Abstract
Cigarette smoking is a possible risk factor for hearing loss. However, the impact of simultaneous exposure to noise and smoke on hearing has remained controversial. This study investigated the combined effect of exposure to cigarette smoking and noise on hearing loss. Three groups of male Wistar rats (275 ± 25 g) were subjected to white noise (102 ± 0.5 dB), cigarette smoking (20 cigarettes), and both cigarette smoking and noise for 8 h and 10 days inside the exposure chamber. The control group was exposed to neither noise nor smoke. Distortion product otoacoustic emissions (DPOAE) were measured before any intervention, and it was repeated 1, 7, and 21 days after the last exposure. One-day postexposure to noise, cigarette smoking, and both cigarette smoking and noise, the mean of DPOAE amplitudes decreased significantly (p < 0.05) between, respectively, 5.7–30.7, 1.5–7.5, and 5.2–32.6 dB within the frequency range of 4620–9960. Temporal DPOAE change in rats exposed to noise or both cigarette smoking and noise was not significantly different (p > 0.05). DPOAE amplitudes returned to the baseline values in the group subjected to smoking 21 days postexposure. The most permanent change was observed in rats exposed to both cigarette smoking and noise. Accordingly, simultaneous subacute exposure to noise and cigarette smoking increases the effect of noise on permanent hearing loss. Therefore, smoking workers exposed to noise might be at a greater risk of developing hearing loss, and it is recommended that authorities in charge take note of this evidence.
Keywords
Introduction
Around half a billion people suffered from disabling hearing loss in 2015. Hearing loss has changed from the 11th leading cause of years lived with disability in 2010 to the fourth leading cause in 2015 (Kasmauski, 2017).
Exposure to noise has been identified as a major risk factor for hearing impairment (Lie et al., 2016). Noise damages cochlear cells not only mechanically but also metabolically. In metabolic mechanisms, free radicals generated by oxidative stress play an important role in noise-induced hearing injury. Free radicals including reactive oxygen species (ROS) and reactive nitrogen species (RNS) are molecules with unpaired electron, which are able to change electron arrangement of stable molecules (Kopke et al., 2007; Le Prell et al., 2007). Noise exposure results in the overriding of the mitochondria, excitotoxicity, ischemia or reperfusion, and the reduction of cochlea’s blood supply, leading in ROS generation. High level of noise exposure induces high demands of energy on the mitochondria, especially in outer hair cells (OHCs). This leads to generating superoxide as an unwanted byproduct that triggers higher levels of ROS generation in the cochlea. Superoxide accumulation is the result of oxygen deficiency due to a noise-induced reduction of blood flow in the cochlea or ischemia. The reperfusion of available oxygen as a result of ischemia generates higher levels of superoxide (Henderson et al., 2005). On the other hand, large amounts of glutamate are released into synapses of inner hair cells (IHCs) with afferent auditory nerve fibers (excitotoxicity) leading to the rise of ROS. The end result of free radical generation is sensory neural hearing loss due to cell death from a combination of necrosis and apoptosis (Henderson et al., 2005; Kopke et al., 2007).
In addition to noise, sensory cells and peripheral nerve endings of cochlea could be injured by occupational and environmental ototoxic chemical agents such as tobacco smoke consumed by approximately 1.3 billion of the world population (Pouryaghoub et al., 2007).
Several possible mechanisms may account for the relationship between smoking and hearing loss. By generating carboxyhemoglobin and exacerbating local hypoxia, smoking is a direct ototoxic (with nicotine effect) factor, which causes ischemia in the cochlea (Palmer et al., 2004). It may also disrupt the blood flow of cochlea by inducing vasospasm, altering blood viscosity, and resulting in blood vessel arteriosclerosis (Fabry et al., 2011; Ferrite and Santana, 2005; Johnson and Morata, 2010; Nomura et al., 2005a, 2005b; Young et al., 1987).
According to a review of 15 observational studies (10 cross-sectional, 4 cohort, and 1 case–control) on the association of smoking with hearing loss, the quality scores of nine studies with a positive association were comparable to those of the remaining six studies with an insignificant association (Nomura et al., 2005a). It has been suggested that smoking alone is not a risk factor of hearing impairment, yet combined with other factors such as elevated blood pressure, use of painkillers, or high cholesterol level, it can trigger hearing problems (Starck et al., 1999).
Cigarette smoking produces many ototoxic agents that interact with the noise effect on hearing loss (Johnson and Morata, 2010). The interaction of noise with organic solvents such as styrene, benzene, and toluene is synergistic, while carbon monoxide (CO) and hydrogen cyanide increase the possibility of noise-induced hearing loss (Campo et al., 2009; Fechter et al., 2000; Johnson et al., 1988; Khavanin et al., 2007; Lataye and Campo, 1997; Lataye et al., 2000; Liu et al., 2011; Motalebi et al., 2010; Rao and Fechter, 2000; Young et al., 1987). Heavy metals are also ototoxic agents that increase the noise effects on hearing loss (George, 2003). However, the combined effect of cigarette smoke agents could be different. For instance, selenium, as a cigarette smoke component, combined with lead ototoxicity can produce an antagonistic interaction that protects auditory function (Chuang et al., 2007). Although previous animal studies have mainly focused on the association of cigarette smoke ototoxic agents with hearing loss, evidence is limited regarding the combined effects of these agents. On the other hand, the effect of the interaction of cigarette smoke and noise on hearing loss has been inconclusive in human studies. In this regard, while some studies have reported an additive interaction (Mizoue et al., 2003; Uchida et al., 2005), another has observed a synergistic interaction (Ferrite and Santana, 2005).
In human studies, the results are affected by confounding factors, which might not be identified or controlled. Therefore, the combined effect of noise and cigarette smoke on hearing function needs to be investigated through various interventions such as exposing groups of experimental animals to moderate levels of noise, cigarette smoke, and their combination (Cappaert et al., 2000). The purpose of the current study is to provide a baseline for exploring the combined effect of noise and cigarette smoke on hearing in rats.
Methods
Animal preparation
This research was conducted in the Animal Research Center of Zahedan University of Medical Sciences (2015) based on the guidelines of the Declaration of Helsinki for animal welfare, care, and use. The study protocol was approved (code 8619, March 2014) by the Ethics Committee for Experimental Medicine of Tarbiat Modares University (Iran).
Prior to beginning the experiment, polypropylene cages (40 × 20 × 15 cm3) were prepared to house male Wistar rats (275 ± 25 g) for acclimatization at a temperature of 21–23°C with a relative humidity of 40–50% for 1 week. Lighting was on between 07:00 and 19:00 under a cycle of 12-h lightness or darkness. Except during the experimental sessions, all rats had access to food (rodent chow; Pars Animal Co., Iran) and water supply ad libitum.
Animals were randomly allocated to four groups of six rats each: control group without exposure, smoke-exposed group, noise-exposed group, and both noise- and smoke-exposed (noise–smoke) group. All rats were transferred to the experiment site and daily returned to their cages.
Whole-body exposure chamber
All experiments were performed inside a ventilated and custom-made sound exposure reverberant chamber. The chamber was fabricated from aluminum sheets with outside dimensions of 60 × 45 × 30 cm3. It was connected to two funnels 30-cm high from up and down with a slope of 52 °C and 30 °C, respectively. All sides of the chamber were made of thick plate glass, and the internal volume of the chamber was 177 L.
The chamber was tangentially connected to a custom-made smoking machine through a 1-cm diameter tube at the top of the chamber. The chamber and smoking machine specifications have been previously explained in more details (Habybabady et al., 2017).
During the experiments, each of the six rats was put in a separate mesh cage (15 × 13 × 17 cm3), and all six cages were carefully positioned on the horizontal parallel shelves inside the chamber.
In order to accelerate the experiment, two chambers located in two separate rooms were used. At the same time, the experiment started with the two groups of six rats inside the two chambers. After this intervention period was over, the chambers were used for the two next groups.
Cigarette smoke exposure
The rats were exposed to smoke for 8 h per day and 10 consecutive days inside the chamber. Main- and sidestream smoke was generated through the smoking machine based on the Federal Trade Commission (FTC) regime of a 2-s puff per minute with a volume of 35 mL and a flow of 1.05L/min (Shopland, 1996). The 8-h exposure consisted of repetitions of two cycles. First, the rats were passively exposed to cigarette smoke once for 8 min (2-s mainstream and 58-s sidestream) including eight puffs per cigarette. Second, fresh air was pumped instead of smoke for 17 min. The two cycles were repeated until a total of 20 cigarettes (9 ± 1 mg of tar, 0.8 ± 0.1 mg of nicotine per cigarette) were burned. The cigarette brand was TIR produced in Iran (Iran Tobacco Company, 2019). The concentration of total suspended particulate and CO, two of the most important indicators for measuring cigarette smoke exposure (Shopland, 1996), were respectively 42.7 ± 5.7 mg/m3 and 300 ppm in the chamber.
Noise exposure
Animals were exposed to a high-pass white noise centered at 8 kHz at 102 decibels sound pressure level (dB SPL) for 8 h per day and 10 consecutive days. Filtered Noise Generator (Timo Esser’s Audio Software, version 1.2) was used to produce noise. The noise was run by Cool Edit Pro v. 2.1 made by Syntrillium Software Corporation. The generated noise was intensified through an amplifier (model: Rock Jw-s317, China) and propagated by four loudspeakers (type: Micro Lab, model: HT 25 tweeter, Italy). The loudspeakers were placed at multiple locations inside the chamber 0.15 m above the animal cages to ensure stimulus uniformity by a maximum of 0.5 dB across the chamber. In order to monitor sound level inside the chamber, four holes with 1.2-cm diameter were drilled and tapped on every side of the chamber. This level was measured by a sound level meter (cel-450, type1, D; Casella-CEL company, UK) attached to an analyzer.
Distortion product otoacoustic emissions test
In order to measure frequency-dependent cochlear function of rats, distortion product otoacoustic emissions (DPOAE) test was performed. In humans, DPOAE measurements are used to diagnose early and differential damage to OHCs. After sending two tones to the normal ear, distortion components are produced at additional frequencies in the hair cell receptor potentials. This results in driving the OHCs’ biological motors, which can move the sensory epithelium at the distortion frequencies. The motion of the epithelium sends pressure waves back to the eardrum in the middle ear and produces DPOAEs (Kujawa and Liberman, 2009). Animal experiments have confirmed that DPOAE is altered by ototoxic drugs, noise exposure, and hypoxia (Emmerich et al., 2000).
As the presence of standing waves in the external meatus introduces measurement errors, the bandwidth of the cubic DPOAE responses (2f2-f1) was restricted to a frequency range of 4.0–10 kHz (referenced to f2) with a sample of 12 points per octave. The primary tone ratio f2/f1 was set to 1.21. To better identify cochlear dysfunction, a nonsymmetrical DPOAE protocol was used with unequal primary tone stimulus intensities (L1 = 60, L2 = 50 dB SPL) to evoke the responses (Hatzopoulos et al., 2002). The sound pressure was not allowed to exceed 42 dB. DPOAE measurements with a signal-to-noise ratio (SNR) of three or more were used for data analysis.
A mixture of 68 mg/kg ketamine and 8 mg/kg xylazine was injected intraperitoneally to anaesthetize the animals before the DPOAE test. A heating blanket was used to keep the body temperature between 37.5°C and 38.5 °C during the test procedure.
DPOAEs amplitudes were measured from the external left canal of the rat a day prior to noise exposure as well as 1, 7, and 21 days after the last session of the experiment. According to previous studies, maximum DPOAE level shifts were observed within 1–7 days and remained relatively constant 21 days after exposure. Therefore, permanent level shifts were observed 21 days after noise exposure (Fetoni et al., 2009; Jianzhong et al., 2014; Lu et al., 2014).
One researcher was responsible for taking all DPOAE tests, and the same DPOAE measurement was repeated three times for each rat at each test, and the average value was considered the DPOAE level. Prior to each test for every ear at each time, calibration was administered automatically.
Statistical analysis
DPOAE measurements were described as mean ± standard deviation (SD). The normal distribution of DPOAE responses was validated by Shapiro–Wilk test. One-way analysis of variance was used for comparing the four groups, and Tukey’s honest significant difference post hoc test was used for comparing every two groups when needed. The significance level was 0.05, and data were analyzed using SPSS (18).
Results
In all groups, the baseline mean of DPOAE amplitudes varied from 23.4 Hz to 32.5 Hz for a frequency range of 4620–9960 Hz, respectively (data not shown). At zero test frequency, there was any significant difference between the pre-exposure DPOAE measurements in the control and exposure groups (p > 0.05).
Temporary hearing changes (THC) (DPOAE difference between the baseline and one-day postexposure) in the experimental groups were more evident at higher frequencies. This was particularly noticeable in the noise–smoke and noise-exposed groups. More specifically, the rats exposed to both noise and smoke underwent a greater THC, compared to those subjected to only noise. The mean of THC significantly differed from that of the noise and noise–smoke groups at all frequency levels in the control group and at frequencies more than 8340 Hz in the smoke group (p < 0.05). Furthermore, except for the frequency of 4620 Hz, the mean of THC was significantly different between rats exposed to noise–smoke and those subjected to smoke alone (p < 0.05). Although the effect of noise–smoke exposure on THC exceeded that of noise alone at higher frequencies, the difference was not significant (p > 0.05; Figure 1).

Mean ± standard deviation of temporary DPOAEs level shift (DPOAE difference between the baseline and one-day postexposure) in terms of frequency in the noise, noise–smoke, smoke, and control groups.
Initial recovery (IR) (DPOAE difference between the 1st and 7th day postexposure) was greater at higher frequency levels in all experimental groups. The most IR was observed in the rats with noise and noise–smoke exposure, respectively. Compared to the control group, the smoke-exposed group experienced a significant IR at frequencies more than 9180 Hz, while the noise and noise–smoke-exposed groups underwent a significant IR at frequencies above 5040 Hz (p < 0.05). The mean of IR was significantly different between the smoke- and noise–smoke-exposed groups at frequency levels exceeding 5040 Hz (p < 0.05). However, such a difference was not found between the noise–smoke- and noise-exposed groups (p > 0.05; Table 1).
Mean ± standard deviation of initial recovery (DPOAE difference between the 1st and 7th day postexposure) in terms of frequency in the noise, noise–smoke, smoke and control groups.
aThe p value compares DPOAE mean between smoke and control group.
bThe p value compares DPOAE mean between noise and control group.
cThe p value compares DPOAE mean between noise–smoke and control group.
dThe p value compares DPOAE mean between noise–smoke and smoke group.
eThe p value compares DPOAE mean between noise–smoke and noise group.
The most subsequent recovery (SR) (DPOAE difference between the 7th and 21st day postexposure) occurred in the noise–smoke- and noise-exposed groups, respectively. The mean of SR was not significantly different in the four groups in all frequencies (p > 0.05; Table 2).
Mean ± standard deviation of subsequent recovery (DPOAE difference between the 7th and 21st day postexposure) in terms of frequency in the noise, noise–smoke, smoke and control groups.
aThe p value compares DPOAE mean between the four groups.
The mean of permanent hearing changes (PHC) (DPOAE difference between the baseline and 21 days postexposure) did not significantly differ between the control and smoke-exposed groups (p > 0.05). From low- to high-frequency levels, an increasing trend in the mean of PHC was revealed at noise- and noise–smoke-exposed groups, with the greater effect being associated with noise–smoke exposure. The control group had a significant difference from the noise–smoke group at all frequencies and the noise-exposed group at frequencies more than 4620 Hz (p < 0.05). Furthermore, at frequencies above 4620 Hz, the mean of PHC in the noise–smoke-exposed group was significantly greater than that in the smoke-exposed groups (p < 0.05). There was any significant difference between the noise- and noise–smoke-exposed groups (Figure 2).

Mean ± standard deviation of permanent DPOAEs level shift (DPOAE difference between the baseline and 21 days postexposure) in terms of frequency in the noise, noise–smoke, smoke, and control groups.
Discussion
In the current study, exposure to both noise and smoke and also noise alone resulted in a significant THC. However, the least total recovery (between the 1st and 21st day postexposure) was observed in animals exposed to both noise and smoke, especially at higher frequencies (6720–9960 Hz). In contrast, the smoke-exposed group experienced decreased THC and almost a full recovery 21 days after the last exposure.
THC was mostly affected at medium and high sound frequencies. This is in line with other studies that have suggested a more noticeable threshold at 6–12 kHz (Fetoni et al., 2009; Lorito et al., 2006). This frequency region corresponds to the 25–40% distance from the apex in cochlear map frequency of rats (Chen and Fechter, 2003). A greater degree of damage can be detected at higher frequencies of 16–24 kHz, corresponding to 60–70% distance from the apex (Chen and Fechter, 2003; Fetoni et al., 2013). This is because OHCs in the cochlear base are more susceptible to oxidative stress and functional changes at high frequencies. This intensified susceptibility may be due to the fact that the glutathione level in the cochlear base is less than its amount in the apex (Mom et al., 1999; Sha et al., 2001). There is also evidence that smoking increases the vulnerability of the most basal portion of cochlea, where higher frequencies are transduced (Fabry et al., 2011; Nakanishi et al., 2000; Paschoal and Azevedo, 2009).
In the current study, the mean of THC was not significantly different in animals exposed to both noise and smoke than those exposed to only noise. As a result, THC was significantly greater in the noise–smoke group than the noise- or smoke-exposed groups at all frequencies of 4, 8, 16, and 32 kHz (Ahn et al., 2011). In the current study, the mean of THC was greater in animals exposed to both noise and smoke than those exposed to only noise; nevertheless, the two groups were not significantly different. The difference between the two studies could be due to using different species, a lower level of cigarette smoke concentration, and a lower level of noise with a more limited frequency range.
IR was greatest in the noise exposed than the noise–smoke exposed animals, followed by a steady improvement until 21 days postexposure. Therefore, rats exposed to only noise experienced a more total recovery (between the 1st and 21st days postexposure). Another study also corroborates the highest recovery rate of DPOAE level to occur between 24 h and 7 days after exposure to noise (Lu et al., 2014). There is report of improvement of hearing loss 3 weeks after the last exposure to noise, but still a full recovery has not occurred in mice exposed to cigarette smoking followed by noise (Ahn et al., 2011). These results may point to some reversible structural changes associated with noise and smoke exposure. These changes include swelling of afferent nerve fibers and endings below the bases of inner hair cells, bent or collapsed pillars, pillar buckling resulting in the decupling of OHC stereocilia from the tectorial membrane, hair cell stimulation reduction, and partially collapsed supporting cells toward the baseline membrane causing height reduction of OHC (Abdel Hafez et al., 2014; Nordmann et al., 2000).
Unlike exposure to smoke alone, exposure to either noise or both noise and smoke resulted in PHC, with more changes occurring in the case of animals subjected to both noise and smoke. This was even more evident at higher frequencies. There was a mean DPOAE level shift of 7.1–11.6 and 11–15.8 at frequencies more than 6720 in the animals exposed to noise and both noise and smoke, respectively. Although the current research has not focused on histopathological examination in the Organ of Corti, previous studies have reported more OHCs damage with higher DPOAEs level shift, indicating clinical importance of the DPOAE level shift. A mean DPOAE level shift of about 16 dB resulted in 10% OHCs permanent loss at apical or middle turns of the rat’s cochlear (Ewert et al., 2012). An average shift of DPOAE iso-response levels of 8–12 dB resulted in 21% OHCs loss in rats (Cappaert et al., 2001), and almost all of OHCs in the third row were missed when DPOAE amplitudes shifted for 6–19 dB (Pouyatos et al., 2002). Furthermore, damage in the Organ of Corti begins with scattered OHC loss during the exposure and OHC degeneration continuous after the exposure. Secondary IHCs, nerve fibers and supporting cells develop afterward (Harding et al., 2002).
The mean DPOAE level shift was almost 5 dB more at higher frequencies in the animals exposed to both noise and smoke than noise only. Although this difference between the two groups was not statistically significant in the current study, it can be clinically important. It has been reported that a decrease of 2–8 dB in the mean DPOAE amplitudes results in greater prevalence of apoptotic cells in the OHC and IHC cells (Ocalan et al., 2014).
PHC results could be indicative of irreversible changes in the cochlear structure and antioxidant defense system after exposure to noise and noise–smoke. Meanwhile, exposure to smoke alone may lead to temporary and reversible biological and physiological changes allowing the endogenous defense system of the cochlea to recover. The glutathione level, oxidative stress markers, and lipid peroxidation products change after acute cigarette smoke exposure. Nevertheless, these parameters return to the normal range within 24 h, suggesting the protective mechanism of cells against oxidative stress from smoke (Van der Vaart et al., 2004). However, the combination of smoke and noise can potentiate PHC and interfere with hearing recovery. This could lead to hearing deterioration (Ahn et al., 2011), which implies the intensified effect of noise–smoke exposure compared with exposure to smoke alone.
Cigarette smoking can increase oxidative stress by producing ROS and RNS, leading to cell structure damage and impairing cellular function. Oxidants may be formed directly by being exposed to cigarette smoke ingredients, or they can be a result of cell inflammation caused by smoke-related oxidative stress (Van der Vaart et al., 2004). Furthermore, cigarette smoke increases blood viscosity and oxygen reduction, which in turn disrupts and damages hair cells in the cochlea (Ferrite and Santana, 2005; Van der Vaart et al., 2004).
There is evidence that noise and cigarette smoke can damage the cells through similar mechanisms. Compared with GSH level, the levels of malondialdehyde (MDA) and 8-isoprostates, as lipid peroxidation indices, have been reported to be greater in smokers (Mahapatra et al., 2008). As similar changes have also been observed for noise exposure (Ohinata et al., 2000a, 2000b), a combination of noise and smoke can disrupt the antioxidant defense system. Moreover, a significant increase in the hypoxia-inducible factor (HIF-1α) has been reported in the cells of the Organ of Corti when exposed to both noise and smoke than exposure to either noise or smoke alone (Ahn et al., 2011).
Nicotinic receptors have been detected on the outer hair cells of cochlea, suggesting the direct impact of nicotine on DPOAE amplitudes (Wild et al., 2005). As the rats exposed to smoke alone did not experience PHC, it is more likely that nicotine exposure makes outer hair cells more vulnerable to external damage caused by factors like noise (Ahn et al., 2011).
Footnotes
Acknowledgement
The authors would like to thank Mostafa Hoseinzadeh and Mohammad Hasan Rezaei for constructing the smoking machine.
Declaration of Conflicting Interests
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This work was partially supported financially by Tarbiat Modares University (Iran) as a PhD student thesis.
