Environmental Pollution 187 (2014) 145e152
Contents lists available at ScienceDirect
Environmental Pollution journal homepage: www.elsevier.com/locate/envpol
Air pollution and decreased semen quality: A comparative study of Chongqing urban and rural areas Niya Zhou a,1, Zhihong Cui a,1, Sanming Yang b, Xue Han a, Gangcai Chen b, Ziyuan Zhou a, f, Chongzhi Zhai b, Mingfu Ma c, d, Lianbing Li c, d, Min Cai d, Yafei Li e, Lin Ao a, Weiqun Shu f, Jinyi Liu a, *, Jia Cao a, * a
Institute of Toxicology, College of Preventive Medicine, Third Military Medical University, Chongqing 400038, People’s Republic of China Chongqing Environmental Science Research Institute, Chongqing, People’s Republic of China Key Laboratory of Birth Defects and Reproductive Health of National Health and Family Planning Commission, Chongqing, People’s Republic of China d Chongqing Population and Family Planning Science and Technology Research Institute, Chongqing, People’s Republic of China e Department of Epidemiology, College of Preventive Medicine, Third Military Medical University, Chongqing, People’s Republic of China f Department of Environmental Hygiene, College of Preventive Medicine, Third Military Medical University, Chongqing, People’s Republic of China b c
a r t i c l e i n f o
a b s t r a c t
Article history: Received 1 October 2013 Received in revised form 27 December 2013 Accepted 31 December 2013
To investigate the association and effects of air pollution level on male semen quality in urban and rural areas, this study examines the outdoor concentrations of particulate matter (PM10), sulfur dioxide (SO2), nitrous dioxide (NO2) and semen quality outcomes for 1346 volunteers in both urban and rural areas in Chongqing, China. We found the urban area has a higher pollution level than the rural area, contrasted with better semen quality in the rural residents, especially for sperm morphology and computer assistant semen analysis (CASA) motility parameters. A multivariate linear regression analysis demonstrates that concentrations of PM10, SO2, and NO2 significantly and negatively are associated with normal sperm morphology percentage (P < 0.001) and sperm kinetic parameters. In conclusion, exposure to higher concentrations of PM10, SO2, and NO2 in urban ambient air may account for worse semen quality in urban males. Ó 2014 Elsevier Ltd. All rights reserved.
Keywords: PM10 SO2 NO2 Regional variation Semen quality Urban area Rural area
1. Introduction During the last two decades, many studies have shown that the various chemical and physical substances in ambient air may have major adverse effects on male fertility (Adamopoulos et al., 1996; Hammoud et al., 2010; Hansen et al., 2010; Sram, 1999). Particulate matter (PM) can carry multiple trace elements and
Abbreviations: PM10, particulate matters with diameters less than 10 mm; SO2, sulfur dioxide; NO2, nitrogen dioxide; WHO, World Health Organization; CASA, computer assistant semen analysis; VCL, curvilinear velocity; VSL, straight-line velocity; VAP, average path velocity; BCF, beat cross frequency; ALH, amplitude of lateral head displacement; LIN, motile sperm linearity; STR, motile sperm straightness; BMI, body mass index; SPSS, statistical package for the social sciences version; PAH, polycyclic aromatic hydrocarbons; PCBs, polychlorinated biphenyls; ER, estrogen receptor. * Corresponding authors. E-mail addresses: [email protected] (J. Liu), [email protected] (J. Cao). 1 These authors contribute equally to this work. 0269-7491/$ e see front matter Ó 2014 Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.envpol.2013.12.030
polycyclic aromatic hydrocarbons (PAHs) that can produce harmful effects on reproductive health. Studies have suggested that certain PAHs or their metabolites can interact with the estrogen receptor (ER) and initiate ER signaling pathways in vitro and in vivo, and exert endocrine and developmental toxicity in males (Ford and Huggins, 1963; MacKenzie and Angevine, 1981; Vinggaard et al., 2000). Similar effects of SO2 and NO2 on the male reproductive system also have been reported. An animal toxicological study has provided evidence that SO2 is a toxin for the reproductive system of mammals, and exposure to SO2 can cause oxidative damage to the testicles of male mice (Meng and Bai, 2004). Dejmek et al. (2000) reported that the association between SO2 and fecundability was more significant in those couples who lived close to a district highly polluted with SO2. Boggia et al. (2009) found subjects occupationally exposed to NO2 have a significant lower sperm total motility than do unexposed workers. However, the biological effects of these air pollutants on the reproductive system have not yet been confirmed in the general population.
146
N. Zhou et al. / Environmental Pollution 187 (2014) 145e152
Fig. 1. Map of study area. Two urban areas (Nanan and Shapingba) and four rural areas (Wanzhou, Wushan, Yunyang, and Zhongxian) were chosen to represent Chongqing in the Three Gorges Reservoir Region. The two urban areas are located in the central part of Chongqing City that has the highest population density, while the four rural areas are not centered around the city and have a lower population density, and typically much of the land there is devoted to agriculture.
Rapid economic development and urbanization in large cities has been reported to lead to severe air pollution in urban areas (Liu et al., 2008; Guéguen et al., 2011; Ma et al., 2011). Because of the increasing traffic, industrial activity, and energy consumption in urban areas, higher emissions of air pollutants usually occur there. Outdoor nitrogen dioxide (NO2) exposure has been used extensively to characterize the exposure to traffic-related air pollutants. SO2 and chemical substances like polychlorinated biphenyls (PCBs) and PAHs are also emitted by industries and other human activities, such as use of vehicles and residential heating facilities (Cooper et al., 1996; Biterna and Voutsa, 2005). Regional variability of air pollutant concentration in urban and rural areas has also been identified (Hoek et al., 2002; Lewné et al., 2004; Van der Zee et al., 1998; Zhu et al., 2002). Further, Pant et al. (2008) reported that the urban population in India was found to have a statistically and significantly lower percentage of sperm motility than the rural population in India and also found that the urban population had statistically significant higher levels of phthalate esters than did the rural areas. Chongqing is a petrochemical, iron and steel, aluminum, and manufacturing industrial center in the southwest of China with more than 30 million residents. For decades, Chongqing City has suffered from substantial air pollution from road and river transport and the presence of different industrial activities along the Yangzi River. These include steel plants, chemical and domestic waste incinerators, and other chemical industries. Venners et al. (2003) used a generalized additive model to estimate the associations of mean daily SO2 and PM2.5 for daily mortality in Chongqing. They found daily ambient SO2 concentration to be positively and
significantly associated with population risks for cardiovascular and respiratory mortality. Our earlier epidemiological study found that 61.1% of male subjects from the general population in the Chongqing area had at least one semen quality parameter below the WHO referenced value (Li et al., 2009). Further, PAHs exposure appeared to cause sperm DNA damage and fertility interference in males (Han et al., 2011). In particular, we are interested in male reproductive health in this region, as its quality may relate to ambient air pollution. In the present study, we chose six sites that included two highly polluted urban districts and four relatively cleaner rural counties. We investigated the distributions of three ambient air pollutants, PM10, SO2 and NO2, in the urban and rural areas in Chongqing and explored the association and effect of air pollution on male semen quality in both regions. 2. Materials and methods 2.1. Study design and subject recruitment Chongqing is the largest of the four direct-controlled municipalities in the People’s Republic of China, and occupies the major reservoir areas known as the Three Gorges Dam project (Three Gorges Reservoir Region). The boundaries of the Chongqing municipality reach much farther into the city’s hinterland than do those of the other three municipalities (Beijing, Shanghai and Tianjin), and most of the administrative area is rural (Fig. 1). This investigation was carried out in 2007. The study proposal was reviewed and approved by the Ethical Committee of the Third Military Medical University. We also worked with the Chongqing Family Planning Commission, the Chongqing Institute of Science and Technology for Population and Family Planning, and six local Family Planning Institutions to recruit volunteers. The inclusion criteria had the following characteristics: Males age 20e40 and permanent residents of the Chongqing area. The exclusion criteria included any reproductive or urological diseases diagnosed by
N. Zhou et al. / Environmental Pollution 187 (2014) 145e152 andrological examination, other known reproductive disorders, and/or an identifiable history of infertility, and prior vasoligation or chronic disease. All study participants were asked to sign an informed consent form to be able to participate. A total of 1976 volunteers were recruited. Of these, 630 were excluded from the analysis for reproductive disorders or other chronic diseases (n ¼ 80), missing or unknown duration of abstinence or reported duration of abstinence of 7 days (n ¼ 210), failing to collect the semen samples (n ¼ 226), and spillage of the sample semen (n ¼ 114). A group of 1346 healthy and eligible volunteers finally agreed to and completed all steps in the study. 2.2. Semen collection and the analyses The methods for semen collection and its analyses are described in detail previously (Li et al., 2009). Administration of the questionnaire and the physical examination, semen collection, and all analyses were carried out at a local Reproductive Health Center at each study site. All participants were asked to be abstinent for 2e7 days before contributing a semen sample. Conventional semen parameters for the original raw samples, including appearance, viscosity, liquefaction time, pH value, semen volume, sperm concentration and motion parameters, were measured using World Health Organization (WHO) guidelines (WHO, 1999). Sperm motion analysis was performed using a computer-aided semen analysis system (CASA, WLJY 9000, Weili New Century Science & Tech Dev., Beijing, China). According to the WHO reference, percent of motile sperm was scored using category A (rapid progressive motility), category B (slow progressive motility), category C (non-progressive motility), and category D (immotility). Sperm motion analysis was conducted for two types of motility: progressive motility (PR, categories A þ B) and non-progressive motility (NP, categories C). CASA outcomes included curvilinear velocity (VCL), straight-line velocity (VSL), average path velocity (VAP), beat cross frequency (BCF) and amplitude of lateral head displacement (ALH). The calculated parameters were linearity (LIN ¼ VSL/VCL 100) and straightness (STR¼VSL/VCL). According to the findings of previous studies (Meeker et al., 2004; Liu et al., 2012), some of the CASA parameters strongly correlated because different aspects of the same movement were described. Therefore, we chose progression (VSL), vigor (VCL), and swimming pattern (LIN) for this statistical analysis. We used the “feathering” method to perform the analysis for sperm morphology assessment. Two slides were utilized for each fresh semen sample. The air-dried smear was fixed in 96% ethanol, stained with a modified Papanicolaou stain, and then analyzed according to the WHO criteria. To reduce the variation in the assessment of sperm characteristics, two technicians performed all analyses of semen quality for all six centers using the same apparatus. These two technicians were well trained in semen analysis and participated in the Continuous Quality Control System (an external quality control system established using the WHO guidelines) under supervision of the Chongqing Science and Technology Commission. 2.3. Air pollution data and exposure assessment Air pollution data were obtained from the Institute of Environmental Science and Technology of Chongqing. Outdoor PM10, SO2, and NO2 concentrations in 2007 were collected from 6 ambient air quality monitoring stations in the studied districts or from counties that belonged to the national air pollutant detection network. The data for PM10, SO2 and NO2 represented a 24 h average and were recorded daily for the Wanzhou, Shapingba, and Nanan districts and every 3 days for the Zhongxian, Wushan, and Yunyang counties, respectively. The daily (24-h) average concentrations of PM10 were measured using the tapered element oscillating microbalance method (China State Environmental Protection Agency, 2000). According to the technical guidelines of the Chinese government, the location of these monitoring stations must not be in the direct vicinity of traffic intersections and not influenced by local pollution sources. They should also be a sufficient distance from any other emitting sources, such as coal-, waste-, or oil-burning boilers; furnaces, and incinerators. To analyze the association between the estimated air pollutants and semen quality outcomes, we employed average exposure data to the pollutants within an exposure period of 90 days before undertaking a semen sampling from each subject. An exposure period of approximately 90 days is generally accepted as a sufficient duration for detecting effects on any stage of spermatogenesis when using semen measures as biological endpoints (Heller and Clermont, 1964; Selevan et al., 2000; Hansen et al., 2010). 2.4. Statistical analysis Demographic information, pollutant levels, and semen parameters of the subjects were described based on the sampling sites. Continuous variables were represented as the median (5e95 percentiles). The significance of the percent distribution of various demographic characteristics of urban and rural study sites was determined using the Chi-Square test and Independent-Samples T test. As semen parameters and air pollution data follow markedly skewed (non-normal) distributions, the ManneWhitney U-test was used to compare the medians of sperm parameters and air pollution levels between two groups.
147
We analyzed the associations between the sperm parameters and air pollution variables using adjusted multivariate regression models. For semen parameters with skewed distribution, we performed statistical transformations on the variables to better approximate the normality assumption of the model. Specifically, we applied log transformation to the sperm concentration and square-root transformation to the sperm progressive motility, total sperm motility, and CASA parameters. The change-in-estimate method (Greenland, 1989) was used to determine which of the potential confounders should be adjusted for in the multivariate models. The coefficient was thought to be a potential confounder if the regression coefficient changed by more than 10% when included one by one in the multivariate models. Finally, age, education, BMI, duration of abstinence, smoking status, alcohol consumption, and season of the year were included in the models as potential confounders. The statistical analysis was performed using the Statistical Package for the Social Sciences Version 16.0 (SPSS, Chicago, IL, USA). All the tests were two-sided, and the level of significance was set at 0.05.
3. Results 3.1. Demographic information The demographical characteristics for the 1346 subjects were categorized into urban and rural groups (Table 1). There were no statistically significant differences in age, alcohol use, body mass index (BMI), duration of abstinence, and analysis time between the urban and rural groups. There was a significant difference (P ¼ 0.000) in the percentage of urban and rural males with different education levels: 68.1% vs. 75.0% subjects had less than a college education; 31.9% vs. 25.0% subjects had college and higher education. For tobacco use, a significant difference (P ¼ 0.039) was
Table 1 Demographic characteristics of the study populations. Characteristics Age (years), n (%) 20e29 30e40 Education, n (%) Less than college College and higher Tobacco use (/day), n (%) No smoking 120 Body mass index (BMI), n (%) 18.5 and
Contents lists available at ScienceDirect
Environmental Pollution journal homepage: www.elsevier.com/locate/envpol
Air pollution and decreased semen quality: A comparative study of Chongqing urban and rural areas Niya Zhou a,1, Zhihong Cui a,1, Sanming Yang b, Xue Han a, Gangcai Chen b, Ziyuan Zhou a, f, Chongzhi Zhai b, Mingfu Ma c, d, Lianbing Li c, d, Min Cai d, Yafei Li e, Lin Ao a, Weiqun Shu f, Jinyi Liu a, *, Jia Cao a, * a
Institute of Toxicology, College of Preventive Medicine, Third Military Medical University, Chongqing 400038, People’s Republic of China Chongqing Environmental Science Research Institute, Chongqing, People’s Republic of China Key Laboratory of Birth Defects and Reproductive Health of National Health and Family Planning Commission, Chongqing, People’s Republic of China d Chongqing Population and Family Planning Science and Technology Research Institute, Chongqing, People’s Republic of China e Department of Epidemiology, College of Preventive Medicine, Third Military Medical University, Chongqing, People’s Republic of China f Department of Environmental Hygiene, College of Preventive Medicine, Third Military Medical University, Chongqing, People’s Republic of China b c
a r t i c l e i n f o
a b s t r a c t
Article history: Received 1 October 2013 Received in revised form 27 December 2013 Accepted 31 December 2013
To investigate the association and effects of air pollution level on male semen quality in urban and rural areas, this study examines the outdoor concentrations of particulate matter (PM10), sulfur dioxide (SO2), nitrous dioxide (NO2) and semen quality outcomes for 1346 volunteers in both urban and rural areas in Chongqing, China. We found the urban area has a higher pollution level than the rural area, contrasted with better semen quality in the rural residents, especially for sperm morphology and computer assistant semen analysis (CASA) motility parameters. A multivariate linear regression analysis demonstrates that concentrations of PM10, SO2, and NO2 significantly and negatively are associated with normal sperm morphology percentage (P < 0.001) and sperm kinetic parameters. In conclusion, exposure to higher concentrations of PM10, SO2, and NO2 in urban ambient air may account for worse semen quality in urban males. Ó 2014 Elsevier Ltd. All rights reserved.
Keywords: PM10 SO2 NO2 Regional variation Semen quality Urban area Rural area
1. Introduction During the last two decades, many studies have shown that the various chemical and physical substances in ambient air may have major adverse effects on male fertility (Adamopoulos et al., 1996; Hammoud et al., 2010; Hansen et al., 2010; Sram, 1999). Particulate matter (PM) can carry multiple trace elements and
Abbreviations: PM10, particulate matters with diameters less than 10 mm; SO2, sulfur dioxide; NO2, nitrogen dioxide; WHO, World Health Organization; CASA, computer assistant semen analysis; VCL, curvilinear velocity; VSL, straight-line velocity; VAP, average path velocity; BCF, beat cross frequency; ALH, amplitude of lateral head displacement; LIN, motile sperm linearity; STR, motile sperm straightness; BMI, body mass index; SPSS, statistical package for the social sciences version; PAH, polycyclic aromatic hydrocarbons; PCBs, polychlorinated biphenyls; ER, estrogen receptor. * Corresponding authors. E-mail addresses: [email protected] (J. Liu), [email protected] (J. Cao). 1 These authors contribute equally to this work. 0269-7491/$ e see front matter Ó 2014 Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.envpol.2013.12.030
polycyclic aromatic hydrocarbons (PAHs) that can produce harmful effects on reproductive health. Studies have suggested that certain PAHs or their metabolites can interact with the estrogen receptor (ER) and initiate ER signaling pathways in vitro and in vivo, and exert endocrine and developmental toxicity in males (Ford and Huggins, 1963; MacKenzie and Angevine, 1981; Vinggaard et al., 2000). Similar effects of SO2 and NO2 on the male reproductive system also have been reported. An animal toxicological study has provided evidence that SO2 is a toxin for the reproductive system of mammals, and exposure to SO2 can cause oxidative damage to the testicles of male mice (Meng and Bai, 2004). Dejmek et al. (2000) reported that the association between SO2 and fecundability was more significant in those couples who lived close to a district highly polluted with SO2. Boggia et al. (2009) found subjects occupationally exposed to NO2 have a significant lower sperm total motility than do unexposed workers. However, the biological effects of these air pollutants on the reproductive system have not yet been confirmed in the general population.
146
N. Zhou et al. / Environmental Pollution 187 (2014) 145e152
Fig. 1. Map of study area. Two urban areas (Nanan and Shapingba) and four rural areas (Wanzhou, Wushan, Yunyang, and Zhongxian) were chosen to represent Chongqing in the Three Gorges Reservoir Region. The two urban areas are located in the central part of Chongqing City that has the highest population density, while the four rural areas are not centered around the city and have a lower population density, and typically much of the land there is devoted to agriculture.
Rapid economic development and urbanization in large cities has been reported to lead to severe air pollution in urban areas (Liu et al., 2008; Guéguen et al., 2011; Ma et al., 2011). Because of the increasing traffic, industrial activity, and energy consumption in urban areas, higher emissions of air pollutants usually occur there. Outdoor nitrogen dioxide (NO2) exposure has been used extensively to characterize the exposure to traffic-related air pollutants. SO2 and chemical substances like polychlorinated biphenyls (PCBs) and PAHs are also emitted by industries and other human activities, such as use of vehicles and residential heating facilities (Cooper et al., 1996; Biterna and Voutsa, 2005). Regional variability of air pollutant concentration in urban and rural areas has also been identified (Hoek et al., 2002; Lewné et al., 2004; Van der Zee et al., 1998; Zhu et al., 2002). Further, Pant et al. (2008) reported that the urban population in India was found to have a statistically and significantly lower percentage of sperm motility than the rural population in India and also found that the urban population had statistically significant higher levels of phthalate esters than did the rural areas. Chongqing is a petrochemical, iron and steel, aluminum, and manufacturing industrial center in the southwest of China with more than 30 million residents. For decades, Chongqing City has suffered from substantial air pollution from road and river transport and the presence of different industrial activities along the Yangzi River. These include steel plants, chemical and domestic waste incinerators, and other chemical industries. Venners et al. (2003) used a generalized additive model to estimate the associations of mean daily SO2 and PM2.5 for daily mortality in Chongqing. They found daily ambient SO2 concentration to be positively and
significantly associated with population risks for cardiovascular and respiratory mortality. Our earlier epidemiological study found that 61.1% of male subjects from the general population in the Chongqing area had at least one semen quality parameter below the WHO referenced value (Li et al., 2009). Further, PAHs exposure appeared to cause sperm DNA damage and fertility interference in males (Han et al., 2011). In particular, we are interested in male reproductive health in this region, as its quality may relate to ambient air pollution. In the present study, we chose six sites that included two highly polluted urban districts and four relatively cleaner rural counties. We investigated the distributions of three ambient air pollutants, PM10, SO2 and NO2, in the urban and rural areas in Chongqing and explored the association and effect of air pollution on male semen quality in both regions. 2. Materials and methods 2.1. Study design and subject recruitment Chongqing is the largest of the four direct-controlled municipalities in the People’s Republic of China, and occupies the major reservoir areas known as the Three Gorges Dam project (Three Gorges Reservoir Region). The boundaries of the Chongqing municipality reach much farther into the city’s hinterland than do those of the other three municipalities (Beijing, Shanghai and Tianjin), and most of the administrative area is rural (Fig. 1). This investigation was carried out in 2007. The study proposal was reviewed and approved by the Ethical Committee of the Third Military Medical University. We also worked with the Chongqing Family Planning Commission, the Chongqing Institute of Science and Technology for Population and Family Planning, and six local Family Planning Institutions to recruit volunteers. The inclusion criteria had the following characteristics: Males age 20e40 and permanent residents of the Chongqing area. The exclusion criteria included any reproductive or urological diseases diagnosed by
N. Zhou et al. / Environmental Pollution 187 (2014) 145e152 andrological examination, other known reproductive disorders, and/or an identifiable history of infertility, and prior vasoligation or chronic disease. All study participants were asked to sign an informed consent form to be able to participate. A total of 1976 volunteers were recruited. Of these, 630 were excluded from the analysis for reproductive disorders or other chronic diseases (n ¼ 80), missing or unknown duration of abstinence or reported duration of abstinence of 7 days (n ¼ 210), failing to collect the semen samples (n ¼ 226), and spillage of the sample semen (n ¼ 114). A group of 1346 healthy and eligible volunteers finally agreed to and completed all steps in the study. 2.2. Semen collection and the analyses The methods for semen collection and its analyses are described in detail previously (Li et al., 2009). Administration of the questionnaire and the physical examination, semen collection, and all analyses were carried out at a local Reproductive Health Center at each study site. All participants were asked to be abstinent for 2e7 days before contributing a semen sample. Conventional semen parameters for the original raw samples, including appearance, viscosity, liquefaction time, pH value, semen volume, sperm concentration and motion parameters, were measured using World Health Organization (WHO) guidelines (WHO, 1999). Sperm motion analysis was performed using a computer-aided semen analysis system (CASA, WLJY 9000, Weili New Century Science & Tech Dev., Beijing, China). According to the WHO reference, percent of motile sperm was scored using category A (rapid progressive motility), category B (slow progressive motility), category C (non-progressive motility), and category D (immotility). Sperm motion analysis was conducted for two types of motility: progressive motility (PR, categories A þ B) and non-progressive motility (NP, categories C). CASA outcomes included curvilinear velocity (VCL), straight-line velocity (VSL), average path velocity (VAP), beat cross frequency (BCF) and amplitude of lateral head displacement (ALH). The calculated parameters were linearity (LIN ¼ VSL/VCL 100) and straightness (STR¼VSL/VCL). According to the findings of previous studies (Meeker et al., 2004; Liu et al., 2012), some of the CASA parameters strongly correlated because different aspects of the same movement were described. Therefore, we chose progression (VSL), vigor (VCL), and swimming pattern (LIN) for this statistical analysis. We used the “feathering” method to perform the analysis for sperm morphology assessment. Two slides were utilized for each fresh semen sample. The air-dried smear was fixed in 96% ethanol, stained with a modified Papanicolaou stain, and then analyzed according to the WHO criteria. To reduce the variation in the assessment of sperm characteristics, two technicians performed all analyses of semen quality for all six centers using the same apparatus. These two technicians were well trained in semen analysis and participated in the Continuous Quality Control System (an external quality control system established using the WHO guidelines) under supervision of the Chongqing Science and Technology Commission. 2.3. Air pollution data and exposure assessment Air pollution data were obtained from the Institute of Environmental Science and Technology of Chongqing. Outdoor PM10, SO2, and NO2 concentrations in 2007 were collected from 6 ambient air quality monitoring stations in the studied districts or from counties that belonged to the national air pollutant detection network. The data for PM10, SO2 and NO2 represented a 24 h average and were recorded daily for the Wanzhou, Shapingba, and Nanan districts and every 3 days for the Zhongxian, Wushan, and Yunyang counties, respectively. The daily (24-h) average concentrations of PM10 were measured using the tapered element oscillating microbalance method (China State Environmental Protection Agency, 2000). According to the technical guidelines of the Chinese government, the location of these monitoring stations must not be in the direct vicinity of traffic intersections and not influenced by local pollution sources. They should also be a sufficient distance from any other emitting sources, such as coal-, waste-, or oil-burning boilers; furnaces, and incinerators. To analyze the association between the estimated air pollutants and semen quality outcomes, we employed average exposure data to the pollutants within an exposure period of 90 days before undertaking a semen sampling from each subject. An exposure period of approximately 90 days is generally accepted as a sufficient duration for detecting effects on any stage of spermatogenesis when using semen measures as biological endpoints (Heller and Clermont, 1964; Selevan et al., 2000; Hansen et al., 2010). 2.4. Statistical analysis Demographic information, pollutant levels, and semen parameters of the subjects were described based on the sampling sites. Continuous variables were represented as the median (5e95 percentiles). The significance of the percent distribution of various demographic characteristics of urban and rural study sites was determined using the Chi-Square test and Independent-Samples T test. As semen parameters and air pollution data follow markedly skewed (non-normal) distributions, the ManneWhitney U-test was used to compare the medians of sperm parameters and air pollution levels between two groups.
147
We analyzed the associations between the sperm parameters and air pollution variables using adjusted multivariate regression models. For semen parameters with skewed distribution, we performed statistical transformations on the variables to better approximate the normality assumption of the model. Specifically, we applied log transformation to the sperm concentration and square-root transformation to the sperm progressive motility, total sperm motility, and CASA parameters. The change-in-estimate method (Greenland, 1989) was used to determine which of the potential confounders should be adjusted for in the multivariate models. The coefficient was thought to be a potential confounder if the regression coefficient changed by more than 10% when included one by one in the multivariate models. Finally, age, education, BMI, duration of abstinence, smoking status, alcohol consumption, and season of the year were included in the models as potential confounders. The statistical analysis was performed using the Statistical Package for the Social Sciences Version 16.0 (SPSS, Chicago, IL, USA). All the tests were two-sided, and the level of significance was set at 0.05.
3. Results 3.1. Demographic information The demographical characteristics for the 1346 subjects were categorized into urban and rural groups (Table 1). There were no statistically significant differences in age, alcohol use, body mass index (BMI), duration of abstinence, and analysis time between the urban and rural groups. There was a significant difference (P ¼ 0.000) in the percentage of urban and rural males with different education levels: 68.1% vs. 75.0% subjects had less than a college education; 31.9% vs. 25.0% subjects had college and higher education. For tobacco use, a significant difference (P ¼ 0.039) was
Table 1 Demographic characteristics of the study populations. Characteristics Age (years), n (%) 20e29 30e40 Education, n (%) Less than college College and higher Tobacco use (/day), n (%) No smoking 120 Body mass index (BMI), n (%) 18.5 and
