Association between perfluoroalkyl substances and reproductive hormones in adolescents and young adults
Meng-Shan Tsaia, Chien-Yu Linb,c, Ching-Chun Lina, Mei-Huei Chend, Sandy H.J.
Hsue, Kuo-Liong Chienf,g, Fung-Chang Sungh, Pau-Chung Chena,i,j,*, Ta-Chen Sug,j,*
a Institute of Occupational Medicine and Industrial Hygiene, College of Public Health,
National Taiwan University, Taipei 100, Taiwan
b Department of Internal Medicine, En Chu Kong Hospital, New Taipei City 237,
Taiwan
c School of Medicine, Fu Jen Catholic University, Taipei County 242, Taiwan
d Department of Pediatrics, National Taiwan University Hospital Yun-Lin Branch
e Department of Laboratory Medicine, National Taiwan University Hospital, Taipei,
Taiwan
f Institute of Epidemiology and Preventive Medicine, College of Public Health,
National Taiwan University, Taipei 100, Taiwan
g Department of Internal Medicine, National Taiwan University Hospital, Taipei 100,
Taiwan
h Department of Public Health, College of Public Health, China Medical University,
Taichung 404, Taiwan 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 1
i Department of Public Health, National Taiwan University College of Public Health,
Taipei, Taiwan
j Department of Environmental and Occupational Medicine, National Taiwan
University Hospital, Taipei 100, Taiwan
*Corresponding author: Ta-Chen Su, MD, PhD
Department of Internal Medicine, National Taiwan University Hospital, 7 Chung-Shan South Road, Taipei 10055, Taiwan.
Tel: +886-2-23123456 ext 66719 Fax: +886-2-23712361 Email: tachensu@ntu.edu.tw
**Co-correspondence:
Pau-Chung Chen, MD, PhD
Institute of Occupational Medicine and Industrial Hygiene, National Taiwan University College of Public Health, #17 Syujhou Road, Taipei 10055, Taiwan.
Telephone: +886-2-3366-8088 Fax: +886-2-3366-8734 Email: pchen@ntu.edu.tw 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38
Running title: Perfluoroalkyl substances and reproductive hormones. Acknowledgements
This study was supported by grants from National Health Research Institute of
Taiwan (EX97-9721PC, EX97-9821PC, X97-9921PC, EX95-9531PI, EX95-9631PI
and EX95-9731PI), from Ministry of Science and Technology
(101-2314-B-002-184-MY3) and (99-2314-B-385-001-MY3 and 102-2314-B-002-166-(101-2314-B-002-184-MY3), and from
Taiwan and the Environmental Medicine Collaboration Center (NTUH 103 A123 and
104 A123). This work was supported in part by the 3rd core facility at National
Taiwan University Hospital.
Conflict of Interest
We declare there is no conflict of interest regarding this manuscript, including financial, consultant, institutional, and other relationships that might lead to bias or a conflict of interest. 39 40 41 42 43 44 45 46 47 48 49 50 51 5
Highlights
Serum concentration of PFAS was associated with reproductive hormone based on a
young population.
The concentrations of PFOA, PFOS, and PFUA associated with reproductive
hormone significantly.
Reproductive hormones of females, ages 12-17, were significantly influenced by
PFAS concentrations. 52 53 54 55 56 57 58
Abstract
Background: Few studies have explored the association between perfluoroalkyl
substances (PFAS) and reproductive hormones in adolescents and young adults.
Objectives: This study aimed to investigate the association of PFAS with
reproductive hormones in adolescents and young adults.
Methods: We recruited 540 subjects aged 12-30 years from a 1992-2000 mass urine
screening population and established a cohort from 2006 to 2008 via invitations by
mail or/and telephone. Serum PFAS levels were analyzed with a Waters ACQUITY
UPLC system coupled with a Waters Quattro Premier XE triple quadrupole mass
spectrometer. Serum reproductive hormone levels were measured by
immunoluminometric assay with an Architect random access assay system. PFAS
levels were divided into different percentiles according to their detection limits in the
multiple regression models to analyze associations between reproductive hormone
levels and exposure with PFAS.
Results: The adjusted mean serum level of sex hormone-binding globulin (SHBG)
decreased significantly in association with t he <50th, 50–75, 75-90 and >90th
percentile categories of perfluorooctanoic acid (PFOA) compared with a reference
category for the females in the 12-17-year-old group. The follicle-stimulating
hormone (FSH) levels were significantly decreased in association with the different
1 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 9
percentile categories of perfluorooctane sulfonate (PFOS) in the male 12-17-year-old
group and the different percentile categories of perfluoroundecanoic acid (PFUA) in
the female 12-17-year-old group. The serum FSH levels in the females aged 12-17
were also decreased in association with the different percentile categories of PFUA.
On the other hand, there was a significantly negative association between the different
percentile categories of PFOS and the serum testosterone level among the female
12-17-year-old group.
Conclusions: We found that the serum concentrations of PFOA, PFOS, and PFUA
were negatively associated with the serum levels of SHBG, FSH, and testosterone in
the young Taiwanese population and that these effects were the strongest in the
females aged 12-17.Further studies are needed to determine whether these
associations are causal.
Key words: perfluoroalkyl substances, reproductive hormone, adolescent, young
adults 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92
Introduction
Perfluoroalkyl substances (PFAS), such as perfluorooctanoic acid (PFOA) and
perfluorooctane sulfonate (PFOS), are highly bio-accumulative environmental
pollutants (Austin et al. 2003) that, which are widely used for both industrial and
everyday purposes (Lau et al. 2004). Although food (including migration from
packaging and cookware) and drinking water are the primary sources of PFAS
exposure in humans, additional exposure routes include air and dust (Haug et al.
2011).Therefore, these substances are widespread in the environment, affecting
wildlife and humans (Kannan et al. 2004). In addition, they have been associated with
adverse health effects. Until now, the majority of animal studies have associated
exposure to PFAS with developmental deficits (Lau et al. 2004), neurotoxicity
(Johansson et al. 2008), and immunotoxicity (Keil et al. 2008). Epidemiological
research has shown an association among PFAS exposure, child development (Chen
et al. 2013), higher thyroid levels (Lin et al. 2013b), and immune system function
(Dong et al. 2013).
PFOS and PFOA are well-recognized endocrine disruptors that have antagonistic
effects on the synthesis of steroid hormone (Zhao et al. 2010). Reproductive
hormones are important for the reproductive system because they play pivotal roles in
both male and female puberty development and are crucial to growth and the
3 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 13
functioning of a broad range of tissues (Kjeldsen and Bonefeld-Jorgensen 2013). In
general, the average age of the onset of puberty is ten years for girls and twelve years
for boys, suggesting that the female onset of puberty occurs earlier than that of males
(Sorensen et al. 2012). Few prior epidemiological studies have investigated the impact
of PFAS on the human reproductive system, and results have been inconsistent. J
Joensen et al. selected 247 healthy young Danish men with a median age of 19 years
to examine the association between reproductive hormones, semen quality and PFAS
in the general population (Joensen et al. 2013). They found that the serum PFOS
concentration was negatively associated with testosterone levels. Additionally,
Lopez-Espinosa et al. have demonstrated that the delay of puberty in children is correlated
with PFOS and PFOA levels based on the level of testosterone or oestradiol and the
self-reported status (Lopez-Espinosa et al. 2011). A nested case-control study reported
that PFAS exposure during pregnancy was not associated with age at menarche in a
British cohort (Christensen et al. 2011). Furthermore, a recent study of a pregnant
Danish cohort has found that higher levels of PFOA exposure in utero may be
associated with a later age of menarche compared with lower levels of exposure
(Kristensen et al. 2013). In animals, PFOA has been associated with decreased serum
testosterone levels in Leydig cell adenomas (Cook et al. 1992) and increased
oestradiol levels in rodents (Biegel et al. 1995). The mechanism of the endocrine-112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130
disrupting activities of PFAS has been discussed by Lau et al., who have reported that
this estrogenic effect is mediated via the oestrogen receptor pathway (Lau et al. 2007).
Studies investigating the impact of PFAS on human reproductive health are limited
and controversial and have mainly focused on PFOA and PFOS. Furthermore, there
are few studies investigating the health effects of PFAS in adolescents. Therefore, the
aim of this study was to assess the association between PFAS and reproductive
hormones in young adolescents and young adults.
5 131 132 133 134 135 136 137 17
Materials and methods
Subjects and data collection
The study groups were established from a 1992-2000 mass urine screening
population of individuals attending grades 1-12 in Taiwan (Wei et al. 2003). From
2006 to 2008, we invited students in the Taipei area to participate in the study and a
follow-up health examination at the National Taiwan University Hospital. Trained
assistants and nurses invited these subjects by mail or/and telephone to undergo the
health examination and complete a questionnaire.The information has been detailed
in previous studies (Lin et al. 2013a; Lin et al. 2013b; Su et al. 2014). Out of 7,097
subjects living in Taipei, 790 students completed the follow-up health examination,
including the collection of a blood sample and the completion of a questionnaire, but
we did not measure the serum PFAS or reproductive hormone levels of all of the
subjects due to the collection of limited serum samples. Among the 790 students, 145
had an insufficient serum sample volume for PFAS measurement, and an additional
105 had an insufficient reproductive hormone measurement. Thus, 540 subjects with
both PFAS and reproductive hormone measurements were included in the final
analysis. This study was approved by the Ethics Committee of the National Taiwan
University Hospital (Research Ethics Committee, NTUH). All of the participants and
their parents (for the child and adolescent participants) signed informed consent 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156
documents upon enrolment in the study.
A questionnaire, including patient age, gender, BMI, lifestyle (drinking, eating, and
exercise), household income, and dietary intake (high fat and high sugar), were
recorded during the follow-up examinations performed from 2006 to 2008. The
subjects were separated by gender and were subdivided by age into 12-17-year-old
and 18-30-year-old groups according to the clinical (Moore et al. 2013) definition of
adolescence as the period from 12 to 17 years of age. Moreover, the general
population has been reported to experience puberty between the ages of 10 and 17
years (Parent et al. 2003). Smoking status (active smoker, passive smoker or has never
smoked) and alcohol intake (current alcohol consumption or no alcohol consumption)
were determined via the questionnaire and categorized. Household income groups
were categorized as above 50,000 New Taiwan dollars (NTD) (equivalent to $1,600
USD) per month or below. Weight and height were measured during the follow-up
health examination. Body mass index (BMI) was determined as the weight (in
kilograms) divided by the square of the height (in metres). Exercise was assessed
based on the presence or absence of current exercise habits. A high-sugar diet was
defined as one in which subjects consumed sweet foods and soft drinks at a frequency
of four times a week or more. Subjects that consumed ham, fatty foods, and fast foods
at a frequency of four times a week or more were categorized as having a high-fat
7 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 21
diet.
Exposure assessment
All of the plasma samples were stored at -80 ℃ prior to PFAS analysis. A previous
study has reported a fast and sensitive ultra-high performance liquid
chromatography/tandem mass spectrometry method to determine PFAS levels (Lien
et al. 2011). We first analysed the levels of 12 PFAS, including potassium
perfluorohexanesulfonate (PFHxS), perfluoroheptanoic acid (PFHpA),
perfluorononanoic acid (PFNA), perfluorooctanoic acid (PFOA), perfluorooctyl
sulfonate (PFOS), perfluorodecanoic acid (PFDeA), perfluoroundecanoic acid
(PFUA), perfluorododecanoic acid (PFDoA), 2-(N-methyl-perfluorooctane
sulfonamido) acetic acid (Me–PFOSA–AcOH), 2-(N-ethylperfluorooctane
sulfonamido) acetic acid (Et–PFOSA–AcOH), perfluorohexanoic acid (PFHxA), and
perfluorooctane sulfonamide (PFOSA). However, the levels of eight of these PFAS
were more than 70% below the limit of quantitation (LOQ). Therefore, only PFOA,
PFOS, PFNA, and PFUA were used for the final analysis. The details of the analytical
methods have been previously described (Lien et al. 2011; Lin et al. 2013b). The
samples were first vortexed for homogeneity for 30 seconds. An additional 30-second
vortex was performed following the addition of 100 µL of 1% formic acid (pH=2.8)
to the 100-µl plasma samples. Eighty microliters of methanol and 20 µl of a 0.375 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194
ng/mL internal standard (13C8-PFOA) in methanol were mixed with the solution, and
the mixture was then sonicated for 20 minutes and centrifuged at 14,000 rpm for 20
minutes. Before analysis, the supernatant was collected and filtered through a 0.22 µm
PVDF syringe filter. One hundred microliters of bovine plasma with standard
calibration solutions were prepared as described above. The concentrations of the
specific analytes were equivalent to 0.05 mL in 300 ng/mL bovine plasma with a
fixed amount of internal standard (75 ng/mL). All samples were analysed using a
Waters ACQUITY UPLC System (Waters Corporation, Milford, MA) coupled with a
Waters Quattro Premier XE triple quadrupole mass spectrometer (Waters
Corporation, Milford, MA). The limit of quantitation (LOQ) for PFOA and PFUA
was 1.5 ng/mL, that for PFOS was 0.22 ng/mL, and that for PFNA was 0.75 ng/mL.
No PFOS and trace background amounts of PFOA (up to 1.5 ng/mL), PFNA (up to
0.75 ng/mL), and PFUA (up to 3 ng/mL) were detected in the blank samples.
Therefore, the reported PFOA, PFNA, and PFUA concentrations were corrected by
subtracting the background levels of the blank. The PFAS concentration was below
the detection limits (37.4% for PFOA, 1.5% for PFOS, 55% for PFNA and 25% for
PFUA); therefore, we used a proxy value of half of the detection limit.
Assessment of reproductive hormones
The samples were stored at -80 ℃ after centrifugation until analysis was performed.
9 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 25
The reproductive hormones and PFAS were analysed at the same time. The serum
concentrations of reproductive hormones, including sex hormone-binding globulin
(SHBG), oestrogen (E2), follicle-stimulating hormone (FSH), luteinizing hormone
(LH), testosterone (T), and free testosterone (free T), were measured by
immunoluminometric assay with an Architect random access assay system (Abbott
Diagnostics, Abbott Park, IL). The intra-assay coefficients of variation of these
measurements were all below 10%, and the inter-assay coefficients of variation were
all below 15%.
Statistical analysis
The PFAS concentrations were described as the geometric mean and geometric
standard deviation. The relationships among the PFAS were assessed using the
Spearman correlation coefficient. The Mann–Whitney U test or Kruskal–Wallis test
was used to evaluation the relationships among the PFAS variables and categorical
variables. We divided each PFAS concentration into different categories for linear
regression analysis due to the high percentage of PFAS below the detection limit
(37.4% for PFOA, 1.5% for PFOS, 55% for PFNA and 25% for PFUA). The PFOA
levels were divided into the 50th (the reference category), 75th, and 90th percentiles,
and the PFOS and PFUA levels were divided into the 25th (the reference category),
50th, and 75th percentiles. Additionally, the PFNA levels were divided into the 60th 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232
(the reference category) and 90th percentiles. The SHBG, E2, FSH, LH, T, and free T
levels were skewed and were thus log-transformed in the regression models.
Covariates, including age, sex, BMI, and high-fat diet intake, were considered to be
significant predictive outcomes in the regression model because they changed the
estimates by >10%. Other covariates, such as household income, alcohol
consumption, smoking status, high-sugar diet, and exercise, were considered but were
not included in the final model.
E Each PFAS model was analysed separately. SAS (version 9.3; SAS Institute Inc.,
Cary, NC, USA) was used to perform all statistical analyses and a p-value of <0.05
was considered to be statistically significant.
11 233 234 235 236 237 238 239 240 241 242 29
Results
The basic characteristics of the sample population are shown in Table 1. The
geometric mean and geometric standard deviation of the concentrations of PFOA,
PFOS, PFNA, and PFUA were 2.74 (2.95) ng/mL, 7.78 (2.40) ng/mL, 1.10 (3.55)
ng/mL, and 5.84 (2.88) ng/mL, respectively. Among the 540 subjects, 330 were
female and 210 were male. The males had a significantly higher mean concentration
of PFOS than the females (p<0.001). The 18- to 30-year-old groups had higher mean
serum concentrations of PFOS than the 12- to 17-year-old groups (p<0.05). The
subjects with a higher BMI (≥24) also had a higher PFOS concentration than those
with a BMI of below 24 (P<0.05). In addition, the PFOS and PFNA serum
concentrations were higher in those consuming a high-fat diet (p<0.05). The
reproductive hormone levels are shown in Supplementary Data (Table 2).
The association between the serum level of PFAS and that of sex hormone-binding
globulin (SHBG) is shown in Table 3. The serum level of SHBG in the females
decreased significantly in association with the percentile categories (<50th, 50–75,
75-90 and >90th percentiles) of PFOA (p for trend <0.05). The association between
FSH and the serum level of PFAS after adjustments for covariates is listed in Table 4.
The mean serum follicle-stimulating hormone (FSH) level was significantly decreased
(p for trend <0.05) in association with the different percentile categories of PFOS in 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261
the male subjects aged 12-17. The female serum FSH level in the 12-17-year-old
group was also decreased (p for trend <0.01) across the different percentile categories
of PFUA. As shown in Table 5, there was a significantly negative association (p for
trend <0.05) among the percentile categories of PFOS with the female serum
testosterone level in the 12-17-year-old group, but the other three PFAS were not
associated with testosterone. However, the serum levels of PFOA, PFOS, PFNA, and
PFUA were not associated with the oestrogen level. The concentration of PFAS was
also not associated with the LH level. Moreover, the serum level of PFAS was not
associated with free testosterone. The association between PFAS and the reproductive
hormones of the subpopulations are shown in Tables S1-S7. There were no
associations between PFAS and the reproductive hormone levels stratified by BMI
and high-fat diet.
13 262 263 264 265 266 267 268 269 270 271 272 273 33
Discussion
In this study, we found that the serum level of PFAS was associated with
reproductive hormones in the younger age groups, especially PFOA, PFOS, and
PFUA. Moreover, the reproductive hormones of the females were significantly
influenced by the PFOS and PFUA concentrations. However, we did not find that an
elevation in the PFAS level was associated with the oestrogen or LH level. To our
knowledge, this is the first study to assess the serum levels of PFAS and reproductive
hormones in a cohort of adolescent and young adults; however, the effects were
determined to be minute and subclinical.
Our results revealed that the serum PFOA, PFOS, and PFUA concentrations were
higher in the 12-17-year-old group compared with the 18-30-year-old group. Due to
the existence of different growth stages, we divided the subjects into two groups and
examined the differences between them. The subjects in the 12-17-year-old group
were within the age range of puberty, which is a complex and immature stage driven
by the endocrine system (Den Hond and Schoeters 2006). One review study has
investigated the effects of endocrine disrupters on puberty and has found that
exposure to endocrine disrupters may be associated with puberty disturbances,
including delayed male and accelerated female puberty (Maranghi and Mantovani
2012). Another study has also revealed that in utero expose to high levels of PFOA 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292
may delay menarche in females (Kristensen et al. 2013). PFAS act as an endocrine
disrupter; therefore, the 12-17-year-old group was more vulnerable than the
18-30-year-old group to this exposure, which resulted in obvious effects. One study has also
found that the age of puberty onset is correlated with PFOS and PFOA concentrations
(Lopez-Espinosa et al. 2011). However, the primary mechanism is unclear and
requires further investigation.
We found that the PFOA, PFOS and PFUA concentrations were negatively
associated with reproductive hormones, except LH, oestrogen, and free testosterone.
Although there are few human studies investigating associations between PFAS
exposure and reproductive hormones in adolescent and young adults, Joensen et al.
have reported negative associations between PFOS exposure and testosterone in
young men (median age of 19 years) (Joensen et al. 2013). A study of puberty
evaluating males of a similar age found a relationship between reduced odds of the
onset of puberty with an increase in the serum level of PFOS in boys, using total
testosterone as a puberty indicator (Lopez-Espinosa et al. 2011). However, additional
studies have found no association between PFOS and reproductive hormones, except
for SHBG in the spouses of pregnant women (Specht et al. 2012), and a relationship
has also been detected between PFOS and PFOA and testosterone levels in males
(Joensen et al. 2013; Raymer et al. 2012). Thus, associations between PFAS exposure
15 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 37
and reproductive hormones have been reported; however, the possible effects of
PFAS exposure are inconsistent. The observations of the current study may provide
novel information in this regard.
Most relevant published studies have focused primarily on male adult populations
and have rarely examined the effects of PFAS on female study populations (Joensen
et al. 2013; Lopez-Espinosa et al. 2011; Raymer et al. 2012). Our study revealed that
female reproductive hormones were affected by PFAS. Female SHBG decreased with
increasing serum PFOA, FSH decreased with increasing serum PFUA, and
testosterone decreased with increasing PFOS. One possible explanation for these
findings is that females usually experience the onset of puberty earlier than males and
are thus more vulnerable to the effects of PFAS than males. An animal study reported
that the levels of ammonium perfluorooctanoate (C8), administered by gavage for 14
days to adult male rats, were associated with increased oestradiol levels (Biegel et al.
1995). However, we did not observe similar results. We found that male FSH
decreased with increasing serum PFOS. The human reproductive system is
complicated and is influenced by the hypothalamus and pituitary glands. FSH is
primarily regulated by the hypothalamus and pituitary glands and responds to
development, whereas LH serves as a secondary trigger of this regulation (Moore et
al. 2013). Therefore, we observed a significant association between FSH and PFUA 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330
and PFOS ; however, we did not observe an association between reproductive
hormones and PFAS. However, more studies, including not only animal studies but
also epidemiological studies of humans, are needed to assess the association between
PFAS and reproductive hormones.
We also found that SHBG in the female subjects was inversely associated with
PFOA. Recently, a review has investigated the bioaccumulation of PFAS and has
concluded that PFAS has a tendency towards protein binding; however, there are
some mechanisms that still need to be explained (Ng and Hungerbuhler 2014). On the
other hand, hormone levels and mechanisms such as enzyme activity (Zhao et al.
2010), cell membrane fluidity and permeability (Hu et al. 2003), and tissue
distribution (Ng and Hungerbuhler 2014) are quite complex; moreover, the
physiology of hormone regulation is rather complicated in adolescents, and most
underlying mechanisms are not well understood. However, the females in our
population did not exhibit variations in their hormone levels due to menstruation,
which is the most important limitation that may have influenced our results. These
results suggest that PFAS have endocrine-disrupting properties. Additional studies are
needed to examine the related health outcomes.
Subgroup analysis did not indicate that the hormone levels varied due to a higher
BMI or high-fat diet intake across the different PFAS concentrations; however, high
17 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 41
BMI and high-fat diet intake are risk factors for hormone imbalance. High doses of
PFOS and PFOA have been associated with weight loss in animal studies (Hines et al.
2009; Thibodeaux et al. 2003). A human epidemiological study has shown that PFOS,
PFOA, and PFNA have few meaningful associations with body weight (Nelson et al.
2010), but other studies have explored prenatal PFOA exposure and body weight later
in life (Halldorsson et al. 2012). An additional study has shown that obesity causes
alterations in the reproductive hormones of male adolescents (Zhang et al. 2013). An
animal study has revealed that high-fat diets induce the early onset of reproductive
function in female rats (Fungfuang et al. 2013). Important contributors to PFOA
intake are derived from different types of dietary routes (Noorlander et al. 2011).
Therefore, more studies are needed to investigate the associations between body
weight and reproductive hormones and between dietary intake and reproductive
hormones across different PFAS concentrations.
The strengths of our study included the use of a complete set of PFAS data and the
measurement of a variety of reproductive hormones in adolescents and young adults.
The measurements of PFAS have been validated (Lin et al. 2013b) and have been
demonstrated to have different health effects in this cohort (Lin et al. 2011; Lin et al.
2013a; Lin et al. 2013b; Su et al. 2014). This study also had several limitations. The
main limitation was that we did not consider menarche status or examine any other 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368
puberty indicators, although they may influence hormone levels and results.
Additionally, although reproductive hormone levels in women vary according to
menarche status, we considered these levels to be random sampling results. Second,
other environmental pollutants that may impact both PFAS and reproductive hormone
levels were not measured in this study. Third, causality cannot be confirmed because
this is a cross-sectional study. Furthermore, we did not take into account medications
that may impact reproductive hormones; however, more than 95% of the participants
self-reported no significant clinical diseases and no medication history. Finally, our
study subjects were obtained from a mass urine screening population in Taipei. We
cannot extrapolate the same association to the general population; however, this fact
does not necessarily negate the clinical relevance of this study.
In conclusion, we found that PFAS levels were associated with reproductive
hormone levels, particularly the higher PFOA and lower SHBG levels, higher PFUA
and PFOS and lower FSH levels, and higher PFOS and lower testosterone levels in
the adolescents and young adults. Although the observed potential biological effects
on humans were low and subclinical in the study population, the effects observed
among the females and the 12-17-year-old group are of particular interest. Further
long-term cohort studies are needed to clarify whether these associations are causal.
19 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 45
References
Austin ME, Kasturi BS, Barber M, Kannan K, MohanKumar PS, MohanKumar SM.
2003. Neuroendocrine effects of perfluorooctane sulfonate in rats. Environmental
health perspectives 111:1485-1489.
Biegel LB, Liu RC, Hurtt ME, Cook JC. 1995. Effects of ammonium
perfluorooctanoate on leydig cell function: In vitro, in vivo, and ex vivo studies.
Toxicology and applied pharmacology 134:18-25.
Chen MH, Ha EH, Liao HF, Jeng SF, Su YN, Wen TW, et al. 2013. Perfluorinated
compound levels in cord blood and neurodevelopment at 2 years of age.
Epidemiology 24:800-808.
Christensen KY, Maisonet M, Rubin C, Holmes A, Calafat AM, Kato K, et al. 2011.
Exposure to polyfluoroalkyl chemicals during pregnancy is not associated with
offspring age at menarche in a contemporary british cohort. Environment
international 37:129-135.
Cook JC, Murray SM, Frame SR, Hurtt ME. 1992. Induction of leydig cell adenomas
by ammonium perfluorooctanoate: A possible endocrine-related mechanism.
Toxicology and applied pharmacology 113:209-217.
Den Hond E, Schoeters G. 2006. Endocrine disrupters and human puberty.
International journal of andrology 29:264-271; discussion 286-290. 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402 403 404 405
polyfluoroalkyl concentrations, asthma outcomes, and immunological markers in
a case-control study of taiwanese children. Environmental health perspectives
121:507-513, 513e501-508.
Fungfuang W, Nakao N, Nakada T, Yokosuka M, Saito TR. 2013. Early onset of
reproductive function in female rats treated with a high-fat diet. The Journal of
veterinary medical science / the Japanese Society of Veterinary Science
75:523-526.
Halldorsson TI, Rytter D, Haug LS, Bech BH, Danielsen I, Becher G, et al. 2012.
Prenatal exposure to perfluorooctanoate and risk of overweight at 20 years of
age: A prospective cohort study. Environmental health perspectives
120:668-673.
Haug LS, Huber S, Schlabach M, Becher G, Thomsen C. 2011. Investigation on per-
and polyfluorinated compounds in paired samples of house dust and indoor air
from norwegian homes. Environmental science & technology 45:7991-7998.
Hines EP, White SS, Stanko JP, Gibbs-Flournoy EA, Lau C, Fenton SE. 2009.
Phenotypic dichotomy following developmental exposure to perfluorooctanoic
acid (pfoa) in female cd-1 mice: Low doses induce elevated serum leptin and
insulin, and overweight in mid-life. Molecular and cellular endocrinology
304:97-105.
Hu W, Jones PD, DeCoen W, King L, Fraker P, Newsted J, et al. 2003. Alterations in
21 407 408 409 410 411 412 413 414 415 416 417 418 419 420 421 422 423 424 425 426 49
cell membrane properties caused by perfluorinated compounds. Comparative
biochemistry and physiology Toxicology & pharmacology : CBP 135:77-88.
Joensen UN, Veyrand B, Antignac JP, Blomberg Jensen M, Petersen JH, Marchand P,
et al. 2013. Pfos (perfluorooctanesulfonate) in serum is negatively associated
with testosterone levels, but not with semen quality, in healthy men. Human
reproduction 28:599-608.
Johansson N, Fredriksson A, Eriksson P. 2008. Neonatal exposure to perfluorooctane
sulfonate (pfos) and perfluorooctanoic acid (pfoa) causes neurobehavioural
defects in adult mice. Neurotoxicology 29:160-169.
Kannan K, Corsolini S, Falandysz J, Fillmann G, Kumar KS, Loganathan BG, et al.
2004. Perfluorooctanesulfonate and related fluorochemicals in human blood
from several countries. Environmental science & technology 38:4489-4495.
Keil DE, Mehlmann T, Butterworth L, Peden-Adams MM. 2008. Gestational
exposure to perfluorooctane sulfonate suppresses immune function in b6c3f1
mice. Toxicological sciences : an official journal of the Society of Toxicology
103:77-85.
Kjeldsen LS, Bonefeld-Jorgensen EC. 2013. Perfluorinated compounds affect the
function of sex hormone receptors. Environmental science and pollution research
international 20:8031-8044. 427 428 429 430 431 432 433 434 435 436 437 438 439 440 441 442 443 444 445
2013. Long-term effects of prenatal exposure to perfluoroalkyl substances on
female reproduction. Human reproduction 28:3337-3348.
Lau C, Butenhoff JL, Rogers JM. 2004. The developmental toxicity of perfluoroalkyl
acids and their derivatives. Toxicology and applied pharmacology 198:231-241.
Lau C, Anitole K, Hodes C, Lai D, Pfahles-Hutchens A, Seed J. 2007. Perfluoroalkyl
acids: A review of monitoring and toxicological findings. Toxicological sciences
: an official journal of the Society of Toxicology 99:366-394.
Lien GW, Wen TW, Hsieh WS, Wu KY, Chen CY, Chen PC. 2011. Analysis of
perfluorinated chemicals in umbilical cord blood by ultra-high performance
liquid chromatography/tandem mass spectrometry. Journal of chromatography B,
Analytical technologies in the biomedical and life sciences 879:641-646.
Lin CY, Wen LL, Lin LY, Wen TW, Lien GW, Chen CY, et al. 2011. Associations
between levels of serum perfluorinated chemicals and adiponectin in a young
hypertension cohort in taiwan. Environmental science & technology
45:10691-10698.
Lin CY, Lin LY, Wen TW, Lien GW, Chien KL, Hsu SH, et al. 2013a. Association
between levels of serum perfluorooctane sulfate and carotid artery intima-media
thickness in adolescents and young adults. International journal of cardiology
168:3309-3316.
Lin CY, Wen LL, Lin LY, Wen TW, Lien GW, Hsu SH, et al. 2013b. The
23 447 448 449 450 451 452 453 454 455 456 457 458 459 460 461 462 463 464 465 466 53
associations between serum perfluorinated chemicals and thyroid function in
adolescents and young adults. Journal of hazardous materials 244-245:637-644.
Lopez-Espinosa MJ, Fletcher T, Armstrong B, Genser B, Dhatariya K, Mondal D, et
al. 2011. Association of perfluorooctanoic acid (pfoa) and perfluorooctane
sulfonate (pfos) with age of puberty among children living near a chemical plant.
Environmental science & technology 45:8160-8166.
Maranghi F, Mantovani A. 2012. Targeted toxicological testing to investigate the role
of endocrine disrupters in puberty disorders. Reproductive toxicology
33:290-296.
Moore KL, Persaud TVN, Torchia MG. 2013. The developing human : Clinically
oriented embryology. 9th ed. Philadelphia, PA:Saunders/Elsevier.
Nelson JW, Hatch EE, Webster TF. 2010. Exposure to polyfluoroalkyl chemicals and
cholesterol, body weight, and insulin resistance in the general u.S. Population.
Environmental health perspectives 118:197-202.
Ng CA, Hungerbuhler K. 2014. Bioaccumulation of perfluorinated alkyl acids:
Observations and models. Environmental science & technology 48:4637-4648.
Noorlander CW, van Leeuwen SP, Te Biesebeek JD, Mengelers MJ, Zeilmaker MJ.
2011. Levels of perfluorinated compounds in food and dietary intake of pfos and
pfoa in the netherlands. Journal of agricultural and food chemistry 59:7496-7505. 467 468 469 470 471 472 473 474 475 476 477 478 479 480 481 482 483 484 485
The timing of normal puberty and the age limits of sexual precocity: Variations
around the world, secular trends, and changes after migration. Endocrine reviews
24:668-693.
Raymer JH, Michael LC, Studabaker WB, Olsen GW, Sloan CS, Wilcosky T, et al.
2012. Concentrations of perfluorooctane sulfonate (pfos) and perfluorooctanoate
(pfoa) and their associations with human semen quality measurements.
Reproductive toxicology 33:419-427.
Sorensen K, Mouritsen A, Aksglaede L, Hagen CP, Mogensen SS, Juul A. 2012.
Recent secular trends in pubertal timing: Implications for evaluation and
diagnosis of precocious puberty. Hormone research in paediatrics 77:137-145.
Specht IO, Hougaard KS, Spano M, Bizzaro D, Manicardi GC, Lindh CH, et al. 2012.
Sperm DNA integrity in relation to exposure to environmental perfluoroalkyl
substances - a study of spouses of pregnant women in three geographical regions.
Reproductive toxicology 33:577-583.
Su TC, Liao CC, Chien KL, Hsu SH, Sung FC. 2014. An overweight or obese status
in childhood predicts subclinical atherosclerosis and
prehypertension/hypertension in young adults. Journal of atherosclerosis and
thrombosis. J. Atheroscler. Thromb. 21 (11), 1170–1182.
Thibodeaux JR, Hanson RG, Rogers JM, Grey BE, Barbee BD, Richards JH, et al.
2003. Exposure to perfluorooctane sulfonate during pregnancy in rat and mouse.
25 487 488 489 490 491 492 493 494 495 496 497 498 499 500 501 502 503 504 505 506 57
I: Maternal and prenatal evaluations. Toxicological sciences : an official journal
of the Society of Toxicology 74:369-381.
Wei JN, Sung FC, Lin CC, Lin RS, Chiang CC, Chuang LM. 2003. National
surveillance for type 2 diabetes mellitus in taiwanese children. JAMA : the
journal of the American Medical Association 290:1345-1350.
Zhang LD, Li HC, Gao M, Wang L, Deng Q, Shi T, et al. 2013. [sexual development
characteristics and sex hormone levels in obese male adolescents]. Zhonghua nan
ke xue = National journal of andrology 19:434-438.
Zhao B, Hu GX, Chu Y, Jin X, Gong S, Akingbemi BT, et al. 2010. Inhibition of
human and rat 3beta-hydroxysteroid dehydrogenase and 17beta-hydroxysteroid
dehydrogenase 3 activities by perfluoroalkylated substances. Chemico-biological
interactions 188:38-43.
Table 1.
Basic demographics of the sample subjects including geometric mean (geometric SD) of PFAS concentrations.
No. PFOA (ng/ml) PFOS (ng/ml) PFNA (ng/ml) PFUA (ng/ml)
Total 540 2.74 (2.95) 7.78 (2.40) 1.10 (3.55) 5.84 (2.88) Sex Male 210 2.75 (2.93) 8.97 (2.59)** 1.21 (3.86) 5.89 (2.95) Female 330 2.73 (2.97) 7.11 (2.25)** 1.03 (3.36) 5.81 (2.83) Age (years) 12-17 95 3.03 (2.98) 7.12 (1.95)* 0.98 (3.53) 6.42 (3.06) 18-30 445 2.68 (2.94) 7.93 (2.49)* 1.12 (3.56) 5.72 (2.84) 507 508 509 510 511 512 513 514 515 516 517 518 519 520
<50000NTD per month 220 2.58 (2.99) 7.43 (2.40) 1.01 (3.50) 5.89 (2.88) ≥50000NTD per month 319 2.84 (2.92) 8.04 (2.41) 1.16 (3.58) 5.78 (2.88) Smoking status Never smoked 452 2.78 (2.97) 7.62 (2.41) 1.04 (3.47) 5.76 (2.85) Passive smoker 13 2.12 (2.89) 9.94 (1.66) 1.46 (4.70) 7.02 (3.21) Active smoker 70 2.65 (2.83) 8.43 (2.51) 1.38 (3.85) 6.18 (3.01)
Current alcohol consumption
No 465 2.76 (2.97) 7.65 (2.40) 1.04 (3.51) 5.78 (2.90)
Yes 67 2.69 (2.86) 8.72 (2.52) 1.40 (3.68) 6.27 (2.72)
Body mass index (kg/m2)
<24 417 2.88 (2.90) 7.47 (2.38)* 1.08 (3.53) 5.76 (2.84)
≥24 123 2.31 (3.09) 8.97 (2.44)* 1.16 (3.64) 6.12 (3.02)
Exercise
No 336 2.74 (2.89) 7.23 (2.49) 1.14 (3.50) 5.72 (2.90)
Yes 144 2.68 (2.88) 8.20 (2.22) 1.23 (3.82) 5.75 (2.95)
High sugar diet
No 342 2.58(2.92) 7.54(2.24) 1.15(3.52) 5.78(2.92)
Yes 138 3.10(2.76) 7.44(2.85) 1.19(3.79) 5.60(2.92)
High fat diet
No 443 2.66 (2.88) 7.36 (2.38)* 1.11 (3.48)* 5.63 (2.90)
Yes 37 3.48 (2.87) 9.60 (2.70)* 1.94 (4.67)* 7.03 (3.01)
PFAS, perfluoroalkyl substances; PFOA, perfluorooctanoic acid; PFOS, perfluorooctane sulfonate; PFNA, perfluorononanoic acid; PFUA, perfluoroundecanoic acid; NTD, new Taiwan dollars.
* p value < 0.05 **p value <0.01
27
Table 2.
Distribution of reproductive hormones in the whole study group.
Mean SD Median Percentile 05 Percentile 95
All E2 (pg/mL) 85.3 84.2 48.2 48.2 290.5 FSH (mIU/mL) 4.70 2.67 4.07 4.07 8.71 LH (mIU/mL) 3.82 4.61 2.48 2.48 11.6 SHBG (nmol/L) 42.7 26.0 36.8 36.8 91.7 Testosterone (ng/dL) 253.9 286.7 54.4 54.4 789 Free Testosterone (ng/dL) 5.40 6.43 0.86 0.86 17.0 Male 12-17 E2 (pg/mL) 36.1 10.0 36.0 36.0 53.3 FSH (mIU/mL) 4.15 1.96 4.02 4.02 8.13 LH (mIU/mL) 1.60 0.87 1.41 1.41 3.54 SHBG (nmol/L) 28.6 14.1 29.1 29.1 50.6 Testosterone (ng/dL) 508.2 181.8 537.5 537.5 820.0 Free Testosterone (ng/dL) 10.7 3.74 10.4 10.4 17.4 Female 12-17 E2 (pg/mL) 89.6 81.8 55.7 55.7 257 FSH (mIU/mL) 5.41 2.05 5.47 5.47 8.63 LH (mIU/mL) 4.58 3.75 3.70 3.70 12.9 SHBG (nmol/L) 45.0 21.2 43.6 43.6 81.8 Testosterone (ng/dL) 42.7 20.0 38.4 38.4 87.8 Free Testosterone (ng/dL) 0.66 0.34 0.59 0.59 1.28
Male 18-30 E2 (pg/mL) 36.9 10.9 35.2 35.2 54.7 FSH (mIU/mL) 3.72 1.91 3.25 3.25 7.53 LH (mIU/mL) 1.75 0.74 1.57 1.57 3.23 SHBG (nmol/L) 27.3 11.9 25.9 25.9 50.9 Testosterone (ng/dL) 599.8 167.2 590.5 590.5 890.5 Free Testosterone (ng/dL) 13.3 3.48 12.9 12.9 18.9 Female 18-30 E2 (pg/mL) 122.7 97.4 89.0 89.0 341.0 FSH (mIU/mL) 5.25 3.08 5.15 5.15 9.38 LH (mIU/mL) 5.29 5.82 3.56 3.56 15.3 SHBG (nmol/L) 54.1 28.9 51.2 51.2 107.0 Testosterone (ng/dL) 42.0 16.2 38.2 38.2 74.2 Free Testosterone (ng/dL) 0.59 0.32 0.52 0.52 1.30 29 521 65
Table 3.
Mean and standard error of natural log-transformed sex hormone-binding globulin (ln-SHBG) across categories of serum PFAS levels in linear regression models (n=540).
ln-SHBG (nmol/L)
Total Male 12–17 Female 12–17 Male 18–30 Female 18–30
No. 540 30 65 180 265 PFOA (ng/ml) <3.63 (<50th) 3.37 (0.07) 3.24 (0.29) 3.50 (0.24)* 3.14 (0.07) 3.83 (0.21) ≤6.78 (50th–75th) 3.43 (0.08) 3.45 (0.29) 3.50 (0.25)* 3.17 (0.09) 3.86 (0.20) ≤9.80 (75th–90th) 3.41 (0.08) 3.67 (0.36) 3.45 (0.29)* 3.20 (0.10) 3.81 (0.22) >9.80 (>90th) 3.33 (0.10) 3.79 (0.39) 2.96 (0.34)* 3.10 (0.14) 3.78 (0.23) PFOS (ng/ml) <5.37 (<25th) 3.47 (0.08) 3.62 (0.29) 3.58 (0.29) 3.13 (0.10) 3.90 (0.21) ≤8.65 (25th–50th) 3.36 (0.08) 3.31 (0.30) 3.36 (0.29) 3.18 (0.10) 3.82 (0.20) ≤13.29 (50th–75th) 3.40 (0.08) 3.47 (0.38) 3.49 (0.24) 3.13 (0.09) 3.89 (0.22) >13.29 (>75th) 3.38 (0.08) 3.46 (0.39) 3.41 (0.44) 3.16 (0.08) 3.80 (0.21) PFNA (ng/ml) <1.64 (<60th) 3.38 (0.07) 3.45 (0.24) 3.46 (0.25) 3.13 (0.08) 3.84 (0.20) ≤6.87 (60th–90th) 3.36 (0.08) 3.51 (0.35) 3.52 (0.26) 3.08 (0.09) 3.79 (0.21) >6.87 (>90th) 3.47 (0.09) 3.59 (0.36) 3.53 (0.41) 3.27 (0.10) 3.92 (0.23) PFUA (ng/ml) <1.53 (<25th) 3.38 (0.08) 3.26 (0.41) 3.81 (0.28) 3.15 (0.09) 3.81 (0.21) ≤6.53 (25th–50th) 3.39 (0.08) 3.62 (0.30) 3.42 (0.32) 3.19 (0.10) 3.80 (0.21) ≤13.41 (50th–75th) 3.40 (0.08) 3.26 (0.37) 3.38 (0.24) 3.17 (0.08) 3.88 (0.21) >13.41 (>75th) 3.40 (0.08) 3.33 (0.36) 3.78 (0.30) 3.12 (0.09) 3.85 (0.20)
Model: adjusted for age, gender, BMI, and high fat diet. PFAS, perfluoroalkyl substances; PFOA, perfluorooctanoic acid; PFOS, perfluorooctane sulfonate; PFNA, perfluorononanoic acid; PFUA, perfluoroundecanoic acid.
*p for trend<0.05
Table 4.
Mean and standard error of natural log-transformed follicle-stimulating hormone (ln-FSH) across categories of serum PFAS levels in linear regression models (n=540).
ln-FSH (mIU/mL)
Total Male 12–17 Female 12–17 Male 18–30 Female 18–30
No. 540 30 65 180 265 PFOA (ng/ml) <3.63 (<50th) 1.41 (0.08) 1.29 (0.28) 1.47 (0.20) 1.29 (0.08) 1.69 (0.24) ≤6.78 (50th–75th) 1.42 (0.08) 1.58 (0.27) 1.38 (0.21) 1.28 (0.10) 1.65 (0.24) ≤9.80 (75th–90th) 1.34 (0.09) 1.21 (0.34) 1.23 (0.25) 1.23 (0.11) 1.64 (0.25) >9.80 (>90th) 1.46 (0.11) 1.49 (0.36) 1.35 (0.29) 1.13 (0.15) 1.79 (0.26) PFOS (ng/ml) <5.37 (<25th) 1.43 (0.09) 1.50 (0.22)* 1.56 (0.23) 1.20 (0.11) 1.71 (0.25) ≤8.65 (25th–50th) 1.42 (0.08) 1.56 (0.22)* 1.67 (0.23) 1.27 (0.11) 1.66 (0.23) ≤13.29 (50th–75th) 1.42 (0.09) 1.26 (0.28)* 1.36 (0.19) 1.34 (0.10) 1.71 (0.25) >13.29 (>75th) 1.36 (0.08) 0.76 (0.29)* 1.23 (0.35) 1.26 (0.08) 1.69 (0.25) PFNA (ng/ml) <1.64 (<60th) 1.44 (0.08) 1.47 (0.21) 1.53 (0.20) 1.30 (0.08) 1.69 (0.23) ≤6.87 (60th–90th) 1.36 (0.08) 1.43 (0.31) 1.26 (0.22) 1.32 (0.10) 1.58 (0.24) >6.87 (>90th) 1.36 (0.10) 1.19 (0.32) 1.37 (0.33) 1.17 (0.11) 1.73 (0.26) PFUA (ng/ml) <1.53 (<25th) 1.44 (0.08) 1.07 (0.33) 1.59 (0.23)* 1.26 (0.09) 1.73 (0.24) ≤6.53 (25th–50th) 1.41 (0.09) 1.73 (0.24) 1.56 (0.26)* 1.22 (0.11) 1.69 (0.24) ≤13.41 (50th–75th) 1.41 (0.08) 1.45 (0.30) 1.37 (0.20)* 1.33 (0.09) 1.64 (0.24) >13.41 (>75th) 1.35 (0.08) 1.05 (0.29) 1.24 (0.24)* 1.21 (0.10) 1.65 (0.24)
Model: adjusted for age, gender, BMI, and high fat diet. PFAS, perfluoroalkyl substances; PFOA, perfluorooctanoic acid; PFOS, perfluorooctane sulfonate; PFNA, perfluorononanoic acid; PFUA, perfluoroundecanoic acid.
*p for trend<0.05
31 Table 5.
Mean and standard error of natural log-transformed testosterone (ln-T) across categories of serum PFAS levels in linear regression models (n=540).
ln-T (ng/dL)
Total Male 12–17 Female 12–17 Male 18–30 Female 18–30
No. 540 30 65 180 265 PFOA (ng/ml) <3.63 (<50th) 4.96 (0.05) 6.23 (0.23) 3.85 (0.20) 6.32 (0.05) 3.70 (0.15) ≤6.78 (50th–75th) 5.00 (0.06) 5.97 (0.22) 3.96 (0.21) 6.32 (0.06) 3.75 (0.15) ≤9.80 (75th–90th) 4.97 (0.06) 6.46 (0.28) 3.95 (0.25) 6.32 (0.07) 3.65 (0.16) >9.80 (>90th) 4.97 (0.07) 6.48 (0.30) 3.84 (0.29) 6.28 (0.09) 3.71 (0.16) PFOS (ng/ml) <5.37 (<25th) 5.00 (0.06) 6.11 (0.23) 3.97 (0.23)* 6.33 (0.06) 3.73 (0.15) ≤8.65 (25th–50th) 5.02 (0.06) 6.25 (0.24) 4.00 (0.23)* 6.29 (0.07) 3.75 (0.15) ≤13.29 (50th–75th) 4.95 (0.06) 6.24 (0.30) 3.87 (0.19)* 6.32 (0.06) 3.64 (0.16) >13.29 (>75th) 4.94 (0.06) 6.34 (0.31) 3.61 (0.36)* 6.33 (0.05) 3.65 (0.15) PFNA (ng/ml) <1.64 (<60th) 5.00 (0.05) 6.18 (0.19) 3.93 (0.20) 6.33 (0.05) 3.74 (0.15) ≤6.87 (60th–90th) 4.96 (0.06) 6.35 (0.27) 3.85 (0.21) 6.30 (0.06) 3.70 (0.15) >6.87 (>90th) 4.93 (0.06) 6.12 (0.28) 3.64 (0.33) 6.32 (0.06) 3.65 (0.17) PFUA (ng/ml) <1.53 (<25th) 4.97 (0.06) 6.61 (0.29) 4.01 (0.24) 6.30 (0.05) 3.70 (0.15) ≤6.53 (25th–50th) 4.99 (0.06) 5.90 (0.21) 3.95 (0.28) 6.34 (0.06) 3.79 (0.15) ≤13.41 (50th–75th) 5.01 (0.05) 6.33 (0.26) 3.85 (0.21) 6.37 (0.05) 3.78 (0.15) >13.41 (>75th) 4.93 (0.06) 6.53 (0.25) 3.99 (0.26) 6.26 (0.06) 3.69 (0.15)
Model: adjusted for age, gender, BMI, and high fat diet. PFAS,perfluoroalkyl substances; PFOA,
perfluorooctanoic acid; PFOS, perfluorooctane sulfonate; PFNA, perfluorononanoic acid; PFUA, perfluoroundecanoic acid. *p for trend<0.05 523 524 525 526 69
Table S1.
Linear regression coefficients (95% CI) of log-transformed E2 (pg/mL), FSH (mIU/mL), LH (mIU/mL), testosterone (mIU/mL), free testosterone (mIU/mL), and SHBG (nmol/L) with PFAAs concentrations (ng/mL) in the sample subjects by increase an unit BMI.
Body mass index (kg/m2)
ln-E2 (pg/mL) ln-FSH (mIU/mL) ln-LH (mIU/mL) ln-T (mIU/mL) ln-freeT (mIU/mL) ln-SHBG (nmol/L) PFOA (ng/ml) -0.004(-0.017,0.01) -0.005(-0.017,0.008) -0.01(-0.025,0.006) -0.01(-0.018,0.002) 0.022(-0.012,0.032) -0.063(-0.074,0.052) PFOS (ng/ml) -0.006(-0.02,0.008) -0.004(-0.016,0.008) -0.01(-0.026,0.006) -0.009(-0.018,0.001) 0.023(-0.013,0.033) -0.063(-0.074,0.053) PFNA (ng/ml) -0.004(-0.018,0.01) -0.005(-0.017,0.007) -0.01(-0.026,0.005) -0.01(-0.018,0.002) 0.021(-0.011,0.031) -0.063(-0.073,0.052) PFUA (ng/ml) -0.004(-0.017,0.01) -0.005(-0.017,0.008) -0.01(-0.025,0.006) -0.01(-0.018,0.002) 0.022(-0.012,0.032) -0.063(-0.073,0.052) Model: adjusted for age, gender, and high fat diet.
PFASs, perfluoroalkyl substances; PFOA, perfluorooctanoic acid; PFOS, perfluorooctane sulfonate; PFNA, perfluorononanoic acid; PFUA, perfluoroundecanoic acid.
*p for trend<0.05 Table S2.
Mean and standard error of natural log sex hormone-binding globulin (SHBG) across categories of serum PFAS levels in subpopulation.
log SHBG (nmol/L)
Body mass index (kg/m2) High fat diet
<24 ≥24 No Yes No. 417 123 443 37 PFOA (ng/ml) <3.63 (<50th) 3.56 (0.09) 3.23 (0.13) 3.29 (0.07) 3.61 (0.18) ≤6.78 (50th–75th) 3.62 (0.09) 3.31 (0.15) 3.3 3(0.08) 4.13 (0.18) ≤9.80 (75th–90th) 3.61 (0.10) 3.30 (0.18) 3.34 (0.09) 3.70 (0.20) >9.80 (>90th) 3.45 (0.12) 3.39 (0.21) 3.28 (0.11) 3.54 (0.27) PFOS (ng/ml) <5.37 (<25th) 3.69 (0.09) 3.35 (0.16) 3.39 (0.08) 4.07 (0.29) ≤8.65 (25th–50th) 3.55 (0.09) 3.20 (0.16) 3.26 (0.08) 3.83 (0.23) ≤13.29 (50th–75th) 3.60 (0.09) 3.28 (0.16) 3.32 (0.08) 3.97 (0.23) >13.29 (>75th) 3.52 (0.09) 3.24 (0.15) 3.27 (0.08) 3.72 (0.18) PFNA (ng/ml) <1.64 (<60th) 3.59 (0.08) 3.21 (0.14) 3.32 (0.07) 3.69 (0.17) ≤6.87 (60th–90th) 3.49 (0.09) 3.23 (0.14) 3.27 (0.07) 4.14 (0.20) >6.87 (>90th) 3.63 (0.10) 3.42 (0.19) 3.41 (0.10) 3.79 (0.18) PFUA (ng/ml) <1.53 (<25th) 3.60 (0.09) 3.23 (0.16) 3.31 (0.08) 3.52 (0.23) ≤6.53 (25th–50th) 3.61 (0.09) 3.10 (0.18) 3.28 (0.08) 3.82 (0.19) ≤13.41 (50th–75th) 3.58 (0.09) 3.27 (0.14) 3.30 (0.08) 4.06 (0.19) >13.41 (>75th) 3.55 (0.09) 3.32 (0.15) 3.34 (0.08) 3.62 (0.20)
Model: adjusted for age and gender.
PFAS, perfluoroalkyl substances; PFOA, perfluorooctanoic acid; PFOS, perfluorooctane sulfonate; PFNA, perfluorononanoic acid; PFUA, perfluoroundecanoic acid.
*p for trend<0.05 Table S3.
Mean and standard error of natural log estrogen (E2) across categories of serum PFAS levels in subpopulation.
log E2 (pg/mL)
Body mass index (kg/m2) High fat diet
<24 ≥24 No Yes No. 417 123 443 37 PFOA (ng/ml) <3.63 (<50th) 3.93 (0.12) 4.00 (0.14) 3.96 (0.09) 3.96 (0.26) ≤6.78 (50th–75th) 3.96 (0.12) 3.99 (0.16) 3.98 (0.10) 4.08 (0.27) ≤9.80 (75th–90th) 4.04 (0.13) 3.84 (0.18) 4.05 (0.11) 3.54 (0.30) >9.80 (>90th) 3.93 (0.16) 3.98 (0.21) 3.95 (0.13) 4.12 (0.39) PFOS (ng/ml) <5.37 (<25th) 3.94 (0.13) 3.88 (0.16) 3.93 (0.10) 3.99 (0.38) ≤8.65 (25th–50th) 3.97 (0.12) 3.89 (0.16) 4.00 (0.10) 3.66 (0.30) ≤13.29 (50th–75th) 3.91 (0.12) 4.03 (0.16) 3.95 (0.10) 3.67 (0.31) >13.29 (>75th) 4.03 (0.13) 4.05 (0.15) 4.08 (0.10) 4.05 (0.24) PFNA (ng/ml) <1.64 (<60th) 4.02 (0.11) 3.91 (0.15) 3.97 (0.09) 3.88 (0.24) ≤6.87 (60th–90th) 4.09 (0.13) 4.02 (0.14) 4.07 (0.09) 3.99 (0.30) >6.87 (>90th) 3.75 (0.14) 3.96 (0.19) 3.76 (0.12) 3.84 (0.26) PFUA (ng/ml) <1.53 (<25th) 3.93 (0.12) 3.85 (0.16) 3.93 (0.09) 3.24 (0.30) ≤6.53 (25th–50th) 3.98 (0.13) 3.64 (0.18) 3.91 (0.11) 4.20 (0.24) ≤13.41 (50th–75th) 4.00 (0.12) 4.04 (0.14) 4.02 (0.10) 4.01 (0.24) >13.41 (>75th) 3.97 (0.12) 4.00 (0.15) 4.03 (0.10) 3.69 (0.26)
Model: adjusted for age and gender.
PFAS, perfluoroalkyl substances; PFOA, perfluorooctanoic acid; PFOS, perfluorooctane sulfonate; PFNA, perfluorononanoic acid; PFUA, perfluoroundecanoic acid.
*p for trend<0.05 Table S4.
Mean and standard error of natural log follicle-stimulating hormone (FSH) across categories of serum PFAS levels in subpopulation.
log FSH (mIU/mL)
Body mass index (kg/m2) High fat diet
<24 ≥24 No Yes No. 417 123 443 37 PFOA (ng/ml) <3.63 (<50th) 1.40 (0.09) 1.42 (0.16) 1.44 (0.08) 1.44 (0.23) ≤6.78 (50th–75th) 1.40 (0.10) 1.42 (0.18) 1.45 (0.08) 1.23 (0.24) ≤9.80 (75th–90th) 1.28 (0.11) 1.49 (0.21) 1.36 (0.09) 1.69 (0.27) >9.80 (>90th) 1.45 (0.13) 1.38 (0.24) 1.49 (0.12) 1.37 (0.36) PFOS (ng/ml) <5.37 (<25th) 1.39 (0.10) 1.49 (0.18) 1.45 (0.08) 1.30 (0.35) ≤8.65 (25th–50th) 1.40 (0.10) 1.42 (0.19) 1.43 (0.08) 1.57 (0.28) ≤13.29 (50th–75th) 1.35 (0.10) 1.56 (0.18) 1.45 (0.09) 1.32 (0.28) >13.29 (>75th) 1.36 (0.10) 1.30 (0.17) 1.37 (0.09) 1.42 (0.22) PFNA (ng/ml) <1.64 (<60th) 1.40 (0.09) 1.51 (0.17) 1.45 (0.07) 1.63 (0.20) ≤6.87 (60th–90th) 1.28 (0.10) 1.49 (0.16) 1.38 (0.08) 1.34 (0.25) >6.87 (>90th) 1.38 (0.11) 1.18 (0.22) 1.41 (0.11) 1.23 (0.22) PFUA (ng/ml) <1.53 (<25th) 1.44 (0.1) 1.37 (0.19) 1.47 (0.08) 1.49 (0.30) ≤6.53 (25th–50th) 1.37 (0.1) 1.49 (0.21) 1.41 (0.09) 1.61 (0.24) ≤13.41 (50th–75th) 1.38 (0.1) 1.44 (0.16) 1.43 (0.08) 1.26 (0.25) >13.41 (>75th) 1.30 (0.1) 1.41 (0.18) 1.38 (0.08) 1.33 (0.26)
Model: adjusted for age and gender.
PFAS, perfluoroalkyl substances; PFOA, perfluorooctanoic acid; PFOS, perfluorooctane sulfonate; PFNA, perfluorononanoic acid; PFUA, perfluoroundecanoic acid.
*p for trend<0.05 527 528 529 530 531 532
33
533
534