The risk of endometriosis and exposure to dioxins and polychlorinated biphenyls: a case–control study of infertile women

A. Pauwels1,7, P.J.C. Schepens1, T. D'Hooghe2, L. Delbeke3, M. Dhont4, A. Brouwer5 and J. Weyler6

1 Antwerp University, Toxicological Centre, Universiteitsplein 1, B-2610 Wilrijk, 2 University Hospital Gasthuisberg, Leuven University Fertility Centre, Herestraat 49, B-3000 Leuven, 3 University Hospital of Antwerp, Centre for Reproductive Medicine, Wilrijkstraat 10, B-2650 Edegem, 4 University Hospital of Ghent, Centre for Infertility, De Pintelaan 185, B-9000 Ghent, Belgium; 5 Institute for Environmental Studies, Free University of Amsterdam, De Boelelaan 1115, 1081 HV Amsterdam, The Netherlands and 6 Antwerp University, Epidemiology/Community Medicine, Universiteitsplein 1, B-2610 Wilrijk, Belgium


    Abstract
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 Acknowledgements
 References
 
BACKGROUND: A case–control study was designed to determine the possible association between chronic exposure to dioxins and polychlorinated biphenyls (PCBs), and the occurrence of endometriosis. The study group consisted of 42 infertile endometriosis cases and 27 mechanical infertile controls, both groups attending one of the collaborating Centres for Reproductive Medicine, enrolled between 1996–1998. METHODS: Exposure assessment to dioxin-like compounds was determined through CALUX (chemical-activated luciferase gene expression)-bioassay to measure dioxin-like total toxic equivalents (dioxins and co-planar PCBs), whereas non-co-planar PCBs were determined through chemical analysis. RESULTS: No association was found between median dioxin-like total toxic equivalents (TEQ) and the occurrence of endometriosis in infertile women [cases (n = 34): 29; controls (n = 27): 24; NS]. When patients were subdivided based on an arbitrary cut-off value of 100 pg TEQ/g serum lipids, no statistically significant association between very high exposure to dioxin-like compounds and endometriosis was found [crude odds ratio (OR) = 4.33; confidence interval (CI) 0.49–38.19; NS]. After adjusting for body mass index, and alcohol consumption, the risk increased slightly to OR = 4.6 (CI 0.48–43.62; NS). There was no confounding by age, ovulatory dysfunction, caffeine intake, smoking or exposure to non-co-planar PCBs. CONCLUSIONS: The study results showed no statistically significant association between exposure to dioxin-like compounds and the occurrence of endometriosis in infertile women.

Key words: CALUX-bioassay/dioxin-like compounds/endometriosis/infertility/polychlorinated biphenyls


    Introduction
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 Acknowledgements
 References
 
Endometriosis is an enigmatic gynaecological disorder that is characterized by the ectopic presence of both endometrial glands and stroma (Olive and Schwartz, 1993Go). The disease is thought to occur through the transtubal spreading and implantation of menstrual endometrium, and is associated with pelvic pain, dyspareunia and infertility. Among the environmental pollutants that have been suggested as being linked to endometriosis are the polyhalogenated aromatic hydrocarbons (PHAH), a class of widespread environmental contaminants which includes polychlorinated dibenzo-p- dioxins (PCDDs), dibenzofurans (PCDFs) and biphenyls (PCBs) (Rier et al., 1993Go; Lebel et al., 1998Go). Because the aetiology of endometriosis seems to be multifactorial, it has been suggested that dioxin exposure may contribute to an imbalance of sex hormones or alter growth factors and the immune response (Mayani et al., 1997Go; Osteen and Sierra-Rivera, 1997Go). Dioxins alter tissue-specific responses to hormones via modulation of steroid receptor expression (Safe et al., 1991Go). Along with inhibition of T-lymphocyte function (Neubert et al., 1991Go) and decreasing natural killer cell activity in plasma and peritoneal fluid, dioxins may stimulate peritoneal fluid macrophages and thus affect angiogenesis and local concentrations of cytokines (e.g. interleukin-1) and growth factors (Clark et al., 1991Go; Koninckx, 1999Go). Alternatively, cellular changes or genetic background may predispose an individual to the immunological modulation caused by dioxin exposure, leading to infiltration and adhesion of endometrial cells in the peritoneum (Koninckx, 1999Go).

Extensive experimental studies have pointed out that most toxic actions induced by 2,3,7,8-tetrachloro-p-dibenzodioxin (TCDD) are mediated via the arylhydrocarbon receptor (AhR) (Safe, 1990Go). Six other PCDDs, 10 PCDFs, and 12 PCB congeners (non-ortho, and to a lesser extent some mono-ortho substituted congeners) can also assume a co-planar configuration. Hence, they also interact with the AhR, and they produce the same spectrum of responses in animal and cell models as TCDD, depending on their binding affinity to the AhR (Van den Berg et al., 1998Go).

PCB congeners possessing two or more chlorine substituents at the ortho positions of the biphenyl rings are non-co-planar, and do not bind with high affinity to the AhR (Battershill, 1994Go). Exposure to these non-co-planar PCB congeners reportedly results in a variety of toxic effects in experimental animals, including neurochemical, neurotoxic, carcinogenic, and endocrine changes (Safe, 1990Go; Brouwer et al., 1995Go). However, the spectrum of activity produced by the latter congeners has not been fully explored and the mechanisms of action remain to be fully elucidated (Fischer et al., 1998Go).

It has been reported recently that in Belgium the incidence and severity of endometriosis in women, as well as the degree of dioxin pollution, is among the highest in the world (Koninckx et al., 1994Go). In view of the accumulating data, we carried out a case–control study aimed to assess whether dioxin-like toxic equivalents in serum are related to endometriosis within a female infertile population.


    Materials and methods
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 Acknowledgements
 References
 
Patients
This prospective case–control study recruited patients undergoing infertility treatment at one of the collaborating Centres for Reproductive Medicine spread over Belgium (University Hospitals of Antwerp, Ghent, and Leuven), enrolled between 1996 and 1998. All women underwent a laparoscopy as part of their infertility work-up. A couple was defined as `infertile' when pregnancy had not been achieved after one year of unprotected sexual intercourse.

The case group consisted of 42 patients with laparoscopy-confirmed endometriosis. Endometriosis was staged as minimal (AFS I) in 21, mild (AFS II) in 7, moderate (AFS III) in 10, and severe (AFS IV) in 4 patients, according to the revised classification of the American Fertility Society (rAFS) (American Fertility Society, 1985Go).

The control group comprised 27 women without endometriosis and with infertility related to tubal disease, tuboperitoneal factors, cervical factors, or uterine factors. In these women no evidence of endometriosis was found at laparoscopy. An association between these causes of female infertility and exposure to PHAHs has not been established.

This study had been accepted by all Ethical Committees (Protocol nos. 96/44/107, 97/100, and ML 536, for UZA, UZG, and UZL respectively). All patients acknowledged their participation by signing an Informed Consent Form.

Medical records and interviews
Medical records of the subjects were reviewed to obtain information on diagnosis (tubal, tuboperitoneal, cervical, uterine factor, endometriosis, ovulatory dysfunction) and anthropometric variables (age, body mass index). A 20-minute telephone interview was conducted, eliciting detailed personal data and documenting factors potentially associated with endometriosis and with PHAH exposure. The enrolment questionnaire was designed primarily to obtain information on the patient's eating, drinking and smoking pattern.

During the interview all subjects were asked `what is your average weekly consumption of glasses of alcohol (including wine, beer, liquors)'. Women consuming >=6 glasses/week were considered to be alcohol drinkers.

We estimated each person's total daily caffeine intake by assuming that there was 107 mg/cup in coffee of unknown preparation method, 34 mg/cup in tea (Pastore and Savitz, 1995Go), and 26 mg/glass (200 ml) in regular or light cola. Furthermore, the patient's current and former smoking pattern was expressed in cumulative dose of exposure, in terms of packages of cigarettes per year.

Peripheral blood analysis
During the preoperative interview, 25 ml of blood per patient was collected in a vacuum system tube, transported in a cooling pail, and centrifuged (15 min, 2000 g) within 24 h after collection. All serum samples were stored at –20°C until analysed.

Lipid determination
Serum cholesterol (free and cholesterol esters), triglycerides, and phospholipids were determined in duplicate as a measurement of lipid content, using enzymatic spectrophotometric determination with commercially available reagents from Elitech Diagnostics (Sées, Normandie, France).

Rationale for the analytical methods used to quantify dioxins and polychlorinated biphenyls
Dioxins and co-planar polychlorinated biphenyls
The TCDD toxic equivalency factor (TEF) concept (Safe, 1990Go), allows conversion of a PHAH chemical data set into the AhR-related toxic potency of a mixture of PHAHs. Concentrations of individual PHAHs are multiplied by their respective TEF-values and added together to give the total TCDD toxic equivalency (TEQ) value. In recent years, bioassays have been developed that can measure the total TEQ-value of complex mixtures directly without the need for extensive clean-up and chemical analysis procedures. One of the novel in-vitro reporter gene assays, the CALUX (chemical-activated luciferase gene expression)-bioassay is based on AhR-mediated firefly (Photinus pyralis) luciferase expression in genetically modified cell lines (Aarts et al., 1995Go).

Based on the CALUX-derived TEQ-levels, the impact of co-planar PCBs on endometriosis can be interpreted, since it has been shown recently that the contribution of dioxin-like PCBs to the total TEQ-value is almost equal in human matrices to the contribution of dioxins and furans (Brouwer et al., 1995Go). The most recently updated TEF-values of dioxin-like PCBs are compiled in Table IGo (Van den Berg et al., 1998Go).


View this table:
[in this window]
[in a new window]
 
Table I. World Health Organization toxic equivalency factors (TEFs) for dioxin-like PCBs for humans (Van den Berg et al., 1998; with permission)
 
Non-co-planar polychorinated biphenyls
Combining the a priori environmental exposure to a mixture of (non)-dioxin like PCBs, and the correlation between CALUX-based TEQ-values and the sum of non-co-planar PCBs (Pauwels et al., 2000Go), the association between TEQ-levels and endometriosis may be feasibly confounded by the effects of non-co-planar PCB congeners. Therefore, four non-co-planar PCB congeners (IUPAC Nos. 118, 138, 153, 180) were determined by chemical analysis through gas chromatography using electron-capture detection (GC-ECD).

CALUX-bioassay
The CALUX-bioassay analysis was performed as described in detail (Murk et al., 1997Go). The extraction and clean-up method involves essentially n-hexane extraction of blood serum aliquots (1–1.5 ml) and removal of acid-labile matrix components by passage through a silica column containing 33% (w/w) concentrated H2SO4. This extract was diluted in dimethylsulphoxide (DMSO) for CALUX measurement using rat H4IIE hepatoma (H4L1.1c4) cells. These cells stably transfected with an AhR-controlled luciferase reporter gene construct (pGudluc1.1) and were grown confluent in 96-well view plates and exposed in triplicate to the PHAH samples and TCDD standards for 24 h, using DMSO (0.5% v/v) as a vehicle.

After removal of the medium, cells were washed twice with phosphate-buffered saline (Oxoid, Hampshire, UK). The cells were harvested in 75 µl cell lysis reagent (Luciferase Assay System; Promega, Leiden, The Netherlands), and centrifuged at 2000 g for 1 min. For measurements of luciferase activity, 20-µl aliquots of the supernatants were pipetted into a 96-well microtitre plate, 100 µl luciferin assay mix at room temperature was added. After thorough mixing, the light production was measured in a Luminoskan RS Luminometer (Labsystems, Helsinki, Finland).

The lipid-corrected CALUX-based TEQs were calculated by comparison of the luciferase activity induced by the sample (three replicates) against a dose-response curve generated from TCDD concentration standards simultaneously analysed. The standard curve was fitted (one-site ligand fit) using SlideWrite 6.00 (Advanced Graphics Software Inc., Encinitas, CA, USA), and the CALUX TEQ-value of an unknown sample was interpolated on this curve.

Chemical analysis of non-co-planar PCBs
All solvents used were purchased pesticide grade from E.Merck (Darmstadt, Germany). The four non-co-planar PCBs were a standard mixture from J.T.Baker at a concentration of 10 ng/µl in iso-octane (J.T.Baker, Deventer, The Netherlands). The 13C12-labelled CB-149 (internal standard) was obtained from Cambridge Isotope Laboratories (Woburn, MA, USA).

Complete details of the extraction were described and evaluated previously (Pauwels et al., 1999Go). A brief description of the method is given below. The sample preparation involved addition of 13C12 CB-149, disruption of protein-binding in an ultrasonic bath using formic acid (1:1, v/v), extraction and concentration of analytes using EmporeTM C18 SPE disk cartridges (3M, St Paul, MN, USA). A mixture of ethyl acetate and hexane was used for elution. Further clean-up of lipid interferences was accomplished using a sulphuric acid wash of the eluate. Two µl per extract were injected in splitless mode on a GC-µECD system (HP 6890, Palo Alto, CA, USA) by an autosampler (HP 7673). A fused silica DB-XLB capillary column (J&W Scientific, Folsom, CA, USA) of 60 mx0.25 mm i.d.x0.25 µm film thickness was used. Details of the analytical conditions receives ample treatment elsewhere (Pauwels et al., 1999Go).

Data analysis
A normal distribution could not be assumed for any of the variables (Shapiro-Wilk's W-test). Therefore, characteristics of cases and controls were compared using the Mann-Whitney U-test for continuous data, and by {chi}2 or Fisher's exact test (when appropriate) for categorical data. Crude associations between exposure to dioxins (after dichotomization), PCB congeners (on a continuous scale), and endometriosis were estimated by logistic regression analysis. The initial selection of potential confounders was guided by the literature on known and suspected determinants of endometriosis (Eskenazi and Warner, 1997Go; Zeyneloglu et al., 1997Go). Those factors included age, body mass index, ovulatory dysfunction, smoking pattern, alcohol and caffeine consumption (Table IIGo). Multiple logistic regression correcting for selected confounders was performed in order to assess the adjusted association between exposure to dioxins, PCBs and endometriosis [STATISTICA version 5, 1997 (StatSoft, Groningen, The Netherlands); EGRET version 0.03, 1991 (Cytel Software Corp., Cambridge, MA, USA)]. Significant associations are indicated by P < 0.05.


View this table:
[in this window]
[in a new window]
 
Table II. Relevant characteristics of the endometriosis cases and mechanical infertile controls, 1996–1998
 

    Results
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 Acknowledgements
 References
 
Table IIGo represents relevant characteristics of the study population. As can be observed, the cases (n = 42) and the controls (n = 27) were similar with respect to age, ovulatory dysfunction, alcohol and caffeine intake, and smoking pattern. A summary of TEQ-values and non-co-planar PCB congener levels is presented in Table IIIGo. Although these figures suggest a trend toward slightly higher exposure for endometriosis patients (median 29 pg TEQ/g lipid) compared with their mechanical infertility counterparts (median 27 pg TEQ/g lipid), none of these differences were statistically significant. For two cases and two controls, no serum was available for PCB-analysis. Consequently, non-co-planar PCB congeners are determined in 40 cases and 25 controls. Endometriosis and mechanical infertile patients are exposed to non-co-planar PCBs in the same range.


View this table:
[in this window]
[in a new window]
 
Table III. CALUX-based TEQ-values and non-co-planar PCB-concentrations in serum of women with endometriosis (cases) and mechanical infertility (controls)
 
The distribution of TEQ-levels in cases and controls is presented in Figure 1Go. These data show clearly a group in the endometriosis-cases with high TEQ-levels. Consequently, statistical analyses on CALUX-based TEQ-values were executed using only two exposure categories. We thereby chose one cut-off point (100 pg TEQ/g serum lipids) to divide the groups into a non-exposed and an exposed group. After dichotomization, the association between exposure to dioxin-toxic equivalents and occurrence of endometriosis in infertile women was non-significant (OR = 4.33; CI = 0.49–38.19; NS). After adjustment for body mass index, and alcohol consumption, the risk increased slightly (OR = 4.56; CI 0.48–43.62; NS).



View larger version (51K):
[in this window]
[in a new window]
 
Figure 1. Distribution of TEQ-levels in cases and controls. All data are expressed in pg TEQ/g serum lipids.

 
Statistical analyses for PCB congeners were executed on a continuous scale. The regression coefficients and corresponding odds ratios did not reveal a significant association between exposure to the four individual congeners and the occurrence of endometriosis. In multiple logistic regression, the association between endometriosis and PCB-138, 153 and 180 even declined (data not shown). The relation of the latter congeners with endometriosis was partly explained by PCB-118, the only congener with a dioxin-like toxic capacity (TEF = 0.0001).

For the different associations, confounding was hardly present. For PCB-118 for instance, the (logistic) regression coefficient changed from 0.027 (crude) to 0.029 in the best fitting model, adjusting for body mass index and caffeine intake (P = not significant).


    Discussion
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 Acknowledgements
 References
 
The association between endometriosis and exposure to dioxins is a highly controversial issue. In non-human primates, 2,3,7,8-tetrachlorodibenzo-p-dioxin (TCDD) has been reported to induce a dose-dependent increase in severity of endometriosis (Rier et al., 1993Go), to facilitate the survival of endometrial implants, and to exert a bimodal effect on endometrial implant growth (Yang et al., 2000Go). A series of human case–control studies have been conducted on the association between endometriosis and PHAH exposure with important differences in study design, patient selection and assay methods, and therefore provide inconsistent results. An association between endometriosis and organochlorine pollutants was confirmed in two reports (Gerhard and Runnebaum, 1992Go; Mayani et al., 1997Go), but refuted in one other publication (Lebel et al., 1998Go).

In several ways, the well designed protocol of our study overcame several shortcomings of the previous publications. Firstly, in our study, the patient population was very well defined: both cases and controls were infertile women undergoing a laparoscopic investigation. In only one other publication (Mayani et al., 1997; 44 cases and 35 controls), the control group consisted strictly of patients with mechanical infertility, whereby the absence of endometriosis was confirmed by laparoscopy. Another report evaluated fertile and infertile women undergoing laparoscopy for pain, infertility or tubal sterilization whereby the controls were indication-matched (Lebel et al., 1998; 68 cases and 70 controls). In one study (Gerhard and Runnebaum, 1992; 24 cases and 484 controls), hormonal disorders were present in both cases and controls, and it was not mentioned whether endometriosis was laparoscopically and histologically confirmed.

Secondly, our study offers a unique combination of measurement of exposure of dioxin-like compounds (dioxins and co-planar PCBs) and non-co-planar PCBs. Dioxin-like compounds were determined using CALUX-bioassay, which has recently been adapted and validated in blood plasma (Murk et al., 1997Go) and bovine milk (Bovee et al., 1998Go), whereas non-co-planar PCBs were determined through chemical analysis. In our study, the association between exposure to dioxin-like compounds and endometriosis seemed not to be confounded by the toxic effects of non-co-planar PCBs, since no strong or consistent relationship between the exposure to non-co-planar PCBs and the occurrence of endometriosis could be observed. In the other published studies, dioxin-like compounds were either not measured (Gerhard and Runnebaum, 1992Go; Lebel et al., 1998Go) or only TCDD was quantified through chemical analysis (Mayani et al., 1997Go).

Thirdly, studies that evaluate the association between exposure to PHAHs and endometriosis are likely to be biased by confounding variables, since it is not possible to conduct a randomized controlled trial on this exposure. Adjustment for these variables (age, body mass index, ovulatory dysfunction, smoking pattern, alcohol and caffeine consumption) was performed in our study but not or only partially in the previously published reports (Gerhard and Runnebaum, 1992Go; Mayani et al., 1997Go; Lebel et al., 1998Go).

Fourthly, the validity of our study results should be considered. To reduce the likelihood of selection bias, the inclusion criterion was a laparoscopy. Thereafter, subjects were defined as cases or controls in this survey. Hence, controls derived from the same recruitment population as the case group.

Fifthly, an important problem that might arise when dealing with case–control studies, is (differential) misclassification of exposure. In order to avoid this, the CALUX-bioassay was done blindly. Similarly, the interview was performed by one single staff member, who was unaware of the woman's case or control status. Recall bias can be ruled out because both cases and controls are both infertile patients. Variations in the recording of operative data cannot be excluded entirely. However, all laparoscopies were performed by gynaecologists familiar with the AFS classification for endometriosis (American Fertility Society, 1985Go).

Because information on alcohol exposure history was obtained by self-report, misclassification of use may have occurred. As we are convinced that these errors are the same for cases and controls, we believe that the misclassification is of the non-differential type, leading to biases towards the `no association' result. Therefore, we are confident that our finding is not a result of information bias.

This pilot study is focused on the relationship between exposure to dioxins and PCBs and the occurrence of endometriosis in infertile females. We plan to extend it using larger populations, and by the inclusion of fertile patients with and without endometriosis. In this preliminary study, the lack of a statistically significant association between exposure to dioxin-like compounds and endometriosis based on an arbitrary cut-off value, can be explained either by the lack of a real association, or by the lack of power to prove such an association. In order to detect a four-fold increase (OR = 4) with a power of 90% and a significance level of 0.05, 85 cases and 85 controls are required. By increasing those figures by 10% (for multiple regression), a sample size of 100 endometriosis patients and 100 mechanically infertile controls is acquired. Therefore, future epidemiological surveys on PHAHs should enrol larger patient samples and biomonitor a well-defined range of potential culprit PHAHs (dioxins, furans, and selected PCB congeners) by chemical analysis, or by bioassays (toxic biological responses), followed by chemical analysis to identify the culprit compounds.


    Acknowledgements
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 Acknowledgements
 References
 
The authors are indebted to P.H. Cenijn for technical help with the CALUX-analyses. The authors gratefully acknowledge the co-operation of the doctors and staff members from the collaborating hospitals for their contribution to the sample collection, and the participating patients for their invaluable contributions.


    Notes
 
7 To whom correspondence should be addressed. Back


    References
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 Acknowledgements
 References
 
American Fertility Society (1985) Revised American Fertility Society classification of endometriosis. Fertil. Steril., 43, 351–352.[Medline]

Aarts, J.M.M.J.G., Denison, M.S., Cox, M.A. et al. (1995) Species-specific antagonism of Ah receptor action by, 2,2',5,5'-tetrachloro- and, 2,2',3,3',4,4'-hexachlorobiphenyl. Eur. J. Pharmacol., 293, 463–774.[Medline]

Battershill, J.M. (1994) Review of the safety assessment of PCBs with particular reference to reproductive toxicity. Hum. Experim. Toxicol., 13, 581–597.

Bovee, T.F.H., Hoogenboom, L.A.P., Hamers, A.R.M. et al. (1998) Validation and use of the CALUX-bioassay for the determination of dioxins and PCBs in bovine milk. Food Addit. Contam., 15, 863–875.[ISI][Medline]

Boyd, J.A., Clark, G.C., Walmer, D.K. et al. (1995) Endometriosis and the environment: Biomarkers of toxin exposure. Conference on endometriosis, 2000, May 15–17.

Brouwer, A., Ahlborg, U.G., Van den Berg, M. et al. (1995) Functional aspects of developmental toxicity of polyhalogenated aromatic hydrocarbons in experimental animals and human infants. Eur. J. Pharmacol., 293, 1–40.[Medline]

Clark, G.C., Taylor, M.J., Tritscher, A.M. et al. (1991) Tumor necrosis factor involvement in, 2,3,7,8-tetrachlorodibenzo-p- dioxin-mediated endotoxin hypersensitivity in C57BL/6J mice congenic at the Ah locus. Toxicol. Appl. Pharmacol., 111, 422–431.[ISI][Medline]

Eskenazi, B. and Warner, L. (1997) Epidemiology of endometriosis. Obstet. Gynecol. Clin. North Am., 24, 235–258.[ISI][Medline]

Fischer, L.J., Seegal, R.F., Gareau, P. et al. (1998) Symposium overview: Toxicity of non-planar PCBs. Toxicol. Sci., 41, 49–61.[Abstract]

Gerhard, I. and Runnebaum, B. (1992) Grenzen der Hormonsubstitution bei Schadstoffbelastung und Fertilitätsstorungen. Zent. Bl. Gynekol., 114, 593–602.

Koninckx, P.R. (1999) The physiopathology of endometriosis: pollution and dioxin. Gynecol. Obstet. Invest., 47 (Suppl. 1), 47–50.[ISI][Medline]

Koninckx, P.R., Braet, P., Kennedy, S.H. et al. (1994) Dioxin pollution and endometriosis in Belgium. Hum. Reprod., 9, 1001–1002.[ISI][Medline]

Lebel, G., Dodin, S., Ayotte, P. et al. (1998) Organochlorine exposure and the risk of endometriosis. Fertil. Steril., 69, 221–228.[ISI][Medline]

Mayani, A., Barel, S., Soback, S. et al. (1997) Dioxin concentrations in women with endometriosis. Hum. Reprod., 12, 373–375.[Abstract]

Murk, A.J., Leonards, P.E.G., Bulder, A.S. et al. (1997) The CALUX assay adapted and validated for measuring TCDD equivalents in blood plasma. Environ. Toxicol. Chem., 16, 1583–1589.[ISI]

Neubert, R., Jacob-Muller, U., Helge, H. et al. (1991) Polyhalogenated dibenzo-p-dioxins and dibenzofurans and the immune system. 2. In vitro effects of, 2,3,7,8-tetrachlorodibenzo-p-dioxin (TCDD) on lymphocytes of venous blood from man and a non-human primate (Callithrix jacchus). Arch. Toxicol., 65, 213–219.[ISI][Medline]

Olive, D.L. and Schwartz, L.B. (1993) Endometriosis. New Engl. J. Med., 328, 1759–1769.[Free Full Text]

Osteen, K.G. and Sierra-Rivera, E. (1997) Does disruption of immune and endocrine systems by environmental toxins contribute to development of endometriosis? Semin. Reprod. Endocrinol., 15, 301–308.

Pastore, L.M. and Savitz, D.A. (1995) Case-control study of caffeinated beverages and preterm delivery. Am. J. Epidemiol., 141, 61–69.[Abstract]

Pauwels, A., Wells, D., Covaci, A. et al. (1999) Improved sample preparation method for selected persistent organochlorine pollutants in human serum cusing solid-phase disk extraction with gas chromatographic analysis. J. Chromatogr. B, 723, 117–125.[ISI]

Pauwels, A., Cenijn, P.H., Schepens, P.J.C. et al. (2000) Comparison of chemical-activated luciferase gene expression (CALUX)-bioassay and gas chromatography for PCB determination in human serum and follicular fluid. Environ. Health Perspect., 108, 553–557.[ISI][Medline]

Rier, S.E., Martin, D.C., Bowman, R.E. et al. (1993) Endometriosis in rhesus monkeys following chronic exposure to, 2,3,7,8-TCDD. Fund. Appl. Toxicol., 21, 433–441.[ISI][Medline]

Safe, S. (1990) PCBs, PCDDs, PCDFs and related compounds: Environmental and mechanistic considerations which support the development of TEFs. Crit. Rev. Toxicol., 21, 51–88.[ISI][Medline]

Safe, S., Astroff, B., Harris, M. et al. (1991), 2,3,7,8-Tetrachlorodibenzo-p-dioxin and related compounds as antiestrogens: Characterization and mechanism of action. Pharmacol. Toxicol., 69, 400–409.[ISI][Medline]

Van den Berg, M., Birnbaum, L.S., Bosveld, A.T.C. et al. (1998) Toxic equivalency factors (TEFs) for PCBs, PCDDs, PCDFs for humans and wildlife. Environ. Health Perspect., 106, 775–792.[ISI][Medline]

Yang, J.Z., Agarwal, S.K. and Foster, W.G. (2000) Subchronic exposure to, 2,3,7,8-tetrachlorodibenzo-p-dioxin modulates the pathophysiology of endometriosis in the cynomolgus monkey. Toxicol. Sci., 56, 374–81.[Abstract/Free Full Text]

Zeyneloglu, H.B., Arici, A. and Olive, D.L. (1997) Environmental toxins and endometriosis. Obstet. Gynecol. Clin. North Am., 24, 307–329.[ISI][Medline]

Submitted on January 23, 2001; accepted on June 15, 2001.