544.

37.

Michikawa

, Mizutari

, Saito

, et al. Glycosylated hemoglobin level is associated with hearing impairment in older Japanese: The Kurabuchi study. J Am Geriatr Soc, 2014; 62:1231–1237.