Phthalate exposure and reproductive parameters in young men from the general Swedish population

Environment International 85 (2015) 54–60

Contents lists available at ScienceDirect

Environment International journal homepage: www.elsevier.com/locate/envint

Phthalate exposure and reproductive parameters in young men from the general Swedish population Jonatan Axelsson a,b,⁎, Lars Rylander b, Anna Rignell-Hydbom b, Bo A.G. Jönsson b, Christian H. Lindh b, Aleksander Giwercman a a b

Molecular Reproductive Medicine, Dept. of Translational Medicine, Lund University, Skåne University Hospital, 205 02 Malmö, Sweden Division of Occupational and Environmental Medicine, Lund University, 221 85 Lund, Sweden

a r t i c l e

i n f o

Article history: Received 12 March 2015 Received in revised form 11 June 2015 Accepted 1 July 2015 Available online xxxx Keywords: Phthalate metabolites Environmental exposure Semen analysis Sperm motility Fertility Reproductive hormones

a b s t r a c t Background: In animals, exposure to certain phthalates negatively affects the male reproductive function. Human results are conflicting and mostly based on subfertile males, in whom the association between exposure and reproductive function may differ from the general population. Objectives: To study if levels of phthalate metabolites were associated with semen quality and reproductive hormones in general Swedish men. Methods: We recruited 314 young men delivering semen, urine and blood samples at the same visit. We analyzed reproductive hormones and several semen parameters including progressive motility and high DNA stainability (HDS)—a marker for sperm immaturity. In urine, we analyzed metabolites of phthalates, including diethylhexyl phthalate (DEHP). We studied associations between urinary levels of the metabolites and seminal as well as serum reproductive parameters, accounting for potential confounders. Results: DEHP metabolite levels, particularly urinary mono-(2-ethyl-5-carboxypentyl) phthalate (MECPP), were negatively associated with progressive sperm motility, which was 11 (95% CI: 5.0–17) percentage points lower in the highest quartile of MECPP than in the lowest. Further, men in the highest quartile of the DEHP metabolite monoethylhexyl phthalate had 27% (95% CI: 5.5%–53%) higher HDS than men in the lowest quartile. Conclusions: DEHP metabolite levels seemed negatively associated with sperm motility and maturation. © 2015 Elsevier Ltd. All rights reserved.

1. Introduction Several phthalates that are common constituents of consumer products (Wittassek et al., 2011) are reported to disturb the reproductive function of male animals (Martino-Andrade and Chahoud, 2010). Those effects, of exposure doses above those encountered in humans, include a reduced sperm production and motility (Glombik et al., 2014) and reduced levels of testosterone (Agarwal et al., 1986; Miura et al., 2007; Oishi, 1985). However, the few studies performed in primates indicate a lower sensitivity, in comparison with rodents (Martino-Andrade and Chahoud, 2010). Testicular toxicity of phthalates is mainly apparent for those with a side-chain of medium length, such as dibutyl phthalate (DBP), or with a longer side-chain that is branched, such as diethylhexyl phthalate (DEHP). The long-chain phthalates are often used as plasticizers such as in PVC, whereas the shorter-chain phthalates have other applications such as in personal-care products, paints, adhesives and entericcoated tablets (Wittassek et al., 2011). Humans are exposed to long⁎ Corresponding author at: Clinical Research Centre, Skåne University Hospital, Malmö, 205 02 Malmö, Sweden. E-mail address: [email protected] (J. Axelsson).

http://dx.doi.org/10.1016/j.envint.2015.07.005 0160-4120/© 2015 Elsevier Ltd. All rights reserved.

chain phthalates mainly through the diet, whereas the shorter-chain phthalates seem to have additional sources of a higher importance (Wittassek et al., 2011). Human phthalate exposure is most often assessed through levels of metabolites in urine, and has in some studies been associated with an impaired semen quality (Kay et al., 2014), but this has mostly been studied in subfertile men. The evidence for associations with the levels of reproductive hormones in humans is weak, or limited for testosterone levels (Kay et al., 2014). In subfertile men, due to a more vulnerable reproductive system, the association between exposure and semen quality may differ from that in the general population. Studies in men who are not subfertile are scarce and rather conflicting. Thus, in men with an occupational exposure to phthalates, DEHP has been linked with a decreased sperm motility and with an increased sperm DNA fragmentation (Huang et al., 2011, 2014) and, further, with reduced levels of free testosterone in one study (Pan et al., 2006) but with increased levels of estradiol in another (Fong et al., 2015). In men without an occupational exposure to phthalates, one study reported a decrease in sperm motility with increasing diethyl phthalate (DEP) metabolite levels (Jönsson et al., 2005). Another study found no associations (Joensen et al., 2012), whereas a third study reported that

J. Axelsson et al. / Environment International 85 (2015) 54–60

metabolite levels of DBP were negatively associated with sperm concentration (Han et al., 2014). In a more recent study, metabolite levels of DEHP and di-isonyl phthalate (DiNP) were negatively associated with semen volume, and DEHP metabolite levels were negatively associated also with the total sperm count (Specht et al., 2014). The available human data based on non-infertile men differ in design, not all being based on measurements of secondary urinary metabolites (Jönsson et al., 2005) considered to be the most reliable exposure markers of longer-chain phthalates (Frederiksen et al., 2010; Wittassek et al., 2011), and which were measured only in the later studies. Furthermore, despite half-lives in hours, metabolite levels were not always sampled the same day as the semen, and not always in urine (Specht et al., 2014). In addition, not all studies included adjustment for time of abstinence and the urinary dilution (Joensen et al., 2012). All these factors may influence potential associations between metabolite levels and semen parameters. Therefore, we wanted to study associations between levels of several (including secondary) phthalate metabolites and markers of male reproductive function, through samples of urine, serum and semen, collected at the same visit in a homogenous group of 314 young Swedish men from the general population. We also wanted to adjust for both abstinence time and the urinary dilution. In addition, in order to validate the use of serum as a matrix for measurements of phthalate metabolites, we wanted to study correlations between metabolite levels in urine and serum, the latter being more frequently available from biobanks. 2. Methods For a previous study of reproductive function in Swedish adolescent men (Axelsson et al., 2011), between 2008 and 2010 we included 314 men (241 men [14% of 1681 invited] from the health board prior to military service and 73 men through announcement in schools). Participation criteria were: a) being 17 to 20 years old; b) living within 60 km from the city of Malmö in Southern Sweden; c) participant and mother born and raised in Sweden. All participants received 500 SEK (55 Euro), signed an informed consent, and answered questionnaires about length and weight, current smoking, parental smoking during pregnancy and previous disease and medications. At the same visit, they delivered samples of urine, blood and semen before noon, and reported their abstinence time. Urinary samples were portioned in 10 mL tubes with 1–2 mL in each. The study was approved by the regional ethical review board of Lund University. 2.1. Semen analysis Semen samples, collected upon masturbation at the hospital, were examined for volume, sperm concentration, total sperm count, proportion of morphologically normal sperm and progressive motility, all according to the WHO guidelines (World Health Organization, 1999). We additionally analyzed the DNA fragmentation index (DFI) and high DNA stainability (HDS) using the Sperm Chromatin Structure Assay (Evenson et al., 2002). 2.2. Analysis of reproductive hormones Blood samples were drawn before noon. Serum was analyzed at the clinical chemistry laboratory of Skåne University Hospital, Sweden by means of ElectroChemiLuminiscenceImmunoassay (Roche Cobas) for concentrations of follicle-stimulating hormone, luteinizing hormone, sex hormone-binding globulin and testosterone and by use of an immunofluorometric method (Delfia, Perkin-Elmer) for estradiol. Coefficients of variation (CV) and level of detection (LOD) were as follows: follicle-stimulating hormone had CV 5.5% at 5.0 IU/L and 6.3% at 75 IU/L and LOD at 0.10 IU/L; luteinizing hormone had CV 3.2% at

55

7.0 IU/L and 3.1% at 82 IU/L and LOD 0.10 IU/L; sex hormone-binding globulin had CV 1.2% at 16 nmol/L and 1.4% at 34 nmol/L and LOD 0.35 nmol/L; testosterone had CV 3.8% at 3.2 nmol/L and 1.6% at 25 nmol/L and LOD 0.087 nmol/L; estradiol had CV 20% at 30 pmol/L and 10% at 280 pmol/L, LOD 8 pmol/L. Concentration of free testosterone was calculated according to Vermeulen et al. (1999). 2.3. Analyses of phthalate metabolites LODs and CVs are shown in the Supplemental Material, Table S1. Levels below LOD were given the value of LOD divided by 2. 2.3.1. Urinary levels Urinary samples were processed through an automated solid-phase extraction technique and analyzed by liquid chromatography–tandem mass spectrometry (LC/MS/MS) according to a previously published method (Toft et al., 2012). In brief, internal standards were added, and samples thereafter treated with glucuronidase, evaporated and dissolved in water containing acetic acid before analysis using the LC/ MS/MS. In each series, an internal control urinary sample was analyzed. We analyzed ten metabolites of five different phthalates. For DEHP, we analyzed the primary metabolite mono-(2ethylhexyl) phthalate (MEHP) and the secondary metabolites: mono(2-ethyl-5-carboxypentyl) phthalate (MECPP), mono-(2-ethyl-5hydroxylhexyl) phthalate (MEHHP) and mono-(2-ethyl-5-oxohexyl) phthalate (MEOHP). For DiNP, we analyzed the secondary metabolites: mono-(carboxy-iso-octyl) phthalate (MCiOP), mono-(hydroxy-isononyl) phthalate (MHiNP) and mono-(oxo-iso-nonyl) phthalate (MOiNP). For DBP, we analyzed monobutyl phthalate (MBP). For butylbenzyl phthalate, we assessed monobenzyl phthalate (MBzP), and for DEP, we measured monoethyl phthalate (MEP). We adjusted for the urinary dilution by dividing the molar concentrations of the metabolites with that of creatinine. 2.3.2. Serum levels In serum, we analyzed the secondary DEHP- and DiNP metabolites listed above according to a previously described LC/MS/MS-based method (Specht et al., 2014). In brief, isotopically labeled internal standards were added to aliquots of 100 μL serum for all the evaluated compounds. Samples were digested with glucuronidase and proteins precipitated with acetonitrile. Analyses were performed in negative ion mode. The quality of the analysis was checked by inclusion of chemical blank samples and an in-house quality control in the analyzed sample batches. Each sample was analyzed three times in three different analytical batches. 2.4. Statistics We used SPSS version 18 to 22 for statistic calculations and considered p b 0.05 as statistically significant. By use of Spearman's rank correlation test, we studied to which degree the different urinary and serum metabolite levels were correlated with each other. Furthermore, for those metabolites for which both serum and urinary levels were available (see above) the correlations between these two measurements were tested. In order to compare our results with those of a previous study (Joensen et al., 2012), we calculated %MEHP. This was defined as the urinary concentration of the primary metabolite MEHP as a proportion of the sum of the urinary levels of the DEHP metabolites (MEHP, MECPP, MEHHP and MEOHP). To achieve a more normal distribution of residuals, and a better statistical prediction (Tabachnick and Fidell, 2013), we transformed sperm concentration and total sperm count by the cubic root (Sanchez-Pozo et al., 2013) and semen volume, HDS, DFI, testosterone, free testosterone, luteinizing hormone, follicle-stimulating hormone, estradiol and

56

J. Axelsson et al. / Environment International 85 (2015) 54–60

exposure markers, by the natural logarithm. To assess normality, linearity and scedasticity, we used histograms, tests of normality, and plots of residuals. Based on the literature, we considered abstinence time, BMI, current own cigarette smoking and maternal and paternal smoking during pregnancy as potential confounders (Axelsson et al., 2013; Hammiche et al., 2012; Vine, 1996; Virtanen et al., 2012) of associations with semen parameters. For associations with reproductive hormones, abstinence time was excluded, whereas the time of day at sampling (as a continuous variable) was included. Abstinence time in hours (missing in 15 men) was categorized into ≤48 (118 men), 49 to 72 (106 men), 73 to 96 (43 men), 97 to 120 (18 men) and N 120 (14 men). Own smoking and maternal and paternal smoking during pregnancy (missing in 2, 8 and 17 men, respectively) were dichotomized (yes/no). In men with data present, 64 men (21%) smoked cigarettes, whereas 57 men (19%) reported that their mother smoked during the pregnancy, and 82 men (30%) that their father did. After adjusting for the potential confounders, regression models were applied to study associations between the exposure marker levels (both as continuous variables and categorized in quartiles) and the semen parameters. Men with missing values of a variable included in the model were excluded. We also performed non-adjusted analyses. For the categorized exposure markers, we calculated the p-value of the overall F test, as well as the difference between the highest and the lowest quartiles (first and fourth). If both were statistically significant, we additionally calculated p-values for trend across the quartiles (ptrend), by entering the categorized metabolite level in the model as if it were a linear variable. Differences in logarithmically transformed variables were back-transformed into ratios. We calculated the effect size eta squared, to decide the proportion of variance in a semen parameter that depended on the level of an exposure marker. This was calculated by dividing the sum of squares for our exposure marker, with the corrected total sum of squares (Grace-Martin, 2008–2015; Levine and Hullett, 2010). Finally, if associations were found between metabolite levels of more than one parent compound in relation to a specific semen variable, we tested the robustness of the associations by including metabolite levels of both of the parent compounds in the model at the same time.

3. Results For background characteristics and levels of exposure markers, see Tables 1 and 2. The highest urinary metabolite levels were found for MEP followed by MBP, MEHHP and MCiOP (Table 2). Previous diseases and medications are shown in Supplemental Material, Table S2.

3.1. Correlation between phthalate metabolites A number of urinary metabolites were statistically significantly correlated (see Supplemental Material, table S3). This was most pronounced among the secondary DEHP metabolites (Spearman's rho: 0.89–0.94) and among the metabolites of DiNP (Spearman's rho: 0.80–0.90). For serum levels, MECPP was correlated with all the other metabolites, and most strongly with MCiOP (see Supplemental material, Table S4). The strongest correlations overall were found between the metabolites of DiNP (Spearman's rho: 0.76–0.88). Levels of all metabolites measured in serum were correlated with the corresponding level in urine (Spearman's rho; 0.84 for MECPP, 0.37 for MEHHP, 0.32 for MEOHP, 0.91 for MCiOP, 0.78 for MHiNP and 0.77 for MOiNP, respectively, all p b 0.001) (Table 3). 3.2. Associations between phthalate metabolite markers and reproductive outcomes Any associations for metabolite levels in serum are shown only in the Supplemental Material, Tables S5–S12. 3.2.1. Associations between continuously treated urinary exposure markers and reproductive outcomes 3.2.1.1. Adjusted. All the DEHP metabolites MEHP, MECPP, MEOHP, MEHHP, as well as MBP were negatively associated with progressive sperm motility (p b 0.001–0.03, with the largest effect size for urinary MECPP [eta squared 4.4%], Supplemental material, Table S5). MECPP was additionally associated with HDS (p = 0.04, eta squared 1.6%). MCiOP and %MEHP were positively associated with semen volume (p = 0.04 and 0.049, with the largest effect size for MCiOP [eta squared 1.5%]). Finally, %MEHP was negatively associated with testosterone (p = 0.022, eta squared 1.7%) and free testosterone (p = 0.027, eta squared 1.7%) (Supplemental material, Table S6). 3.2.1.2. Unadjusted. MEHP, MBP, MECPP and MHiNP were negatively associated with progressive motility (see Supplemental Material, Table S7). Further, all DEHP metabolites and MBzP were positively associated with HDS. Finally, MCiOP and MOiNP were positively associated with semen volume. Finally, a negative association was found between MOiNP and FSH, and between %MEHP and free testosterone (Supplemental material, Table S8). 3.2.2. Associations between categorized urinary exposure markers and reproductive outcomes

Table 1 Background characteristics.

Age (years) BMI (kg/m2) Semen volume (mL) Sperm concentration (×106/mL) Total sperm count (×106) Progressive motility (%) Morphologically normal (%) DNA fragmentation index (%) High DNA stainability (%) Abstinence time (h) Own smoking Maternal smoking Paternal smoking

N

Mean ± SD or %

Minimum Median Maximum

314 314 314 312

18.4 ± 0.36 23 ± 3.1 2.9 ± 1.5 70 ± 60

17.5 16 0.1 0.0

18.3 23 2.7 56

20.5 37 14 410

313 312 314 305 305 299 312 306 297

200 ± 210 53 ± 17 8.5 ± 5.7 11 ± 6.1 12 ± 7.5 60 ± 34 21% 19% 30%

0.0 0.0 0.0 3.0 3.0 8 N. A. N. A. N. A.

150 58 8.0 10 10 58 N. A. N. A. N. A.

1700 86 29 39 47 230 N. A. N. A. N. A.

Abbreviation: N. A., not applicable; SD, standard deviation.

3.2.2.1. Adjusted. All the secondary DEHP metabolites were associated with differences in progressive motility (overall F test with p b 0.05) between quartiles of metabolite levels (Table 4). Thus, when comparing men in the highest quartile with those in the lowest, progressive motility was: 11 (95% CI: 5.0–17) percentage points lower in the highest quartile of MECPP (ptrend b 0.001), 8.7 (95% CI: 2.8–15) percentage points lower in the highest quartile of MEHHP (ptrend = 0.002), and 6.9 (95% CI: 1.1–13) percentage points lower in the highest quartile MEOHP (ptrend = 0.009). Largest effect size was found for MECPP (eta squared 6.6%) with 57% progressively motile sperm in the lowest exposure quartile vs 46% in the highest. Further, quartiles of MEHP differed in HDS, which was 26% (95% CI: 5.2–52%) higher in the highest than in the lowest quartile (12% vs 9.6%, eta squared 4.1%, ptrend = 0.02). Despite additional differences between the highest and lowest quartiles of MCiOP and MOiNP in relation to semen volume; of MEHP in

J. Axelsson et al. / Environment International 85 (2015) 54–60

57

Table 2 Concentrations of phthalate metabolites in urine (unadjusted and adjusted for creatinine) and serum. Parent compound

Diethyl phthalate (DEP) Dibutyl phthalate (DBP) Butylbenzyl phthalate (BBzP) Diethylhexyl phthalate (DEHP)

Metabolite or marker

Matrix

MEP MBP MBzP MEHP MECPP

Urine Urine Urine Urine Urine Serum Urine Serum Urine Serum Urine Serum Urine Serum Urine Serum Urine

MEOHP MEHHP Di-isononyl phthalate (DiNP)

MCiOP MOiNP MHiNP

Diethylhexyl phthalate (DEHP)

%MEHP (%)

N

Adjusted (nmol/mmol creatinine)a

Unadjusted (ng/mL)

314 314 314 314 314 306 314 306 314 306 314 306 314 306 314 306 314

Arithmetic mean ± SD

Min

Median

Max

% ≥ LOD

Arithmetic mean ± SD

Min

Median

Max

190 ± 670 64 ± 64 22 ± 30 4.1 ± 6.3 22 ± 38 0.56 ± 0.44 16 ± 62 0.20 ± 0.14 35 ± 110 0.36 ± 0.47 30 ± 61 2.2 ± 5.0 12 ± 34 0.059 ± 0.19 19 ± 62 0.38 ± 1.5 –

2.0 1.0 0.5 bLOD 1.6 0.13 0.5 0.03 1.1 0.07 1.4 0.17 0.2 bLOD 0.5 0.01 –

41 47 13 2.8 15 0.45 9.6 0.16 21 0.21 16 1.1 5.0 0.028 8.4 0.17 –

6900 690 260 97 600 5.0 1100 1.2 1900 3.9 810 77 500 2.9 980 25 –

100 100 100 97 100 100 100 100 100 100 100 100 100 98 100 100 –

57 ± 170 17 ± 15 5.0 ± 5.5 0.88 ± 1.2 4.1 ± 6.7 – 3.2 ± 12 – 6.9 ± 20 – 5.7 ± 12 – 2.4 ± 7.9 – 3.9 ± 14 – 6.8a ± 4.3a

1.5 1.0 0.3 0.01 0.3 – 0.2 – 0.5 – 0.4 – 0.1 – 0.2 – 0.2a

14 13 3.0 0.6 3.1 – 2.1 – 4.5 – 3.2 – 0.9 – 1.6 – 6.1a

1600 180 37 19 110 – 200 – 340 – 180 – 110 – 220 – 28a

Abbreviations: LOD, level of detection; SD, standard deviation. a %MEHP in %.

relation to progressive motility; and of MECPP in relation to HDS, the overall F tests were not statistically significant. We found no difference in reproductive hormones between men in the highest quartile and those in the lowest quartile of any of the metabolite markers (Table 5).

CI: 3.8–16] percentage points lower in the highest than in the lowest quartile), whereas the association between MBP and progressive motility was no longer statistically significant (data not shown).

3.2.2.2. Unadjusted. All the DEHP metabolites and MBP were associated with an inter-quartile difference in progressive motility (see Supplemental Material, Table S9). Similarly, an inter-quartile difference in HDS was seen for all the DEHP metabolites and MBzP. For these metabolites, motility was lower or HDS higher in the highest quartile than in the lowest, and trends across the quartiles were also statistically significant (data not shown). Finally, quartiles of MBP differed with statistical significance with respect to sperm morphology, but not between the highest and the lowest quartiles. As considers reproductive hormones, no metabolite or marker differed with statistical significance, both between the highest and lowest quartiles and in the overall test (see Supplemental material, Table S10).

In adolescent men from the general Swedish population, levels of DEHP metabolites were negatively associated with progressive sperm motility. A similar association for the metabolite of DBP was not robust to adjustment for DEHP metabolite levels. Furthermore, levels of DEHP metabolites were positively associated with HDS, suggesting a negative effect on sperm maturation (Evenson et al., 2002). Other associations with the reproductive outcomes were less consistent between the categorized and continuously treated metabolite markers. These include a positive association between DiNP metabolite levels, as well as %MEHP, and semen volume, and a negative association between %MEHP and testosterone as well as free testosterone. Levels of several metabolites were correlated with each other, including metabolites of different parent compounds such as DEHP and DBP. This might explain the association found between MBP and progressive motility, which was not robust to adjustment for levels of DEHP metabolite levels. Finally, moderate to strong correlations were found between metabolite levels in serum and urine. This study was based on men from the general population. Despite a low participation rate, we believe this cohort fairly well represents the general population, since, in a similar Danish study men delivering semen samples had equal levels of reproductive hormones to those refusing semen sampling (Andersen et al., 2000). In addition, we previously found similar semen parameters in men recruited through the health board prior to military service and those enrolled through schools (Axelsson et al., 2011). Levels of DEHP metabolites vary considerably within individuals (Wittassek et al., 2011), and the primary metabolite MEHP has been suggested to be the most bioactive (Chauvigne et al., 2009). Therefore, %MEHP in urine, as a measure of low MEHP metabolism, was proposed as a more stable marker (Adibi et al., 2008; Meeker et al., 2012), but might also indicate a poor metabolism of toxicants in general (Hauser, 2008). Nonetheless, MEHP has been reported to decrease testosterone production in the human testis in vitro (Desdoits-Lethimonier et al., 2012), why we cannot exclude that the men with a high %MEHP in our study had lower testosterone due to generally high MEHP levels. A similar association as ours has previously been reported (Joensen

3.2.3. Adjustment for urinary metabolite level of another parent compound Since metabolites of both DEHP (both as continuous and categorized) and DBP (as continuous) were negatively associated with progressive motility, which for DEHP metabolites was strongest for MECPP, we included both MECPP (separately as continuous and as categorized) and MBP (continuous) along with the potential confounders in the same model. Through this procedure, the negative association between MECPP and progressive motility remained statistically significant (p = 0.005 for continuous MECPP, progressive motility being 9.9 [95%

Table 3 Correlation coefficients (Spearman's rho) between DEHP- and DiNP metabolites in men's serum an urine (N = 306). Parent compound

Urinary levels DEHP

DiNP

MECPP MEHHP MEOHP MCiOP MHiNP MOiNP Levels in DEHP MECPP serum MEHHP MEOHP DiNP MCiOP MHiNP MOiNP ⁎⁎ p b 0.01.

0.84⁎⁎ 0.32⁎⁎ 0.33⁎⁎ 0.27⁎⁎ 0.23⁎⁎ 0.29⁎⁎

0.74⁎⁎ 0.37⁎⁎ 0.34⁎⁎ 0.21⁎⁎ 0.21⁎⁎ 0.26⁎⁎

0.74⁎⁎ 0.36⁎⁎ 0.32⁎⁎ 0.18⁎⁎ 0.17⁎⁎ 0.24⁎⁎

0.41⁎⁎ 0.03 0.01 0.91⁎⁎ 0.82⁎⁎ 0.73⁎⁎

0.43⁎⁎ 0.09 0.04 0.82⁎⁎ 0.86⁎⁎ 0.80⁎⁎

0.41⁎⁎ 0.07 0.06 0.74⁎⁎ 0.78⁎⁎ 0.77⁎⁎

4. Discussion

58

J. Axelsson et al. / Environment International 85 (2015) 54–60

Table 4 Adjusteda difference in semen parameters between the highest and lowest quartiles of phthalate metabolite markers in urine. Parent compound

Metabolite or marker

DEP DBP BBzP DEHP

MEP MBP MBzP MEHP MECPP MEOHP MEHHP MCiOP MOiNP MHiNP %MEHP

DiNP

DEHP

Sperm concentration (×106/mL)b

Total sperm count (×106)b

Semen volume (mL)c

Progressively motile sperm (%)

MD (95% CI)

MD (95% CI)

MD (95% CI)

MD (95% CI)

MD (95% CI)

MD (95% CI)

MD (95% CI)

0.17 (−0.24, 0.59) 0.04 (−0.38, 0.45) −0.29 (−0.71, 0.13) −0.08 (−0.49, 0.33) −0.11 (−0.53, 0.31) −0.06 (−0.48, 0.36) 0.01 (−0.41, 0.43) −0.08 (−0.49, 0.33) −0.15 (−0.56, 0.27) −0.06 (−0.49, 0.36) −0.02 (−0.44, 0.39)

0.19 (−0.45, 0.82) 0.10 (−0.54, 0.74) −0.44 (−1.1, 0.21) 0.14 (−0.49, 0.77) −0.29 (−0.93, 0.35) −0.07 (−0.72, 0.57) 0.08 (−0.56, 0.73) 0.05 (−0.58, 0.67) 0.04 (−0.60, 0.68) 0.06 (−0.59, 0.70) 0.23 (−0.41, 0.86)

−1.0 (−18, 20) −1.7 (−19, 19) −0.09 (−18, 20) 14 (−5.4, 39) −6.7 (−23, 13) 0.2 (−17, 22) 4.4 (−14, 27) 21 (0.1, 46) 23 (1.4, 49) 17 (−3.3, 43) 19 (−2.1, 44)

2.6 (−3.3, 8.5) −7.7 (−14, −1.8) −5.4 (−11, 0.56) −7.8 (−14, −2.0) −11 (−17, −5.0)⁎ −6.9 (−13, −1.1)⁎ −8.7 (−15, −2.8)⁎

−0.73 (−2.7, 1.3) 0.65 (−1.3, 2.6) −0.79 (−2.8, 1.2) −1.3 (−3.3, 0.66) −1.0 (−3.0, 0.98) −0.42 (−2.4, 1.6) −1.2 (−3.2, 0.80) −1.2 (−3.1, 0.82) −1.4 (−3.4, 0.61) −1.3 (−3.3, 0.77) 0.03 (−2.0, 2.0)

−14 (−29, 4.4) 11 (−8.4, 34) 19 (−1.8, 44) 27 (5.5, 53)⁎

−8.6 (−24, 9.4) 15 (−3.8, 37) 3.0 (−14, 23) 11 (−7.4, 32) 1.2 (−15, 21) −7.1 (−22, 11) −5.2 (−21, 13) −9.5 (−24, 8.0) −12 (−26, 5.5) −14 (−28, 3.5) 6.4 (−11, 27)

−1.3 (−7.2, 4.6) −4.4 (−10, 1.6) −3.8 (−9.8, 2.2) −2.1 (−7.9, 3.8)

Normal morphology (%)

High DNA stainability (%)c

25 (3.3, 51) 11 (−8.2, 34) 18 (−2.5, 43) 7.3 (−11, 30) 6.2 (−13, 29) 12 (−7.7, 36) 2.0 (−16, 23)

DNA fragmentation index (%)c

Abbreviations: CI, confidence interval; MD, mean difference highest minus lowest quartile; %MEHP, percentage of measured DEHP metabolites excreted as MEHP. a Adjusted for abstinence time, BMI and own and parental smoking. b Transformed by the cubic root. c Back-transformed to relative difference (%) from ln-transformation. ⁎ Overall F test p b 0.05.

et al., 2012), albeit with inconsistency (Meeker et al., 2009). The additional positive association we found for %MEHP, as well as for MCiOP, with semen volume seems hard to explain. They are also in opposition to previously reported negative associations between DEHP- and DiNP metabolite levels and semen volume (Specht et al., 2014). Previous human studies relating phthalate exposure to semen parameters have shown somewhat contradictory results, although exposure to DEHP and DBP have been the most consistently reported to be associated with a lower semen quality, according to a recent review (Kay et al., 2014). Among the previous studies, two occupational (Huang et al., 2011, 2014), and some non-occupational studies (Duty et al., 2003; Hauser et al., 2006; Jurewicz et al., 2013; Kranvogl et al., 2014; Pant et al., 2008, 2011, 2014) reported negative associations between exposure to (or metabolite levels of) either DEHP or DBP and sperm motility. However, the seven latter studies were either based on subfertile men or applied the parent phthalates as exposure markers. In subfertile men, the link between exposure and semen parameters may be influenced by an underlying pathological condition in the reproductive system. Further, the use of the parent compounds as markers of exposure has been questioned due to the risk of contamination of the samples from phthalates in the surrounding environment (Calafat et al., 2013). Therefore, to assess any potential effects of environmental phthalate exposure, it seems preferable to measure the metabolites of phthalates (Wittassek et al., 2011) in men from the general population.

Four such studies have been performed (Han et al., 2014; Joensen et al., 2012; Jönsson et al., 2005; Specht et al., 2014), but none of them found any association between levels of metabolites of DEHP, or DBP, and sperm motility. However, in only one (Joensen et al., 2012) of these studies the most representative (Wittassek et al., 2011) secondary DEHP metabolites were studied in samples of urine collected the same day as the semen. This may be critical given the short half-lives of the metabolites of less than a day (Wittassek et al., 2011) and since urinary levels may be more representative than those in serum (Frederiksen et al., 2010). In the study by Joensen et al. (Joensen et al., 2012), however, no adjustment was made for the urinary dilution which can affect metabolite levels (Fromme et al., 2007; Hoppin et al., 2002; Peck et al., 2010) and no adjustment was performed for abstinence time, reported to have an impact on sperm motility (Elzanaty et al., 2005; Levitas et al., 2005). Therefore, the previous studies may have underestimated potential associations between exposure to phthalates and a decreased sperm motility. Two potential explanations to our results are chance findings and residual confounding. Including a total of 18 exposure markers, 7 semen parameters and 6 reproductive hormones, we made 234 comparisons, and found 10 associations for continuous exposure and 4 for categorized. Still, to calculate the number of expected associations is difficult because of the relatively strong correlations between the exposure markers, as well as between the outcomes (Aschengrau and Seage,

Table 5 Adjusteda difference in reproductive hormones between the highest and lowest exposure quartiles of urinary phthalate metabolite markers. Parent compound

Metabolite or marker

DEP DBP BBzP DEHP

MEP MBP MBzP MEHP MECPP MEOHP MEHHP MCiOP MOiNP MHiNP %MEHP

DiNP

DEHP

Testosterone (nmol/L)b

Free testosterone (nmol/L)b

Luteinizing hormone (IU/L)b

Follicle-stimulating hormone (IU/L)b

Estradiol (pmol/L)b

MD (95% CI)

MD (95% CI)

MD (95% CI)

MD (95% CI)

MD (95% CI)

MD (95% CI)

1.0 (−7.9, 11) 2.1 (−11, 7.6) 2.3 (−2.8, 6.0) −2.7 (−11, 6.7) 4.7 (−4.5, 15) −1.0 (−9.7, 8.7) −0.9 (−9.7, 8.7) 4.3 (−4.9, 14) −2.1 (−11, 7.8) 9.0 (−8.4, 11) −6.9 (−15, 2.4)

1.7 (−6.5, 11) −1.5 (−9.4, 7.1) −2.8 (−11, 6.0) −0.9 (−8.8, 7.7) 2.7 (−5.4, 12) 0.4 (−7.7, 9.2) 1.1 (−6.9, 10) 3.1 (−5.2, 12) 0.0 (−8.3, 9.0) 2.0 (−6.4, 11) −4.7 (−12, 3.9)

3.1 (−7.6, 15) −1.1 (−11, 10)⁎ −7.3 (−17, 3.8) 6.3 (−4.7, 19) −6.9 (−17, 4.0) −7.2 (−17, 3.6) −4.8 (−15, 6.3) 5.2 (−5.7, 17) 0.9 (−10, 13) 3.4 (−7.9, 16) 7.8 (−3.6, 21)

4.5 (−11, 23) 0.6 (−15, 19)⁎ −4.0 (−19, 14) 3.3 (−12, 22) −8.9 (−23, 7.3) −9.1 (−23, 7.1) −10 (−24, 5.9) −8.4 (−22, 7.9) −8.5 (−23, 8.4) −12 (−26, 4.2) 9.0 (−7.9, 29)

0.8 (−9.3, 8.7) 6.0 (−3.2, 16) 2.3 (−6.7, 12) 4.8 (−4.2, 15) 4.4 (−4.6, 14) 0.8 (−7.9, 10) 2.3 (−6.5, 12) −3.4 (−12, 5.7) −2.3 (−11, 7.4) −3.5 (−12, 6.0) 1.4 (−7.5, 11)

−0.2 (−3.6, 3.3) −0.2 (−3.7, 3.2) 0.3 (−3.2, 3.9) −1.3 (−4.7, 2.1) 2.1 (−1.3, 5.6) −1.0 (−4.4, 2.4) −1.5 (−4.9, 2.0) 1.5 (−2.0, 4.9) −1.4 (−4.9, 2.2) −0.9 (−4.4, 2.7) −2.0 (−5.6, 1.5)

Sex hormone-binding globulin (nmol/L)

Abbreviations: CI, confidence interval; MD, mean difference highest minus lowest quartile; %MEHP, percentage of measured DEHP metabolites excreted as MEHP. a Adjusted BMI, own and parental smoking and time of day. b Back-transformed to relative difference (%) from ln-transformation. ⁎ Overall F test p b 0.05.

J. Axelsson et al. / Environment International 85 (2015) 54–60

2014). Residual confounding may have played a role if environmental, life style related or metabolic factors were associated with both the phthalate metabolite markers and the reproductive outcomes. Nevertheless, the hypothesis of a biological link between exposure to certain phthalates and a decreased semen quality or altered reproductive hormones is strengthened by the fact that DEHP, as well as DBP, are considered to belong to the most toxic phthalates regarding effects on the male reproductive system (Martino-Andrade and Chahoud, 2010). Furthermore, more specific negative effects have been reported for DEHP, and DBP, on human sperm motility in vitro (Fredricsson et al., 1993; Pant et al., 2010), as well as on the sperm motility of rodents (Agarwal et al., 1986; Aly et al., 2015; Erkekoglu et al., 2011; Kwack et al., 2009; Lamb et al., 1987; Madkour, 2014; Zhou et al., 2010). In addition, we found the strongest association with a lower sperm motility for levels of the metabolite MECPP which is considered to be the most reliable exposure marker for DEHP (Frederiksen et al., 2010; Wittassek et al., 2011). The associations with sperm motility were, overall, stronger for the secondary metabolites than for the primary metabolite MEHP. This might be explained by the longer half-lives of the secondary metabolites, making them more suitable to capture the average exposure (Wittassek et al., 2011). The strongest correlation between serum and urinary levels was found for MECPP, indicating the suitability of this specific metabolite as exposure marker. It remains unexplained why %MEHP was not associated with a lower progressive motility, but this picture might indicate that the association between DEHP exposure and lower motility was not causal, or on the other hand, that %MEHP has a limited suitability as an exposure marker. Nonetheless, one potential biological mechanism behind a reduced sperm motility could be a disturbed acquisition of sperm motility in the epididymis, as due to increased oxidative stress (Zhou et al., 2010, 2011), given that DEHP has been associated with oxidative stress in humans (Ferguson et al., 2011, 2012; Kim et al., 2013). Another possibility is a preterm detachment of the germ cells from the supporting Sertoli cells. Such an effect has been reported in animal testicular tissue for DEHP (Breslin et al., 2013; Erkekoglu et al., 2011; Gray, 1986; Gray and Beamand, 1984; Guibert et al., 2013). Such a mechanism might lead to a higher proportion of immature sperm (in line with the association we found with a higher HDS) which indeed has been associated with a reduced sperm motility. (Giwercman et al., 2003; Sills et al., 2004). Providing a negative effect of DEHP on human sperm motility, the clinical consequence of a high exposure to DEHP might be a reduced fertility since sperm motility seems linearly associated with pregnancy rates (Larsen et al., 2000; van der Steeg et al., 2011). Although it has been hypothesized that the early fetal life represents the most sensitive period of the human male reproductive system in respect to adverse effects of the environment (Skakkebaek et al., 2001), our data might indicate that exposure during adulthood may negatively affect the male reproductive function, including fertility. 5. Conclusions In young Swedish men from the general population, levels of DEHP metabolites were associated with a lower proportion of progressively motile and mature spermatozoa. These findings might indicate that exposure in adulthood to DEHP has a negative impact on male fertility. Competing financial interests None. Acknowledgments Sources of financial support: Skåne County Council's Research and Development Foundation (Grant No: REGSKANE-58821, 120081, 192641, 270931 and 350291); Skåne University Hospital Funds;

59

Swedish Research Council (Grant No: 2014–3185), Swedish Governmental Funding for Clinical Research; and The Swedish Environmental Protection Agency (235-1780-08). The sponsors were not involved in the study design, collection, analyses or the interpretation of the data, writing or in decision to submit the article. We thank Cecilia Tingsmark, Mania Winitsky and others for the semen analyses, Åsa Amilon and Agneta Kristensen for analyzing exposure markers, and the National Service Administration in Sweden, especially Erna Jeppson-Stridsberg in Kristianstad for recruiting study subjects. Appendix A. Supplementary data Supplementary data to this article can be found online at http://dx. doi.org/10.1016/j.envint.2015.07.005. References Adibi, J.J., Whyatt, R.M., Williams, P.L., Calafat, A.M., Camann, D., Herrick, R., Nelson, H., Bhat, H.K., Perera, F.P., Silva, M.J., Hauser, R., 2008. Characterization of phthalate exposure among pregnant women assessed by repeat air and urine samples. Environ. Health Perspect. 116, 467–473. Agarwal, D.K., Eustis, S., Lamb, J.C.T., Reel, J.R., Kluwe, W.M., 1986. Effects of di(2ethylhexyl) phthalate on the gonadal pathophysiology, sperm morphology, and reproductive performance of male rats. Environ. Health Perspect. 65, 343–350. Aly, H.A., Hassan, M.H., El-Beshbishy, H.A., Alahdal, A.M., Osman, A.M., 2015. Dibutyl phthalate induces oxidative stress and impairs spermatogenesis in adult rat. Toxicol. Ind. Health http://dx.doi.org/10.1177/0748233714566877. Andersen, A.G., Jensen, T.K., Carlsen, E., Jorgensen, N., Andersson, A.M., Krarup, T., Keiding, N., Skakkebaek, N.E., 2000. High frequency of sub-optimal semen quality in an unselected population of young men. Hum. Reprod. 15, 366–372. Aschengrau, A., Seage, G.R., 2014. Random Error. Essentials of Epidemiology in Public Health. Jones & Bartlett Learning, Burlington, Mass. Axelsson, J., Rylander, L., Rignell-Hydbom, A., Giwercman, A., 2011. No secular trend over the last decade in sperm counts among Swedish men from the general population. Hum. Reprod. 26, 1012–1016. Axelsson, J., Rylander, L., Rignell-Hydbom, A., Silfver, K.A., Stenqvist, A., Giwercman, A., 2013. The impact of paternal and maternal smoking on semen quality of adolescent men. PLoS ONE 8, e66766. Breslin, W.J., Paulman, A., Sun-Lin, D., Goldstein, K.M., Derr, A., 2013. The inhibin B (InhB) response to the testicular toxicants mono-2-ethylhexyl phthalate (MEHP), 1,3 dinitrobenzene (DNB), or carbendazim (CBZ) following short-term repeat dosing in the male rat. Birth Defects Res. B Dev. Reprod. Toxicol. 98, 72–81. Calafat, A.M., Koch, H.M., Swan, S.H., Hauser, R., Goldman, L.R., Lanphear, B.P., Longnecker, M.P., Rudel, R.A., Teitelbaum, S.L., Whyatt, R.M., Wolff, M.S., 2013. Misuse of blood serum to assess exposure to bisphenol A and phthalates. Breast Cancer Res. 15, 403. Chauvigne, F., Menuet, A., Lesne, L., Chagnon, M.C., Chevrier, C., Regnier, J.F., Angerer, J., Jegou, B., 2009. Time- and dose-related effects of di-(2-ethylhexyl) phthalate and its main metabolites on the function of the rat fetal testis in vitro. Environ. Health Perspect. 117, 515–521. Desdoits-Lethimonier, C., Albert, O., Le Bizec, B., Perdu, E., Zalko, D., Courant, F., Lesne, L., Guille, F., Dejucq-Rainsford, N., Jegou, B., 2012. Human testis steroidogenesis is inhibited by phthalates. Hum. Reprod. 27, 1451–1459. Duty, S.M., Silva, M.J., Barr, D.B., Brock, J.W., Ryan, L., Chen, Z., Herrick, R.F., Christiani, D.C., Hauser, R., 2003. Phthalate exposure and human semen parameters. Epidemiology 14, 269–277. Elzanaty, S., Malm, J., Giwercman, A., 2005. Duration of sexual abstinence: epididymal and accessory sex gland secretions and their relationship to sperm motility. Hum. Reprod. 20, 221–225. Erkekoglu, P., Zeybek, N.D., Giray, B., Asan, E., Arnaud, J., Hincal, F., 2011. Reproductive toxicity of di(2-ethylhexyl) phthalate in selenium-supplemented and seleniumdeficient rats. Drug Chem. Toxicol. 34, 379–389. Evenson, D.P., Larson, K.L., Jost, L.K., 2002. Sperm chromatin structure assay: its clinical use for detecting sperm DNA fragmentation in male infertility and comparisons with other techniques. J. Androl. 23, 25–43. Ferguson, K.K., Loch-Caruso, R., Meeker, J.D., 2011. Urinary phthalate metabolites in relation to biomarkers of inflammation and oxidative stress: NHANES 1999–2006. Environ. Res. 111, 718–726. Ferguson, K.K., Loch-Caruso, R., Meeker, J.D., 2012. Exploration of oxidative stress and inflammatory markers in relation to urinary phthalate metabolites: NHANES 1999–2006. Environ. Sci. Technol. 46, 477–485. Fong, J.P., Lee, F.J., Lu, I.S., Uang, S.N., Lee, C.C., 2015. Relationship between urinary concentrations of di(2-ethylhexyl) phthalate (DEHP) metabolites and reproductive hormones in polyvinyl chloride production workers. Occup. Environ. Med. 72, 346–353. Frederiksen, H., Jorgensen, N., Andersson, A.M., 2010. Correlations between phthalate metabolites in urine, serum, and seminal plasma from young Danish men determined by isotope dilution liquid chromatography tandem mass spectrometry. J. Anal. Toxicol. 34, 400–410. Fredricsson, B., Moller, L., Pousette, A., Westerholm, R., 1993. Human sperm motility is affected by plasticizers and diesel particle extracts. Pharmacol. Toxicol. 72, 128–133.

60

J. Axelsson et al. / Environment International 85 (2015) 54–60

Fromme, H., Bolte, G., Koch, H.M., Angerer, J., Boehmer, S., Drexler, H., Mayer, R., Liebl, B., 2007. Occurrence and daily variation of phthalate metabolites in the urine of an adult population. Int. J. Hyg. Environ. Health 210, 21–33. Giwercman, A., Richthoff, J., Hjollund, H., Bonde, J.P., Jepson, K., Frohm, B., Spano, M., 2003. Correlation between sperm motility and sperm chromatin structure assay parameters. Fertil. Steril. 80, 1404–1412. Glombik, K., Basta-Kaim, A., Sikora-Polaczek, M., Kubera, M., Starowicz, G., Styrna, J., 2014. Curcumin influences semen quality parameters and reverses the di(2ethylhexyl)phthalate (DEHP)-induced testicular damage in mice. Pharmacol. Rep. 66, 782–787. Grace-Martin, K., 2008–2015. How to calculate effect size statistics. The Analysis Factor. Gray, T.J., 1986. Testicular toxicity in vitro: sertoli-germ cell co-cultures as a model system. Food Chem. Toxicol. 24, 601–605. Gray, T.J., Beamand, J.A., 1984. Effect of some phthalate esters and other testicular toxins on primary cultures of testicular cells. Food Chem. Toxicol. 22, 123–131. Guibert, E., Prieur, B., Cariou, R., Courant, F., Antignac, J.P., Pain, B., Brillard, J.P., Froment, P., 2013. Effects of mono-(2-ethylhexyl) phthalate (MEHP) on chicken germ cells cultured in vitro. Environ. Sci. Pollut. Res. Int. 20, 2771–2783. Hammiche, F., Laven, J.S., Twigt, J.M., Boellaard, W.P., Steegers, E.A., Steegers-Theunissen, R.P., 2012. Body mass index and central adiposity are associated with sperm quality in men of subfertile couples. Hum. Reprod. 27, 2365–2372. Han, X., Cui, Z., Zhou, N., Ma, M., Li, L., Li, Y., Lin, H., Ao, L., Shu, W., Liu, J., Cao, J., 2014. Urinary phthalate metabolites and male reproductive function parameters in Chongqing general population, China. Int. J. Hyg. Environ. Health 217, 271–278. Hauser, R., 2008. Urinary phthalate metabolites and semen quality: a review of a potential biomarker of susceptibility. Int. J. Androl. 31, 112–117. Hauser, R., Meeker, J.D., Duty, S., Silva, M.J., Calafat, A.M., 2006. Altered semen quality in relation to urinary concentrations of phthalate monoester and oxidative metabolites. Epidemiology 17, 682–691. Hoppin, J.A., Brock, J.W., Davis, B.J., Baird, D.D., 2002. Reproducibility of urinary phthalate metabolites in first morning urine samples. Environ. Health Perspect. 110, 515–518. Huang, L.P., Lee, C.C., Hsu, P.C., Shih, T.S., 2011. The association between semen quality in workers and the concentration of di(2-ethylhexyl) phthalate in polyvinyl chloride pellet plant air. Fertil. Steril. 96, 90–94. Huang, L.P., Lee, C.C., Fan, J.P., Kuo, P.H., Shih, T.S., Hsu, P.C., 2014. Urinary metabolites of di(2-ethylhexyl) phthalate relation to sperm motility, reactive oxygen species generation, and apoptosis in polyvinyl chloride workers. Int. Arch. Occup. Environ. Health 87, 635–646. Joensen, U.N., Frederiksen, H., Jensen, M.B., Lauritsen, M.P., Olesen, I.A., Lassen, T.H., Andersson, A.M., Jorgensen, N., 2012. Phthalate excretion pattern and testicular function: a study of 881 healthy Danish men. Environ. Health Perspect. 120, 1397–1403. Jönsson, B.A., Richthoff, J., Rylander, L., Giwercman, A., Hagmar, L., 2005. Urinary phthalate metabolites and biomarkers of reproductive function in young men. Epidemiology 16, 487–493. Jurewicz, J., Radwan, M., Sobala, W., Ligocka, D., Radwan, P., Bochenek, M., Hawula, W., Jakubowski, L., Hanke, W., 2013. Human urinary phthalate metabolites level and main semen parameters, sperm chromatin structure, sperm aneuploidy and reproductive hormones. Reprod. Toxicol. 42, 232–241. Kay, V.R., Bloom, M.S., Foster, W.G., 2014. Reproductive and developmental effects of phthalate diesters in males. Crit. Rev. Toxicol. 1–32. Kim, J.H., Park, H.Y., Bae, S., Lim, Y.H., Hong, Y.C., 2013. Diethylhexyl phthalates is associated with insulin resistance via oxidative stress in the elderly: a panel study. PLoS ONE 8, e71392. Kranvogl, R., Knez, J., Miuc, A., Voncina, E., Voncina, D.B., Vlaisavljevic, V., 2014. Simultaneous determination of phthalates, their metabolites, alkylphenols and bisphenol a using GC–MS in urine of men with fertility problems. Acta Chim. Slov. 61, 110–120. Kwack, S.J., Kim, K.B., Kim, H.S., Lee, B.M., 2009. Comparative toxicological evaluation of phthalate diesters and metabolites in Sprague–Dawley male rats for risk assessment. J. Toxic. Environ. Health A 72, 1446–1454. Lamb, J.C.T., Chapin, R.E., Teague, J., Lawton, A.D., Reel, J.R., 1987. Reproductive effects of four phthalic acid esters in the mouse. Toxicol. Appl. Pharmacol. 88, 255–269. Larsen, L., Scheike, T., Jensen, T.K., Bonde, J.P., Ernst, E., Hjollund, N.H., Zhou, Y., Skakkebaek, N.E., Giwercman, A., 2000. Computer-assisted semen analysis parameters as predictors for fertility of men from the general population. The Danish First Pregnancy Planner Study Team. Hum. Reprod. 15, 1562–1567. Levine, T.R., Hullett, C.R., 2010. Partial eta-squared. In: Salkind, N.J. (Ed.), Encyclopedia of Research Design. Sage Publications, Inc. Levitas, E., Lunenfeld, E., Weiss, N., Friger, M., Har-Vardi, I., Koifman, A., Potashnik, G., 2005. Relationship between the duration of sexual abstinence and semen quality: analysis of 9489 semen samples. Fertil. Steril. 83, 1680–1686. Madkour, N.K., 2014. Beneficial role of celery oil in lowering the di(2-ethylhexyl) phthalate-induced testicular damage. Toxicol. Ind. Health 30, 861–872.

Martino-Andrade, A.J., Chahoud, I., 2010. Reproductive toxicity of phthalate esters. Mol. Nutr. Food Res. 54, 148–157. Meeker, J.D., Calafat, A.M., Hauser, R., 2009. Urinary metabolites of di(2-ethylhexyl) phthalate are associated with decreased steroid hormone levels in adult men. J. Androl. 30, 287–297. Meeker, J.D., Calafat, A.M., Hauser, R., 2012. Urinary phthalate metabolites and their biotransformation products: predictors and temporal variability among men and women. J. Expo. Sci. Environ. Epidemiol. 22, 376–385. Miura, Y., Naito, M., Ablake, M., Terayama, H., Yi, S.Q., Qu, N., Cheng, L.X., Suna, S., Jitsunari, F., Itoh, M., 2007. Short-term effects of di-(2-ethylhexyl) phthalate on testes, liver, kidneys and pancreas in mice. Asian J. Androl. 9, 199–205. Oishi, S., 1985. Reversibility of testicular atrophy induced by Di(2-ethylhexyl) phthalate in rats. Environ. Res. 36, 160–169. Pan, G., Hanaoka, T., Yoshimura, M., Zhang, S., Wang, P., Tsukino, H., Inoue, K., Nakazawa, H., Tsugane, S., Takahashi, K., 2006. Decreased serum free testosterone in workers exposed to high levels of di-n-butyl phthalate (DBP) and di-2-ethylhexyl phthalate (DEHP): a cross-sectional study in China. Environ. Health Perspect. 114, 1643–1648. Pant, N., Shukla, M., Kumar Patel, D., Shukla, Y., Mathur, N., Kumar Gupta, Y., Saxena, D.K., 2008. Correlation of phthalate exposures with semen quality. Toxicol. Appl. Pharmacol. 231, 112–116. Pant, N., Pant, A., Shukla, M., Mathur, N., Gupta, Y., Saxena, D., 2010. Environmental and experimental exposure of phthalate esters: the toxicological consequence on human sperm. Hum. Exp. Toxicol. 30, 507–514. Pant, N., Pant, A., Shukla, M., Mathur, N., Gupta, Y., Saxena, D., 2011. Environmental and experimental exposure of phthalate esters: the toxicological consequence on human sperm. Hum. Exp. Toxicol. 30, 507–514. Pant, N., Kumar, G., Upadhyay, A.D., Patel, D.K., Gupta, Y.K., Chaturvedi, P.K., 2014. Reproductive toxicity of lead, cadmium, and phthalate exposure in men. Environ. Sci. Pollut. Res. Int. 21, 11066–11074. Peck, J.D., Sweeney, A.M., Symanski, E., Gardiner, J., Silva, M.J., Calafat, A.M., Schantz, S.L., 2010. Intra- and inter-individual variability of urinary phthalate metabolite concentrations in Hmong women of reproductive age. J. Expo. Sci. Environ. Epidemiol. 20, 90–100. Sanchez-Pozo, M.C., Mendiola, J., Serrano, M., Mozas, J., Bjorndahl, L., Menkveld, R., Lewis, S.E., Mortimer, D., Jorgensen, N., Barratt, C.L., Fernandez, M.F., Castilla, J.A., 2013. Proposal of guidelines for the appraisal of SEMen QUAlity studies (SEMQUA). Hum. Reprod. 28, 10–21. Sills, E.S., Fryman, J.T., Perloe, M., Michels, K.B., Tucker, M.J., 2004. Chromatin fluorescence characteristics and standard semen analysis parameters: correlations observed in andrology testing among 136 males referred for infertility evaluation. J. Obstet. Gynaecol. 24, 74–77. Skakkebaek, N.E., Rajpert-De Meyts, E., Main, K.M., 2001. Testicular dysgenesis syndrome: an increasingly common developmental disorder with environmental aspects. Hum. Reprod. 16, 972–978. Specht, I.O., Toft, G., Hougaard, K.S., Lindh, C.H., Lenters, V., Jonsson, B.A., Heederik, D., Giwercman, A., Bonde, J.P., 2014. Associations between serum phthalates and biomarkers of reproductive function in 589 adult men. Environ. Int. 66, 146–156. Tabachnick, B.G., Fidell, L.S., 2013. Multiple regression. In: Tabachnick, B.G., Fidell, L.S. (Eds.), Using Multivariate Statistics. Boston, Mass.Pearson, London Toft, G., Jonsson, B.A., Lindh, C.H., Jensen, T.K., Hjollund, N.H., Vested, A., Bonde, J.P., 2012. Association between pregnancy loss and urinary phthalate levels around the time of conception. Environ. Health Perspect. 120, 458–463. van der Steeg, J.W., Steures, P., Eijkemans, M.J., JD, F.H., Hompes, P.G., Kremer, J.A., van der Leeuw-Harmsen, L., Bossuyt, P.M., Repping, S., Silber, S.J., Mol, B.W., van der Veen, F., 2011. Role of semen analysis in subfertile couples. Fertil. Steril. 95, 1013–1019. Vermeulen, A., Verdonck, L., Kaufman, J.M., 1999. A critical evaluation of simple methods for the estimation of free testosterone in serum. J. Clin. Endocrinol. Metab. 84, 3666–3672. Vine, M.F., 1996. Smoking and male reproduction: a review. Int. J. Androl. 19, 323–337. Virtanen, H.E., Sadov, S., Toppari, J., 2012. Prenatal exposure to smoking and male reproductive health. Curr. Opin. Endocrinol. Diabetes Obes. 19, 228–232. Wittassek, M., Koch, H.M., Angerer, J., Bruning, T., 2011. Assessing exposure to phthalates — the human biomonitoring approach. Mol. Nutr. Food Res. 55, 7–31. World Health Organization, 1999. WHO Laboratory Manual for the Examination of Human Semen and Sperm-Cervical Mucus Interaction. Published on Behalf of the World Health Organization [by] Cambridge University Press, Cambridge, UK; New York, NY. Zhou, D., Wang, H., Zhang, J., Gao, X., Zhao, W., Zheng, Y., 2010. Di-n-butyl phthalate (DBP) exposure induces oxidative damage in testes of adult rats. Syst. Biol. Reprod. Med. 56, 413–419. Zhou, D., Wang, H., Zhang, J., 2011. Di-n-butyl phthalate (DBP) exposure induces oxidative stress in epididymis of adult rats. Toxicol. Ind. Health 27, 65–71.