Endometriosis is the third leading cause of gynecologic hospitalization in the United States.1 Despite the high morbidity and health care cost associated with endometriosis, the etiology of endometriosis has not been fully delineated. The pathophysiology likely includes hormonal, anatomical, genetic, and immune factors. Risk may be associated with factors that increase the volume, frequency, and duration of retrograde menstruation and promote implantation and growth of endometrial plaques.2 There is also strong circumstantial evidence that endometriosis is influenced by steroid hormones.3–5
Case-control and cross-sectional studies have been unable to establish a clear temporal relation between menstrual cycle characteristics and incident endometriosis versus symptoms of present disease. Analyses of the relation with reproductive history have been conducted primarily within infertile populations and those undergoing tubal ligation, complicating interpretation of effects. Using data collected from the Nurses' Health Study II, an ongoing, prospective cohort study of premenopausal U.S. nurses that began in 1989, we have attempted to clarify these associations with laparoscopically confirmed endometriosis.
MATERIALS AND METHODS
Data for these analyses were collected in the Nurses' Health Study II cohort from September 1989 to June 1, 1999. Questionnaires requesting information on incident diseases and demographic, biologic, environmental, and lifestyle risk factors are updated and mailed biennially. A total of 116,678 female registered nurses—ranging in age from 25 to 42 years and residing in one of 14 states in the United States—completed the baseline questionnaire. Follow-up of this cohort in each 2-year interval has consistently been 90% or higher. The Institutional Review Board of the Harvard School of Public Health approved this research.
In 1993, the women were first asked if they had “ever had physician-diagnosed endometriosis.” If “yes,” they were asked to report when the diagnosis had occurred (before September 1989, September 1989 through May 1991, and June 1991 through May 1993, which correspond to the follow-up periods) and if it had been confirmed by laparoscopy—a standard surgical method for diagnosing endometriosis.6,7 These questions were asked again in each subsequent questionnaire.
In March 1994, we conducted a study to validate self-reported endometriosis diagnosis within the Nurses' Health Study II prospective cohort. Supplementary questionnaires were mailed to 200 women who were randomly selected from the then 1,766 cases who had reported incident diagnosis. Among those who reported laparoscopic confirmation and for whom records were received and reviewed (n = 105), a diagnosis of endometriosis was confirmed in 88.6%. However, among those women without laparoscopic confirmation (n = 26), evidence of clinical diagnosis was found in only 53.8% of the records. Requests for permission to review medical records were also sent to any woman who indicated that she had had a hysterectomy during the period of reported endometriosis diagnosis. A diagnosis of endometriosis at time of surgical procedure was confirmed in 79.6% (144 of 181) of the records received. However, endometriosis was the primary indication for hysterectomy in only 5.5% (9 of 163) of women for whom information regarding indication was available.
Based on these validation results, self-reported physician-diagnosed endometriosis without laparoscopic confirmation may be misclassified. Indeed, when these cases (1,080 reported from 1989–1999) were included in our analyses, all effect estimates were attenuated modestly. In addition, allowing women who report endometriosis and a hysterectomy in the same follow-up period to be cases might yield spurious results, because it would be unclear if the associated risk factors are related to endometriosis or to the pathology for which the hysterectomy was performed. Therefore, to reduce the magnitude of misclassification and prevent confounding by indication for hysterectomy, analyses of incident diagnosis of endometriosis were restricted to those women who reported laparoscopic confirmation of their diagnosis.
Within this restricted case definition, the relation between endometriosis and infertility status is complex. At baseline, the prevalence of infertility (defined as attempting to become pregnant for more than 1 year without success) was greater among women with laparoscopic confirmation (20%) than among those who were clinically diagnosed without laparoscopic confirmation (4%), potentially resulting in oversampling those with “asymptomatic” disease. Approximately 20% of all infertile women are found to have endometriosis.8 Had these women not attempted to become pregnant, a large proportion may never have received a laparoscopic diagnosis of endometriosis. We may also assume that women with no infertility who have had a laparoscopic diagnosis are “symptomatic,” otherwise a surgical evaluation would not have been conducted. Because endometriosis with infertility may be indicative of asymptomatic disease secondary to other primary causes of infertility, the risk factors for endometriosis with infertility could differ from those for endometriosis without concurrent infertility. Hence, we looked at risk factors separately by these two “subtypes” of endometriosis: 1) cases with neither past nor concurrent infertility, and 2) cases with concurrent infertility. Within this cohort, self-reported infertility was validated in a study of 100 randomly selected women who reported ovulatory infertility—95% of the self-reports were confirmed through medical record review.9
In 1989, detailed information was collected on age at menarche, usual menstrual cycle length and pattern during high school and college, and years after menarche until cycles became regular. Current menstrual cycle length and pattern were assessed in 1993.
A parity history (defined as the total number of pregnancies lasting 6 months or more) including age at first-term pregnancy was collected at baseline and updated biennially. Time since the last term birth was computed by subtracting the age of the participant at her most recently reported term birth from her current age. Data regarding lactation were collected in 1993 and updated in 1997. On the 1993 questionnaire, the total number of months of breastfeeding for all births combined was reported. In 1997, a detailed pregnancy-specific lactation history was collected, with women reporting if they had breastfed for at least 1 month and, if so, at what postpartum month they had stopped completely.
A history of oral contraceptive (OC) use since age 13 was recorded at baseline, and information about subsequent use was updated biennially. Women who had used OCs for 2 months or longer were classified as ever users. For past and current OC users, we calculated age at first use, cumulative duration of use, and, among parous women, duration of use before first pregnancy. Time since last use was calculated for past users. The validity of self-reported OC history was evaluated by comparing the questionnaire responses to information obtained through a detailed telephone interview. The correlation between questionnaire and telephone reported ever use of an OC was 99%. The Spearman correlation for reliability of duration of use was 0.94 (P < .001).10
A crude estimate of lifetime number of ovulatory cycles was calculated by subtracting age at menarche from current age (months). Lifetime months of OC use was subtracted from this number as was 11 months for each pregnancy to incorporate postpartum amenorrhea. This number was modified by adding 12 months if the woman reported an average menstrual cycle length of less than 26 days between ages 18–22 years but subtracting 20 months if she reported a length of more than 50 days during this 4-year interval. Finally, because duration of lactation was reported as a lifetime (not pregnancy-specific) number of months, we subtracted 4 months for women who reported lactation for 3–17 months, 8 months for 18–35 months, and 12 months for more than 35 lifetime months of lactation. Lifetime number of ovulatory cycles was not estimated for women missing either age at menarche, parity, or duration of OC use data.
Those who reported the diagnosis of endometriosis or a history of infertility before September 1989 were excluded from all analyses. Analyses were also restricted to those who were premenopausal and had intact uteri, because the occurrence of endometriosis after hysterectomy or in postmenopausal women is rare. Women with prior cancer diagnoses other than nonmelanoma skin cancer also were excluded.
Woman-months at risk were calculated from entry into the cohort until independently confirmed death or cancer diagnosis (other than nonmelanoma skin cancer), or self-reported laparoscopically confirmed endometriosis diagnosis, hysterectomy, or the onset of menopause. Women who reported physician-diagnosed endometriosis with no laparoscopic confirmation were censored at the time of that report but were allowed to reenter the analysis population if they reported laparoscopic confirmation on a subsequent questionnaire. In addition, because infertility is so strongly correlated with diagnosis of endometriosis via laparoscopy, we censored at self-report of infertility. Therefore, in all analyses our comparison group consists of women with neither diagnosed endometriosis nor infertility.
Incidence rates for each exposure category were computed as the number of incident cases divided by the woman-time accumulated. Time-varying Cox proportional hazards models treating age in months and 2-year questionnaire period as the time scale were used to estimate multivariate incidence rate ratios and to calculate 95% confidence intervals (CI), after adjusting simultaneously for confounding variables. Tests for trend in ordinal categorical exposures were calculated by creating a variable in which the median value or midpoint of each category was assigned to all participants in that group. Tests for heterogeneity comparing the effect estimates among cases with no past or current infertility to cases with concurrent infertility were calculated with a Wald statistic referred to a χ2 distribution with 1 degree of freedom.11 To evaluate effect modification, stratified analyses were conducted, and likelihood ratio tests comparing the model with both the main effects and the interaction terms to that with the main effects only were calculated.
We considered other possible risk factors for endometriosis as potential confounders if addition of that variable to the model changed the rate ratio by 10% or greater.12 If a factor was identified as a confounder of any estimated main effect, it was kept in all models. Based on these criteria, risk factors not observed to confound relations with any menstrual or reproductive exposures included race, age at menarche, time to menstrual cycle regularity, menstrual cycle length during college, menstrual cycle pattern during college, duration of lactation, current body mass index, current alcohol use, current smoking status, woman's birthweight, if the woman was breastfed as an infant, if she was one of a multiple gestation, or health care use through a proxy variable created from the answers to several questions that asked if the nurse had a clinical physical examination, Papanicolaou smear, pelvic examination, or breast examination in the past 2 years (categorized as no examination, examination for screening, examination for symptoms).
After baseline exclusions, a total of 90,065 women contributed 726,205 woman-years to these analyses; 1,721 incident cases of laparoscopically confirmed endometriosis with no past infertility were reported. These included 1,340 cases with no past or current infertility and 361 women who reported an infertility evaluation during the same follow-up period as laparoscopic confirmation of endometriosis. The overall incidence rate among women with no past infertility (the population for analysis) was 237/100,000 person-years and did not begin to decrease significantly until women were in their late 30s to early 40s (Missmer et al, American Journal of Epidemiology, in press).
Age at menarche was linearly associated with the rate of laparoscopically confirmed endometriosis (P value, test for trend < .001) (Table 1). Among women with no past or concurrent infertility we observed an increased risk with earlier age at menarche (rate ratio of 1.4 comparing menarche earlier than age 10 to menarche at age 12, 95% CI 0.9–2.0), while among cases with concurrent infertility, later age at menarche was associated with a reduction in risk (P value, test for heterogeneity = .05). An average menstrual cycle length of 25 or fewer days during college was associated with a 30% increase in risk (95% CI 1.1–1.5) compared with women with a cycle length of 26–31 days (Table 1); findings for cycle length during high school (data not shown) and current cycle length were similar. Again, this relation appeared to be more pronounced among women with no infertility, but the difference was not statistically significant. Regularity of menstrual cycle pattern during college and time to regularity were not associated with the rate of endometriosis diagnosis, whereas a 2-fold increase in risk was observed among those with current cycle irregularity compared with women with very regular cycles. To evaluate if the association between current cycle pattern and endometriosis may reflect prediagnostic symptoms of disease, we repeated our analysis excluding 4 years of follow-up. Although the observed effects were attenuated, an increase in risk with increasing cycle irregularity was still observed (compared with “very regular [± 4 days],” rate ratio of 0.7; 95% CI 0.5–1.0 for “extremely regular [± 2 days]”; however, rate ratio of 1.4, 95% CI 0.6–3.4 for “always irregular”). The effects of menstrual cycle characteristics were not modified by nulliparity, OC use, or having had a recent gynecologic examination (data not shown).
Independent of infertility status, women with 4 or more pregnancies lasting greater than 6 months were 50% less likely to be diagnosed, compared with women with 2 live births (P value, test for trend < .001) (Table 2). The rate of diagnosis was lower among never-infertile women who had given birth within the past 10 years, although this effect was not observed among cases with concurrent infertility (P value, test for heterogeneity = 0.10). The effects of parity, age at first birth, and age at last birth were not modified by time since last birth (data not shown). Nulliparity was strongly associated with increased risk among women in whom infertility was diagnosed during the same follow-up period of endometriosis diagnosis.
Among parous women, longer lifetime duration of lactation was associated with a linear decrease in risk, however this relation was observed primarily among those who had given birth within the past 5 years (rate ratio of 0.2 for breastfeeding > 23 months compared with never; 95% CI 0.1–0.4) (Table 3). Age-adjusted effects were very similar to those of the multivariate model (data not shown). The effect did not differ by infertility status, although the data among cases with concurrent infertility were sparse.
Compared with never users, the rate of laparoscopically confirmed diagnosis of endometriosis was decreased among current OC users (rate ratio of 0.8; 95% CI 0.6–1.0) but increased among past users (rate ratio of 1.7; 95% CI 1.5–2.0) (data not shown). However, the decreased risk among current users was evident only among nulliparous women (P value, test for heterogeneity = .01). A shorter time period since last use was associated with increased risk (P value, test for trend < .001), although lifetime duration of OC use was not related to the rate of endometriosis diagnosis (P value, test for trend = .28). None of these relations were modified by having had a recent gynecologic examination (data not shown).
Among all women with no past infertility, the risk of endometriosis increased with greater lifetime number of ovulatory cycles (P value, test for trend = .004) (Table 4). However the effect differed by case infertility status (P value, test for heterogeneity < .001) and was modified by OC use. Among women who never used OCs, the risk of endometriosis within the highest quartile of ovulatory cycles was 6-fold larger than that of women in the lowest quartile of lifetime cycles (95% CI 2.0–17.5).
In this prospective study among premenopausal women, we observed an inverse association between age at menarche, menstrual cycle length, parity, and lifetime duration of lactation and the incidence of diagnosis of laparoscopically-confirmed endometriosis.
Previous studies have been conducted in populations presenting for treatment of infertility or chronic pelvic pain or for tubal sterilization.13–18 These populations make evaluation of the effect of reproductive factors complex. Our prospective design, restriction of the comparison group to those without past infertility, and ability to define cases by current infertility status and stratify by potential effect modifiers strengthened our ability to clarify these relations with the risk of endometriosis.
Because it is impossible to know the exact onset of disease as opposed to date of diagnosis, our results are based on analyses of incidence of endometriosis diagnosis rather than incidence of disease onset. Therefore, the possibility that menstrual characteristics may represent underlying disease that precedes diagnosis should be considered. However, our prospective analyses of characteristics earlier in life (such as menstrual cycle length during late adolescence) should not be influenced by prediagnosis symptoms. Also the effect of current cycle characteristics was not altered when we stopped updating exposure level 4 years before diagnosis.
Menstrual cycle characteristics and reproductive history may represent between-person variation in hormonal milieu. In some but not all studies,19,20 higher levels of estradiol21–23 and estrone22 have been observed among adult women who experienced earlier age at menarche. Estradiol levels have also been observed to be higher among nulliparous women than among parous women,24,25 whereas androgen levels have an opposite association with parity.26 In addition, menstrual cycle characteristics and reproductive history may influence the total volume of endometrial cells released into the peritoneal cavity.27
Our data support hypotheses that the risk of endometriosis increases with menstrual cycle and reproductive factors that are associated with increased exposure to menstruation, such as earlier age at menarche, shorter menstrual cycle length, and reduced parity.1 Although several studies have found that an early menarche, often defined as age 11 years, increases the risk for endometriosis, sample sizes were small and a significant linear trend has not been reported.13,15,18,28–30 In a hospital-based case-control study of 286 women with primary infertility and endometriosis and 3,794 who had delivered a liveborn infant, endometriosis cases were more likely than controls to have shorter menstrual cycle lengths (≤ 27 days versus ≥ 38 days; 2.1 odds ratio; 95% CI 1.5–2.9), and a nonsignificant increase in risk was found with cycle irregularity.18 We observed a more modest effect of cycle length (rate ratio of 1.3 comparing < 26 days with 26–31 days; 95% CI 1.1–1.5) and no association with cycle regularity or time to regularity, suggesting that the effect of menstruation cannot be explained by ovulatory dysfunction.
The epidemiologic data have consistently observed a protective effect of current OC use whereas former use seems to increase risk.1,14,31,32 Our results agree and also confirm that risk among past users was greatest with more recent time since last use.1,31–33 However, because OCs are often prescribed as a first line of treatment, it is impossible to tell if those who are taking OCs and are subsequently diagnosed with endometriosis developed the disease before or after the exposure. In addition, it is probable that diagnosis of endometriosis is delayed in OC users. Oral contraceptives have been shown to decrease symptoms in the short-term, but symptoms may reemerge once use is discontinued or the disease progresses in severity.1 This intractable confounding by indication makes interpretation of these analyses difficult.1,32 We observed that the protective effect of current OC use was specific to nulliparous women, and that age at first use was associated with the rate of laparoscopically confirmed endometriosis only among those who have never been infertile. Therefore, OC treatment and subsequent laparoscopic diagnosis are more likely to have been prompted by endometriosis-specific symptoms. A true underlying effect of OCs is also less likely because no study has reported a dose-response effect for lifetime duration of use.1,14,32 Future studies addressing this association must include data that explicitly describe the reasons for initiation of OC use and preceding gynecologic symptoms.
A case-control study among multiparous women in whom endometriosis was diagnosed during a tubal ligation suggested that higher parity is associated with a lower risk of endometriosis.14 This relation was also observed in a case-control study comparing cases with women admitted for acute conditions at the same hospital.15 Our data suggest that among women with no past or current infertility, the more often a woman is pregnant for more than 6 months, the lower her risk of developing endometriosis. Because each pregnancy decreases the lifetime number of months during which a woman is exposed to menstrual fluid, the mechanism of risk may be similar to that of women with longer menstrual cycles who are exposed to the sloughing of endometrial tissue less frequently during any given calendar period as compared with women with shorter cycles.
Although we did not have detailed delivery data, we further hypothesize that dilation of the cervix during vaginal child birth may improve menstrual outflow once menses resume postpartum.34 Permanent cervical dilation occurs with labor and vaginal delivery, possibly reducing the resistance to menstrual flow and decreasing the likelihood of retrograde menstruation.35 In addition, the decidual reaction that occurs on the ovarian and pelvic surface attributed to high hormone levels during pregnancy might result in decreased susceptibility to endometriosis implantation or growth.
We observed an inverse relation with lactation, although the effect was attenuated overtime, appearing primarily among those who had given birth within the past 5 years. Prolonged lactation is associated with prolonged amenorrhea, and lactational steroid hormone suppression continues beyond the return of menses. A study of surgically induced endometriosis among rats reported that during pregnancy, 4% of plaques regressed while 30% progressed. However, lactation among these animals resulted in regression of 69% of plaques and no progression.36
Together, the observed protective effect of parity, independent of time since last birth, but of lactation only among those who had given birth within the past 5 years may help to elucidate underlying biologic mechanisms. It is possible that the decreased lifetime exposure to menstruation and changes in the hormonal milieu that suppress tissue implantation and progression associated with both parity and lactation may protect against endometriosis in the short-term. However, the additional permanent postpartum changes to the uterus and cervix may convey long-term protection by improving menstrual flow and making the surfaces of the peritoneal cavity less hospitable to lesion progression.
We attempted to estimate the combined effect of these menstrual and reproductive factors in a crude summary calculation of lifetime number of ovulatory cycles. Although we were not able to incorporate exact measures of postpartum and lactational amenorrhea or variations in menstrual cycle length, our estimate captures the major contributors to between-woman differences in exposure to menstruation. Our observation that the ovulatory cycle-associated risk of endometriosis was greatest among never users of OCs may suggest that prescription of OCs before disease onset is a valid public health intervention. Indeed, as discussed above, the ever user population (and consequently the combined analytic population given the prevalence of OC use) is likely enriched for women who suffered from endometriosis symptoms, therefore blunting the effect of lifetime ovulatory cycles.
Our prospective analyses among premenopausal U.S. registered nurses suggest that the rate of endometriosis is greatest among nulliparous women with earlier age at menarche and shorter menstrual cycle length. Among parous women, risk decreased with greater parity and lifetime duration of lactation. The large sample size and prospective design of the Nurses' Health Study II offers a unique opportunity to clarify the temporal relations underlying and add to the limited knowledge of menstrual and reproductive characteristics and endometriosis.
1. Eskenazi B, Warner ML. Epidemiology of endometriosis. Obstet Gynecol Clin North Am 1997;24:235–58.
2. Oral E, Arici A. Pathogenesis of endometriosis. Obstet Gynecol Clin North Am 1997;24:219–33.
3. Witz CA. Current concepts in the pathogenesis of endometriosis. Clin Obstet Gynecol 1999;42:566–85.
4. Witz CA, Schenken RS. Pathogenesis. Semin Reprod Endocrinol 1997;15:199–208.
5. Barbieri RL. Endometriosis and the estrogen threshold theory. Relation to surgical and medical treatment. J Reprod Med 1998;43:287–92.
6. Duleba AJ. Diagnosis of endometriosis. Obstet Gynecol Clin North Am 1997;24:331–46.
7. Pardanani S, Barbieri RL. The gold standard for the surgical diagnosis of endometriosis: visual findings or biopsy results? J Gynecol Techn 1998;4:121–4.
8. Tanahatoe S, Hompes PG, Lambalk CB. Accuracy of diagnostic laparoscopy in the infertility work-up before intrauterine insemination. Fertil Steril 2003;79:361–6.
9. Rich-Edwards JW, Goldman MB, Willett WC, Hunter DJ, Stampfer MJ, Colditz GA, et al. Adolescent body mass index and infertility caused by ovulatory disorder. Am J Obstet Gynecol 1994;171:171–7.
10. Hunter DJ, Manson JE, Colditz GA, Chasan-Taber L, Troy L, Stampfer MJ, et al. Reproducibility of oral contraceptive histories and validity of hormone composition reported in a cohort of US women. Contraception 1997;56:373–8.
11. Prentice RL, Kalbfleisch JD, Peterson AV Jr, Flournoy N, Farewell VT, Breslow NE. The analysis of failure times in the presence of competing risks. Biometrics 1978;34:541–54.
12. Greenland S. Modeling and variable selection in epidemiologic analysis. Am J Public Health 1989;79:340–9.
13. Signorello LB, Harlow BL, Cramer DW, Spiegelman D, Hill JA. Epidemiologic determinants of endometriosis: a hospital-based case-control study. Ann Epidemiol 1997;7:267–74.
14. Sangi-Haghpeykar H, Poindexter AN III. Epidemiology of endometriosis among parous women. Obstet Gynecol 1995;85:983–92.
15. Parazzini F, Ferraroni M, Fedele L, Bocciolone L, Rubessa S, Riccardi A. Pelvic endometriosis: reproductive and menstrual risk factors at different stages in Lombardy, northern Italy. J Epidemiol Community Health 1995;49:61–4.
16. Darrow SL, Selman S, Batt RE, Zielezny MA, Vena JE. Sexual activity, contraception, and reproductive factors in predicting endometriosis. Am J Epidemiol 1994;140:500–9.
17. Darrow SL, Vena JE, Batt RE, Zielezny MA, Michalek AM, Selman S. Menstrual cycle characteristics and the risk of endometriosis. Epidemiology 1993;4:135–42.
18. Cramer DW, Wilson E, Stillman RJ, Berger MJ, Belisle S, Schiff I, et al. The relation of endometriosis to menstrual characteristics, smoking, and exercise. JAMA 1986;255:1904–8.
19. Verkasalo PK, Thomas HV, Appleby PN, Davey GK, Key TJ. Circulating levels of sex hormones and their relation to risk factors for breast cancer: a cross-sectional study in 1092 pre- and postmenopausal women (United Kingdom). Cancer Causes Control 2001;12:47–59.
20. Hill P, Garbaczewski L, Kasumi K, Wynder EL. Plasma hormones in parous, nulliparous and postmenopausal Japanese women. Cancer Lett 1986;33:131–6.
21. Apter D, Reinila M, Vihko R. Some endocrine characteristics of early menarche, a risk factor for breast cancer, are preserved into adulthood. Int J Cancer 1989;44:783–7.
22. MacMahon B, Trichopoulos D, Brown J, Andersen AP, Cole P, deWaard F, et al. Age at menarche, urine estrogens and breast cancer risk. Int J Cancer 1982;30:427–31.
23. Moore JW, Key TJ, Wang DY, Bulbrook RD, Hayward JL, Takatani O. Blood concentrations of E2 and sex hormone-binding globulin in relation to age at menarche in premenopausal British and Japanese women. Breast Cancer Res Treat 1991;18(suppl 1):S47–50.
24. Dorgan JF, Reichman ME, Judd JT, Brown C, Longcope C, Schatzkin A, et al. Relationships of age and reproductive characteristics with plasma estrogens and androgens in premenopausal women. Cancer Epidemiol Biomarkers Prev 1995;4:381–6.
25. Bernstein L, Pike MC, Ross RK, Judd HL, Brown JB, Henderson BE. Estrogen and sex hormone-binding globulin levels in nulliparous and parous women. J Natl Cancer Inst 1985;74:741–5.
26. Musey VC, Collins DC, Brogan DR, Santos VR, Musey PI, Martino-Saltzman D, et al. Long term effects of a first pregnancy on the hormonal environment: estrogens and androgens. J Clin Endocrinol Metab 1987;64:111–8.
27. Cramer DW. Epidemiology of endometriosis. In: Wilson EA, editor. Endometriosis. New York (NY): Liss, 1987. p. 5–22.
28. Candiani GB, Danesino V, Gastaldi A, Parazzini F, Ferraroni M. Reproductive and menstrual factors and risk of peritoneal and ovarian endometriosis. Fertil Steril 1991;56:230–4.
29. Moen MH, Schei B. Epidemiology of endometriosis in a Norwegian county. Acta Obstet Gynecol Scand 1997;76:559–62.
30. Matorras R, Rodiquez F, Pijoan JI, Ramon O, Gutierrez de Teran G, Rodriguez-Escudero F. Epidemiology of endometriosis in infertile women. Fertil Steril 1995;63:34–8.
31. Parazzini F, Ferraroni M. Epidemiology of endometriosis. BMJ 1993;306:930–1.
32. Vessey MP, Villard-Mackintosh L, Painter R. Epidemiology of endometriosis in women attending family planning clinics. BMJ 1993;306:182–4.
33. Parazzini F, Di Cintio E, Chatenoud L, Moroni S, Mezzanotte C, Crosignani PG. Oral contraceptive use and risk of endometriosis. Italian Endometriosis Study Group. Br J Obstet Gynaecol 1999;106:695–9.
34. Cramer DW, Missmer SA. The epidemiology of endometriosis. Ann N Y Acad Sci 2002;955:11–22; discussion 34–6, 396–406.
35. Krantz KE. The anatomy of the cervix, gross and histologic. In: Blandau RJ, Moghissi K, editors. The biology of the cervix. Chicago: University of Chicago, 1973.
36. Barragan JC, Brotons J, Ruiz JA, Acien P. Experimentally induced endometriosis in rats: effect on fertility and the effects of pregnancy and lactation on the ectopic endometrial tissue. Fertil Steril 1992;58:1215–9.