Effect of occupational exposure to whole-body vibration and noise on Sex Hormone Levels: A Case Study in an Automobile Parts Manufacturing Plant

Background: The present study investigated the effects of exposure to noise and whole body vibration (WBV) on the levels of sex hormones in an automobile parts manufacturing plant. Methods: The level of workers' exposure (n=162) to each of the mentioned stressors, was measured through standard methods and for each person the time-weighted average (TWA) of exposure was calculated. In order to determine serum sex hormones (free testosterone, LH and FSH), blood samples were taken from all participants after 8-10 hours of fasting between 7-9 am and then the blood samples were analyzed by ELISA method. Results : In general, regarding testosterone as the main male sex hormone, only 49% of the participants were in the normal range. In total of three sections, the lowest mean testosterone levels was observed in the third exposure group (WBV>1.93 m/s 2 ; noise >92.69 dB) of the studied stressor, however, only the difference in testosterone levels between the three different groups of exposure to noise was statistically significant (P = 0.001). The relationship between demographic variables and levels of noise and WBV exposure with sex hormones was not linear and only the relationship between noise exposure and testosterone levels was statistically significant (R = -0.201, P = 0.013). Conclusion: According to the results of Logistic Regression, the WBV had the greatest effect on testosterone levels as the main male hormone. However, according to the results of the correlation test, only the relationship between noise exposure and testosterone levels is statistically significant.

and is currently a problem for human society [1]. Sex hormones play an important role in the male reproductive system. Sex hormones or gonadal steroids are hormones that interact with vertebrate androgen or estrogen receptors. The most important sex hormones are testosterone and estrogen which play a key role in the emergence of male and female sex characteristics [2].Testosterone is regulated in the thyroid and exists in both males and females but due to its levels being 50 times higher in males, it has been dubbed the "male sex hormone". This hormone is responsible for secondary sex characteristics in men and measuring its concentration can be helpful in evaluating the health of the gonads [3].
The secretion of sex hormones is regulated via the gonadotropin-releasing hormones which starts in the hypothalamus. This hormone controls the levels of two other hormones called the luteinizing hormone (LH) and the follicle-stimulating hormone (FSH) via the pituitary gland. [4] LH causes the activation, synthesis and secretion of testosterone by binding to receptors on the Leydig cells within the testes. FSH is responsible for the regulation of spermatogenesis in men [5].
Studies on the effects of occupational exposures on the reproductive system has expanded in recent decades [6]. A study conducted in the USA regarding the effects of occupational exposure on fertility and birth defects showed one in seven couples having problems with fertility, 7% of live births being underweight and 3% suffering from birth defects [7]. Evidence indicates that the quality of sperm has suffered a significant reduction in the past 50 years. Increasing environmental pollution in industrialized nations in recent decades has raised questions regarding the effects of harmful environmental and occupational hazards on sperm quality [8]. The main known causes of infertility in occupational environments are: exposure to heavy metals, solvents, insecticides or chemical agricultural products, noise, vibration, radiation and heat. Combined exposure to various harmful hazards, such as in welding or driving occupations is also a cause [9].
Noise is generally defined as "unwanted sound". Exposure to high levels of noise is reported as being among the top ten occupational hazards [10]. Noise is considered as a form of stress and can cause biochemical and physiological changes in humans and animals. The harmful effects of noise on the auditory system has been studied extensively, but very few studies have considered the effects of noise on other organs of the human body. Similar to other forms of stress, noise induced stress can cause elevated levels of stress hormones such as norepinephrine and cortisol [11]. Studies have shown that cortisol can activate the secretion of the leptin hormone which is an important hormone related to metabolism, reproduction and food consumption. Since noise exposure elevates corticosterone, it is possible that noise exposure leads to changes in serum leptin levels [12]. An in vivo study conducted on adult male mice showed that exposure to noise (100 dBA, 1 kHz) over a period of 60 days (3 hours/day) caused a reduction in testosterone levels [13]. Studies on rats also showed that exposure to noise (90 to 130 dBA) over a period of 50 days (12hours/day) caused a significant decrease in testosterone, FSH and LH levels along with negative effects on fertility [14].
Whole body vibration (WBV) as an another physical hazards exists in the majority of occupational environments as almost all terrestrial, airborne and naval machines used in industry, agriculture or logistics expose humans to vibration [15]. Vibration can increase the secretion of stimulating hormones and cause neurostimulation resulting in elevated blood pressure and pulse rate. Vibration can also disrupt in the secretion of certain bodily enzymes [16]. Although very few studies have been conducted regarding the direct effects of vibration exposure on reproductive indices, vibration has been reported as being an environmental stressor with the potential of being able to directly influence the reproductive system [17,18]. A prevalent inability to produce natural sperm among drivers occupied in industrial and agricultural sectors as compared to other occupations has been observed [19]. Similar results have been observed among taxi drivers and those exposed to whole body vibration [20,21]. In a cohort study, the relationship between occupational exposure to harmful physical factors and the quality of sperm among those seeking help from infertility clinics was assessed. This study found that those with a history of exposure to whole body vibration suffered from a change in their sperm parameters [22].
Generally, it can be said that so far, many in vivo and in vitro studies have been conducted regarding the independent effects of WBV and noise hazards on reproductive indices, however, few studies have evaluated these effects on workers in occupational environments by measuring the level of occupational exposure. A limited number of studies merely used retrospective evaluation of subjects with a history of exposure to these stressors (usually with questionnaires) without considering environmental monitoring [23]. Some studies only assessed reproductive disorders in employees regardless of specifying a particular stressor or its monitoring [24,25].
Based on a review of the relative literature, it can be stated that no study has attempted a concurrent evaluation of both the level of exposure to these stressors and the relative effect on sex hormones. The results of the present study can, as both a basic-applied research and a field study, determine the effects of each of these stressor on the reproductive indices among workers.
Thus, this study is an essential step in the prioritization of control procedures regarding these types of occupational hazards of the reproductive system.

Study Population
The study population consists of the entire workforce employed in three manufacturing sections of Aluminum, Cast Iron and Grinding in an automobile parts manufacturing plant which were exposed to studied environmental stressors (WBV and noise). Participants were admitted according to sample size calculations, meeting entry criteria and signing of consent forms. The minimum sample size for each estimate (exposure type) was determined to be 121 participants, though after considering entry criteria, 162 participants were eventually enrolled in the study. The participants were all male, between 20 to 50 years of age [26] and with two or more years of employment [27]. Participants who had used steroids, testosterone, prednisolone, anti-oxidants [28]such as selenium, vitamin C, E or B supplements [29] and body-building supplements [30] were not enrolled. Those with a family history of infertility or organic disorders affecting fertility such as diabetes, kidney disorders, angina pectoris, heart disorders, arterial hypertension, pituitary disorders or a chronic pulmonary obstruction disorder were not enrolled. Also, those with testicular infections, orchitis, varicocele [31], history of chemotherapy or radiotherapy [32]and those with abnormal hearing (according to medical records) were not enrolled.
Relevant participant background information was obtained from demographic questionnaires and also by referring to the participants' medical records. Participants who failed to complete the questionnaire or those who wished to leave the program at any stage were dismissed. All participants were required to sign informed consent forms approved by the ethics committee of the Tehran University of Medical Sciences (ethics code: IR.TUMS.SPH.REC.1398.297).
The research environment in the studied industry included Aluminum, Cast Iron and Grinding sections [33] . Their process flowchart has been shown in Figure 1.All participants depending on their job, had appropriate personal protective equipment such as earplug, goggles, helmets, safety shoes, and protective clothing.

Environmental Measurement
In order to measure the exposure level to occupational noise, a calibrated noise dosimeter (Casella, Cell-320; USA) was used. Since the workers' exposure pattern had a constant cycle, the short-term dosimetry method was used. For this, the dosimeter microphone was placed 10 to 30 cm from the participant's ear and attached to their collar [34].

Hormone measurements
Blood samples were taken from each participant between 7 and 9 AM after 8 to 10 hours of hunger [38]. Samples were taken by laboratory professionals at the studied plant and were immediately taken to the laboratory under controlled conditions (ice box) and kept at -20 °C.
According to the kit manufacturer's guidelines, free serum testosterone levels (AccuBind ELISA test system, Monobind Inc., USA) and LH/FSH levels (Padtan Gostar Isar Inc., Iran) were measured using the Enzyme-Linked Immunosorbent Assay (ELISA) method and read using a

Data Analysis
Data analysis was performed using the SPSS v22 software (Chicago Il, USA). Descriptive statistics such as percentage, percentile, minimum, maximum, average, median and standard deviation were used to present demographic data, environmental monitoring results and hormone analysis. The Shapiro-Wilk Test was used to determine data distribution normality. The one-way analysis of variance (ANOVA) or corresponding non parametric test was used to compare mean hormone levels at different exposure levels. Partial correlation tests were used in order to remove or control influencing factors (background variables) on hormone levels. Also, in order to analyze the effect of, or find the relationship between, occupational exposure and sexual hormone levels by sdjusting the effect of other variables, linear (or nonlinear) regression was used. Logistic regression was used to analyze the relationship between the studied variables. A probability value (p-value) of 0.05 was considered as statistically significant. The participants were classified into separate groups based on the level of noise and WBV exposure. The cutoff point for separating the exposure groups was the 33 rd and 66 th percentile. highest percentage of exposure to WBV (49%) and to noise (80%) was observed in the first and second groups, respectively. In the Cast Iron section, highest percentage of exposure to WBV (46%) and to noise (72%) was found in the third and first groups, respectively. In the Grinding section, highest percentage of exposure to WBV (49%) and to noise (85%) was found in the third and third groups, respectively. Table 2. Level of exposure to noise and WBV among participants.    It was observed that among the two age groups, lowest serum testosterone levels (6.58 pg/ml ±2.16 for those with 20 to 39 years of age and 6.70 pg/ml ±2.87 for those above 39 years of age)

RESULTS
were observed under high levels of noise exposure (>92.69 dBA), although only those that were 20 to 39 years old ( Figure 2) showed a statistically significant difference in hormone levels within the three noise exposure groups (P=0.001).   The results of the correlation test and subsequent scatter plots showed no linear relation between demographic characteristics, exposure levels and sex hormone levels. Only the relation between noise exposure and testosterone levels (Fig.3) was statistically significant (R=-0.201, P=0.013). For this, a simple logistic regression was initially performed and then, for the multiple logistic regression, variables with sig>0.3 were removed from the model. The results are presented in Table 6 and 7. Table 6 shows the results of the logistic regression of the demographic and exposure variables with testosterone levels for those aged 20 to 39. Based on the results obtained, no significant relationship was found between input variables in the simple logistic regression analysis. The WBV had the most effect on testosterone levels, so for every one unit increase in level of WBV exposure, the probability of testosterone levels falling below 9.2 pg/ml was increased by 30%.
Among the exposure variables, noise had the smallest confidence interval range and is close to significant levels. For every one unit increase in noise level, the probability of testosterone levels falling below 9.2 pg/ml was increased by 5%. Among the demographic variables, marital status had the most effect on testosterone levels as married participants were 2.3 times more probable to have a lower than 9.2 pg/ml testosterone level. Again, with this range, a confidence level of 95% is rather large for this exposure variable. Considering the confidence interval, BMI was closer to meaningful levels among the demographic variables.  Table 7 shows the results of the logistic regression of the demographic and exposure variables with testosterone levels for those aged 40 and above. Based on the results obtained, no significant relationship was found between input variables and the results of the simple logistic regression analysis, although WBV had the most effect on testosterone levels. For every one unit increase in WBV acceleration, the probability of testosterone levels falling below 6.1 pg/ml was increased by 18%. Among the exposure variables, noise is closer to significant levels considering the confidence interval. Among the demographic variables, employment duration was closer to meaningful levels considering the confidence interval. For every single unit increase in employment duration, the probability of testosterone levels falling below 6.1 pg/ml was increased by 7%. Table 7. Results of the logistic regression test for demographic and exposure variables with testosterone levels for participants aged 40 and above.

DISCUSSION
Knowledge regarding the effects of occupational hazardous agents on male fertility is limited.
Workers around the world are increasingly exposed to hazardous physical and chemical hazards which can affect the male reproductive system. Expanding the knowledge regarding types of exposure that effect male fertility is especially important for preserving the health of workers and their family life since unlike cancer, they become apparent shortly before imposing damage.
Therefore, protecting workers against exposures that effect the reproductive system can prevent other negative effects on their overall health [40].
The present study investigated the effects of exposure to noise and WBV on the levels of sex hormones influential in the performance of the male reproductive system in an automobile parts manufacturing plant. According to the results, overall, testosterone levels were normal among 49% of participants with 51% having lower than normal levels of testosterone. Sex hormone levels can be affected by genes, age, background illnesses, diet, medication, or drug use as well as exposure to occupational hazardous agents. Although certain entry criteria were set for enrollment and efforts were made to control influential factors, it is impossible to account for and control all influential factors which is a common problem with field studies. Even so, the theories tested in these types of studies are valuable since they reflect real world situations [33].

Noise
According to the results of the present study, mean exposure level to noise in all manufacturing sections was above TLV and the difference between them was statistically meaningful. The difference between all three exposure groups in the various manufacturing sections was also statistically meaningful with the highest average exposure to noise being observed in the dBA. In casting, repeated impacts of the chisel causes vibration at resonance frequencies and produces loud noise [41]. The lowest average testosterone levels were observed in the Grinding section (7.01 pg/ml ±3.09) followed by the Cast Iron section (7.76 pg/ml) and the Aluminum section (8.19 pg/ml ±2.96) but the differences between them were not statistically significant.
The highest average noise exposure for the third exposure group (99.95 dBA ±3.15) was observed in the Grinding section with 85% of participants being exposed to higher than noise 92.69 dBA. Around 65% of participants in this section had lower than normal testosterone levels.
Serum testosterone levels were measured under various exposure scenarios for those between 20 to 39 years of age and also those above 39 years of age. The results showed that in both age groups, lowest testosterone levels belonged to those exposed to noise >96.69 dBA with the differences in testosterone levels among those 20 to 39 years of age being statistically meaningful. Correlation tests and scatter plots determined that among the demographic and exposure variables, only noise exposure and testosterone levels had a statistically significant relationship. According to the simple logistic regression, among those 20 to 39 years of age, noise had the smallest confidence interval range and was close to meaningful levels. For each one unit increase in noise levels, the probability of testosterone levels falling below 9.2 pg/ml was increased by 5%.
Chamkoory et al. conducted a study in 2016 titled "Effect of noise pollution on the hormonal and semen analysis parameters in industrial workers of Bushehr" where a group of 27 men were continually exposed to 107 to 119 dBA noise while another group of 27 men worked in a quite environment. The men were then sent to fertility clinics where blood samples were taken in order to determine levels of adrenocorticotropic (ACTH) hormones, cortisol, testosterone, prolactin, follicle stimulating hormone (FSH), luteinizing hormone (LH), thyroxine hormones (T3, T4) and thyroid stimulating hormones (TSH). The results of that study showed that noise stress had significantly reduced the concentration of testosterone, prolactin, LH, FSH, thyroxin hormones and TSH while causing a meaningful increase in ACTH and cortisol levels [42]. Results of the present study agrees with Chamkoory et al, in the case of reduced testosterone, LH and FSH levels in those exposed to high levels of noise.
In a review study conducted in 2016 by Nadry et al. the effects of noise on male fertility (human and lab animals) was investigated [43]. The study concluded that exposure to noise can affect testicular weight, sperm parameters (sperm count, vitality, mobility and morphology), sex hormone levels, testicular tissue and oxidative stress levels which in turn can be influential in male infertility. The present study [43] also found noise to be the main exposure variables that affects sex hormone levels.
An in vivo study showed that exposure to noise (100 dBA) during a 50-day period (3 hours per day) caused reduced testosterone levels in adult male mice [44]. Exposure to noise (90 to 130 dBA) during a 50-day period (12 hours per day) also caused a considerable reduction in testosterone, FSH, LH and Lutropin (LH) hormone levels in lab rats which had negative effects on fertility [45].
The results of Farzadnia et al. (2016) showed reduced testosterone levels in lab rats when exposed to noise at 115 dBA while increasing adrenocorticotropic (ACTH) hormone and cortisol levels in lab rats [46]. Testicular histology showed reduced average seminiferous tubule diameter and reduced thickness of the germinal epithelium compared to the control group. These in vivo results are in agreement with the findings of the present field study. sperm quality parameters such as sperm concentration, mobility and sperm chromatin structure were investigated using occupational questionnaires. Their results showed a meaningful relation between the existence of occupational noise and reduced sperm motility and increased DNA breakup [47].
Considering the relevant literature in this regard, it seems that among all physical hazardous agents, the effects of noise exposure on male fertility are better understood. Noise exposure is considered to be among the top ten occupational hazards and is a known stressor that can cause physiological and biochemical changes in the human body. Certain studies report that exposure to high levels of noise cause reductions in testosterone levels, sperm motility, average seminiferous tubule diameter and thickness of the germinal epithelium while increasing DNA breakup [14,44,47]. These studies state that the mechanism of effect is dependent on the increase in stress hormone levels such as corticosterone and norepinephrine due to noise exposure and the resulting changes in the secretion of the leptin hormone which is an important hormone in metabolism and reproduction [12].

Vibration
The results of the present study show that mean exposure to whole body vibration was above  [22]. In this study, a total of 501 couples who had stopped using contraceptives and were willing to conceive were observed for a period of 1 year. During this time, 473 men (94% of male participants) provided one sperm sample while 80% provided two sperm samples. The type of occupational exposure was determined via questionnaires. Based on the self-reported questionnaires, 23% of participants had been exposed to WBV and 27% had been exposed to noise in their occupational environments. Regression analysis had revealed that average ejaculate volume, overall sperm count and the DNA fragmentation index (DFI) was lower in the vibration exposure group compared to the control group though this was not statistically meaningful. Average ejaculate concentration, overall sperm count, abnormal sperm morphology and DFI was also higher in the sound exposure group compared to the control group though this was not statistically meaningful [22].
In a review article titled "Effect of vibrations on male reproductive system and function" Penkov and Tzetkove state that based on experimental and clinical studies, exposure to vibration results in reduced sperm count, lower spermatogenesis index, shrinking of the testes, changes in motility, chronic venous stagnation, dystrophic alterations in tubuli contorti, changes in ScDH, ATP-ase, LDH, GL-6-PDH activity and structural damage to the germ cell [48]. They also point out that oligozoospermia, asthenozoospermia teratozoospermia is prevalent among workers exposed to vibration while sexual disorders have also been observed [48].
In the aformentioned study by Jurewicz et al. (2014), it was shown that among the participants, 17% had been exposed to vibration but no significant relationship was found between sperm parameters and occupational exposure to vibration [47]. A cohort study investigated the relationship between occupational exposure to hazardous physical agents and sperm quality among clients seeking help from infertility clinics. Their results showed changes in sperm parameters in those who had a history of WBV exposure [22].
As is apparent, considering the lack of adequate information and the existence of contradictory results from other studies, it is difficult to arrive at a definitive conclusion, however, there is no doubt that that chronic exposure to WBV, especially in occupations involving working while sitting down, can have negative effects on the male reproductive system [47]. Despite its limitations, the presents study has shown how exposure to WBV can effect sex hormone levels.
Vibration has been identified as an environmental stressor with the potential to affect the male reproductive system [17,18] and this is due to the prevalence of disorders related to sperm production among drivers occupied in industry and agriculture and also among taxi drivers compared to other occupations [19][20][21].
Despite the limited available scientific sources, it seems that the possible effect mechanism of vibration exposure on the reproductive system is related to changes in hormonal and enzymatic levels, vascular disruptions within the testicular tissue, entropy and temperature changes [49]. It must be noted that in vitro studies involving short term exposure to vibration have reported increased sperm motility without accompanying changes in sperm count or morphology [50].
In total, considering the wide range of normal testosterone levels, daily hormone fluctuations and influencing parameters such as psychological stressors, quality of sleep, background diseases and etc., it is suggested that a follow-up study be conducted from the beginning of employment with a large enough sample size and a control group in order to definitively evaluate the effects of these physical hazards on sex hormone levels and reproductive indices. Due to a limited budget, the present study was conducted using cross sectional measurements along with exposure group categorizations. Also due to cultural and religious reasons, sperm samples were not taken from participants and sex hormones were only measured using blood samples in the preliminary phase of the study. This does not however reduce the value of the present study as it is hard to find any study that investigates the effects of physical stressors on reproductive indices in occupational environments while also attempting to calculate the amount of occupational exposure received.

CONCLUSION
The aim of the presents study was to evaluate the effects of exposure to noise and WBV on sex

Availability of data and materials
Not applicable.

Consent for publication
Not applicable.

Ethics approval
Ethical approval for this study was obtained from School of Public         * Mean differences of testosterone levels between categorized values of exposure (20-39 years old) ** Mean differences of testosterone levels between categorized values of exposure for (40 year old and above) *** Mean differences of testosterone levels between different manufacturing sections (20-39 years old) **** Mean differences of testosterone levels between different manufacturing sections (40 year old and above) -Normal range of testosterone for 20-39 years of age is 9.2-34.6 pg/ml -Normal range of testosterone for 40 years and above is 6.1-30.3 pg/ml  Grindindg part •pieces are inspected before they are transferred to it and are removed from the cycle if there is a problem. •is to separate the additional components from the produced pieces