Noise-induced hearing loss among workers in textile factory

Introduction Noise-induced hearing loss ( NIHL) is one of the most common chronic health problems, which produces gradual progressive impairment and disturbs the patient’s quality of life. Industries in general and especially textile ones produce noise levels that, if intensi ed, can cause damage to worker’s hearing. Hence, premature hearing loss is a well-known outcome of noise exposure at work in industrial workers. Aims of the work The aim of the study was to assess hearing threshold levels among exposed patients and to compare them with the nonexposed control group and to evaluate other variables such as outer hair cell and medial olivocochlear bundle function represented with transient evoked otoacoustic emissions (TEOAEs) testing with and without suppression and to nd the relationship with duration of exposure if any. Patients and methods The study included 145 patients and same number of controls who were exposed to hazardous levels of noise for variable durations. All participants were subjected to audiological examination including basic audiological evaluation, otoacoustic emissions, and otoacoustic emissions with contralateral suppression ( CAS). Results Of the 145 patients (290 ears), 214 ears showed sensorineural hearing loss (73.8%). Sensorineural hearing loss was mild in 63 (43.44%) ears, moderate in 82 (56.55%) ears, and severe in 69 (47.58%) ears. TEOAEs were found in those with mild hearing loss with signi cantly lower amplitudes. Intact suppression (CAS) was signi cantly lower for the study group than for the control group. However, there was no signi cant difference in level of suppression for different duration of exposure. Conclusion A high incidence for NIHL is present among workers in textile factories, which indicate the mandatory use of different protective measures. CAS can be used as a predictor for the susceptibility to NIHL.


Introduction
About 30 million workers are exposed to hazardous noise, with an additional nine million exposed to solvents and metals that put them at risk for hearing loss (HL). Occupational HL is one of the most common occupational diseases. In all, 49% of male miners have HL by the age of 50 years. By the age of 60 years, this number goes up to 70% [1].
Th is problem is faced by a large sector of the working force; worldwide, about 16% of the disabling HL in adults (over four millions) results from occupational noise [2].
HL due to chronic noise exposure or noise-induced hearing loss (NIHL) has been associated with industry for many years [3]. Most of the western countries have their own regulations and rules for the protection of workers in noise-producing factories [4].

The
Occupational Safety and Health Administration ( OSHA) describes standards for occupational noise exposure in articles 1910.95 and 1926.52 [5]. OSHA states that an employer must implement hearing conservation programs for employees, if the noise level of the workplace is equal to or above 85 dB(A) for an averaged 8-h time period [6].
OSHA also states that exposure to impulsive or impact noise should not exceed 140 dB sound pressure level ( SPL) peak. Th e United States Department of Defense ( DoD) instruction 605512 has some diff erences from the OSHA 1910.95 standard -for example, OSHA 1910.95 uses a 5 dB SPL exchange rate and DoD instruction 605512 uses a 3 dB SPL exchange rate.

Introduction
Noise-induced hearing loss ( NIHL) is one of the most common chronic health problems, which produces gradual progressive impairment and disturbs the patient's quality of life. Industries in general and especially textile ones produce noise levels that, if intensi ed, can cause damage to worker's hearing. Hence, premature hearing loss is a well-known outcome of noise exposure at work in industrial workers.

Aims of the work
The aim of the study was to assess hearing threshold levels among exposed patients and to compare them with the nonexposed control group and to evaluate other variables such as outer hair cell and medial olivocochlear bundle function represented with transient evoked otoacoustic emissions (TEOAEs) testing with and without suppression and to nd the relationship with duration of exposure if any.

Patients and methods
The study included 145 patients and same number of controls who were exposed to hazardous levels of noise for variable durations. All participants were subjected to audiological examination including basic audiological evaluation, otoacoustic emissions, and otoacoustic emissions with contralateral suppression ( CAS).
Employees are required to wear hearing protection when it is identifi ed that their 8-h time weighted average is above the exposure action value of 90 dB SPL. If subsequent monitoring shows that 85 dB SPL is not surpassed for an 8-h time weighted average, the employee is no longer required to wear hearing protection [7].
Occupational health diseases generally are diffi cult to diagnose early because they often have a long latency period [8]. Hence, it is important to monitor worker's hearing for early diagnosing and preventing NIHL through a program of hearing conservation [9].
Th is study was designed to assess hearing threshold levels among exposed patients and to compare them with the nonexposed control group and to evaluate other variables such as outer hair cell and medial olivocochlear bundle function in this vulnerable group.

Patients and methods
Th is was a case-control study conducted among workers of Zefta textile factory (the exposed group) and among normal persons of the offi cial clerks of Zagazig University of Medicine and relatives of the patients after their consent (the control group). Evaluation was performed in Audiology Units, Otolaryngology Department, Zagazig University Hospitals, Egypt. Occupational HL among textile workers is 30% according to Farouk et al. (2000) [10]; hence, the sample size required was calculated using Epi-Info program (Epi Info™ Help Desk Centers for Disease Control and Prevention, Atlanta, USA) to be 145 workers.
All 145 patients were men with age range 35-50 years (43.4 ± 10.1) to avoid HL due to aging. Interviewing questionnaire for included patients was developed by the researchers acquiring about sociodemographic characteristics, audiological symptoms, and brief occupational history.
Setting for interviewing patients and detecting level of noise was at Zefta textile factory within both textile and weaving sectors for the exposed group and at administration building of Zagazig Faculty of Medicine for the nonexposed group. Audiological examination was performed at the Audiology Unit, ENT Department, Zagazig University.
Noise level detection was performed using EXTECH noise level meter model (Omni Controls, Tampa, Florida), 407764, with available range of 30-130 dB, A and C frequency weighting. Display is with dB(A).
All patients were clinically examined (otoscopical examination). Th ose who had suggestive history of HL, such as diabetes, hypertension, and family history of HL or head trauma, were excluded from the study. Any pathology of middle or external ear, such as impacted wax or otitis media, was detected and treated fi rst to avoid fallacies in audiological tests.
Tympanometry was performed in all patients using Tympanometer Ampalid 724 (Amplifon, Milano, Italy); only patients with normal middle ear pressure were involved in this study.
Audiometric assessment by standard pure-tone audiometry, using Audiometer Orbiter 922 (GN Otometrics, Taastrup, Denmark), was performed by the audiology consultants; bone and air conduction for both ears were individually performed from 250 up to 8000 Hz. HL was categorized according to Clark [11] into the following: Otoacoustic emissions testing was performed using ILO version 6 (Otodynamics Ltd, Hertfordshire, UK). Contralateral suppression (CAS) of otoacoustic emissions was performed using Amplaid 309 audiometer (Amplifon), delivering white noise of 70 dB SPL.

Statistical analysis
Data were analyzed using SPSS version 17 (SPSS Inc., Chicago, Illinois, USA). Comparison between the study and the control group was performed using the t-test for two independent means. Comparison among the subgroups of the study group was carried out using one-way analysis of variance test, and comparison for nonparametric data was carried out using the Fisher exact test.

Results
Noise levels in the factory departments using the portable noise level meter taken from diff erent spots with diff erent number of machines in work at the same time revealed that textile (new machines) noise level equals 95-104 dB(A), textile sections (old machines) noise level equals 98-109 dB(A), and in the weaving sections bed sheets and coverings using manual weaving machines noise level equals 80-84 dB(A).
Among 145 patients (290 ears), 214 ears showed sensorineural hearing loss (SNHL) (73.8%). SNHL was found to be nearly equal bilaterally ranging from mild to severe and mostly involving the high-frequency region in most of the aff ected patients. Involvement of other frequency regions was associated with longer time of exposure (Table 1).
Reviewing the degrees of severity of SNHL, SNHL was mild in 63 (43.44%) ears, moderate in 82 (56.55%) ears, and severe in 69 (47.58%) ears (Fig. 1). Pure-tone audiometry thresholds revealed statistically signifi cant diff erence between the study and control groups in the high-frequency region (Table 2). Similarly, statistically signifi cant diff erence was found between both the study and control groups with respect to speech discrimination scores (Table 3).
To study the factor of duration of exposure, the study group was divided into three subgroups: those exposed for less than 1 year (subgroup A) included 32 cases, those exposed for 1-5 years (subgroup B) included 60 cases, and those exposed for 5-10 years (subgroup C) included 53 cases (Figs. 2-4).
One-way analysis of variance test was performed to study the eff ect of duration of HL on the hearing threshold levels across diff erent frequencies. A statistically signifi cant diff erence was found among the three subgroups frequencies starting from 1 kHz and above (Table 4).
Once it is determined that diff erences exist among the means, post-hoc range tests and pairwise multiple comparisons can determine which means diff er. Range tests can identify homogeneous subsets of means that are not diff erent from each other. Pairwise multiple comparisons can test the diff erence between each pair of means and yield a matrix where asterisks indicate signifi cantly diff erent group means at an -level of 0.05 (Table 5). It showed signifi cant diff erence between each pair of means except between groups B and C at 1 kHz.
Regarding prevalence, 53.125% had HL in subgroup A and 70% in group B, whereas 90.567% had HL in group C (Table 6). Th is diff erence was found to be statistically signifi cant as proved by the Fisher exact test.

Figure 1
Degrees of severity of hearing loss.

Figure 2
TEOAEs testing for a patient with mild high-frequency hearing loss showing preserved OAEs. Comparing degree of HL among subgroups of noise exposure was found to be signifi cantly diff erent (Table 7).
TEOAE testing was performed for patients with mild HL (126 ears) and for the control group and revealed absent emissions in three patients in the study group and in 11 patients in the control group. Th ose with preserved TEOAEs had signifi cantly lower amplitudes in all frequencies ( Table 8).
CAS of otoacoustic emissions as a measure for intact medial olivocochlear bundle was performed. Th e incidence of intact suppression was signifi cantly higher for the study group than for the control group (Table 9). However, there was no signifi cant diff erence for diff erent duration of exposure (Table 10).

Discussion
NIHL is a growing health issue, with costly treatment and lost quality of life [12]. Exposure to high levels of noise can damage the inner ear and cause SNHL. Th ese losses may be temporary or permanent. Some studies using animal models have suggested that these temporary and permanent HLs may be a result of diff erent mechanisms rather than diff erent stages of HL [13]. Audiogram of a case showing bilateral mild to severe sloping sensorineural hearing loss.

Figure 3
TEOAEs testing for the same case after contralateral suppression.   In this study, 214 ears showed SNHL (73.8%). Th is is consistent with the typical clinical picture of chronic noise exposure, which is always bilateral. High-frequency losses rarely exceed 75 dB. Loss is always greater at the frequencies 3000-6000 Hz than at 500-2000 Hz. Loss is usually greatest at 4000 Hz, with the 4000-Hz notch often preserved even in advanced stages. In stable exposure conditions, losses at 3000, 4000, and 6000 Hz usually reach a maximum level in 10-15 years [14].
Th ere was a statistically higher pure-tone audiometry thresholds found among the subgroups of noise exposure (Table 4). Similarly, there was a signifi cant diff erence between each pair of means except between groups B and C at 1 kHz (Table 5).
Th is is supported by the principal characteristics of chronic, occupational NIHL as specifi ed by the American College of Occupational Medicine Noise and Hearing Conservation Committee, which described it as bilateral and symmetric SNHL, rarely produces a profound HL, does not progress once noise exposure is stopped, rate of HL decreases as the threshold increases, the 4 kHz frequency is the most severely aff ected, and the higher frequencies (3-6 kHz) are more aff ected than the lower frequencies (500 Hz to 2 kHz). Finally, maximum losses typically occur after 10-15 years of chronic exposure [15].    [15] found a positive correlation of hearing impairment with the duration of job when analyzed by linear regression analysis and correlation coeffi cient.
With respect to prevalence, the Fisher exact test was performed and revealed that the longer the duration of exposure by years, the higher the incidence of HL (Table 6). Similarly, permanent NIHL is related to the SPL and frequency distribution of the noise, the time pattern and duration of exposure, and individual susceptibility as stated by Johnson [16].
Speech discrimination scores were signifi cantly higher for the study group than the control group (Table 3).
Th is is consistent with elevated hearing thresholds. Th is is also consistent with the assumption that the most aff ected frequencies by noise are the speech frequencies. Speech aff ection is explained by the general agreement that the hearing level at 3000 Hz is related to the hearing and understanding of speech, particularly in the presence of noise. In 1978, in the summary of an investigation by Suter [17], it was reported that 'Correlation tests revealed that frequency combinations that included frequencies above 2000 Hz were signifi cantly better predictors of speech discrimination scores than the combination of 500, 1000, and 2000 Hz'.
Otoacoustic emission results revealed signifi cantly lower amplitudes at the high-frequency regions in the study group than the control group (Table 8).
As generally known, the outer hair cells in the inner ear are thought to be responsible for transient evoked otoacoustic emissions (EOAE) generation and they are one of the fi rst structures damaged by noise [18].
Healthy ears have emissions [19]; however, noisedamaged ears have fewer, smaller, or no emissions [20]. Th us, it seems plausible that, by monitoring EOAEs, we can indirectly monitor the health and status of the inner ear. Furthermore, in laboratory studies, temporary threshold shift (TTSs) and emission shifts after exposure to 105 dB SPL noise are negatively correlated (i.e. the increase in hearing level is correlated with a decrease in emission level) and follow the same recovery [21].
In contrast, in fi eld settings, a number of studies have also shown that emissions may undergo temporary changes after exposure to noise. However, changes in emissions are not necessarily associated with TTSs. In addition, in some cases, we have found that emissions do not shift with hearing decrements, and in other cases the emissions do not recover with hearing improvements [22]. Finally, they stated that they have noticed that, in people who are regularly exposed to hazardous levels of noise, their emissions may shift before their hearing shifts.
Finally, CAS of otoacoustic emission is a measure of state of olivocochlear bundle integrity. A statistically significant difference was found between the number of patients with intact suppression and the control group (Table 9). This reflects defective function of olivocochlear bundle in the noise-exposed group. When comparing the three subgroups of duration of noise exposure, there was a nonstatistically significant difference (Table 10). This reflects the absence of duration of noise exposure as a factor affecting olivocochlear bundle function among noise exposure patients, which may be a clue that defective olivocochlear bundle function is a cause for NIHL, not a result.
Other fi ndings were stated by Mariola and Kowalska [23] who found a signifi cant decrease of OAE in response to contralateral noise stimulation for the level of 70 dB SPL for a group of metal factory workers as compared with healthy nonexposed control group. However, eff erent suppression was weaker for the metal-exposed group compared with the other group. Finally, they concluded that OAEs, particularly distortion produce otoacoustic emissions and otoacoustic emissions (DPOAEs) CAS, could be a promising method for early identifi cation of auditory damage in workers at risk of developing industrial NIHL.

Conclusion
NIOSH recommends the use of quieter equipment, better work practices, and hearing protection devices and implementation of eff ective HL prevention programs to prevent NIHL in fi refi ghters. In addition, audiological assessment including CAS should be performed pre-employment to fi nd out vulnerable patients for NIHL.