Original Article
Amniotic Fluid Phthalate Levels and Male Fetal Gonad Function Morten Søndergaard Jensen,a,b Ravinder Anand-Ivell,c Bent Nørgaard-Pedersen,d Bo A. G. Jönsson,e Jens Peter Bonde,f David M. Hougaard,d Arieh Cohen,d Christian H. Lindh,e Richard Ivell,g and Gunnar Tofth Background: Prenatal exposure to phthalates may pose a threat to human male reproduction. However, additional knowledge about the in vivo effect in humans is needed, and reported associations with genital abnormalities are inconclusive. We aimed to study prenatal di(2-ethylhexyl) phthalate (DEHP) and diisononyl phthalate (DiNP) exposure in relation to cryptorchidism, hypospadias, and human fetal Leydig cell function. Methods: We studied 270 cryptorchidism cases, 75 hypospadias cases, and 300 controls. Second-trimester amniotic fluid samples were available from a Danish pregnancy-screening biobank (n = 25,105) covering 1980–1996. We assayed metabolites of DEHP and DiNP (n = 645) and steroid hormones (n = 545) by mass spectrometry. We assayed insulin-like factor 3 by immunoassay (n = 475) and analyzed data using linear or logistic regression. Results: Mono(2-ethyl-5-carboxypentyl) phthalate (5cx-MEPP, DEHP metabolite) was not consistently associated with cryptorchidism or hypospadias. However, we observed an 18% higher (95% confidence interval [CI] = 5%–33%) testosterone level, and a 41% lower (−56% to −21%) insulin-like factor 3 level in the highest 5cx-MEPP tertile compared with the lowest. Mono(4methyl-7-carboxyheptyl) phthalate (7cx-MMeHP, DiNP metabolite) showed elevated odds ratio point estimates for having
Submitted 07 April 2014; accepted 15 July 2014; posted 7 November 2014. From the aDepartment of Pediatrics, Regional Hospital of Randers; bPerinatal Epidemiology Research Unit, Department of Pediatrics, Aarhus University Hospital, Skejby, Denmark; cSchool of Bioscience, University of Nottingham, United Kingdom; dDanish Center for Neonatal Screening, Department of Clinical Biochemistry and Immunology, Statens Serum Institute, Copenhagen, Denmark; eDivision of Occupational and Environmental Medicine, Department of Laboratory Medicine, Lund University, Lund, Sweden; fDepartment of Occupational and Environmental Medicine, Bispebjerg Hospital, University of Copenhagen, Copenhagen, Denmark; g Reproductive Endocrinology, Leibniz Institute for Farm Animal Biology, Dummerstorf, Germany; and hDanish Ramazzini Center, Department of Occupational Medicine, Aarhus University Hospital, Aarhus, Denmark. M.S.J. and R.A.-I. are joint first authors Supplemental digital content is available through direct URL citations in the HTML and PDF versions of this article (www.epidem.com). This content is not peer-reviewed or copy-edited; it is the sole responsibility of the authors. Correspondence: Morten Søndergaard Jensen, Department of Pediatrics, Regional Hospital of Randers, Skovlyvej 1, DK-8930, Randers NØ, Denmark. E-mail: [email protected]. Copyright © 2014 by Lippincott Williams & Wilkins ISSN: 1044-3983/15/2601-0091 DOI: 10.1097/EDE.0000000000000198
Epidemiology • Volume 26, Number 1, January 2015
cryptorchidism (odds ratio = 1.28 [95% CI = 0.80 to 2.01]) and hypospadias (1.69 [0.78 to 3.67]), but was not consistently associated with the steroid hormones or insulin-like factor 3. Conclusions: Data on the DEHP metabolite indicate possible interference with human male fetal gonadal function. Considering the DiNP metabolite, we cannot exclude (nor statistically confirm) an association with hypospadias and, less strongly, with cryptorchidism. (Epidemiology 2015;26: 91–99)
F
or more than a decade the prevailing view in the literature has been that certain phthalates may induce lack of masculinization and male reproductive disorders by reducing fetal testicular testosterone or insulin-like factor 3 production.1,2 Such views have been fueled by abundant experimental data from studies on rats.3 Most recently, however, studies using human ex vivo fetal testis xenograft methodology have questioned the hypothesized effects on testosterone production in humans,4,5 highlighting the important issue of possible species-specificity.6,7 The potential hazards posed to male reproduction during fetal life by phthalates must be fully characterized considering the virtually inevitable exposure. Phthalates such as di(2-ethylhexyl) phthalate (DEHP) and diisononyl phthalate (DiNP) are high-volume commercial chemicals used for production of certain building materials, car interiors, clothing, food packaging, toys, and medical devices.8 Biomonitoring has documented their pervasive nature by detecting metabolites in samples from pregnant women, cord blood, and amniotic fluid.9–13 Given this exposure scenario, knowledge about mechanisms of action in humans and high-quality epidemiologic studies of reproductive outcomes are indispensable. However, as mentioned, our mechanistic insight may be hampered by species-specific mechanisms and effects. Moreover, the few existing epidemiologic studies of male reproductive disorders using measurements of prenatal (or newborn) DEHP or DiNP levels have been of limited size and are contradictory; of the 3 studies that have investigated cryptorchidism and hypospadias, 1 reported a borderline significant positive association between the sum of DEHP metabolites and cryptorchidism,14 whereas the other 2 observed no association.15,16 www.epidem.com | 91
Epidemiology • Volume 26, Number 1, January 2015
Jensen et al
The aim of this study was to investigate potential effects of prenatal DEHP and DiNP metabolite levels on the occurrence of cryptorchidism and hypospadias in a large human population. We measured the phthalate metabolites in amniotic fluid from about pregnancy week 16 to obtain contaminant levels from a fetal compartment close to the hypothesized period (weeks 8–14) critical to fetal testis development.17,18 Second, we targeted in vivo endocrine effects of phthalate levels on human fetal Leydig cells by studying amniotic fluid insulin-like factor 3 and steroid hormones, including testosterone and androstenedione.19,20
METHODS Study Population and Amniotic Fluid Samples The study population has previously been described in detail.13 Briefly, we used amniotic fluid samples from a Danish biobank maintained at the State Serum Institute in Copenhagen. The biobank holds samples from a pregnancy screening registry, including both amniotic fluid and maternal serum samples from more than 100,000 pregnancies covering the period 1979 through 2004 (amniotic fluid samples 1980–1996). The amniotic fluid samples were centrifuged before routine diagnostic analyses and the supernatants were kept frozen at –20°C until the present analyses were carried out. The main indication for amniocentesis was age ≥35 years, but some samples were from women with increased risk of severe malformations or Down syndrome based on results from maternal serum analyses. We calculated the gestational week of amniocentesis based on the date of amniocentesis, the date of birth, and the estimated gestational age at birth, as previously described.13 Each sample in the pregnancy screening registry was identified by the personal identification number (unique to each Danish citizen) of the pregnant woman. We used these unique identifiers to obtain obstetric data on the pregnancies from the Danish Medical Birth Registry,21 including gestational age at birth, singleton or multiple birth, maternal parity, and birth weight and Apgar score of the infant. In addition, we used the unique identifiers in the Danish Civil Registration System to identify male offspring from the pregnancies.22 The Danish Regional Ethics Committee, the Danish National Board of Health, and the Danish Data Protection Agency approved the study. The use of the biobank for research purposes has been approved, and additional informed consent from the study subjects for this specific project was neither recommended nor required.
Case–control Definitions and Ascertainment Controls were randomly selected from 25,105 amniotic fluid samples belonging to singleton live-born male offspring pregnancies in the screening database, with complete obstetric data in the Danish Medical Birth Registry. The number of controls (n = 412) was chosen to roughly equal the largest case group, which consisted of 404 boys with cryptorchidism. Case ascertainment was possible using the unique identifiers 92 | www.epidem.com
of the male offspring in the Danish National Patient Registry.23 Cryptorchidism cases had both a diagnosis of undescended testis according to the International Classification of Diseases (ICD8: 75210, 75211, 75219; ICD10: Q53, Q531[A], Q532[A], Q539) and a corrective surgical procedure according to the Surgery and Treatment Classification of the Danish National Board of Health (STC: 55600, 55640) or the Nordic Classification of Surgical Procedures (NCSP: KKFH00, KKFH01, KKFH10) recorded in the Danish National Patient Registry. Boys with a registry entry of inguinal hernia repair (STC: 40620, 40640; NCSP: KJAB00-KJAB97) were excluded from this case group to avoid iatrogenic cryptorchidism secondary to hernia repair. We included all boys who fulfilled these criteria (404 of 25,105; 1.61%) to maximize the cryptorchidism case group and overall study size. For the case group of hypospadias (abnormal opening of the urethral meatus on the ventral penis or perineum), we included all 109 of the 25,105 boys (0.43%) with a diagnosis of hypospadias in the Danish National Patient Registry (ICD8: 75220, 75221, 75222, 75228, 75229; ICD10: Q540, Q541, Q542, Q548, Q549). All boys were followed for the aforementioned diagnoses and surgery entries in the Danish National Patient Registry from birth until November 2008.
Chemical Assays During all chemical analyses, laboratory technicians were blinded to case–control status, and to levels of analytes from other laboratories. We assayed environmental pollutants including phthalate metabolites and detected the DEHP metabolite mono(2-ethyl-5-carboxypentyl) phthalate (5cxMEPP), the DiNP metabolite mono(4-methyl-7-carboxyheptyl) phthalate (7cx-MMeHP), and cotinine, as described in detail elsewhere.13 The assay did not include monoester phthalate metabolites, and additional DEHP metabolites as well as DiNP metabolites were assayed but not detected.13 We use the carboxylated phthalate metabolites because they are the most abundant phthalate metabolites in amniotic fluid and can therefore be detected in a large fraction of the analyzed samples,13 and because they are considered unaffected by external contamination. Briefly, coefficients of variation (analyzing duplicate samples) were 16% for 5cx-MEPP, 12% for 7cx-MMeHP, and 9% for cotinine; the limits of detection were 0.05 ng/ml for 5cx-MEPP, 0.02 ng/ml for 7cx-MMeHP, and 0.20 ng/ml for cotinine. We analyzed the samples using a liquid chromatograph (model UFLCXR, Shimadzu Corp., Kyoto, Japan) connected to a hybrid triple quadrupole linear ion trap tandem mass spectrometer equipped with a turbo ion spray source (QTRAP 5500; AB Sciex, Foster City, CA). We assayed the steroid hormones testosterone, androstenedione, progesterone, 17-OH-progesterone, and cortisol in amniotic fluid at the Danish Center for Neonatal Screening, the State Serum Institute, Copenhagen, Denmark. We used a volume of 50 μl amniotic fluid for the assay. The limit of detection and the intra-assay and interassay coefficients of variation for each of the steroid hormones were as follows: © 2014 Lippincott Williams & Wilkins
Epidemiology • Volume 26, Number 1, January 2015
for testosterone, 0.1 nmol/l, 9% and 10%; androstenedione, 0.3 nmol/l, 8% and 9%; progesterone, 0.4 nmol/l, 10% and 11%; 17-OH-progesterone, 0.4 nmol/l, 9% and 10%; and cortisol, 4.4 nmol/l, 10% and 12%. Further technical details of the applied liquid chromatography/mass spectrometry system are available in the supplemental material (see eAppendix 1, http://links.lww.com/EDE/A843). Insulin-like factor 3 in amniotic fluid was measured using a semicompetitive time-resolved fluorometric immunoassay slightly modified from that described in detail by AnandIvell and colleagues.24 The laboratory work was performed at Leibniz Institute for Farm Animal Biology, Dummerstorf, Germany. The modifications involved the replacement of the original rat antiserum by a new polyclonal antiserum (no. #RIA5) raised in rabbits (IMVS antibody services, Adelaide, Australia) against the same chemically synthesized human insulinlike factor 3 as previously used, at a final dilution of 1:10,000. As tracer, we used the same Europium-labeled human insulinlike factor 3 as described.24 Accordingly, also the secondary goat-anti-rabbit antibody (Rockland Immunochemicals Inc., Gilbertsville, PA) used to coat the plates substituted for the original anti-rat antibody. All other conditions were similar. Standard curves for measurement in amniotic fluid were constructed using serial dilutions of human insulin-like factor 3 in ethylenediaminetetraacetic acid phosphate-buffered saline.25 A volume of 100 μl amniotic fluid was used for the assay, and the limit of detection for the modified assay was 0.01 ng/ml, with interplate and intraplate coefficients of variation of
Amniotic Fluid Phthalate Levels and Male Fetal Gonad Function Morten Søndergaard Jensen,a,b Ravinder Anand-Ivell,c Bent Nørgaard-Pedersen,d Bo A. G. Jönsson,e Jens Peter Bonde,f David M. Hougaard,d Arieh Cohen,d Christian H. Lindh,e Richard Ivell,g and Gunnar Tofth Background: Prenatal exposure to phthalates may pose a threat to human male reproduction. However, additional knowledge about the in vivo effect in humans is needed, and reported associations with genital abnormalities are inconclusive. We aimed to study prenatal di(2-ethylhexyl) phthalate (DEHP) and diisononyl phthalate (DiNP) exposure in relation to cryptorchidism, hypospadias, and human fetal Leydig cell function. Methods: We studied 270 cryptorchidism cases, 75 hypospadias cases, and 300 controls. Second-trimester amniotic fluid samples were available from a Danish pregnancy-screening biobank (n = 25,105) covering 1980–1996. We assayed metabolites of DEHP and DiNP (n = 645) and steroid hormones (n = 545) by mass spectrometry. We assayed insulin-like factor 3 by immunoassay (n = 475) and analyzed data using linear or logistic regression. Results: Mono(2-ethyl-5-carboxypentyl) phthalate (5cx-MEPP, DEHP metabolite) was not consistently associated with cryptorchidism or hypospadias. However, we observed an 18% higher (95% confidence interval [CI] = 5%–33%) testosterone level, and a 41% lower (−56% to −21%) insulin-like factor 3 level in the highest 5cx-MEPP tertile compared with the lowest. Mono(4methyl-7-carboxyheptyl) phthalate (7cx-MMeHP, DiNP metabolite) showed elevated odds ratio point estimates for having
Submitted 07 April 2014; accepted 15 July 2014; posted 7 November 2014. From the aDepartment of Pediatrics, Regional Hospital of Randers; bPerinatal Epidemiology Research Unit, Department of Pediatrics, Aarhus University Hospital, Skejby, Denmark; cSchool of Bioscience, University of Nottingham, United Kingdom; dDanish Center for Neonatal Screening, Department of Clinical Biochemistry and Immunology, Statens Serum Institute, Copenhagen, Denmark; eDivision of Occupational and Environmental Medicine, Department of Laboratory Medicine, Lund University, Lund, Sweden; fDepartment of Occupational and Environmental Medicine, Bispebjerg Hospital, University of Copenhagen, Copenhagen, Denmark; g Reproductive Endocrinology, Leibniz Institute for Farm Animal Biology, Dummerstorf, Germany; and hDanish Ramazzini Center, Department of Occupational Medicine, Aarhus University Hospital, Aarhus, Denmark. M.S.J. and R.A.-I. are joint first authors Supplemental digital content is available through direct URL citations in the HTML and PDF versions of this article (www.epidem.com). This content is not peer-reviewed or copy-edited; it is the sole responsibility of the authors. Correspondence: Morten Søndergaard Jensen, Department of Pediatrics, Regional Hospital of Randers, Skovlyvej 1, DK-8930, Randers NØ, Denmark. E-mail: [email protected]. Copyright © 2014 by Lippincott Williams & Wilkins ISSN: 1044-3983/15/2601-0091 DOI: 10.1097/EDE.0000000000000198
Epidemiology • Volume 26, Number 1, January 2015
cryptorchidism (odds ratio = 1.28 [95% CI = 0.80 to 2.01]) and hypospadias (1.69 [0.78 to 3.67]), but was not consistently associated with the steroid hormones or insulin-like factor 3. Conclusions: Data on the DEHP metabolite indicate possible interference with human male fetal gonadal function. Considering the DiNP metabolite, we cannot exclude (nor statistically confirm) an association with hypospadias and, less strongly, with cryptorchidism. (Epidemiology 2015;26: 91–99)
F
or more than a decade the prevailing view in the literature has been that certain phthalates may induce lack of masculinization and male reproductive disorders by reducing fetal testicular testosterone or insulin-like factor 3 production.1,2 Such views have been fueled by abundant experimental data from studies on rats.3 Most recently, however, studies using human ex vivo fetal testis xenograft methodology have questioned the hypothesized effects on testosterone production in humans,4,5 highlighting the important issue of possible species-specificity.6,7 The potential hazards posed to male reproduction during fetal life by phthalates must be fully characterized considering the virtually inevitable exposure. Phthalates such as di(2-ethylhexyl) phthalate (DEHP) and diisononyl phthalate (DiNP) are high-volume commercial chemicals used for production of certain building materials, car interiors, clothing, food packaging, toys, and medical devices.8 Biomonitoring has documented their pervasive nature by detecting metabolites in samples from pregnant women, cord blood, and amniotic fluid.9–13 Given this exposure scenario, knowledge about mechanisms of action in humans and high-quality epidemiologic studies of reproductive outcomes are indispensable. However, as mentioned, our mechanistic insight may be hampered by species-specific mechanisms and effects. Moreover, the few existing epidemiologic studies of male reproductive disorders using measurements of prenatal (or newborn) DEHP or DiNP levels have been of limited size and are contradictory; of the 3 studies that have investigated cryptorchidism and hypospadias, 1 reported a borderline significant positive association between the sum of DEHP metabolites and cryptorchidism,14 whereas the other 2 observed no association.15,16 www.epidem.com | 91
Epidemiology • Volume 26, Number 1, January 2015
Jensen et al
The aim of this study was to investigate potential effects of prenatal DEHP and DiNP metabolite levels on the occurrence of cryptorchidism and hypospadias in a large human population. We measured the phthalate metabolites in amniotic fluid from about pregnancy week 16 to obtain contaminant levels from a fetal compartment close to the hypothesized period (weeks 8–14) critical to fetal testis development.17,18 Second, we targeted in vivo endocrine effects of phthalate levels on human fetal Leydig cells by studying amniotic fluid insulin-like factor 3 and steroid hormones, including testosterone and androstenedione.19,20
METHODS Study Population and Amniotic Fluid Samples The study population has previously been described in detail.13 Briefly, we used amniotic fluid samples from a Danish biobank maintained at the State Serum Institute in Copenhagen. The biobank holds samples from a pregnancy screening registry, including both amniotic fluid and maternal serum samples from more than 100,000 pregnancies covering the period 1979 through 2004 (amniotic fluid samples 1980–1996). The amniotic fluid samples were centrifuged before routine diagnostic analyses and the supernatants were kept frozen at –20°C until the present analyses were carried out. The main indication for amniocentesis was age ≥35 years, but some samples were from women with increased risk of severe malformations or Down syndrome based on results from maternal serum analyses. We calculated the gestational week of amniocentesis based on the date of amniocentesis, the date of birth, and the estimated gestational age at birth, as previously described.13 Each sample in the pregnancy screening registry was identified by the personal identification number (unique to each Danish citizen) of the pregnant woman. We used these unique identifiers to obtain obstetric data on the pregnancies from the Danish Medical Birth Registry,21 including gestational age at birth, singleton or multiple birth, maternal parity, and birth weight and Apgar score of the infant. In addition, we used the unique identifiers in the Danish Civil Registration System to identify male offspring from the pregnancies.22 The Danish Regional Ethics Committee, the Danish National Board of Health, and the Danish Data Protection Agency approved the study. The use of the biobank for research purposes has been approved, and additional informed consent from the study subjects for this specific project was neither recommended nor required.
Case–control Definitions and Ascertainment Controls were randomly selected from 25,105 amniotic fluid samples belonging to singleton live-born male offspring pregnancies in the screening database, with complete obstetric data in the Danish Medical Birth Registry. The number of controls (n = 412) was chosen to roughly equal the largest case group, which consisted of 404 boys with cryptorchidism. Case ascertainment was possible using the unique identifiers 92 | www.epidem.com
of the male offspring in the Danish National Patient Registry.23 Cryptorchidism cases had both a diagnosis of undescended testis according to the International Classification of Diseases (ICD8: 75210, 75211, 75219; ICD10: Q53, Q531[A], Q532[A], Q539) and a corrective surgical procedure according to the Surgery and Treatment Classification of the Danish National Board of Health (STC: 55600, 55640) or the Nordic Classification of Surgical Procedures (NCSP: KKFH00, KKFH01, KKFH10) recorded in the Danish National Patient Registry. Boys with a registry entry of inguinal hernia repair (STC: 40620, 40640; NCSP: KJAB00-KJAB97) were excluded from this case group to avoid iatrogenic cryptorchidism secondary to hernia repair. We included all boys who fulfilled these criteria (404 of 25,105; 1.61%) to maximize the cryptorchidism case group and overall study size. For the case group of hypospadias (abnormal opening of the urethral meatus on the ventral penis or perineum), we included all 109 of the 25,105 boys (0.43%) with a diagnosis of hypospadias in the Danish National Patient Registry (ICD8: 75220, 75221, 75222, 75228, 75229; ICD10: Q540, Q541, Q542, Q548, Q549). All boys were followed for the aforementioned diagnoses and surgery entries in the Danish National Patient Registry from birth until November 2008.
Chemical Assays During all chemical analyses, laboratory technicians were blinded to case–control status, and to levels of analytes from other laboratories. We assayed environmental pollutants including phthalate metabolites and detected the DEHP metabolite mono(2-ethyl-5-carboxypentyl) phthalate (5cxMEPP), the DiNP metabolite mono(4-methyl-7-carboxyheptyl) phthalate (7cx-MMeHP), and cotinine, as described in detail elsewhere.13 The assay did not include monoester phthalate metabolites, and additional DEHP metabolites as well as DiNP metabolites were assayed but not detected.13 We use the carboxylated phthalate metabolites because they are the most abundant phthalate metabolites in amniotic fluid and can therefore be detected in a large fraction of the analyzed samples,13 and because they are considered unaffected by external contamination. Briefly, coefficients of variation (analyzing duplicate samples) were 16% for 5cx-MEPP, 12% for 7cx-MMeHP, and 9% for cotinine; the limits of detection were 0.05 ng/ml for 5cx-MEPP, 0.02 ng/ml for 7cx-MMeHP, and 0.20 ng/ml for cotinine. We analyzed the samples using a liquid chromatograph (model UFLCXR, Shimadzu Corp., Kyoto, Japan) connected to a hybrid triple quadrupole linear ion trap tandem mass spectrometer equipped with a turbo ion spray source (QTRAP 5500; AB Sciex, Foster City, CA). We assayed the steroid hormones testosterone, androstenedione, progesterone, 17-OH-progesterone, and cortisol in amniotic fluid at the Danish Center for Neonatal Screening, the State Serum Institute, Copenhagen, Denmark. We used a volume of 50 μl amniotic fluid for the assay. The limit of detection and the intra-assay and interassay coefficients of variation for each of the steroid hormones were as follows: © 2014 Lippincott Williams & Wilkins
Epidemiology • Volume 26, Number 1, January 2015
for testosterone, 0.1 nmol/l, 9% and 10%; androstenedione, 0.3 nmol/l, 8% and 9%; progesterone, 0.4 nmol/l, 10% and 11%; 17-OH-progesterone, 0.4 nmol/l, 9% and 10%; and cortisol, 4.4 nmol/l, 10% and 12%. Further technical details of the applied liquid chromatography/mass spectrometry system are available in the supplemental material (see eAppendix 1, http://links.lww.com/EDE/A843). Insulin-like factor 3 in amniotic fluid was measured using a semicompetitive time-resolved fluorometric immunoassay slightly modified from that described in detail by AnandIvell and colleagues.24 The laboratory work was performed at Leibniz Institute for Farm Animal Biology, Dummerstorf, Germany. The modifications involved the replacement of the original rat antiserum by a new polyclonal antiserum (no. #RIA5) raised in rabbits (IMVS antibody services, Adelaide, Australia) against the same chemically synthesized human insulinlike factor 3 as previously used, at a final dilution of 1:10,000. As tracer, we used the same Europium-labeled human insulinlike factor 3 as described.24 Accordingly, also the secondary goat-anti-rabbit antibody (Rockland Immunochemicals Inc., Gilbertsville, PA) used to coat the plates substituted for the original anti-rat antibody. All other conditions were similar. Standard curves for measurement in amniotic fluid were constructed using serial dilutions of human insulin-like factor 3 in ethylenediaminetetraacetic acid phosphate-buffered saline.25 A volume of 100 μl amniotic fluid was used for the assay, and the limit of detection for the modified assay was 0.01 ng/ml, with interplate and intraplate coefficients of variation of
