Parabens effects on female reproductive health - Review of evidence from epidemiological and rodent-based studies.

Several epidemiological studies reported an association between paraben exposure and changes in infant birth weight, anthropometric parameters, and ponderal index, and the ratio of infant body weight to body length 96 . Most of these studies, with a few noted exceptions, utilized maternal urine samples collected during pregnancy. Notably, in many of these studies, results were moderated by infant sex and will be indicated below. Overall, body length at birth appeared to be negatively associated with methylparaben in male and female newborns, whereas ethyl-, and propylparaben were associated with shorter length at birth only in female infants. Some conflicting results were also reported regarding body length at birth, and these may be due to differences in study design/data analysis. Unfortunately, only a limited number of experimental studies reported birthweights of the offspring. Methylparaben levels were associated with greater body length at birth regardless of infant sex 87 . Likewise, ethylparaben was positively associated with body length at birth but only in unadjusted models 97 . Wu et al. also reported ethylparaben and sum parabens (methyl-, ethyl-, propyl-, butyl-, and benzylparaben) were positively associated with body length at birth in male infants, yet only in unadjusted models 97 . Conversely, Uldbjerg et al. 98 found methylparaben was negatively associated with body length at birth in male infants. Other studies reported that methyl- 99 , propyl- and ethylparaben 100 were negatively associated with body length at birth in female infants. In addition, Pacyga et al. 100 reported associations between ethyl- and propylparaben with lower birth weight, smaller birth weight z-scores, shorter body length, and smaller ponderal index in female newborns while noting these associations were stronger among mothers with poorer diet quality. Still, Philippat et al. 101 reported no association between maternal paraben exposure and infant birth length. These conflicting results could be explained by differences in study design and sample characteristics. For example, urine specific gravity is a commonly used to adjust paraben levels in urine samples yet, some researchers use creatinine values for this adjustment. Further, differences in samples characteristics often lead to variations in the factors included in the adjusted statistical models. Interestingly, infant head circumference was overwhelmingly negatively associated with paraben exposure. Zhang et al. 102 reported maternal methyl- and propylparaben were negatively associated with newborn head circumference in all newborns. However, when stratified by sex, propylparaben and sum parabens (methyl-, ethyl-, propyl-, butyl-, and benzylparaben) were negatively associated with head circumference in male newborns only 102 . Hajizadeh et al. 103 reported third trimester methyl- and butylparaben exposure were negatively associated with head circumference in female infants only. Bräuner et al. 104 reported second trimester prenatal urinary butylparaben was negatively associated with head circumference in female infants. Similarly, Chang et al. 96 found maternal propylparaben levels in the highest two quartiles were associated with smaller head circumference and shorter thoracic circumference of female infants compared to the lowest quartiles of maternal exposure. Additional studies support these results on head circumference for other parabens, reporting ethylparaben 96 and butylparaben 105 were negatively associated with head circumference in female infants. Similarly, for male infants Fadaei et al. 105 reported propyl- and butylparaben were positively associated with head circumference. Only one study reported no association between gestational paraben exposure and infant head circumference 101 . Still, gestational paraben exposure was generally associated with greater head circumference in male infants but smaller in female infants. Overall, gestational paraben exposure was generally associated with smaller head circumference in female infants, while two studies reported conflicting results regarding parabens and male infant head circumference. Notably, these conflicting studies utilized vastly different samples sizes, with Fadaei et al. recruiting only 171 mother-infant pairs, while Zhang et al. 102 recruited 942. Furthermore, Zhang et al. 102 reported a narrow IQR for propylparaben exposure (1.12, 4.49) μg/g in their population, compared to the wide IQR propylparaben range (<LOD, 535) μg/g reported by Fadaei et al. 105 . These differences suggest a potential negative association between propylparaben exposure and male infant head circumference, yet additional studies are needed to confirm these findings that will include additional parabens in the analyses. Two studies investigated paraben’s effects on birth size using samples other than maternal urine. One study reported methylparaben concentration in amniotic fluid samples was associated with smaller newborn head circumference, chest circumference, and hip circumference, while ethyl- and propylparaben were associated with smaller arm circumference in infants, regardless of sex 106 . Last, Vrijens et al. 107 reported that sum placental parabens and ethylparaben only were associated with smaller head circumference. Of note, ethylparaben was also associated with lower placenta weight, but only for female newborns 107 . Overall, results regarding paraben exposure and birth weight were mixed, suggesting the effects may be dependent on exposure levels, infant sex, and timing of exposure and measurements throughout gestation. Most epidemiologic studies suggest butylparaben may be associated with greater birth weight in male and female infants 105 , 108 . Gestational exposure to ethylparaben was associated with lower birth weight in female infants 100 and all infants 99 , whereas conflicting results were reported in male infants 108 . Methylparaben was associated with greater birth weight in females at high exposure levels 103 and lower birth weight at lower exposure levels 96 , 98 . Whereas, in males methylparaben was associated with greater birth weight 96 . The direction of the association between propylparaben exposure and birth weight varied depending on levels and timing of exposure 87 , 100 , 105 . Experimental studies investigated only butylparaben and mostly reported null results. However, co-exposure with butylparaben and DEHP did reduce birth weight in Wistar rats 53 . Future experimental studies necessitating inclusion of additional parabens showing effects in epidemiologic studies along with co-exposures of paraben and other EDCs such as phthalates. Several studies examined the associations between gestational paraben levels and birth weight. Specifically, Fadaei et al. 105 found a positive association between butylparaben levels and birth weight only in female infants, while Li et al. 108 found the association was only significant in male infants. Still, Bräuner et al. 104 reported maternal butylparaben concentration was negatively associated with the birth weight of female infants. Of note, the butylparaben concentrations detected in this cohort were very low, with many women having concentrations below the limit of detection. Maternal ethylparaben levels (on delivery day) were associated with greater birth weight but only for male infants 108 . In contrast, maternal ethylparaben concentrations (3rd trimester) were associated with lower birth weight z-scores for all infants 99 . These findings were partially supported by Pacyga et al. 100 , who reported a negative association between ethylparaben levels (pooled samples throughout pregnancy), birth weight, and birth weight z-scores in female infants. Differences in the timing of ethylparaben exposure between these studies, may explain the conflicting results, especially as paraben levels taken on delivery day may not accurately reflect paraben exposure throughout the duration of fetal development. Similarly, gestational methylparaben levels were associated with greater birth weight only in female infants 103 . In contrast, Chang et al. 96 found that female infants of mothers with the highest methylparaben concentrations had lower birth weight than those in the lowest exposure quartiles. Notably, Chang et al. 96 also reported maternal methylparaben levels were positively correlated with ponderal index in male infants, suggesting the association’s direction may be sex dependent. In contrast, Uldbjerg et al. 98 reported associations between moderate and high methylparaben levels and lower birth weight. The conflicting results between these studies may be related to the vastly different methylparaben levels in the populations studied, suggesting much higher levels of methylparaben exposure may result in greater birth weight whereas lower exposure levels may result in lesser birth weight, in females. In female infants, propylparaben was negatively associated with birth weight 100 . In contrast, Hajizadeh et al. 87 reported propylparaben levels were positively associated with birth weight in female infants only (depending on the statistical model). Fadaei et al. 105 also reported that propylparaben levels were positively associated with birth weight, only in male infants. Notably, the two studies finding positive associations reported much higher propylparaben concentrations 87 , 105 than Pacyga et al. which reported a negative association 100 . Only one study reported that heptylparaben levels were correlated with greater birth weight in all infants regardless of sex 11 . Aker et al. reported maternal methyl-, butyl-, and propylparaben were selectively associated with decreased odds ratio of infants being small for gestational age (defined as being below the 10th percentile for birth weight z-score) 93 . Last, Vrijens et al. 107 reported that sum placental parabens was associated with lower birth weight in all infants and in female infants only. Finally, Philippat et al. 101 found no association between gestational paraben exposure and infant birth weight. While these results are mixed, and appear to depend on the timing of exposure, specific parabens present, and infant sex, they suggest that gestational paraben exposure may have significant effects on infant birthweight. Gestational butylparaben exposure (10–1000 mg/kg/day) of Wistar, Holtzman, or Sprague Dawley rats did not affect pups’ birth weight 49 , 51 , 75 , 78 , 80 . Interestingly, in utero co-exposure of butylparaben (100 mg/kg/day) and DEHP (500 mg/kg/day) of Wistar rats resulted in reduced pups’ birth weight compared to butylparaben alone 53 . In Swiss albino mice, in utero co-exposure of bisphenol A and iso-butylparaben did not affect pups’ birth weight 82 . Last, gestational ethylparaben exposure (400 mg/kg/day) of Wistar rats had no effect on pup’s birth weight 78 . Overall, the limited experimental data indicate that butylparaben did not affect birth weight in contrast to the reports in human infants 11 , 103 , 105 , 108 . Future studies should examine the potential effects of methyl- or propylparaben, which are associated with human birth weight 103 and are most prevalently detected in human samples 20 , 28 , 83 – 90 .

Future studies should include human-relevant paraben mixtures in their analyses and experimental designs The scientific literature on parabens is limited, with most studies focusing on butylparaben in rodent models. Butylparaben has shown significant results in these studies, leading some countries to restrict its use, resulting in low detection levels in certain epidemiologic cohorts 20 , 28 , 83 – 90 . Further, studies have shown that mixtures of parabens can produce significant results, unlike single paraben investigations. Thus, future research should focus on parabens more relevant to human exposure (e.g., methyl- and propylparaben) and consider the combined effects of mixtures of parabens and other EDCs. Future research should use doses closer to human EDI values and consider duration of exposure. Most rodent studies use high paraben doses (mg/kg) compared to human biological samples (μg/kg). Experimental dosages should be calculated by adjusting human estimated daily intake (EDI) values for species differences such as body surface area and metabolism. Studies should focus on chronic exposure rather than acute exposure to better reflect human lifetime exposure. Last, to the best of our knowledge there are no studies that examined the multi- or trans- generational effects of paraben exposures. Finally, several epidemiologic and experimental topics remain underexplored. Only one study has examined the effects of parabens on human menstrual cycles, despite experimental studies showing effects on estrus cyclicity. Similarly, only two studies have investigated paraben effects on ovarian and uterine disorders. Moreover, many studies on parabens and fertility have focused on couples seeking fertility treatments, which may not reflect the impact on a healthy population. Thus, future research should examine paraben effects on menstrual cyclicity and include participants that are otherwise healthy. These experimental models will significantly enhance the ability to evaluate potential risks associated with parabens exposures.

This literature review is based on an electronic literature search utilizing the National Library of Medicine (PubMed) database of epidemiological and rodent-based studies published in the English language by May 2023. Initial search was as follows: 2000/01/01:3000/12/31[Date - Create] NOT (review OR cancer OR allergy) AND ‘English’[Language] AND ‘paraben’[Text Word] AND (‘female’[Text Word] OR hormone OR ovary OR uterus OR fertility OR reproduction) which resulted in 394 publications matching the search terms. Then, the titles and abstracts were screened while focusing on epidemiological and rodent-based studies. The identified articles were read in full, and their references lists were hand-searched for additional articles that could be included. Overall, 51 epidemiological and 34 rodent-based studies were included in the review. The topics we reviewed include paraben effects on cyclicity, fertility and pregnancy, newborn and pubertal development, reproductive-related hormones, and ovarian and uterine specific outcomes. In supplemental material , we provide detailed tables (S1 – 14) complementing the details provided in the text (e.g., study characteristics, study design, population characteristics, methods, covariates in epidemiological studies, sample size, and the specific paraben examined in epidemiological and in in vivo studies).

Limited studies investigated the associations between paraben exposure and uterine and ovarian diseases. Significant associations were reported between higher paraben levels and ovarian benign tumors, adenomyosis, endometriosis, and uterine leiomyoma 143 , 144 . Experimental studies focused on ovarian histomorphology, folliculogenesis, and steroidogenesis. These studies suggested paraben exposure may induce follicular arrest or enhance the transition from primordial to primary follicles which may lead to sub/infertility 51 , 53 , 56 , 57 , 145 potentially by altering Amh levels 56 , 145 . Yet, due to the limited studies on the topic and conflicting results additional studies are needed on paraben effects on folliculogenesis. Only two case-control studies investigated the associations between paraben exposure and risk of uterine and ovarian diseases. Specifically, higher levels of OH-methylparaben and ethylparaben were reported in cases of adenomyosis and ovarian benign tumors, respectively 143 . Ethylparaben was also associated with higher odds ratio of ovarian benign tumors, and 3,4-DHB was associated with greater odds ratio of uterine leiomyomas and having at least one gynecological disorder compared to generally healthy women 143 . Peinado et al. 144 found no difference in paraben concentrations between women with endometriosis and controls even when evaluated by the stage of endometriosis (Stage I/II compared to Stage III/IV) except for methylparaben in the second tertile of exposure that was associated with greater odds ratio of endometriosis. Several studies examined folliculogenesis by ovarian histology to examine if parabens affect the ovary. In Holtzman rats, butylparaben exposure (GD 6 – PND 21, 10–1000 mg/kg/day) selectively increased the numbers of secondary follicles and atretic follicles compared to controls, on PND 30 49 . However, on PND 45, butylparaben increased the number of primary follicles compared to controls, whereas on PND 75, butylparaben selectively increased the numbers of primordial follicles and corpora lutea 49 . Overall, butylparaben exposure altered folliculogenesis, enhanced follicle death, and increased ovulation, which may lead to premature ovarian failure. In Wistar rats, butylparaben exposure (GD 7 – PND 22, 10–500 mg/kg/day) resulted in similar outcomes to controls 75 . Of note, the authors examined only one histological section per ovary rather than conducting a comprehensive quantitative analysis 75 . Similarly, though, in Wistar rats, exposure (GD12 – PND 21) of butylparaben (10–200 mg/kg/day), DEHP (500 mg/kg/day), or a mixture of both resulted in similar numbers of total follicles, corpora lutea, primordial follicles, preantral follicles, antral follicles, or atretic follicles compared to controls 51 , 53 . In Sprague Dawley rats, methylparaben, iso-propylparaben, butylparaben, and iso-butylparaben exposure (PND 21 – 40, 62.5–1000 mg/kg/day) selectively increased the number of cystic follicles and decreased the number of corpora lutea, whereas ethyl- and propylparaben had similar outcomes of controls 55 . These results suggest paraben exposure may reduce ovulation by inhibiting the development of pre-ovulatory follicles. In Sprague Dawley rats, butyl- or propylparaben exposure but not methylparaben (PND 1 - PND 7, 62.5–1000 mg/kg/day) selectively increased the number of primordial follicles and decreased the number of early primary follicles compared to controls 145 . This suggests paraben exposure may induce follicular arrest at the primordial stage which may lead to sub/infertility. Similarly, in adult Sprague-Dawley rats, methyl-, propyl-, or butylparaben exposure (5 weeks, 100 mg/kg/day) either alone or following a two-week pre-treatment with 4-vinylcyclohexene diepoxide (VCD, 40 mg/kg/day to induce premature ovarian failure) resulted in reduced numbers of secondary/preantral and pre-ovulatory follicles with all treatments excluding butylparaben with VCD compared to controls 57 . Interestingly, in C57BL/6J mice, propylparaben exposure (8 months, 7.5 mg/kg/day) reduced the number of primordial follicles and increased the number of secondary follicles compared to controls 54 . Further, in adult Balb/c mice, propylparaben exposure (21 days, 100, 1000 mg/kg/day) decreased the numbers of corpora lutea following superovulation compared to controls 56 . However, the numbers of preantral, antral, and large antral follicles were similar to controls 56 . These findings suggest that propylparaben may affect folliculogenesis by enhancing the transition from primordial to primary follicles and may suppress ovulation. The effects of paraben on folliculogenesis may involve various factors that were previously shown to play a central role in folliculogenesis. Anti-Müllerian hormone gene ( Amh ), kit ligand/stem cell factor ( Kitl ), and forkhead box protein l2 ( Foxl2 ) are involved in early folliculogenesis. Anti-Müllerian hormone is produced by mature follicles and plays a role in sex differentiation and suppressing follicle development 146 , 147 . Amh is an estrogen-sensitive gene that can be affected by EDCs 148 . In Sprague Dawley rats, Amh expression was upregulated by propyl- and butylparaben and downregulated by methylparaben exposures (5 weeks, PND 1 – PND 7, 250 and 1000 mg/kg/day) 145 . Similarly, Lee et al. 57 found propyl- and butylparaben (100 mg/kg/day) upregulated Amh expression in Sprague Dawley rats. This upregulation of Amh by propyl- and butylparaben may inhibit folliculogenesis of primary follicles, offering a mechanism to explain why fewer primary, secondary, and pre-ovulatory follicles were observed in rodents treated with these parabens 57 , 145 . However, given methylparaben treatment downregulated Amh expression 145 while still increasing the number of primordial follicles 57 suggests other factors may be involved in mediating paraben effects on folliculogenesis. Kit ligand/stem cell factor (KITL) is involved in various stages of ovarian development including primordial follicle development and initiation of folliculogenesis 149 , 150 . Ahn et al. 145 reported ovarian KITL gene expression was upregulated by butylparaben exposure (PND 1 – PND 7, 250 and 1000 mg/kg/day) but not by methyl- or propylparaben, in Sprague Dawley rats. In contrast, Lee et al. 57 reported Kitl expression was upregulated by propylparaben exposure (100 mg/kg/day, 5 weeks), as well as propyl-, or butylparaben combined with VCD treatment. The upregulation in Kitl may explain why several studies found an increase in primordial follicles in paraben-treated rodents 49 , 145 . Forkhead box protein l2 (FOXL2) is a transcription factor involved in cellular differentiation and myriad developmental processes, possibly including ovarian follicle maturation 151 . In Sprague Dawley rats, neonatal ovarian Foxl2 expression was selectively upregulated by butyl- and propylparaben exposures and downregulated by methylparaben exposure at all doses (PND 1 – PND 7, 62.5–1000 mg/kg/day) 145 . While in adult Sprague Dawley rats propylparaben exposure (5 weeks, 100 mg/kg/day) downregulated Foxl2 expression, but not by methyl- or butylparaben 57 . Notably, Foxl2 expression was not affected in rats treated with estradiol (positive control), suggesting the effect is not estrogen dependent 145 . The discrepancy in the results between the two studies may suggest paraben effects on Foxl2 vary by the timing and levels of exposure and may act via non-estrogenic pathways. Further, FOXL2 may regulate genes playing an essential role in late folliculogenesis and ovulation 151 , and thus, Foxl2 is potentially an interesting target of paraben that should be considered when further examining paraben effects on ovarian function. Ovarian steroidogenesis is one of the central functions of the ovary. Steroidogenesis involves key proteins including steroidogenic acute regulatory protein (StAR), cytochrome P450 family 11 subfamily A member 1 (CYP11A1), cytochrome P450 family 19 subfamily A member 1 (CYP19A1, aromatase), and hydroxy-delta-5-steroid dehydrogenase, 3 beta- and steroid delta-isomerase 1 (HSD3B1). StAR is a rate-limiting transport protein that shuttles cholesterol from the outer to the inner mitochondrial membrane. In the mitochondria, cholesterol is converted to pregnenolone by CYP11A1, which is further converted to dehydroepiandrosterone or progesterone by HSD3B1 152 , while aromatase catalyzes the conversion of androgens to estrogens 153 . In Sprague-Dawley neonatal rats, propyl-, methyl, and butylparaben exposures (PND 1 – PND 7, 62.5–1000 mg/kg/day) decreased Star and increased or decreased Cyp11a1 expression 145 . In contrast, in Holtzman rats, butylparaben exposure (GD 6 – PND21, 100 mg/kg/day) increased Star expression 49 . Further, in 4-week-old Balb/c mice, propylparaben exposure (21 days, 100 mg/kg/day) increased StAR and CYP19A1 compared to controls 56 . Last, in 8-week-old Sprague-Dawley rats, methyl-, propyl-, and butylparaben exposure (5 weeks, 100 mg/kg/day) did not impact ovarian Star expression, Cyp11a1 expression was upregulated only by methylparaben exposure, Hsd3b1 expression decreased only by propyl- and butylparaben exposure, and Cyp19a1 expression was decreased by all parabens 57 . Results from these studies suggest a complex relationship between paraben exposure and steroidogenesis, potentially modulated by the specific paraben, species, dose, exposure route (i.e., in utero and lactation versus direct exposure), and duration and age at which exposure occurs. More research is needed to elucidate the specific mechanisms while considering these moderators. In Holtzman rats, butylparaben exposure (GD 6 – PND 21, 100, 1000 mg/kg/day) selectively upregulated ERα and ERβ expression 49 . In contrast, in Wistar rat fetuses, ethyl- or butylparaben exposure (GD 7 – GD 21, 200, 400 mg/kg/day) did not affect ERα expression and downregulated ERβ expression only by butylparaben, compared to controls 78 . Last, in Wistar rat fetuses, butylparaben exposure (GD 7 – GD 20 or GD 21, 100 mg/kg/day) did not change the expression of ERα and ERβ genes 115 . Overall, the timing of gestational exposure and age at ovarian analyses may change how parabens affect hormone receptor expression. The uterotrophic assay in rodents is regarded as the “gold standard” for determining estrogenicity of chemicals, through comparison of uterine weight 122 , 154 , 155 . Experimental studies utilized the uterotrophic assay to measure the relative estrogenic effects of parabens in several models, including immature and adult rodents, ovariectomized rodents, and in rodents following gestational paraben exposure. Overall, in immature and adult rodents, paraben appeared to increase uterus weight. However, effects were variable depending on mouse strain, type of paraben, timing, and duration of exposure. Last, gestational paraben did not appear to appreciably affect uterine weight. In Sprague Dawley rats, butylparaben (PND 1 - PND 7, 1000 mg/kg/day) increased uterus weight but not propyl- or methylparaben 145 . In Sprague Dawley rats, exposure (PND 14 – PND 16, 250 or 1000 mg/kg/day) to iso-propyl-, butylparaben, and iso-butylparaben, but not propylparaben, selectively increased uterus weight 156 . Furthermore, pre-treatment with the antiestrogen ICI 182,780 reversed paraben effects on uterus weight, suggesting the effect was estrogen-dependent 156 . In Wistar rats, butylparaben exposure (PND 20 – PND 22, 10–200 mg/kg/day) had no effect on uterine wet weight or dry weight 51 . Notably, Wistar rats required higher doses of estradiol to produce an effect on uterus weight when compared to Sprague Dawley rats, suggesting the uterotrophic assay is less sensitive in Wistar rats 157 . Last, in older Sprague Dawley rats, methyl-, ethyl-, propyl-, iso-propyl-, or iso-butylparaben exposure (PND 21 – PND 40, 62.5–1000 mg/kg/day) did not increase uterus weight 55 , suggesting that timing of exposure may result in different outcomes. In Sprague-Dawley rats, benzylparaben exposure (PND 21 – PND 24, 0.16–100 mg/kg/day) selectively increased relative uterus weight compared to controls 157 . In CD-1 mice, iso-butylparaben exposure (PND 18 – PND 20, 12 mg/day) increased uterus wet weight 158 . Furthermore, benzylparaben exposure (PND 18 – PND 20, 10, 33 mg/day) selectively increased uterine weight 159 . Lemini et al. 160 also found an increase in uterus weight in immature (PND 21) CD-1 mice, as well as immature Wistar rats (PND 21) exposed to methyl-, ethyl-, propyl-, and butylparaben for three days, but only at specific doses. Last, Hossaini et al. treated B6D2 mice with individual parabens and mixtures (e.g., butyl-, methyl-, ethyl-, butyl-, propylparaben, or mixtures) and found no effect on uterus weight following three days of exposure 161 . These null results may be related to lesser uterine estrogen sensitivity in this strain of mice. In ovariectomized adult CD-1 mice (age not specified), uterus weight was greater after three days of exposure to ethyl-, methyl-, propyl-, and butylparaben at various doses 160 . In CF-1 and CD-1 ovariectomized mice, butylparaben exposure (PND 27 – PND 29, 0.735 – 35 mg/day) did not affect uterus weight 81 . In ovariectomized Sprague Dawley rats, propylparaben (400 mg/kg/day) and co-exposure of propylparaben and BPA increased uterine weight after three days of exposure 162 . These findings suggest that parabens may increase uterine weight, regardless of estrogen contributions from the ovary. In Wistar rats, butylparaben exposure (GD 7 – PND 21 or 22, 100, 500 mg/kg/day) did not affect uterus weight 75 . In Wistar rats, butylparaben exposure (GD 12 – PND 21, 10 – 200 mg/kg/day) alone or when combined with DEHP resulted in similar uterus weight as controls 51 , 53 . Similarly, in Sprague-Dawley rats, butylparaben exposure (GD 6 – PND 20, 100, 200 mg/kg/day) did not increase uterus weight at PND 21- PND 90 80 . In Holtzman rats, butylparaben exposure (GD 6 – PND 21, 1000 mg/kg/day) resulted in increased uterus weight only on PND 75 49 . In Sprague Dawley rats, iso-butylparaben exposure (3 weeks before mating through lactation, 4.4. mg/day) increased uterus weight of the dams but not of the pups on PND 21 52 . Finally, in Swiss albino mice, gestational exposure (GD 1- GD 21) to a mixture of iso-butylparaben and BPA did not change uterus weight compared to controls 82 . Overall, these results suggest paraben exposure during gestation and lactation does not affect uterine weight and that the critical period for exposure on this measure likely occurs post-weaning. A handful of studies investigated how paraben exposure affected uterine histology. Guerra et al. 51 analyzed the thickness of luminal and glandular epithelium, endometrial height, and the number of glands in uteri from Wistar rats treated with butylparaben (GD 12 – PND 21) and found no significant changes. In Holtzman rats, butylparaben exposure (GD 6 – PND 21, 10–1000 mg/kg/day) resulted in selective inadequate and irregular development of uterine stroma on PND 30 – 75 and degenerative changes in uterine epithelium on PND 45 and PND 75 49 . In ovariectomized Sprague Dawley rats, only co-exposure of propylparaben and BPA (400 mg/kg/day) resulted in increased numbers of endometrial glands and a change from cuboidal to columnar endometrial epithelial cells, compared to controls 162 . In Sprague Dawley rats, butyl-, propyl-, iso-propyl, and iso-butylparaben exposure (PND 21 – PND 40, 62.5–1000 mg/kg/day) selectively increased myometrial hypertrophy 55 . Similarly, Vo et al. 82 found disorganization and hypertrophy of endometrial glands in Swiss Albino mice co-exposed to BPA and iso-butylparaben (GD 1 – GD 21, 5–500 mg/kg/day). Calbindin-9k ( CaBP-9k ) expression is frequently used as a biomarker of estrogenicity for various EDCs in rodents’ uteri 163 – 165 . In Sprague Dawley rats, CaBP-9k expression was upregulated following exposure to propyl- and butylparaben, (PND 1 – PND 7, 250, 1000 mg/kg/day) but not methylparaben 145 . Similarly, in Sprague Dawley rats, CaBP-9k mRNA and protein levels were selectively increased following exposure of iso-butyl-, iso-propyl-, and butylparaben (PND 14 – PND 16, 250–1000 mg/kg/day) 156 . These results agree with the results of uterotrophic assays, suggesting parabens exert estrogenic effects in uterine tissue, yet the doses that resulted in significant increase of Calbindin-9K levels were relatively supra-physiologic. Estrogen receptors α and β (ERα, ERβ) and progesterone receptor (PGR) are nuclear receptors that play a central role in regulating the function of female reproductive tissues. In Sprague-Dawley rats, exposure to ethyl-, iso-propyl-, and butylparaben (PND 14 – PND 16, 250–1000 mg/kg/day) selectively downregulated uterine Erα mRNA and protein levels 156 . Furthermore, paraben effects were only partially reversed by a pre-treatment with ICI 182,780, suggesting paraben effects may be partially estrogen-mediated 156 . Additionally, uterine Pgr mRNA and protein levels were selectively downregulated following exposure to iso-, propyl-, and butylparaben (250–1000 mg/kg/day, PND 14 – PND 16) 156 . These results suggest parabens may selectively impact estrogen and progesterone signaling in the uterus, which may have implications for uterine disease, cyclicity, fertility, and pregnancy maintenance.

Paraben exposure is ubiquitous for women worldwide, with myriad environmental and lifestyle exposure sources. The ubiquitous exposures, combined with the endocrine-disrupting effects of parabens, has raised concerns regarding the effects of parabens on female reproductive health. In this review we extensively searched and summarized the existing literature. In the first section on cyclicity, studies suggest that paraben exposure during female reproductive years may impact estrous cyclicity. While several studies indicated paraben exposure shortened cycle length in both women 47 and experimental models 49 , 55 , others found paraben exposure resulted in more general cycle perturbations 54 , 56 . Paraben effects on fertility were limited in epidemiologic studies. Only methylparaben exposure was associated with a lower ovarian sensitivity index 61 , and only one study found an association between paraben exposure and lower AFC 74 . Despite this, parabens seemed to have only a moderate impact on fertility and fecundity rates, with some associations reported between higher paraben exposure and increased risk of infertility 69 , 72 or decreased fecundity odds ratios 68 . The limited experimental studies also largely reported null results between paraben exposure and fertility outcomes. As for pregnancy outcomes, only a handful of epidemiologic studies investigated paraben’s effects on gestation length which reported some associations with altered gestation length 11 , 91 , 95 , 102 , 129 . Fewer experimental studies investigated paraben effects on gestation length, all finding null results 75 , 81 . In epidemiologic studies, gestational paraben exposure was associated with fetal size (e.g., weight, length, head circumference). Many of these effects on fetal growth were dependent upon the sex of the fetus 104 96 – 100 , 102 , 103 , 105 – 107 . Interestingly, experimental studies failed to report any effect of gestational paraben exposure on birth size 49 , 51 , 53 , 75 , 78 , 80 , 82 , yet, experimental studies were limited in numbers and examined end points. Findings on gestational paraben levels and postnatal development were also limited and these indicated that parabens were selectively associated with lower height and weight among children of younger ages (under 3 years) 99 , 117 whereas in older children (up to 8 years) parabens were associated with higher weight 116 , 118 . Experimental studies on postnatal growth reported mostly null results 49 , 80 , 116 . Further, of the limited available literature, only peripubertal paraben exposure was associated with earlier age of pubertal onset, while in experimental studies, findings were inconsistent among studies 49 , 80 . Paraben exposure was associated with altered levels of reproductive and thyroid hormones and signaling 80 . This included derangements in serum DHEA-S, FSH, 17-OHP, LH, and FSH in young girls 110 . In adult non-pregnant women, paraben exposure was associated with derangements in serum FSH, LH, and estradiol 56 , 66 , 126 , 137 . In pregnant women, paraben exposure was associated with changes in serum estradiol, SHBG, testosterone, CRH, T3, free-T4, total T4, and the ratio of T3 to T4 93 , 128 , 132 . Interestingly, a handful of experimental data supported some of the findings of epidemiological studies 49 , 51 , 52 , 54 – 56 , 133 . The effects on reproductive-related hormones may be mediated by paraben effects on steroidogenic factors and hormone receptors 49 , 75 , 78 . There is a major gap in the scientific literature related to ovarian and uterine pathologies. There were only two epidemiology-based studies reporting associations between paraben and ovarian pathologies and null associations with endometriosis 143 , 144 . Few experimental studies indicate paraben selectively alters folliculogenesis and ovulation by potentially altering expression of Amh, Kitl , and Foxl2 57 , 145 . Additionally, paraben exposure during gestation/lactation does not seem to affect uterine weight of the offspring unless examined post-weaning and/or at supra-physiologic levels 49 , 52 , 53 , 55 , 75 , 80 , 81 , 145 , 156 – 158 , 160 , 162 . Last, paraben selectively increased endometrial glands and myometrial hypertrophy potentially by altering levels of estrogen and progesterone receptors 49 , 51 , 82 , 156 .

In women, the menstrual cycle ranges from 26–34 days, consisting of a period of menses, a follicular phase induced by rising follicle stimulating hormone (FSH) levels, ovulation following a luteinizing hormone (LH) surge, and a luteal phase post-ovulation 45 . The menstrual cycle is controlled by reproductive-related hormones and its characteristics overall represent endocrine functioning in females of reproductive age 45 . Comparatively, the estrous cycle has four phases, proestrus, estrus, metestrus, and diestrus, with the entire cycle lasting 4–5 days in rats and mice. Physiologically, the proestrus stage corresponds to the follicular phase in humans. The estrus phase correlates with ovulation in the human menstrual cycle, while metestrus and diestrus approximate the luteal phase in humans 46 . As in humans, perturbations in the estrous cycle of rodents are indicative of endocrine dysfunction. Given the hormone-mimicking properties of parabens, an investigation regarding the specific effects of parabens on cyclicity is warranted. Interestingly, despite myriad in vivo studies on the topic, only one epidemiologic study investigated the associations between paraben exposure and menstrual cyclicity. Overall, this single epidemiologic study suggested paraben exposure was associated with shorter menstrual cycle length. However, given this study recruited only young women of a limited age range, the results are difficult to extrapolate. Thus, more studies are warranted to investigate the effects of parabens on menstrual cyclicity in more diverse populations. This is especially of concern, as the experimental studies reviewed below reported some effects of paraben on cyclicity. A study of 19–22-year-old women in Japan found an association between urinary paraben concentrations and menstrual cycle variations 47 . The researchers reported that butylparaben levels were associated with shorter menstrual cycle length. Additionally, sum parabens weighted by their relative estrogenic activity (i.e., butylparaben > propylparaben > ethylparaben > methylparaben) was associated with shorter menstrual cycle length 47 . Further, propylparaben and the weighted sum parabens were associated with decreased intra-individual variability of menstrual cycle length, which could be due to the shortening of cycle length seen with paraben exposure 47 . These results may have specific implications on women’s fertility as shorter menstrual cycle length has been associated with a decreased probability of pregnancy 46 and a longer time to pregnancy 48 . In Holtzman rats, butylparaben exposure (GD 6 to PND 21) resulted in a shorter estrus stage and a shorter estrous cycle length, but only with exposure at 10 mg/kg/day and 1000 mg/kg/day and not at 100 mg/kg/day, suggesting the effect followed a non-monotonic dose-response 49 . In Sprague-Dawley rats, butylparaben exposure (13 weeks 2–50 mg/kg/day, or GD 12 to PND 21 10–200 mg/kg/day) had no effect on estrous cyclicity 50 , 51 . Similarly, no changes were seen in Sprague Dawley rats exposed to iso-butylparaben (4.4 mg/day, pre-gestation through lactation) 52 . Likewise, no effect was seen on estrous cyclicity in Wistar rats co-exposed to butylparaben (GD 12 – PND 21, 100 mg/kg/day) and di(2-ethylhexyl)phthalate (DEHP, 500 mg/kg/day) 53 . These data suggest that butylparaben effects on estrous cyclicity may be strain-specific in rats. Other parabens were reported to have more pronounced effects on estrous cyclicity. C57BL/6J mice exposed to propylparaben (8 months, 7.5 mg/kg/day) had irregular estrous cycles, with significantly more days in metestrus 54 . Similarly, Sprague-Dawley rats dosed with iso-propylparaben or methylparaben (PND 21 to PND 40, 62.5–1000 mg/kg/day) had shorter estrous cycles, while propyl-, butyl-, and iso-butylparaben did not affect estrus length 55 . Balb/c mice exposed to propylparaben (21 days, 100 mg/kg/day) had longer estrus and shorter proestrus periods than controls 56 . In another study, Sprague-Dawley rats were exposed to methyl-, propyl-, or butylparaben (100 mg/kg/day) for 5 weeks alone or following a 2-week pre-treatment with 4-vinylcyclohexane diepoxide (VCD), which induced premature ovarian failure 57 . Results indicate that cotreatment with methylparaben and VCD induced a persistent diestrus after 16 days from first administration (similar to VCD treatment alone). Treatment with butylparaben alone shortened the estrous cycle, while cotreatment with butylparaben and VCD induced a persistent diestrus earlier than VCD alone (i.e., after 13 days). Interestingly, treatment with propylparaben alone induced persistent diestrus after 15 days, whereas cotreatment with propylparaben and VCD induced diestrus after only 12 days. Overall, these results suggest paraben exposure during female reproductive years may impact estrous cyclicity, with iso/propylparaben exposure resulting in the most profound effect on estrous cyclicity. Propylparaben is ubiquitous in the environment and has been frequently detected in air and soil 13 , dust 14 , pharmaceuticals 7 , PCPs 8 – 10 , and subsequently in biological samples from women. Thus, it may represent a potential disruptor to female reproduction. More studies are needed to examine the potential implications of propylparaben exposure on women’s reproductive health.

Due to the endocrine-disrupting effects of parabens, many have sought to quantify how these chemicals may affect fecundity and fertility. Fecundity is defined as the capability of an individual or population to produce offspring, whereas fertility is defined as the number of offspring produced by a population or individual. The summarized studies used several different endpoints to assess fertility and fecundity, yet the terms were often used interchangeably. Hence, the terminology used in the original papers have been retained in the text below. Of the epidemiologic studies reviewed, the data suggest parabens may selectively impact fertility outcomes depending on the causes of infertility (e.g., maternal versus paternal causes of infertility), timing of paraben exposure during gestation, and levels of exposure. Notably, these studies primarily utilized populations of couples seeking treatment for infertility. Utilizing a population already experiencing infertility may confound extrapolation of these results to a healthy population. Regardless, additional studies are needed to better understand the specific targets and mechanisms of action. This critical gap in knowledge may be more closely approximated by experimental studies. However, this extrapolation to a human population is difficult, given species variation in paraben metabolism, limitations in mimicking real-life exposures of multiple EDCs, and differences in the chronicity of exposure in humans compared to animal models. Though women are susceptible to paraben exposure 20 , 58 , only a handful of studies have investigated the effects of parabens on maternal fertility in humans. These studies primarily used a prospective cohort design, including populations of couples seeking fertility treatments. The results are mixed, yet mostly report null associations. Specifically, maternal paraben concentrations 59 or paraben combined with phthalate and bisphenol A 60 were not associated with the number of live births or clinical pregnancies. Similarly, methylparaben concentration in follicular fluid was not associated with live births or clinical pregnancies 61 . Yet, methylparaben was associated with a lower ovarian sensitivity index 61 , which is a measure of ovarian response during fertility treatments calculated by the ratio of the number of oocytes retrieved/total dose of FSH administered 62 . Interestingly, this finding was significant in an Estonian cohort but not in a Swedish cohort nor the combined analysis of both cohorts 61 . Further, the association between methylparaben and lower ovarian sensitivity index remained statistically significant when controlling for male infertility causes but not when female-related causes of infertility (e.g., endometrial thickness, AMH levels) were included as potential mediators 61 . Antral follicle count (AFC) is a quantitative marker of ovarian reserve and responsiveness in fertility treatments 63 , 64 . Smith et al. reported a marginal association between propylparaben and lower AFC 65 , while Jurewicz et al. 66 reported an inverse association between propylparaben concentration in the 3rd quartile and AFC. Nevertheless, other measures such as total and mature oocyte counts, rates of fertilization, and implantation rates were not associated with urinary paraben concentrations when assessed alone 59 or when combined with phthalate and bisphenol A levels 60 . Similarly, urinary paraben concentrations were not associated with the number of recovered high-quality embryos determined by morphology or the number of blastomeres 59 . Last, the sum of urinary paraben, phthalate, and bisphenol A concentrations were not associated with endometrial wall thickness 60 , a measure correlated with pregnancy loss and percentage of live births during IVF treatments 67 . Smarr et al. 68 reported several associations between parabens and couples’ fecundity odds ratio (FOR), which evaluates the odds of conception per menstrual cycle. Specifically, higher methylparaben urinary levels (4th quartile, ≥104 ng/mL) were associated with a 34% reduction in FOR than lower levels (1st quartile, <12 ng/mL). Likewise, higher levels of ethylparaben (3rd quartile, ≥5.62 ng/mL) were associated with a 34% reduction in FOR compared to lower levels (1st quartile, <0.27 ng/mL). Ao et al. 69 studied the effects of parabens on infertility and fecundity in couple-based models. They reported that only in maternal-specific models, propylparaben (2nd tertile, 0.23 – 19.9 ng/mL), butylparaben (2nd and 3rd tertiles, >0.001 ng/mL), and heptylparaben (> LOD) were associated with decreased fecundity and increased risk of infertility 69 . Similarly, Dodge et al. 70 and Mínguez-Alacrón et al. 71 reported that paternal urinary paraben concentrations were not associated with IVF outcomes. However, paternal urinary methylparaben concentration (2nd quartile, 10.47 – 28.95 ng/mL) was associated with decreased odds of live birth after artificial insemination 70 . Last, Arya et al. 72 reported that ethylparaben concentrations at or above the LOD were associated with self-reported female infertility in NHANES cohorts. The same study reported null associations between self-reported infertility and butylparaben, methylparaben, and propylparaben concentrations 72 . Two other studies assessed ovarian volume, which is another predictor of response to superovulation during fertility treatments 73 , and a measure of ovarian reserve and subsequently status of reproductive aging. Specifically, in a population of younger women (21 – 46 years), ovarian volume was not associated with paraben concentrations 65 , but in a population of mid-life women (45 – 54 years), ovarian volume was marginally inversely associated with ethylparaben, propylparaben, methylparaben, and sum parabens 74 . Overall, these data suggest that parabens may selectively impact fertility outcomes depending on the causes of infertility (e.g., maternal/paternal, geographical location) and timing and levels of exposure. Paraben may also impact the timing of reproductive senescence, yet additional studies are needed to better understand the specific targets and mechanisms of action. Several studies investigated the effect of parabens on fertility in various experimental models. Similar to the epidemiological studies discussed above, the results are mixed, likely due to differences in the study design and examined paraben. In Wistar rats, butylparaben exposure (gestation through weaning, 100 and 500 mg/kg/day) did not affect the percentage of post-implantation fetal loss, percentage of perinatal pup survival, nor the number of live births 51 , 53 , 75 – 78 even with concurrent exposure to DEHP 53 . Gestational methylparaben or butylparaben exposure (100 and 200 mg/kg/day) did not affect pup survival 77 , and ethylparaben exposure (GD 7 – GD 21, 400 mg/kg/day) did not affect the number of live births, number of implantation sites, the percentage of post-implantation loss, nor the percentage of late fetal resorptions 78 . In Sprague Dawley rats, butylparaben exposure (GD 6 – PND 4, 1500–15000 ppm) did not affect the number of live births nor pups surviving until PND 4 79 . Similarly, in Sprague Dawley rats, butylparaben exposure (GD 6 – GD 20, 100 and 200 mg/kg/day) did not affect the number of implantation sites 80 . Yet, smaller proportions of live births and pups surviving to weaning were found when compared to controls 80 . Further, in Holtzman rats, exposure to butylparaben (10–1000 mg/kg/day) had no effect on pre- or post-implantation fetal survival 49 . In Sprague Dawley rats, iso-butylparaben exposure (pre-gestation through lactation, 4.4 mg/day) did not affect litter size 52 . In CF-1 mice, butylparaben (GD 1 – GD 4, 35, 40 mg/kg/day) or propylparaben exposure (PND 27 – PND 29, 0.735, 7.35, and 35 mg/kg/day) did not affect the number of live or dead pups or the number of implantation sites 81 . Last, in Swiss albino mice, co-exposure to bisphenol A and iso-butylparaben (GD 1 – GD 21, 5, 50, 500 mg/kg/day) had no effect on the number of live births or dead pups 82 . Overall, only a limited number of experimental studies investigated the effects of paraben exposure on fertility outcomes in rodent models. Most studies found no effect when compared to controls. Yet, additional studies are needed to examine the effects of parabens on fertility and postnatal survival.

Given that parabens have been detected in samples of human amniotic fluid 24 , umbilical cord blood 25 , 39 , placentas 36 , 37 , and breast milk 29 – 32 it is likely that fetuses and infants are ubiquitously exposed during early stages of development. The following sections will summarize key findings of epidemiological and in vivo studies that examined the associations and potential effects of paraben exposure and pregnancy outcomes. The limited available epidemiologic studies suggest that gestational exposure of ethyl-, methyl-, and propylparaben may extend gestation length, but at relatively high doses or specific windows of exposure during gestation. Interestingly, butylparaben alone was not reported to affect gestation length except when included in the calculation of sum paraben levels. It is noteworthy that some countries have limited the use of butylparaben. Thus, it is detected in low frequencies/levels in some cohorts 20 , 28 , 83 – 90 and this may explain some of the null associations reported in the literature. In experimental rodent models, only two studies specifically investigated how butylparaben affected gestation length, finding null results. Thus, null results could be attributed to the specific paraben chosen, and further studies should include parabens more relevant to human exposure. Several studies investigated the association between paraben exposure and gestation length or the risk of pre-term birth. In a case-control study of full-term pregnancies, spontaneous pre-term pregnancies, and placental pre-term pregnancies maternal urinary ethylparaben (but not methyl-, propyl-, or butylparaben) was associated with a greater odds ratio [OR = 1.5; 95 %CI (1.16, 1.97)] of placental pre-term births 91 . Whereas Zhang et al. reported that first-trimester summation of maternal methyl-, propyl-, and butylparaben concentrations were associated with a higher risk of pre-term birth 92 . In contrast, Aker et al. 93 reported maternal methyl- and propylparaben levels in the second trimester (16–20 weeks) were associated with increased gestational age and decreased odds ratio of pre-term birth [methylparaben OR = 0.58; 95% CI (0.38, 0.89), propylparaben OR = 0.51; 95% CI (0.32, 0.80)]. Maternal propylparaben levels at 24–28 weeks were also associated with increased gestational age at birth 93 . Yet, other studies found no correlation between maternal paraben levels and gestation length. Specifically, Varghese et al. 11 reported no association between levels of methyl-, ethyl-, propyl-, butyl-, benzyl-, or heptylparaben in preconception maternal samples and gestational age at birth. Zhang et al. 94 reported no association between methyl-, propyl-, and ethylparaben in combination with other estrogenic compounds and pre-term birth. Last, Mustieles et al. 95 found no association between sum parabens in preconception maternal samples and pre-term birth, whereas a marginal association was reported between preconception paternal sum parabens and an elevated risk of pre-term birth [RR = 1.36; 95% CI (0.94, 1.96)]. The limited available studies suggest that gestational exposure of ethyl-, methyl-, and propylparaben may extend gestation length, but at relatively high doses or specific windows of exposure during gestation. Interestingly, butylparaben alone was not reported to affect gestation length except when included in the calculation of sum paraben levels. It is noteworthy that some countries have limited the use of butylparaben. Thus, it is detected in low frequencies/levels in some cohorts 20 , 28 , 83 – 90 and this may explain some of the null associations reported in the literature. Only two studies investigated the effects of paraben exposure on gestation length in vivo . In Wistar rats, butylparaben exposure (GD 7 to birth, 10–500 mg/kg/day) resulted in gestation lengths comparable to the control group 75 . Similarly, in CF-1 mice exposed to butylparaben (GD 1 – GD 4, 0.05–35 mg/day) gestation length was similar across treatment groups 81 . More experimental research should focus on other parabens more relevant to human exposure (e.g., methyl-, propylparaben, and mixtures). Also, given the data on paternal preconception paraben exposure, more research can expand on how paternal paraben exposure may affect gestation length. Future studies could further investigate how low-dose paraben exposure (individual and mixture of parabens) affects gestation length during different developmental windows.

Paraben exposure is ubiquitous for women throughout the world and lifespan, with myriad environmental and lifestyle exposure sources. The ubiquitous exposure and scientific findings suggesting endocrine-disrupting effects of parabens raise some concerns regarding potential implications on female reproductive health. While this review includes substantial research in many areas related to female reproduction, it also highlights significant gaps in the literature. More epidemiological and experimental studies are needed to investigate if and how human-relevant paraben exposures may affect female reproductive health.

Anogenital distance (AGD) of newborns is considered a sensitive marker for fetal androgen activity 109 . Generally, reduced AGD in males may indicate anti-androgenic effects. Epidemiologic findings indicate that AGD was inconsistently associated with gestational paraben exposure. Yet, significant associations were reported between maternal paraben and shorter AGD in male infants 28 , 110 . It is possible that matrix (e.g., serum, urine) and timing of collection of samples may explain some of the differences in the reported results. Collectively, these mostly null results are interesting given that other EDCs, such as BPA, have been implicated in changes to infant AGD 111 , 112 . Thus, it is possible that parabens have a limited anti-androgenic activity compared to other EDCs, leading to fewer significant findings. Differences in cohorts’ characteristics and limited ability in examining simultaneous exposures may also explain some of the differences in the reported results. Further, only a few experimental studies evaluated the effect of paraben on AGD, while most studies focused on butylparaben. Though most studies found no association between paraben exposure and AGD, mid-gestation butylparaben exposure decreased AGD in both male and female pups. Overall, these limited findings suggest timing of paraben exposure during pregnancy may affect postnatal AGD suggesting the effects may depend on the timing of evaluation and sex of the offspring. Anogenital distance (AGD) of newborns is considered a sensitive marker for fetal androgen activity 109 . Generally, reduced AGD in males may indicate antiandrogenic effects. Findings indicate that AGD was inconsistently associated with gestational paraben exposure. Specifically, Shirai et al. 113 found no associations between gestational paraben concentrations and AGD in newborn males. Similarly, AGD was not associated with parabens measured in maternal urine at 28 weeks of gestation in three-month-old infants, excluding maternal methylparaben which was associated with shorter AGD in males 110 . Last, Fisher et al. 28 reported an association between maternal serum propylparaben concentration measured at 10–17 weeks of gestation and shorter AGD in male infants at 24 months. Collectively, these mostly null results are interesting given that other EDCs, such as BPA, have been implicated in changes to infant AGD 111 , 112 . One possible explanation for these null results is due to timing of maternal sample-collection for paraben analysis which tended towards mid- to late pregnancy. Last, it is possible that parabens have a limited anti-androgenic activity compared to other EDCs, leading to fewer significant findings 114 . In Sprague Dawley rats, butylparaben exposure (GD 6 – birth, 100–200 mg/kg/day) did not affect AGD in pups on PND1 80 . Further, butylparaben exposure around mid-gestation to birth (10–400 mg/kg/day) of Wistar rats resulted in similar AGD compared to controls 51 , 78 , 115 . Similarly, butylparaben exposure (4.4 mg/kg/day) preconception through gestation and lactation had no impact on the AGD of Sprague Dawley rats at PND 7 52 . Yet, mid-gestation butylparaben exposure (100, 500 mg/kg/day) decreased AGD in female and male PND 1 neonates 75 . Interestingly, co-exposure of butylparaben (100 mg/kg/day) and DEHP (GD 12 to birth, 500 mg/kg/day) of Wistar rats increased AGD of PND 1 female pups 53 . Last, ethylparaben exposure (GD 7 to birth, 400 mg/kg/day) of Wistar rats did not affect AGD at birth 78 . These limited findings suggest paraben exposure during early pregnancy may affect postnatal AGD measurement suggesting androgenic or antiandrogenic effects depending on the timing of evaluation and sex of pups. Yet, only a few studies evaluated the effect of paraben on AGD, while most studies focused on butylparaben. Epidemiologic results suggest inverse associations between gestational paraben exposure and early childhood growth and potentially in a sex-specific manner. Interestingly, in children older than 7 years, gestational paraben exposure was associated with overweight 116 . Yet, the results of rodent-based studies on body weight are mixed and seem to partly be impacted by the examined dose, duration of exposure, and strain. Notably, all studies only investigated butylparabens. As epidemiologic studies found associations between methyl- 99 and ethylparaben 99 , 117 with postnatal growth, more in vivo studies could add valuable information to the limited scientific literature. In children at 0–29 months, placental ethylparaben (but not methyl-, propyl-, or butylparaben) concentration was associated with lower body mass index (BMI) z-scores 117 . Similarly, Wu et al. 99 reported maternal ethylparaben concentration was associated with lower weight z-scores and methylparaben with lower height z-scores in 1- and 2-year-olds, yet only in boys. These results suggest inverse associations between gestational paraben exposure and early childhood growth and potentially in a sex-specific manner. Interestingly, in older children, the direction of the association appeared to be reversed. Specifically, higher maternal (gestation week 28) urinary butylparaben concentrations were positively associated with higher fat and android (abdominal including visceral) fat percentages in seven-year-old children 118 . Furthermore, higher gestational iso-butylparaben and butylparaben were associated with overweight in children up to 8 years old 116 . In contrast, a longitudinal study following children until the age of 11 years reported null associations between gestational paraben concentrations and BMI 119 . Differences in study designs and analytical data methods may explain some discrepancies in the relatively limited abovementioned results. Butylparaben exposure (GD 7 to PND 22, 100 and 500 mg/kg/day) of Wistar rats did not affect pups’ body weight and weight of retroperitoneal fat pad evaluated at various ages 75 . Similarly, butylparaben exposure (GD 12 to PND 21, 10–200 mg/kg/day) did not affect body weight of female offspring at various ages 51 , 53 . In contrast, butylparaben exposure (GD 6 to PND 20, 100 mg/kg/day) in Sprague-Dawley rats lowered offspring weights as young adults compared to controls 80 . Whereas butylparaben exposure (4.4 mg/kg/day) preconception through gestation and lactation of Sprague-Dawley rats resulted in similar weights to controls at PND 7 52 . Last, butylparaben alone (100 mg/kg/day) significantly altered the weights of the offspring compared to co-exposure of butylparaben and DEHP (500 mg/kg/day) 53 . Despite the conflicting results in Sprague-Dawley and Wistar rats, butylparaben affected postnatal body weight in other experimental models. Maske et al. found that Holtzman rat female pups exposed to butylparaben from GD 6 to PND 21 had higher body weight at various ages 49 . However, the effects were observed only at 10 and 100 mg/kg/day and not at 1000 mg/kg/day 49 . Furthermore, butylparaben exposure (bi-weekly during GD 0 to PND 28, 1.75 μg) of BALBcByJ mice increased body weight, lowered lean body mass, increased percentage fat mass, and increased food intake of female offspring 116 . Interestingly, postnatal co-treatment with DNA methyltransferase inhibitor 5-Aza-2’-deoxycytidine attenuated the effects of n-butylparaben on body weight and food intake, suggesting DNA hypermethylation may be involved in these effects 116 . Finally, co-exposure of bisphenol A and iso-butylparaben (GD 1 to GD 21, 5–500 mg/kg/day) of Swiss Albino mice decreased the body weight of the offspring however, it is unclear if the effects were due to bisphenol A, iso-butylparaben, or both because the study did not include a treatment group of iso-butylparaben alone 82 . The experimental results on body weight are mixed and seem to partly be impacted by the examined dose. Notably, none of these studies investigated parabens besides butylparabens. As epidemiologic studies found associations between methyl- 99 and ethylparaben 99 , 117 with postnatal growth, more in vivo studies could add valuable information to the scientific literature. Pubertal onset is marked by the development of external sexual characteristics and reproductive function as part of the transition to adulthood. Various factors such as nutritional status, genetics, and emotional well-being may impact the timing of pubertal onset. In recent years, several EDCs, including paraben, were examined in the context of altered timing of pubertal onset. In epidemiologic studies, gestational paraben exposure was not associated with pubertal onset. However, in peripubertal boys and girls parabens were associated with earlier pubertal onset. In experimental studies, butylparaben was shown to affect pubertal onset depending on the dose and timing of exposure. Additional studies are needed to expand the limited data on paraben effects on pubertal onset. Pubertal onset is marked by the development of external sexual characteristics and reproductive function as part of the transition to adulthood. Various factors such as nutritional status, genetics, and emotional well-being may impact the timing of pubertal onset. In recent years, several EDCs, including paraben, were examined in the context of altered timing of pubertal onset. Harley et al. 120 reported that gestational paraben exposure was not associated with puberty timing in males or females. However, in peripubertal girls (9 years), urine methylparaben concentration was associated with earlier thelarche, pubarche, and menarche, whereas propylparaben was associated with earlier pubarche 120 . In boys (9 years), propylparaben concentration was associated with earlier gonadarche 120 . Thus, paraben levels during the peripubertal period may be a better predictor of the timing of pubertal onset compared to gestational paraben exposure. However, more studies are needed to confirm these findings. In rodents, in vivo assessment of vaginal opening combined with the age at first estrus is used to determine reproductive maturation 121 . In Wistar rats, butylparaben exposure (GD 7 to PND 21–22, 10–500 mg/kg/day) had no effect on the timing of vaginal opening 75 . Similarly, butylparaben exposure (GD 12 to PND 21, 10–200 mg/kg/day) had no effect on timing of vaginal opening or age at first estrus compared to controls 51 . Interestingly, co-exposure of DEHP (500 mg/kg/day) and butylparaben (100 mg/kg/day) during GD 12 to PND 21 delayed the timing of vaginal opening but not by butylparaben alone 53 . In Sprague-Dawley rats, butylparaben exposure (GD 6 to PND 20, 100 mg/kg/day) advanced the timing of vaginal opening 122 . In contrast, butylparaben exposure at higher doses (GD 6 to PND 21, 100 and 1000 mg/kg/day) delayed timing of vaginal opening in Holtzman rats, whereas no effect was seen at a lower dose (10 mg/kg/day) 49 . Last, in Sprague Dawley rats, iso-butylparaben exposure (4.4 mg/kg/day) did not affect the age at vaginal opening 52 . Overall, these results suggest butylparaben may affect the timing of pubertal onset, but the effect may be dependent on dose, route of exposure, and timing of exposure.

Reports of ubiquitous environmental exposure to endocrine disrupting chemicals (EDCs) have led to a growing concern regarding their potential effects on human health 1 . Reports of global lifetime infertility prevalence reaching 17.5% 2 , along with reports of the hormone-mimicking characteristics of some EDCs have added to this concern, thus prompting research regarding how EDC exposure may affect fertility and reproductive health 3 , 4 . The deleterious effects of EDCs can begin in-utero during embryonic development of the female reproductive system, continue throughout pubertal development and into adult life, culminating towards the timing of reproductive senescence 4 . While much of the research on the reprotoxic effects of EDCs predominantly focused on bisphenols, phthalates, pesticides, and industrial chemicals, research on parabens is relatively limited 5 . Parabens are a group of chemicals synthesized by esterification of p-hydroxybenzoic acid. For nearly a century, parabens have been used as antimicrobial agents in various products including processed foods and drinks 6 , pharmaceuticals 7 , and personal care products (PCPs) 8 – 12 . The ubiquitous use of parabens results in detection of parabens and their metabolites in soil, indoor dust, water sources, and outdoor air samples 13 , 14 . Exposure to paraben can be via dermal, inhalation, and ingestion/dietary routes. Once absorbed, paraben undergo biotransformation by hydrolysis to 4-hydroxybenzoic acid (PHBA) and alcohol, and conjugation with glucuronide acid, sulfate acid, and glycine (reviewed by Nowak et al.,) 15 . In humans, parabens are primarily excreted in the urine within 12–72 16 , 17 depending on route of exposure and specific paraben (reviewed by Penrose and Cob, 2021) 18 . Urine paraben concentration has been used most frequently to evaluate paraben exposure in human subjects, similar to evaluations of other EDCs. In rodents, paraben biotransformation and metabolism are less efficient than in humans. After dermal paraben exposure, as much as 50% was collected in tissues 19 . In rats, oral and subcutaneous exposure led to highest paraben concentrations in the urine, whereas dermal exposure resulted in highest concentration measured in the serum 20 . Interestingly, in humans and rats, females have higher dermal absorption rates than males 20 . Thus, when evaluating paraben levels researchers should consider species differences, sex, and feasibility (urine collection is none-invasive and most cost-effective). Though parabens have a relatively short half-life, they have been detected in various biological samples from women including blood 21 – 26 , breast milk 27 – 32 , breast tissue 33 , menstrual blood 34 , 35 , placental tissue 36 , 37 , ovarian follicular fluid 38 , and umbilical cord blood 25 , 39 . Collectively, these studies suggest that women across their lifespan are continuously exposed to parabens. This is worrying because studies have reported hormone-mimicking effects of parabens including anti-androgenic, estrogenic, and anti-estrogenic activity 40 , 41 . Downstream effects on hormone signaling include interference with steroidogenesis, metabolism of hormones, and morphological and physiological alterations to female reproduction 42 . Among the parabens detected, methylparaben is generally found at the highest concentration in biological samples as it is the most prevalently used paraben 21 , 27 , 29 – 31 , 34 , 35 , 39 , 43 . However, propyl, ethyl-, butyl-, benzyl-, and heptylparaben have also been detected. Notably, several parabens are often used in combination in PCPs 44 , thus creating additional concerns about compounded effects. In this review, we aim to summarize paraben effects on female reproductive health while focusing on findings of epidemiological and rodent-based experimental models. We also aim to identify potential knowledge gaps and directions of future research on parabens as emerging chemicals of concern.

Given the endocrine-disrupting properties of parabens, studies have investigated their effects on reproductive hormones. Many of these studies indicated that paraben exposure is associated with changes in reproductive hormone levels throughout the lifespan. In male infants, parabens were associated with higher dehydroepiandrosterone-sulfate (DHEA-S) levels 110 . Conversely, in female infants, parabens were associated with lower DHEA-S, FSH, and 17-hydroxyprogesterone 110 . In girls 6–17 years, paraben exposure has been associated with decreased estradiol, LH, and FSH 123 . Some of these associations appeared to be confounded by population characteristics such as race and weight 124 , 125 . In non-pregnant, adult women parabens (single or mixtures) were associated with changes in estradiol, progesterone, LH, among others 74 , 126 . Interestingly, results in adult women suggested that paraben co-exposures may have greater effects on hormone levels 74 , 126 , yet these associations may be impacted by having other reproductive disturbances 56 , 66 , 127 . In pregnant women, the timing of paraben exposure during gestation appeared to affect whether reproductive hormones were altered 128 – 132 . In experimental studies, gestational butylparaben exposure resulted in increased FSH and decreased estradiol, testosterone, and progesterone. Whereas adult paraben exposure demonstrated significant alterations in estradiol, triiodothyronine (T3), thyroxine (T4), thyroid stimulating hormone (TSH), estradiol, testosterone, and FSH. However, hormone perturbations were dependent on the species/strain, paraben studied, and doses 54 – 56 , 133 – 135 . Reproductive development in children, including pubertal onset, and the development of secondary sex characteristics, is highly regulated by the endocrine system and thus may be impacted by EDCs exposure. Jensen et al. 110 investigated the association between serum hormone levels in 3-month-old infants and gestational urinary paraben concentrations. They found the highest tertile of propylparaben was associated with higher DHEA-S levels in male infants 110 . In female infants, propylparaben was associated with lower levels of DHEA-S, FSH, and 17-hydroxyprogesterone, and methylparaben was associated with lower FSH levels 110 . No association was seen between infants’ gestational paraben levels and testosterone, LH, or androstenedione 110 . These results suggest gestational paraben exposure may affect hormone concentrations during early stages of development in both male and female infants. More research is needed as the specific effects of paraben-induced hormone perturbations during this period have not been fully elucidated. In girls, serum estradiol is low until LH and FSH surges at puberty to stimulate the ovary to make estradiol 136 . Altered hormone levels may have implications on reproductive development in girls, potentially including a delayed pubertal onset and menarche. Thus, Guth et al. 123 investigated the effects of paraben exposure on hormone levels in girls aged 6 – 17 years and found associations between the 2-fold increase in methyl-, ethyl-, propyl-, or butylparaben concentrations and a 4–5% decrease in estradiol, LH, and FSH concentrations. Similarly, every two-fold increase in sum parabens was associated with a 4% decrease in serum FSH and an 11 % decrease in serum LH 123 . Finally, in a study of adolescents (12 – 19 years), Aimuzi et al. found no association between paraben and testosterone, SHBG, or testosterone to estradiol ratio in female adolescents 124 . In another study of children and adolescents (6 – 19 years), sum paraben was not associated with testosterone levels in females 125 . It is noteworthy that the authors reported some significant associations between paraben and male participants suggesting sex-dimorphic sensitivities. Further, the study by Aimuzi et al. 124 mainly included Hispanic participants, and >42 % were overweight or obese, while Scinicariello & Buser 125 , included primarily non-Hispanic, white and 93.8 % normal or underweight participants, suggesting parabens effects may vary due to cohorts’ characteristics (e.g., race, weight). Several studies investigated the associations between paraben and reproductive hormones in women at various life stages. In premenopausal women (18 – 44 years), 4-hydroxybenzoic acid (4-HB, a paraben metabolite) was associated with higher FSH and progesterone, and 3,4-dihydroxybenzoic acid (3,4-DHB) was associated with higher estradiol levels, depending on the statistical model 126 . Further, neither of the parent parabens was associated with any of the examined hormones that included a “paraben factor” (butyl-, ethyl-, methyl-, and propylparaben) and a “paraben metabolite and BPA factor” (3,4-DHB, 4-HB, and BPA) were associated with higher estradiol, progesterone, and LH 126 . These results suggest that exposure to multiple parabens (which is more relevant to human exposure) may be more detrimental to hormone regulation than exposure to individual parabens. In American midlife women (45 – 54 years), every two-fold increase in methyl-, propylparaben, and the molar sum of parabens (methyl-, ethyl, propyl-, butylparaben) were associated with up to 14.5 % higher FSH, whereas no association was reported for estradiol, testosterone, progesterone, SHBG, or AMH 74 . Notably, during menopause, FSH levels begin to rise in response to estradiol levels falling. Therefore, the positive association of paraben and FSH in midlife women may be indicative of earlier onset of menopause, though additional studies are needed. Last, in women who used paraben-free PCPs for 28 days, urinary paraben levels were significantly reduced post-intervention while estradiol, progesterone, or SHBG levels were unchanged 127 . Thus, paraben may not be the only factor influencing hormone levels, and additional studies can further examine these outcomes. Several studies investigated the effects of paraben exposure on reproductive hormones also in non-pregnant women who were seeking fertility treatment. In a study of Chinese women (18 – 45 years), propylparaben was associated with lower estradiol (but not FSH, LH, progesterone, testosterone, or AMH) 56 . None of the other studied parabens (methyl-, ethyl-, or butylparaben) were associated with the measured serum hormones 56 . Similarly, in Polish women (24 – 39 years), only propylparaben (third quartile) was associated with lower estradiol and higher serum FSH, but not AMH 66 . In American women (32 – 38 years), no associations were seen between methyl-, propyl-, or butylparaben and estradiol or FSH 59 . Similarly, in American women (18 – 46 years), no association was seen between parabens and FSH 65 . Last, in Iranian women experiencing abnormal menstruation, methyl- and butylparaben were negatively associated with FSH and LH concentrations, whereas ethyl- and propylparaben were positively associated with FSH and LH concentrations 137 . Furthermore, methyl- or propylparaben (2nd versus 3rd tertiles) were associated with the odds ratio for an overall change in reproductive hormone concentrations 137 . These results suggest parabens may alter hormone levels in women with abnormal menstruation or infertility, yet the effects may be limited to specific parabens. Several studies examined the relationship between paraben exposure and reproductive hormones during pregnancy, as reproductive hormones are essential for maintaining pregnancy to term and fetal development. In pregnant women (16 – 28 weeks gestation) in Puerto Rico, an increase in butylparaben levels was associated with an 8.5% decrease in estradiol and a 9.3% decrease in the ratio of estradiol to progesterone 128 . Furthermore, increased methylparaben levels was associated with a 7.7% increase in SHBG 128 . Yet, when the data were stratified by gestation time point the associations appeared to be moderated by gestation age 128 . Specifically, methyl- and propylparaben were associated with lower estradiol at 16 – 20 weeks gestation and associated with higher estradiol at 24 – 28 weeks gestation 128 . In a larger cohort, no association was seen between paraben concentrations and estriol or the progesterone-to-estriol ratio 129 . Ethylparaben was associated with higher corticotropin-releasing hormone (CRH) and testosterone, whereas methylparaben was associated with lower testosterone 129 . In contrast to previous findings, butyl-, methyl-, and propylparaben were associated with lower SHBG 129 . Overall, results suggest that paraben may impact reproductive hormone levels during pregnancy depending on the timing of evaluation and specific paraben. Interestingly, Kolatorova et al. 138 reported that umbilical cord paraben concentrations (methyl-, propyl-, and sum parabens of methyl-, ethyl-, butyl-, and benzylparaben) were negatively associated with umbilical cord blood testosterone levels even though maternal plasma parabens were not associated with cord blood hormones. Furthermore, umbilical cord blood propylparaben concentration was negatively associated with umbilical cord blood DHEA 138 . Thus, paraben exposure through umbilical cord blood may potentially risk fetal development, especially the masculinization of male fetuses, which is dependent on higher testosterone levels in utero . These results are especially impactful given parabens’ rapid transplacental transfer rate 139 . Circulating thyroid hormone concentrations are critical during pregnancy. During the first 10 – 12 weeks of gestation, the fetus is unable to produce thyroid hormones 130 and required for the development and growth of the fetus 140 . EDCs, such as BPA, have previously been reported to be associated with altered thyroid hormone levels in pregnant women 130 . Recently, several studies investigated if parabens had similar effects. In a study of pregnant women (18 – 39 years) in Puerto Rico, butylparaben was associated with a 5.6% increase in free T4 128 . In a follow-up study, only methylparaben was marginally (p = 0.07) associated with a 6.9% reduction in TSH 129 . No other associations were found between parabens and serum TSH, free T4, total T3, total T4, or the T3 to T4 ratio in either study 128 , 129 . In pregnant American women, each increase in butylparaben was associated with a 2.8% decrease in T3 and a 3.7% decrease in the T3 to T4 ratio 131 . Furthermore, every increase in methylparaben was associated with a 2.4% increase in T3 and a 1.2% increase in total T4, while each increase in ethylparaben was associated with a 1.9% reduction in the T3 to T4 ratio 131 . Propylparaben was associated with a 3.1% reduction in free T4 131 . Furthermore, ethylparaben at 15 – 21 weeks gestation was associated with a marginal (4%) decrease in the T3 to T4 ratio 131 . Butylparaben was associated with a 6 – 7% decrease in T3 at 15 – 20 and > 30 weeks of gestation, a 5 – 6% decrease in the T3 to T4 ratio at 15 – 21 and > 30 weeks of gestation, and a 14.7% increase in TSH at > 30 weeks of gestation 131 . An increase in propylparaben was associated with a 5 – 6 % decrease in free T4 at < 15 weeks and 15 – 21 weeks gestation 131 . Finally, an increase in methylparaben was associated with a 4.2% decrease in free T4, a 3.7 % increase in T3, and a 2.7% increase in the T3 to T4 ratio, as well as a marginal decrease (7 %) in TSH at 21 – 30 weeks 131 . These findings are important because T4 is a critical hormone for fetal neurodevelopment, especially before 10 – 12 weeks gestation 141 . Berger et al. 132 reported that methyl- and propylparaben concentrations were associated with lower TSH in mothers, depending on the statistical model. Interestingly, the authors reported no association between maternal urinary paraben concentrations and hormone levels in their 3-month-old infants 132 . Last, Li et al. 108 investigated the association between maternal paraben levels and thyroid hormone levels, thyroid peroxidase antibodies, and thyroglobulin antibodies in umbilical cord blood. Evaluation of thyroid peroxidase and thyroglobulin antibodies are used to diagnose autoimmune/thyroid disorders. The researchers found that increased maternal ethylparaben levels were associated with increased cord blood total T3 and thyroid peroxidase antibody 108 . Similarly, increased propylparaben levels were associated with increased cord blood thyroid peroxidase antibody 108 . Overall, parabens, especially propylparaben, may potentially affect thyroid hormone levels in pregnant women. Several studies investigated the relationship between gestational paraben exposure and hormone concentrations in fetuses and pups throughout their development. In Wistar rats, butylparaben exposure (GD 7 to PND 21–22, 10–500 mg/kg/day) resulted in similar levels of estradiol and testosterone 75 , 115 . Similarly, in fetuses of Wistar rats, ethyl- or butylparaben exposure (GD 7 - GD 21) resulted in similar levels of testosterone and progesterone in testis, estradiol in ovaries, cortisol in adrenal glands, and plasma T4 across treatment groups 78 . Further, there were no differences in levels of testosterone, progesterone, T3, or T4 in the dams 78 . Butylparaben exposure in Holtzman rats (GD 6 - PND 21, 100 mg/kg/day) resulted in lower estradiol at PND 30 and PND 75 and increased testosterone at PND 45 in female offspring 49 . Additionally, butylparaben at 1000 mg/kg/day decreased progesterone in female offspring at PND 30 and PND 45 49 . In female pups of Wistar rats, butylparaben (GD 12 - PND 21, 10 mg/kg/day) increased FSH while LH, progesterone, and estradiol were similar compared to controls 51 . Last, in Sprague Dawley rats, iso-butylparaben exposure (pre-gestation through weaning, 4.36 mg/day) resulted in similar levels of LH, FSH, prolactin, and estradiol in the pups 52 . However, in the dams, corticosterone was significantly reduced while LH, FSH, estradiol, testosterone, total T3, and total T4 were similar to controls 52 . Overall, most studies focused on butylparaben and found it altered some hormone levels mostly at PND30 and older. Several studies investigated hormone levels following paraben exposure in rodents post-weaning age. In adult Sprague Dawley rats, co-exposure of iso-propylparaben and iso-butylparaben (100–600 mg/kg/day, 20 days) reduced levels of T3, TSH, estradiol, testosterone, and FSH, while exposure to individual parabens (50–600 mg/kg/day) did not alter hormone levels compared to controls 133 . In female Sprague Dawley rats, propyl-, iso-propyl-, butyl-, or iso-butylparaben (but not methyl- or ethylparaben) exposure (62.5–1000 mg/kg/day, PND 21– 40) reduced estradiol levels 55 . Furthermore, T4 was reduced after exposure to methyl-, propyl-, iso-butyl-, and iso-propylparaben 55 . In C57BL/6J female mice, propylparaben exposure (7.5. mg/kg/day, 8 months) reduced estradiol and progesterone levels without affecting FSH levels 54 . Further, in Balb/c adult female mice, propylparaben exposure (100, 1000 mg/kg/day, 21 days) reduced testosterone and increased estradiol levels, while there was no difference in progesterone levels compared to controls 56 . Last, CF-1 female mice given a single dose of butyl- or propylparaben had higher urine estradiol concentrations than controls 6–10 hours post-butylparaben exposure (3 mg) 134 . Similarly, CF-1 female mice exposed to a single dose of EDC mixture that included propylparaben (1 mg) had higher urine estradiol levels than controls 8 hours post-exposure 142 . Overall, these results suggest a potential disruption to the hormonal functioning of the animal and, therefore, a potential impact on their reproductive health/function. Lack of consistency in the reported results could be due to differences in paraben doses and experimental design. These hormone levels alterations potentially significantly impact female reproductive health despite not being observed at all doses, nor for all parabens, as EDCs do not always have the same effects at all doses 135 .