ORIGINAL ARTICLE: GYNECOLOGY AND MENOPAUSE

M€ ullerian duct anomalies diagnosed by saline contrast sonohysterography: prevalence in a general population Eva Dreisler, M.D., Ph.D.,a and Søren Stampe Sørensen, M.D., D.M.Sc.b a b

Department of Gynecology, Juliane Marie Centre, University of Copenhagen Rigshospitalet, Copenhagen, Denmark; and Department of Gynecology and Obstetrics, Herlev Hospital, University of Copenhagen, Herlev, Denmark

Objective: To estimate the prevalence of uterine m€ ullerian duct anomalies in a general population. Design: Cross-sectional study. Setting: University hospital in collaboration with the Danish Civil Registry. Patient(s): A total of 1,654 women randomly recruited from a general population; 686 women were eligible and accepted inclusion (429 pre- and 257 postmenopausal). Saline contrast sonohysterography (SCSH) was finally performed in 622 women (aged 20–74 years) (the procedure was impossible owing to cervical stenosis in 58, contraindicated in 2, other patient-related factors in 4). Intervention(s): The shape of the uterus was dynamically evaluated in the transversal and longitudinal planes during SCSH and classified in accordance with American Fertility Society as normal, arcuate, septate (partial, complete), bicorn (partial, complete), or unicorn. History of previous miscarriage and menstrual cycle was obtained by a questionnaire. Main Outcome Measure(s): Prevalence of m€ ullerian anomalies, miscarriage, and oligomenorrhea. Result(s): The overall prevalence of m€ ullerian anomalies was 9.8% (61 of 622) (95% confidence interval [CI] 7.5–12.1). The majority had arcuate uteri (n ¼ 42, 6.8%), 17 partial septate (2.7%), 1 complete septate (1.6%), and 1 unicorn uterus (1.6%). M€ ullerian anomalies were significantly more frequently diagnosed in nulliparous (20% [26 of 128]) compared with parous women (7% [35 of 494]). M€ ullerian anomalies were more frequent in women with oligomenorrhea compared with women with normal menstrual periods (19% [15 of 79] vs. 10% [34 of 339]). One first-trimester miscarriage or multiple miscarriages (more than one) were not significantly more frequent in premenopausal women with m€ ullerian anomalies compared with women with normal-shaped uteri (24% [6 of 25] vs. 22% [57 of 265]). Conclusion(s): In a general population examined by SCSH the prevalence of M€ ullerian anomalies is estimated at 9.8%. M€ ullerian anomalies were more frequent in nulliparous women. Both impaired fertility and a pregnancy-associated modulation of the uterine corpus are among explanations. M€ ullerian anomalies were associated with oligomenorrhea in premenopausal women. In this general population there was no association between miscarriage and m€ ullerian anomalies; however, the number of cases was limited. (Fertil SterilÒ 2014;-:-–-. Ó2014 by American Use your smartphone Society for Reproductive Medicine.) to scan this QR code Key Words: M€ ullerian duct anomalies, congenital uterine malformations, saline contrast and connect to the sonohysterography, prevalence, oligomenorrhea, reproductive medicine, miscarriage Discuss: You can discuss this article with its authors and with other ASRM members at http:// fertstertforum.com/dreislere-mullerian-duct-anomalies-saline-contrast-sonohysterography/

T

he female genitals are derived from the paramesonephric duct (m€ ullerian) by a lateral fusion of the two paired m€ ullerian ducts, which conjoin and form the uterus

and the upper part of the cervix. Septal absorption by apoptosis results in a single cavity. Congenital uterine anomalies arise from either agenesia, fusion defects, or

Received December 12, 2013; revised April 24, 2014; accepted April 25, 2014. E.D. has nothing to disclose. S.S.S. has nothing to disclose. Supported by grants from The Martha Margrethe and Christian Hermansen Foundation, The Ludvig and Franciska Andersen Foundation, and Organon (Organon funds for gynecologic research). Ultrasound and hysteroscopy equipment donated by Toyota Foundation. E.D. and S.S.S. contributed equally to this work. Reprint requests: Eva Dreisler, M.D., Ph.D., University of Copenhagen Rigshospitalet, Department of Gynecology, Juliane Marie Centre, Blegdamsvej 9, DK-2100 København Ø, Denmark (E-mail: [email protected]). Fertility and Sterility® Vol. -, No. -, - 2014 0015-0282/$36.00 Copyright ©2014 American Society for Reproductive Medicine, Published by Elsevier Inc. http://dx.doi.org/10.1016/j.fertnstert.2014.04.043 VOL. - NO. - / - 2014

discussion forum for this article now.*

* Download a free QR code scanner by searching for “QR scanner” in your smartphone’s app store or app marketplace.

absorption defects (modulation defects). Various classification systems have been developed to describe congenital uterine anomalies (1–4). The American Fertility Society classification is in widespread use, but a new exhaustive classification system has been suggested (5). Congenital uterine anomalies have been related clinically to several reproductive consequences (6, 7): oligomenorrhea (8), early miscarriage (9, 10), preterm birth (11), and fetal malpresentation, depending on the type and severity of the malformation. 1

ORIGINAL ARTICLE: GYNECOLOGY AND MENOPAUSE Estimates of prevalence in a general population have been derived from studies of women scheduled for laparoscopic tubal ligation (prevalence 15%) (12) or transcervical tubal sterilization (prevalence 1.9%) (13). In 1997, using three-dimensional (3D) ultrasound technology, Jurkovic et al. (14) examined a population of women undergoing ovarian cancer screening or with suspected pelvic abnormalities (prevalence 5.4%) (14). Ilan Tur-Kaspa (15) evaluated women with abnormal uterine bleeding (prevalence 9.5%) or infertility (prevalence 20%) with saline contrast sonohysterography (SCSH). The prevalence will vary depending on the definition of m€ ullerian anomalies (MA), the diagnostic method used, and the population studied. The diagnostic methods have been classified in three groups according to diagnostic accuracy. The best diagnostic methods with accuracy >90% are combined hysteroscopy and laparoscopy, sonohysterography, and 3D ultrasound. The best methods are capable of accurately identifying uterine anomalies and classifying them to appropriate subtypes (16). The prevalence of MA in unselected populations was estimated at 5.5% in a recent systematic review that included studies using optimal tests (7). The prevalence among very young women has not been elucidated in the previous studies, and women seeking sterilization are assumed to have an obstetric history before the decision for sterilization. The information available on MA in a general population is sparse. The primary aim of our study was to document the prevalence of MA in a general population using SCSH and the

American Fertility Society classification criteria. The secondary aim was to relate the observed MA to history of miscarriage, to elucidate whether a possible impact on early pregnancy is detected and explore the previously described (8) relation between MA and oligomenorrhea.

MATERIALS AND METHODS Subjects Sixteen hundred sixty women were invited after random selection from the Danish Civil Registration System, which identifies every Danish citizen by a 10-digit number. Women born outside Denmark were not included. The women were living in seven municipalities comprising a mixture of socioeconomic areas: low-, middle-, and high-income areas were represented. The selected women were recruited by letter, including a questionnaire; 1,261 responded (76%). Three hundred two women were excluded (Fig. 1), and another 273 refused to participate. Finally, 686 women accepted examination with both transvaginal scan and SCSH. The age range was 20–74 years (median 45 years). The included women were slightly younger than those who refused to participate (median 51 years, range 20–74; P< .05). Of the included women, 429 were premenopausal and 257 were postmenopausal (defined as no natural bleeding for 12 months). Saline contrast sonohysterography was successfully performed in 622 women, aged 20–74 years, median 44 years

FIGURE 1

The population studied. Dreisler. Prevalence of m€ ullerian anomalies. Fertil Steril 2014.

2

VOL. - NO. - / - 2014

Fertility and Sterility® (2.5th percentile 21 years, 97.5th percentile 70 years). Saline contrast sonohysterography was not completed in 64 women with age range 21–74 years, median 64 (2.5th percentile 26 years, 97.5th percentile 74 years). In the study population 418 were premenopausal and 204 postmenopausal.

Questionnaire An 85-item questionnaire was developed for the study (17). The questionnaire was completed by 1,114 women (67%), 686 included and 428 nonincluded. The information obtained by the questionnaire was verified at the clinical examination for the included women. The questionnaire was validated by interview and test–retest analysis. The questionnaire also included questions on menstrual cycle (oligomenorrea defined as cycle length R35 days), pregnancies, miscarriage(s), and gestational age at miscarriages. The women were asked whether and when they had previous operations in the genital area, but no questions regarding infertility were included. No data on fetal presentation/malpresentation or information on preterm labor or other labor-associated complications were obtained.

Scanning Procedures Ultrasound examinations were performed using a Hitachi EUB-6000 ultrasound system (Hitachi Medical Corporation) with a 5–7.5-MHz vaginal probe, as described in reference 17. In the evaluation of the uterine shape, SCSH was used. A speculum was inserted and the cervix washed with antiseptic solution. Uterine cavities were visualized by installing saline with a baby feeding tube (Unomedical): 100 cm, charrie 8, connected to a 20-mL syringe with isotonic saline. In women with a narrow cervical orifice, we used a 5-charrie baby feeding tube. Generally a tenaculum was not used, but 25 women with cervical stenosis accepted a slight dilatation, leading to successful saline contrast sonohysterography in 20 of these women. The procedure was contraindicated in 2 (endometrial cancer), impossible because of cervical stenosis in 58, and impossible owing to other patient-related factors in 4. Uterine cavities were examined systematically in the longitudinal plane from the right to the left uterine corner and in the transversal plane from fundus to cervix. E.D. and S.S.S. performed all ultrasound examinations. To optimize the quality and interpretation of the ultrasonographic examinations, both investigators were present at the first 100 examinations, and in difficult cases. The investigators achieved consensus during examinations and by reviewing video clips. The shape of the uterus was dynamically evaluated in the transversal and longitudinal planes during SCSH. On the basis of the two-dimensional impression, the shape was classified in accordance with American Fertility Society criteria as normal, arcuate, septate (partial, complete), bicorn (partial, complete), or unicorn.

formed. A P level of < .05 was considered statistically significant. Statistical analysis was performed using the SPSS software package (release 22; IBM).

Ethics The Scientific-Ethical Committee of the County of Copenhagen (KA02083) and The Danish Data Protection Agency approved the study. All participants gave written consent according to Helsinki declarations.

RESULTS The overall prevalence of MA was 9.8% (61 of 622) (95% CI 7.5– 12.1). The majority had arcuate uteri (n ¼ 42), 17 partial septate, 1 complete septate, and 1 unicorn uterus. The complete septate uterus was previously diagnosed and had been corrected by metroplasty. In premenopausal women the prevalence of MA was 11.7% (49 of 418) (95% CI 8.6–14.8). The prevalence according to age is illustrated in Figure 2. In univariate analysis MA were significantly more frequently diagnosed both in nulligravid women (28% [23 of 105]) compared with gravid women (7% [38 of 517]) (P< .0001) and nulliparous women (20% [26 of 128]) compared with parous women (7% [35 of 494]) (P< .0001). In a linear regression model including age, menopause, and parity, only parity was significantly inversely associated with MA (B ¼ 0.03, 95% CI 0.005–055, P¼ .02). In premenopausal women (n ¼ 418) oligomenorrhea was reported by 79. M€ ullerian anomalies were more frequent (19% [15 of 79]) in women with oligomenorrhea compared with women with normal periods (10% [34 of 339]) (P¼ .0326). Among young premenopausal women (