- Email: [email protected]
Postoperative oral contraceptive exposure and risk of endometrioma recurrence
Research
www. AJOG.org
GENERAL GYNECOLOGY
Postoperative oral contraceptive exposure and risk of endometrioma recurrence Paolo Vercellini, MD; Edgardo Somigliana, MD, PhD; Raffaella Daguati, MD, PhD; Paola Vigano, PhD; Francesca Meroni, MD; Pier Giorgio Crosignani, MD OBJECTIVE: The purpose of this study was to compare the postoperative risk of endometrioma recurrence in women using oral contraception and in those undergoing simple observation. STUDY DESIGN: After laparoscopic excision of ovarian endometriotio-
mas, a cyclic, low-dose, monophasic oral contraceptive pill (OCP) was offered to women not seeking pregnancy. One month after surgery, and every 6 months afterward, the patients underwent clinical and ultrasonographic assessment. RESULTS: Of the 277 patients who entered the study, 102 used OCP
for the entire follow-up period (always users), 129 used OCP dis-
continuously (ever users), and 46 declined treatment (never users). The median follow-up was 28 months. Recurrent endometriotic cysts were detected in 74 subjects (27%). The 36-month cumulative proportion of subjects free from endometrioma recurrence was 94% in the always users compared with 51% in the never users (P ⬍ .001); adjusted incidence rate ratio (IRR) ⫽ 0.10 (95% CI, 0.04-0.24). CONCLUSION: Regular postoperative use of OCP effectively prevents endometrioma recurrence.
Key words: endometrioma, oral contraceptive pill, recurrence
Cite this article as: Vercellini P, Somigliana E, Daguati R, et al. Postoperative oral contraceptive exposure and risk of endometrioma recurrence. Am J Obstet Gynecol 2008;198:504.e1-504.e5.
T
he postoperative recurrence rate of ovarian endometriotic cysts has been indicated to be around 12-30% after 2-5 years of follow-up.1-6 However, several methodologic drawbacks (eg, postal or telephonic vs clinical followup, variable diagnostic modalities such as transabdominal vs transvaginal ultraFrom the Department of Obstetrics and Gynecology (Drs Vercellini, Daguati, Vigano, Meroni, and Crosignani), University of Milan; the Center for Research in Obstetrics and Gynecology (Drs Vercellini, Somigliana, Daguati, and Vigano); and the Infertility Unit (Dr Somigliana), Fondazione Ospedale Maggiore Policlinico, Mangiagalli, Regina Elena, Milan, Italy. This study was supported by the University of Milan School of Medicine Research Grant FIRST no. 12-01-5068118-00067. Received Nov. 6, 2006; revised June 12, 2007; accepted Nov. 6, 2007. Reprints: Paolo Vercellini, MD, Clinica Ostetrica e Ginecologica II, University of Milan, Istituto “Luigi Mangiagalli,” Via Commenda, 12 - 20122 Milan, Italy. [email protected]. 0002-9378/$34.00 © 2008 Mosby, Inc. All rights reserved. doi: 10.1016/j.ajog.2007.11.010
504.e1
sonography vs histology, exclusion of dropouts, unknown proportion of subjects using medical treatment) undermine the reliability of the available evidence, and the risk is potentially higher than previously reported. While there is a general consensus that surgery for endometrioma increases the chance of conception,7,8 recent evidence also suggests that the presence of these cysts and/or their surgical excision may impair ovarian reserve.9 The pathophysiologic mechanisms leading to this impairment have not been fully clarified. Surgery may directly affect ovarian reserve. Although rare, few cases of postsurgical ovarian failure in patients operated for bilateral endometriomas have been reported recently.9 Ovarian vascularization may also be compromised by surgical procedures,7 as the excision of the so-called cyst wall involves removal of ovarian cortex and follicles.10 However, impaired vascularization may precede surgery. During in vitro fertilization (IVF) cycles unoperated ovaries with endometriomas showed a reduced response to ovarian hyperstimulation.11 A reduced follicular number and activity have been demonstrated in ovarian
American Journal of Obstetrics & Gynecology MAY 2008
stroma adjacent to an endometrioma.12 Whatever the precise mechanisms causing damage to ovarian reserve are, and whether or not the latter precedes or follows surgery, prevention of recurrence and repeat surgery should be mainstays in long-term therapeutic strategies for women with endometriosis, especially in those seeking pregnancy in the future. It has been hypothesized that ovulation is causally related to endometriotic cyst development.13,14 If this were true, ovarian suppression after conservative surgery for endometrioma would greatly reduce the risk of lesion recurrence. We therefore set up a cohort study with the aim of defining the risk of endometrioma recurrence after conservative surgery in women using an oral contraceptive pill (OCP) and in those undergoing simple postoperative observation.
M ATERIALS AND M ETHODS Patients undergoing conservative surgery for endometriotic ovarian cysts evaluated in our tertiary care referral center for women with endometriosis were considered for inclusion in this cohort study. Local institutional review board approval was obtained and the pa-
General Gynecology
www.AJOG.org tients gave informed consent to the study. Selected subjects were under 40 years of age at surgery and underwent laparoscopic excision of endometriotic cysts. Subjects were excluded if they had undergone previous surgery for endometriosis, if they were seeking pregnancy, if precise details of the intervention or histologic diagnosis of endometriomas were not available, and if the usual contraindications to the use of estrogenprogestin combinations were present. Conservative surgery at laparoscopy was performed using mechanical instruments and electrosurgery only. Adhesions were sectioned with microscissors, the ovaries were completely mobilized, endometriomas were evacuated, rinsed with normal saline, and excised by means of countertraction applied to the pseudocapsule and normal gonadal cortex with atraumatic microforceps. Hemostasis was achieved by selective application of bipolar current. Disease was staged according to the American Society for Reproductive Medicine (ASRM) classification.15 At hospital discharge, patients not seeking pregnancy were offered long-term oral contraception. They were informed that ovulation inhibition induced by estrogenprogestin combinations could theoretically reduce the risk of endometriotic cyst recurrence, although no definitive data on this issue were available. Counseling was based on a structured and previously tested shared-decision making process.16 No written or web-based material was offered routinely to women. A cyclic, low-dose, monophasic oral contraceptive containing ethinyl-estradiol (0.02 mg) and desogestrel (0.15 mg) was prescribed to women who chose ovarian suppression. One month after surgery, and every 6 months afterward, patients underwent clinical assessment, vaginal and rectal examination, and transvaginal ultrasonography. Women were excluded if ovarian endometriomas were identified at the first postoperative evaluation to avoid inclusion of subjects with persistent, instead of recurrent, endometriotic cysts. Ovarian endometrioma recurrence was diagnosed when a round-shaped cystic mass with a minimum diameter of 2 cm,
with thick walls, regular margins, homogeneous low echogenic fluid content with scattered internal echoes, and without papillary proliferations was observed.17 Two highly qualified sonographers with extensive experience in the detection of endometriotic lesions performed all the scannings blinded to OCP use. When immediate surgery was not deemed necessary (eg, in the presence of a complex adnexal cyst), transvaginal ultrasonography was repeated after 2 months to confirm the diagnosis of endometrioma recurrence. Data were archived using Excel 2003 and then exported into SPSS 14.0 (SPSS, Inc, Chicago, IL) for analysis. Quantitative variables were compared by means of unpaired Student t test and analysis of variance (ANOVA), whereas qualitative variables were analyzed with Fisher exact test. Time to endometrioma recurrence was analyzed with the product limit method and the curves obtained were compared by the log-rank test. Subjects deciding to seek conception were censored. The event data used in computing time to endometrioma recurrence were the date of surgery and the date of ultrasonographic demonstration of recurrence of an ovarian endometriotic cyst, or last follow-up visit, when the patient started to seek pregnancy, or last menstrual cycle in cases of unexpected conception. As the product limit method allows for comparison of subjects with different lengths of follow-up, the women who became pregnant unexpectedly were not excluded from evaluation. Logistic regression and Cox’s proportional hazards models were used to estimate the effect of several covariates simultaneously. Included in these equations were terms for age at surgery, body mass index (BMI), previous pregnancies, ASRM classification, and whether endometriomas were monoor bilateral. P values below .05 were considered statistically significant.
R ESULTS Of the 277 patients who entered the study, 102 (37%) used OCP for the entire follow-up period (always users), 129 (46%) used OCP discontinuously (ever
Research
users), and 46 (17%) declined treatment (never users). The median follow-up was 28 months (interquartile range, 17-45 months). The baseline clinical characteristics of the women enrolled according to study groups are shown in the Table. Never users were significantly older than always and ever users. At ultrasonography performed 1 month after surgery, a unilateral endometrioma with a mean diameter of 2.9 ⫾ 0.6 cm was detected in 7 women. According to the study selection criteria, these patients were not included in the analysis as their cysts were defined as persistent instead of recurrent. Recurrent endometriotic cysts, detected in 74 subjects (27%), were unilateral in 51 cases and bilateral in 23, and had a mean diameter of 4.9 ⫾ 2.1 cm. The ultrasonographic diagnosis of recurrent endometrioma was confirmed by histology in each of the 53 patients who underwent second-line surgery. The main indications for surgery were pain (n ⫽ 27), increasing cyst diameter over time (n ⫽ 21), and psychological or personal reasons (n ⫽ 5). To determine the overall effect of OCP, we compared the populations with definite exposure, namely, those who used OCP for the entire follow-up period, and never users. The crude recurrence rate was 9% (9/102) in the former group and 56% (26/46) in the latter (Fisher exact test, P ⬍ .001; odds ratio [OR], 0.07; 95% confidence interval [CI], 0.03-0.18). Logistic regression showed that only OCP use was associated with a significant modification of the risk of recurrence. The adjusted OR for OCP use was 0.04 (95% CI, 0.02-0.13). Recurrence-free survival analysis is shown in Figure 1. An impressive difference in favor of the always OCP users was observed. The 36-month cumulative proportion of subjects free from endometrioma recurrence was 94% in the always users, compared with 51% in the never users (log-rank test, 21 ⫽ 36.2; P ⬍ .001). Cox regression analysis confirmed that OCP use is the sole factor influencing recurrence (P ⬍ .001). The adjusted incidence rate ratio (IRR) was 0.10 (95% CI, 0.04-0.24). The absolute risk reduction of endometrioma recurrence in al-
MAY 2008 American Journal of Obstetrics & Gynecology
504.e2
Research
General Gynecology
www.AJOG.org
TABLE
Characteristics of 277 women who underwent conservative laparoscopic surgery for ovarian endometriotic cysts according to postoperative oral contraception use Always users (n ⴝ 102)
Ever users (n ⴝ 129)
Never users (n ⴝ 46)
P value
Age (y)a
30.6 ⫾ 5.8
30.2 ⫾ 5.5
33.0 ⫾ 5.8
.01
BMI (kg/m )
21.3 ⫾ 3.1
20.8 ⫾ 2.3
20.5 ⫾ 2.1
.14
Women with previous pregnancies
27 (26)
33 (26)
................................................................................................................................................................................................................................................................................................................................................................................ 2a ................................................................................................................................................................................................................................................................................................................................................................................ b
8 (17)
.46
................................................................................................................................................................................................................................................................................................................................................................................ b
ASRM classification
.06
.......................................................................................................................................................................................................................................................................................................................................................................
Stage III
79 (78)
82 (64)
30 (65)
Stage IV
23 (22)
47 (36)
16 (35)
....................................................................................................................................................................................................................................................................................................................................................................... ................................................................................................................................................................................................................................................................................................................................................................................ b
Laterality
.08
.......................................................................................................................................................................................................................................................................................................................................................................
Monolateral
80 (78)
94 (73)
28 (61)
Bilateral
22 (22)
35 (27)
18 (39)
....................................................................................................................................................................................................................................................................................................................................................................... ................................................................................................................................................................................................................................................................................................................................................................................ a
5.4 ⫾ 2.0
Largest cyst diameter (cm)
5.5 ⫾ 2.5
5.2 ⫾ 2.2
.72
................................................................................................................................................................................................................................................................................................................................................................................ a
Values are means ⫾ SD.
b
Figures are numbers with percentages in parentheses.
Vercellini. Postoperative OC use and endometrioma. Am J Obstet Gynecol 2008.
FIGURE 1
Cumulative recurrence of endometriomas according to OC use 1.0
Recurrence-free subjects
0.8
0.6
0.4
0.2
0.0 0
6
12
18
24
30
36
Time after surgery (months)
Thirty-six-month endometrioma recurrence-free survival analysis after conservative laparoscopic surgery according to the treatment modality adopted: (____) oral contraception for the entire follow-up period (n ⫽ 102); (- - -) expectant management (n ⫽ 46) (log-rank test, 21 ⫽ 36.2; P ⬍ .001). Vertical tick marks are censored observations. Vercellini. Postoperative OC use and endometrioma. Am J Obstet Gynecol 2008.
504.e3
vented endometrioma recurrence in 1 out of 2 patients (95% CI, 0.2-7) 3 years after surgery, with a relative risk reduction of 80%. To determine whether duration of OCP use had an effect on the study outcome, we analyzed the 129 women exposed to OCP discontinuously separately. We compared the 62 subjects who used the estrogen-progestin combination for 12 months or more with the 67 who used OCP for less than 12 months. The baseline characteristics of women in these 2 groups did not differ significantly (data not shown). Median (interquartile range) duration of OCP use in the 2 groups was 20 (15-34) and 6 (3-6) months, respectively. The crude recurrence rate was 23% (14/62) in the former group and 37% (25/67) in the latter. As duration of follow-up was predictably quite different in the 2 groups, analysis of a simple contingency table was not appropriate. The 36-month cumulative proportion of subjects free from endometrioma recurrence was 78% in the 12 months or more users compared with 51% in the less than 12 months users (log-rank test, 21 ⫽ 11.9; P ⬍ .001). Recurrence-free survival analysis is shown in Figure 2. Cox regression analysis confirmed that longer OCP use significantly affected recurrence rate (P ⫽ .001; adjusted IRR ⫽ 0.29; 95% CI, 0.14-0.62).
American Journal of Obstetrics & Gynecology MAY 2008
FIGURE 2
Cumulative recurrence of endometriomas according to duration of postoperative OCP use 1.0
0.8
Recurrence-free subjects
ways users compared with never users was 47% (95% CI, 37-57). This means that regular postoperative use of an estrogen-progestin combination pre-
0.6
0.4
0.2
0.0 0
6
12
18
24
30
36
Time after surgery (months)
Thirty-six--month endometrioma recurrencefree survival analysis after conservative laparoscopic surgery according to duration of postoperative oral contraception use: (____) 12 months or more (n ⫽ 62); (- - -) less than 12 months (n ⫽ 67) (log-rank test, 21 ⫽ 11.9; P ⬍ .001). Vertical tick marks are censored observations. Vercellini. Postoperative OC use and endometrioma. Am J Obstet Gynecol 2008.
www.AJOG.org The analysis also confirmed that OCP use was the only factor associated with a significant modification of the risk of recurrences. Mean ⫾ SD BMI was, respectively, 20.9 ⫾ 2.3 and 21.0 ⫾ 2.7 in patients with and without recurrent endometriomas (P ⫽ .890).
C OMMENT The pathogenesis of endometriomas is controversial, as they might originate from inversion and progressive invagination of the ovarian cortex,18 secondary involvement of functional ovarian cysts by superficial endometrial implants,13,14 or metaplasia of the coelomic epithelium covering the gonad.19 Our data support the view that ovulation is crucial for the development of endometriotic cysts.14 This is in line with the demonstrated increase in risk of endometriosis with greater lifetime number of ovulatory cycles and with the hypothesized 6-fold increase in risk in never OCP users compared with ever users.20 The findings of the present cohort study strongly suggest that continuous exposure to oral contraceptives after conservative surgery for ovarian endometriosis is associated with a major reduction in the risk of long-term endometrioma recurrence. The gradient effect observed with regard to duration of use in women who took the estrogenprogestin combination discontinuously supports the consistency of the overall results. The shape and slope of recurrence-free survival curves in women who used OCP for more vs less than 12 months after surgery support the view that the protective effect of estrogenprogestin combination tends to vanish rapidly after discontinuation.20 Some drawbacks undermine the strength of our data. The allocation of subjects to treatment groups was not random but based on patient preference after detailed counseling. This obviously exposed the study to selection bias, as the profile of women who chose different options may vary in several and undefined aspects. Moreover, as the available scientific evidence on the risk of endometrioma recurrence in relation to OCP use is scanty, inconsistent, and probably
General Gynecology confusing for patients, no written or web-based material was offered routinely to women. Hypothetically, verbal counseling might have facilitated an undue provider’s influence on individual choices. However, had this type of bias been introduced, women taking OCP could have been considered at higher risk of endometrioma recurrence because the suggestion to use postoperative medical treatment would be expected to be more probable after difficult or nonradical surgery than after routine procedures. However, multivariate logistic and Cox regression analyses, used to allow for several covariates potentially associated with the study outcome, confirmed the strong protective effect of OCP. Consequently, it seems improbable that methodologic limits could explain per se such an impressive difference in outcome. The diagnosis of endometrioma was mainly based on ultrasonographic findings. However, the very high overall accuracy of vaginal ultrasonography in the detection of endometriotic cysts has been repeatedly demonstrated,17,21 and histology always confirmed the sonographic diagnosis of cyst type in women who underwent repeat surgery. The particularly good diagnostic capabilities of our sonographers are probably due to extensive experience with endometriosis patients at our large referral center. Moreover, ultrasonography as a test for the identification of endometriomas is expected to perform better in a selected group of high-risk subjects with increased prevalence of ovarian endometriotic cysts than in the general female population. The high recurrence rate observed in the never users group compares unfavorably with the results of previously published studies1-3 but is in line with the findings of a more recent report on this issue.6 Although we cannot exclude that this could be due to suboptimal technical capabilities, the laparoscopists in our department have a long-standing and extensive experience in conservative surgery for endometriosis. Indeed, several factors could influence available evidence, such as the surgical modality adopted,5 the diagnostic cutoff in cyst di-
Research
ameter, the duration of follow-up of a chronic disease with a high relapse rate, and the exclusion of dropouts, a subgroup with a potentially worse prognosis. Publication bias seems probable, as the estimate of endometrioma recurrence rate derives mainly from noncomparative, retrospective trials, which tend to overrepresent optimistic results. Surgeons with suboptimal long-term outcomes may be less willing to submit their data or have them published. The real incidence of endometriotic cyst recurrence may thus be undefined due to misdiagnosis and underreporting. Two highly qualified sonographers with extensive experience in the detection of endometriotic lesions performed all the scannings. Erroneous selection of persistent instead of recurrent cysts should have been avoided with the exclusion of patients with evidence of endometriomas at first assessment 1 month after surgery. Our detection limit of only 2 cm could have inflated the incidence of recurrence observed in our study and our results may be only apparently at odds with findings reported in the literature.1-5 Interestingly, using a similar cut off value of 2 cm, a 2-year endometrioma recurrence rate as high as 30% has been recently identified in a series of 224 patients.6 Data on the possible association between OCP use and endometriosis are conflicting, but any risk reduction appears to affect only current or recent users.20,22,23 The findings of 2 Italian case-control studies suggest that oral contraception is associated with an increased risk of endometriosis, but several biases impede generalizability of these results.24,25 For example, dysmenorrhea is the most frequent symptom of endometriosis and a major indication for OCP use. It is impossible to clarify if patients with a diagnosis of endometriosis who previously used OCP developed the disease before or after exposure. In the only published RCT comparing adjuvant OCP use for 6 months only vs no medical treatment after laparoscopic excision of ovarian endometriomas, no significant difference in long-term cyst recurrence rate was observed between groups.26 However, this is not unex-
MAY 2008 American Journal of Obstetrics & Gynecology
504.e4
Research
General Gynecology
pected given the protective effect of OCP on current but not previous users. Indeed, we never detected de novo endometrioma formation during 1- to 2-year estrogen-progestin assumption in women with surgically diagnosed endometriosis.27 The real incidence of endometrioma recurrence after conservative surgery may be higher than previously reported. This finding, which must be verified by further large cohort studies, could have a major impact on prognosis and treatment strategies. In addition to the inherent increases in morbidity and health care costs, repeat excisional surgery for ovarian endometriotic cysts might be detrimental in terms of decrease in follicular reserve and worsening of reproductive performance.10 Current OCP use appears to be associated with a dramatic reduction in the risk of ovarian endometriotic cyst recurrence. If randomized controlled trials confirm the observations of the present cohort study, postoperative long-term ovarian suppression with estrogen-progestins could be routinely offered, especially to women seeking conception in the future. f REFERENCES 1. Busacca M, Marana R, Caruana P, et al. Recurrence of ovarian endometrioma after laparoscopic excision. Am J Obstet Gynecol 1999;180:519-23. 2. Saleh A, Tulandi T. Reoperation after laparoscopic treatment of ovarian endometriomas by excision and by fenestration. Fertil Steril 1999;72:322-4. 3. Ghezzi F, Beretta P, Franchi M, Parissis M, Bolis P. Recurrence of ovarian endometriosis and anatomical location of the primary lesion. Fertil Steril 2001;75:136-40.
504.e5
4. Jones KD, Sutton CJ. Recurrence of chocolate cysts after laparoscopic ablation. J Am Assoc Gynecol Laparosc 2002;9:315-20. 5. Vercellini P, Chapron C, De Giorgi O, Consonni D, Frontino G, Crosignani PG. Coagulation or excision of ovarian endometriomas? Am J Obstet Gynecol 2003;188:606-10. 6. Koga K, Takemura Y, Osuga Y, et al. Recurrence of ovarian endometrioma after laparoscopic excision. Hum Reprod 2006;21:2171-4. 7. Chapron C, Vercellini P, Barakat H, Vieira M, Dubuisson JB. Management of ovarian endometriomas. Hum Reprod Update 2002;8: 591-7. 8. Kikuchi I, Takeuchi H, Kitade M, Shimanuki H, Kumakiri J, Kinoshita K. Recurrence rate of endometriomas following a laparoscopic cystectomy. Acta Obstet Gynecol 2006;85: 1120-4. 9. Busacca M, Riparini J, Somigliana E, et al. Postsurgical ovarian failure after laparoscopic excision of bilateral endometriomas. Am J Obstet Gynecol 2006;195:421-5. 10. Hachisuga T, Kawarabayashi T. Histopathological analysis of laparoscopically treated ovarian endometriotic cysts with special reference to loss of follicles. Hum Reprod 2002;17:432-5. 11. Somigliana E, Infantino M, Benedetti F, Arnoldi M, Calanna G, Ragni G. The presence of ovarian endometriomas is associated with a reduced responsiveness to gonadotropins. Fertil Steril 2006;86:192-6. 12. Maneschi F, Marasa L, Incandela S, Mazzarese M, Zupi E. Ovarian cortex surrounding benign neoplasms: a histologic study. Am J Obstet Gynecol 1993;169:388-93. 13. Nezhat F, Nezhat C, Allan CJ, Metzger DA, Sears DL. Clinical and histologic classification of endometriomas. Implications for a mechanism of pathogenesis. J Reprod Med 1992;37: 771-6. 14. Jain S, Dalton ME. Chocolate cysts from ovarian follicles. Fertil Steril 1999;72:852-6. 15. Revised American Society for Reproductive Medicine classification of endometriosis: 1996. Fertil Steril 1997;67:817-21. 16. Vercellini P, Pietropaolo G, De Giorgi O, Daguati R, Pasin R, Crosignani PG. Reproductive performance in infertile women with rectovaginal endometriosis: is surgery worthwhile? Am J Obstet Gynecol 2006;195:1303-10.
American Journal of Obstetrics & Gynecology MAY 2008
www.AJOG.org 17. Mais V, Guerriero S, Ajossa S, Angiolucci M, Paoletti AM, Melis GB. The efficiency of transvaginal ultrasonography in the diagnosis of endometrioma. Fertil Steril 1993;60:776-80. 18. Brosens IA, Puttemans PJ, Deprest J. The endoscopic localization of endometrial implants in the ovarian chocolate cyst. Fertil Steril 1994;61:1034-8. 19. Nisolle M, Donnez J. Peritoneal endometriosis, ovarian endometriosis, and adenomyotic nodules of the rectovaginal septum are three different entities. Fertil Steril 1997;68:585-96. 20. Missmer SA, Hankinson SE, Spiegelman D, et al. Reproductive history and endometriosis among premenopausal women. Obstet Gynecol 2004;104:965-74. 21. Eskenazi B, Warner M, Bonsignore L, Olive D, Samuels S, Vercellini P. Validation study of nonsurgical diagnosis of endometriosis. Fertil Steril 2001;76:929-35. 22. Vessey MP, Villard-Mackintosh L, Painter R. Epidemiology of endometriosis in women attending family planning clinics. BMJ 1993; 306:182-4. 23. Vercellini P, Ragni G, Trespidi L, Oldani S, Crosignani PG. Does contraception modify the risk of endometriosis? Hum Reprod 1993; 8:547-51. 24. Parazzini F, Ferraroni M, Bocciolone L, Tozzi L, Rubessa S, La Vecchia C. Contraceptive methods and risk of pelvic endometriosis. Contraception 1994;49:47-55. 25. Parazzini F, Di Cintio E, Chatenoud L, Moroni S, Mezzanotte C, Crosignani PG. Oral contraceptive use and risk of endometriosis. Italian Endometriosis Study Group. BJOG 1999;106: 695-9. 26. Muzii L, Marana R, Caruana P, Catalano GF, Margutti F, Panici PB. Postoperative administration of monophasic combined oral contraceptives after laparoscopic treatment of ovarian endometriomas: a prospective, randomised trial. Am J Obstet Gynecol 2000; 183:588-92. 27. Vercellini P, Frontino G, De Giorgi O, Pietropaolo G, Pasin R, Crosignani PG. Continuous use of an oral contraceptive for endometriosis-associated recurrent dysmenorrhea that does not respond to a cyclic pill regimen. Fertil Steril 2003;80:560-3.
www. AJOG.org
GENERAL GYNECOLOGY
Postoperative oral contraceptive exposure and risk of endometrioma recurrence Paolo Vercellini, MD; Edgardo Somigliana, MD, PhD; Raffaella Daguati, MD, PhD; Paola Vigano, PhD; Francesca Meroni, MD; Pier Giorgio Crosignani, MD OBJECTIVE: The purpose of this study was to compare the postoperative risk of endometrioma recurrence in women using oral contraception and in those undergoing simple observation. STUDY DESIGN: After laparoscopic excision of ovarian endometriotio-
mas, a cyclic, low-dose, monophasic oral contraceptive pill (OCP) was offered to women not seeking pregnancy. One month after surgery, and every 6 months afterward, the patients underwent clinical and ultrasonographic assessment. RESULTS: Of the 277 patients who entered the study, 102 used OCP
for the entire follow-up period (always users), 129 used OCP dis-
continuously (ever users), and 46 declined treatment (never users). The median follow-up was 28 months. Recurrent endometriotic cysts were detected in 74 subjects (27%). The 36-month cumulative proportion of subjects free from endometrioma recurrence was 94% in the always users compared with 51% in the never users (P ⬍ .001); adjusted incidence rate ratio (IRR) ⫽ 0.10 (95% CI, 0.04-0.24). CONCLUSION: Regular postoperative use of OCP effectively prevents endometrioma recurrence.
Key words: endometrioma, oral contraceptive pill, recurrence
Cite this article as: Vercellini P, Somigliana E, Daguati R, et al. Postoperative oral contraceptive exposure and risk of endometrioma recurrence. Am J Obstet Gynecol 2008;198:504.e1-504.e5.
T
he postoperative recurrence rate of ovarian endometriotic cysts has been indicated to be around 12-30% after 2-5 years of follow-up.1-6 However, several methodologic drawbacks (eg, postal or telephonic vs clinical followup, variable diagnostic modalities such as transabdominal vs transvaginal ultraFrom the Department of Obstetrics and Gynecology (Drs Vercellini, Daguati, Vigano, Meroni, and Crosignani), University of Milan; the Center for Research in Obstetrics and Gynecology (Drs Vercellini, Somigliana, Daguati, and Vigano); and the Infertility Unit (Dr Somigliana), Fondazione Ospedale Maggiore Policlinico, Mangiagalli, Regina Elena, Milan, Italy. This study was supported by the University of Milan School of Medicine Research Grant FIRST no. 12-01-5068118-00067. Received Nov. 6, 2006; revised June 12, 2007; accepted Nov. 6, 2007. Reprints: Paolo Vercellini, MD, Clinica Ostetrica e Ginecologica II, University of Milan, Istituto “Luigi Mangiagalli,” Via Commenda, 12 - 20122 Milan, Italy. [email protected]. 0002-9378/$34.00 © 2008 Mosby, Inc. All rights reserved. doi: 10.1016/j.ajog.2007.11.010
504.e1
sonography vs histology, exclusion of dropouts, unknown proportion of subjects using medical treatment) undermine the reliability of the available evidence, and the risk is potentially higher than previously reported. While there is a general consensus that surgery for endometrioma increases the chance of conception,7,8 recent evidence also suggests that the presence of these cysts and/or their surgical excision may impair ovarian reserve.9 The pathophysiologic mechanisms leading to this impairment have not been fully clarified. Surgery may directly affect ovarian reserve. Although rare, few cases of postsurgical ovarian failure in patients operated for bilateral endometriomas have been reported recently.9 Ovarian vascularization may also be compromised by surgical procedures,7 as the excision of the so-called cyst wall involves removal of ovarian cortex and follicles.10 However, impaired vascularization may precede surgery. During in vitro fertilization (IVF) cycles unoperated ovaries with endometriomas showed a reduced response to ovarian hyperstimulation.11 A reduced follicular number and activity have been demonstrated in ovarian
American Journal of Obstetrics & Gynecology MAY 2008
stroma adjacent to an endometrioma.12 Whatever the precise mechanisms causing damage to ovarian reserve are, and whether or not the latter precedes or follows surgery, prevention of recurrence and repeat surgery should be mainstays in long-term therapeutic strategies for women with endometriosis, especially in those seeking pregnancy in the future. It has been hypothesized that ovulation is causally related to endometriotic cyst development.13,14 If this were true, ovarian suppression after conservative surgery for endometrioma would greatly reduce the risk of lesion recurrence. We therefore set up a cohort study with the aim of defining the risk of endometrioma recurrence after conservative surgery in women using an oral contraceptive pill (OCP) and in those undergoing simple postoperative observation.
M ATERIALS AND M ETHODS Patients undergoing conservative surgery for endometriotic ovarian cysts evaluated in our tertiary care referral center for women with endometriosis were considered for inclusion in this cohort study. Local institutional review board approval was obtained and the pa-
General Gynecology
www.AJOG.org tients gave informed consent to the study. Selected subjects were under 40 years of age at surgery and underwent laparoscopic excision of endometriotic cysts. Subjects were excluded if they had undergone previous surgery for endometriosis, if they were seeking pregnancy, if precise details of the intervention or histologic diagnosis of endometriomas were not available, and if the usual contraindications to the use of estrogenprogestin combinations were present. Conservative surgery at laparoscopy was performed using mechanical instruments and electrosurgery only. Adhesions were sectioned with microscissors, the ovaries were completely mobilized, endometriomas were evacuated, rinsed with normal saline, and excised by means of countertraction applied to the pseudocapsule and normal gonadal cortex with atraumatic microforceps. Hemostasis was achieved by selective application of bipolar current. Disease was staged according to the American Society for Reproductive Medicine (ASRM) classification.15 At hospital discharge, patients not seeking pregnancy were offered long-term oral contraception. They were informed that ovulation inhibition induced by estrogenprogestin combinations could theoretically reduce the risk of endometriotic cyst recurrence, although no definitive data on this issue were available. Counseling was based on a structured and previously tested shared-decision making process.16 No written or web-based material was offered routinely to women. A cyclic, low-dose, monophasic oral contraceptive containing ethinyl-estradiol (0.02 mg) and desogestrel (0.15 mg) was prescribed to women who chose ovarian suppression. One month after surgery, and every 6 months afterward, patients underwent clinical assessment, vaginal and rectal examination, and transvaginal ultrasonography. Women were excluded if ovarian endometriomas were identified at the first postoperative evaluation to avoid inclusion of subjects with persistent, instead of recurrent, endometriotic cysts. Ovarian endometrioma recurrence was diagnosed when a round-shaped cystic mass with a minimum diameter of 2 cm,
with thick walls, regular margins, homogeneous low echogenic fluid content with scattered internal echoes, and without papillary proliferations was observed.17 Two highly qualified sonographers with extensive experience in the detection of endometriotic lesions performed all the scannings blinded to OCP use. When immediate surgery was not deemed necessary (eg, in the presence of a complex adnexal cyst), transvaginal ultrasonography was repeated after 2 months to confirm the diagnosis of endometrioma recurrence. Data were archived using Excel 2003 and then exported into SPSS 14.0 (SPSS, Inc, Chicago, IL) for analysis. Quantitative variables were compared by means of unpaired Student t test and analysis of variance (ANOVA), whereas qualitative variables were analyzed with Fisher exact test. Time to endometrioma recurrence was analyzed with the product limit method and the curves obtained were compared by the log-rank test. Subjects deciding to seek conception were censored. The event data used in computing time to endometrioma recurrence were the date of surgery and the date of ultrasonographic demonstration of recurrence of an ovarian endometriotic cyst, or last follow-up visit, when the patient started to seek pregnancy, or last menstrual cycle in cases of unexpected conception. As the product limit method allows for comparison of subjects with different lengths of follow-up, the women who became pregnant unexpectedly were not excluded from evaluation. Logistic regression and Cox’s proportional hazards models were used to estimate the effect of several covariates simultaneously. Included in these equations were terms for age at surgery, body mass index (BMI), previous pregnancies, ASRM classification, and whether endometriomas were monoor bilateral. P values below .05 were considered statistically significant.
R ESULTS Of the 277 patients who entered the study, 102 (37%) used OCP for the entire follow-up period (always users), 129 (46%) used OCP discontinuously (ever
Research
users), and 46 (17%) declined treatment (never users). The median follow-up was 28 months (interquartile range, 17-45 months). The baseline clinical characteristics of the women enrolled according to study groups are shown in the Table. Never users were significantly older than always and ever users. At ultrasonography performed 1 month after surgery, a unilateral endometrioma with a mean diameter of 2.9 ⫾ 0.6 cm was detected in 7 women. According to the study selection criteria, these patients were not included in the analysis as their cysts were defined as persistent instead of recurrent. Recurrent endometriotic cysts, detected in 74 subjects (27%), were unilateral in 51 cases and bilateral in 23, and had a mean diameter of 4.9 ⫾ 2.1 cm. The ultrasonographic diagnosis of recurrent endometrioma was confirmed by histology in each of the 53 patients who underwent second-line surgery. The main indications for surgery were pain (n ⫽ 27), increasing cyst diameter over time (n ⫽ 21), and psychological or personal reasons (n ⫽ 5). To determine the overall effect of OCP, we compared the populations with definite exposure, namely, those who used OCP for the entire follow-up period, and never users. The crude recurrence rate was 9% (9/102) in the former group and 56% (26/46) in the latter (Fisher exact test, P ⬍ .001; odds ratio [OR], 0.07; 95% confidence interval [CI], 0.03-0.18). Logistic regression showed that only OCP use was associated with a significant modification of the risk of recurrence. The adjusted OR for OCP use was 0.04 (95% CI, 0.02-0.13). Recurrence-free survival analysis is shown in Figure 1. An impressive difference in favor of the always OCP users was observed. The 36-month cumulative proportion of subjects free from endometrioma recurrence was 94% in the always users, compared with 51% in the never users (log-rank test, 21 ⫽ 36.2; P ⬍ .001). Cox regression analysis confirmed that OCP use is the sole factor influencing recurrence (P ⬍ .001). The adjusted incidence rate ratio (IRR) was 0.10 (95% CI, 0.04-0.24). The absolute risk reduction of endometrioma recurrence in al-
MAY 2008 American Journal of Obstetrics & Gynecology
504.e2
Research
General Gynecology
www.AJOG.org
TABLE
Characteristics of 277 women who underwent conservative laparoscopic surgery for ovarian endometriotic cysts according to postoperative oral contraception use Always users (n ⴝ 102)
Ever users (n ⴝ 129)
Never users (n ⴝ 46)
P value
Age (y)a
30.6 ⫾ 5.8
30.2 ⫾ 5.5
33.0 ⫾ 5.8
.01
BMI (kg/m )
21.3 ⫾ 3.1
20.8 ⫾ 2.3
20.5 ⫾ 2.1
.14
Women with previous pregnancies
27 (26)
33 (26)
................................................................................................................................................................................................................................................................................................................................................................................ 2a ................................................................................................................................................................................................................................................................................................................................................................................ b
8 (17)
.46
................................................................................................................................................................................................................................................................................................................................................................................ b
ASRM classification
.06
.......................................................................................................................................................................................................................................................................................................................................................................
Stage III
79 (78)
82 (64)
30 (65)
Stage IV
23 (22)
47 (36)
16 (35)
....................................................................................................................................................................................................................................................................................................................................................................... ................................................................................................................................................................................................................................................................................................................................................................................ b
Laterality
.08
.......................................................................................................................................................................................................................................................................................................................................................................
Monolateral
80 (78)
94 (73)
28 (61)
Bilateral
22 (22)
35 (27)
18 (39)
....................................................................................................................................................................................................................................................................................................................................................................... ................................................................................................................................................................................................................................................................................................................................................................................ a
5.4 ⫾ 2.0
Largest cyst diameter (cm)
5.5 ⫾ 2.5
5.2 ⫾ 2.2
.72
................................................................................................................................................................................................................................................................................................................................................................................ a
Values are means ⫾ SD.
b
Figures are numbers with percentages in parentheses.
Vercellini. Postoperative OC use and endometrioma. Am J Obstet Gynecol 2008.
FIGURE 1
Cumulative recurrence of endometriomas according to OC use 1.0
Recurrence-free subjects
0.8
0.6
0.4
0.2
0.0 0
6
12
18
24
30
36
Time after surgery (months)
Thirty-six-month endometrioma recurrence-free survival analysis after conservative laparoscopic surgery according to the treatment modality adopted: (____) oral contraception for the entire follow-up period (n ⫽ 102); (- - -) expectant management (n ⫽ 46) (log-rank test, 21 ⫽ 36.2; P ⬍ .001). Vertical tick marks are censored observations. Vercellini. Postoperative OC use and endometrioma. Am J Obstet Gynecol 2008.
504.e3
vented endometrioma recurrence in 1 out of 2 patients (95% CI, 0.2-7) 3 years after surgery, with a relative risk reduction of 80%. To determine whether duration of OCP use had an effect on the study outcome, we analyzed the 129 women exposed to OCP discontinuously separately. We compared the 62 subjects who used the estrogen-progestin combination for 12 months or more with the 67 who used OCP for less than 12 months. The baseline characteristics of women in these 2 groups did not differ significantly (data not shown). Median (interquartile range) duration of OCP use in the 2 groups was 20 (15-34) and 6 (3-6) months, respectively. The crude recurrence rate was 23% (14/62) in the former group and 37% (25/67) in the latter. As duration of follow-up was predictably quite different in the 2 groups, analysis of a simple contingency table was not appropriate. The 36-month cumulative proportion of subjects free from endometrioma recurrence was 78% in the 12 months or more users compared with 51% in the less than 12 months users (log-rank test, 21 ⫽ 11.9; P ⬍ .001). Recurrence-free survival analysis is shown in Figure 2. Cox regression analysis confirmed that longer OCP use significantly affected recurrence rate (P ⫽ .001; adjusted IRR ⫽ 0.29; 95% CI, 0.14-0.62).
American Journal of Obstetrics & Gynecology MAY 2008
FIGURE 2
Cumulative recurrence of endometriomas according to duration of postoperative OCP use 1.0
0.8
Recurrence-free subjects
ways users compared with never users was 47% (95% CI, 37-57). This means that regular postoperative use of an estrogen-progestin combination pre-
0.6
0.4
0.2
0.0 0
6
12
18
24
30
36
Time after surgery (months)
Thirty-six--month endometrioma recurrencefree survival analysis after conservative laparoscopic surgery according to duration of postoperative oral contraception use: (____) 12 months or more (n ⫽ 62); (- - -) less than 12 months (n ⫽ 67) (log-rank test, 21 ⫽ 11.9; P ⬍ .001). Vertical tick marks are censored observations. Vercellini. Postoperative OC use and endometrioma. Am J Obstet Gynecol 2008.
www.AJOG.org The analysis also confirmed that OCP use was the only factor associated with a significant modification of the risk of recurrences. Mean ⫾ SD BMI was, respectively, 20.9 ⫾ 2.3 and 21.0 ⫾ 2.7 in patients with and without recurrent endometriomas (P ⫽ .890).
C OMMENT The pathogenesis of endometriomas is controversial, as they might originate from inversion and progressive invagination of the ovarian cortex,18 secondary involvement of functional ovarian cysts by superficial endometrial implants,13,14 or metaplasia of the coelomic epithelium covering the gonad.19 Our data support the view that ovulation is crucial for the development of endometriotic cysts.14 This is in line with the demonstrated increase in risk of endometriosis with greater lifetime number of ovulatory cycles and with the hypothesized 6-fold increase in risk in never OCP users compared with ever users.20 The findings of the present cohort study strongly suggest that continuous exposure to oral contraceptives after conservative surgery for ovarian endometriosis is associated with a major reduction in the risk of long-term endometrioma recurrence. The gradient effect observed with regard to duration of use in women who took the estrogenprogestin combination discontinuously supports the consistency of the overall results. The shape and slope of recurrence-free survival curves in women who used OCP for more vs less than 12 months after surgery support the view that the protective effect of estrogenprogestin combination tends to vanish rapidly after discontinuation.20 Some drawbacks undermine the strength of our data. The allocation of subjects to treatment groups was not random but based on patient preference after detailed counseling. This obviously exposed the study to selection bias, as the profile of women who chose different options may vary in several and undefined aspects. Moreover, as the available scientific evidence on the risk of endometrioma recurrence in relation to OCP use is scanty, inconsistent, and probably
General Gynecology confusing for patients, no written or web-based material was offered routinely to women. Hypothetically, verbal counseling might have facilitated an undue provider’s influence on individual choices. However, had this type of bias been introduced, women taking OCP could have been considered at higher risk of endometrioma recurrence because the suggestion to use postoperative medical treatment would be expected to be more probable after difficult or nonradical surgery than after routine procedures. However, multivariate logistic and Cox regression analyses, used to allow for several covariates potentially associated with the study outcome, confirmed the strong protective effect of OCP. Consequently, it seems improbable that methodologic limits could explain per se such an impressive difference in outcome. The diagnosis of endometrioma was mainly based on ultrasonographic findings. However, the very high overall accuracy of vaginal ultrasonography in the detection of endometriotic cysts has been repeatedly demonstrated,17,21 and histology always confirmed the sonographic diagnosis of cyst type in women who underwent repeat surgery. The particularly good diagnostic capabilities of our sonographers are probably due to extensive experience with endometriosis patients at our large referral center. Moreover, ultrasonography as a test for the identification of endometriomas is expected to perform better in a selected group of high-risk subjects with increased prevalence of ovarian endometriotic cysts than in the general female population. The high recurrence rate observed in the never users group compares unfavorably with the results of previously published studies1-3 but is in line with the findings of a more recent report on this issue.6 Although we cannot exclude that this could be due to suboptimal technical capabilities, the laparoscopists in our department have a long-standing and extensive experience in conservative surgery for endometriosis. Indeed, several factors could influence available evidence, such as the surgical modality adopted,5 the diagnostic cutoff in cyst di-
Research
ameter, the duration of follow-up of a chronic disease with a high relapse rate, and the exclusion of dropouts, a subgroup with a potentially worse prognosis. Publication bias seems probable, as the estimate of endometrioma recurrence rate derives mainly from noncomparative, retrospective trials, which tend to overrepresent optimistic results. Surgeons with suboptimal long-term outcomes may be less willing to submit their data or have them published. The real incidence of endometriotic cyst recurrence may thus be undefined due to misdiagnosis and underreporting. Two highly qualified sonographers with extensive experience in the detection of endometriotic lesions performed all the scannings. Erroneous selection of persistent instead of recurrent cysts should have been avoided with the exclusion of patients with evidence of endometriomas at first assessment 1 month after surgery. Our detection limit of only 2 cm could have inflated the incidence of recurrence observed in our study and our results may be only apparently at odds with findings reported in the literature.1-5 Interestingly, using a similar cut off value of 2 cm, a 2-year endometrioma recurrence rate as high as 30% has been recently identified in a series of 224 patients.6 Data on the possible association between OCP use and endometriosis are conflicting, but any risk reduction appears to affect only current or recent users.20,22,23 The findings of 2 Italian case-control studies suggest that oral contraception is associated with an increased risk of endometriosis, but several biases impede generalizability of these results.24,25 For example, dysmenorrhea is the most frequent symptom of endometriosis and a major indication for OCP use. It is impossible to clarify if patients with a diagnosis of endometriosis who previously used OCP developed the disease before or after exposure. In the only published RCT comparing adjuvant OCP use for 6 months only vs no medical treatment after laparoscopic excision of ovarian endometriomas, no significant difference in long-term cyst recurrence rate was observed between groups.26 However, this is not unex-
MAY 2008 American Journal of Obstetrics & Gynecology
504.e4
Research
General Gynecology
pected given the protective effect of OCP on current but not previous users. Indeed, we never detected de novo endometrioma formation during 1- to 2-year estrogen-progestin assumption in women with surgically diagnosed endometriosis.27 The real incidence of endometrioma recurrence after conservative surgery may be higher than previously reported. This finding, which must be verified by further large cohort studies, could have a major impact on prognosis and treatment strategies. In addition to the inherent increases in morbidity and health care costs, repeat excisional surgery for ovarian endometriotic cysts might be detrimental in terms of decrease in follicular reserve and worsening of reproductive performance.10 Current OCP use appears to be associated with a dramatic reduction in the risk of ovarian endometriotic cyst recurrence. If randomized controlled trials confirm the observations of the present cohort study, postoperative long-term ovarian suppression with estrogen-progestins could be routinely offered, especially to women seeking conception in the future. f REFERENCES 1. Busacca M, Marana R, Caruana P, et al. Recurrence of ovarian endometrioma after laparoscopic excision. Am J Obstet Gynecol 1999;180:519-23. 2. Saleh A, Tulandi T. Reoperation after laparoscopic treatment of ovarian endometriomas by excision and by fenestration. Fertil Steril 1999;72:322-4. 3. Ghezzi F, Beretta P, Franchi M, Parissis M, Bolis P. Recurrence of ovarian endometriosis and anatomical location of the primary lesion. Fertil Steril 2001;75:136-40.
504.e5
4. Jones KD, Sutton CJ. Recurrence of chocolate cysts after laparoscopic ablation. J Am Assoc Gynecol Laparosc 2002;9:315-20. 5. Vercellini P, Chapron C, De Giorgi O, Consonni D, Frontino G, Crosignani PG. Coagulation or excision of ovarian endometriomas? Am J Obstet Gynecol 2003;188:606-10. 6. Koga K, Takemura Y, Osuga Y, et al. Recurrence of ovarian endometrioma after laparoscopic excision. Hum Reprod 2006;21:2171-4. 7. Chapron C, Vercellini P, Barakat H, Vieira M, Dubuisson JB. Management of ovarian endometriomas. Hum Reprod Update 2002;8: 591-7. 8. Kikuchi I, Takeuchi H, Kitade M, Shimanuki H, Kumakiri J, Kinoshita K. Recurrence rate of endometriomas following a laparoscopic cystectomy. Acta Obstet Gynecol 2006;85: 1120-4. 9. Busacca M, Riparini J, Somigliana E, et al. Postsurgical ovarian failure after laparoscopic excision of bilateral endometriomas. Am J Obstet Gynecol 2006;195:421-5. 10. Hachisuga T, Kawarabayashi T. Histopathological analysis of laparoscopically treated ovarian endometriotic cysts with special reference to loss of follicles. Hum Reprod 2002;17:432-5. 11. Somigliana E, Infantino M, Benedetti F, Arnoldi M, Calanna G, Ragni G. The presence of ovarian endometriomas is associated with a reduced responsiveness to gonadotropins. Fertil Steril 2006;86:192-6. 12. Maneschi F, Marasa L, Incandela S, Mazzarese M, Zupi E. Ovarian cortex surrounding benign neoplasms: a histologic study. Am J Obstet Gynecol 1993;169:388-93. 13. Nezhat F, Nezhat C, Allan CJ, Metzger DA, Sears DL. Clinical and histologic classification of endometriomas. Implications for a mechanism of pathogenesis. J Reprod Med 1992;37: 771-6. 14. Jain S, Dalton ME. Chocolate cysts from ovarian follicles. Fertil Steril 1999;72:852-6. 15. Revised American Society for Reproductive Medicine classification of endometriosis: 1996. Fertil Steril 1997;67:817-21. 16. Vercellini P, Pietropaolo G, De Giorgi O, Daguati R, Pasin R, Crosignani PG. Reproductive performance in infertile women with rectovaginal endometriosis: is surgery worthwhile? Am J Obstet Gynecol 2006;195:1303-10.
American Journal of Obstetrics & Gynecology MAY 2008
www.AJOG.org 17. Mais V, Guerriero S, Ajossa S, Angiolucci M, Paoletti AM, Melis GB. The efficiency of transvaginal ultrasonography in the diagnosis of endometrioma. Fertil Steril 1993;60:776-80. 18. Brosens IA, Puttemans PJ, Deprest J. The endoscopic localization of endometrial implants in the ovarian chocolate cyst. Fertil Steril 1994;61:1034-8. 19. Nisolle M, Donnez J. Peritoneal endometriosis, ovarian endometriosis, and adenomyotic nodules of the rectovaginal septum are three different entities. Fertil Steril 1997;68:585-96. 20. Missmer SA, Hankinson SE, Spiegelman D, et al. Reproductive history and endometriosis among premenopausal women. Obstet Gynecol 2004;104:965-74. 21. Eskenazi B, Warner M, Bonsignore L, Olive D, Samuels S, Vercellini P. Validation study of nonsurgical diagnosis of endometriosis. Fertil Steril 2001;76:929-35. 22. Vessey MP, Villard-Mackintosh L, Painter R. Epidemiology of endometriosis in women attending family planning clinics. BMJ 1993; 306:182-4. 23. Vercellini P, Ragni G, Trespidi L, Oldani S, Crosignani PG. Does contraception modify the risk of endometriosis? Hum Reprod 1993; 8:547-51. 24. Parazzini F, Ferraroni M, Bocciolone L, Tozzi L, Rubessa S, La Vecchia C. Contraceptive methods and risk of pelvic endometriosis. Contraception 1994;49:47-55. 25. Parazzini F, Di Cintio E, Chatenoud L, Moroni S, Mezzanotte C, Crosignani PG. Oral contraceptive use and risk of endometriosis. Italian Endometriosis Study Group. BJOG 1999;106: 695-9. 26. Muzii L, Marana R, Caruana P, Catalano GF, Margutti F, Panici PB. Postoperative administration of monophasic combined oral contraceptives after laparoscopic treatment of ovarian endometriomas: a prospective, randomised trial. Am J Obstet Gynecol 2000; 183:588-92. 27. Vercellini P, Frontino G, De Giorgi O, Pietropaolo G, Pasin R, Crosignani PG. Continuous use of an oral contraceptive for endometriosis-associated recurrent dysmenorrhea that does not respond to a cyclic pill regimen. Fertil Steril 2003;80:560-3.