View metadata, citation and similar papers at core.ac.uk brought to you by CORE

provided by Elsevier - Publisher Connector The Egyptian Journal of Radiology and Nuclear Medicine (2010) 41, 517–523

Egyptian Society of Radiology and Nuclear Medicine The Egyptian Journal of Radiology and Nuclear Medicine

www.elsevier.com/locate/ejrnm www.sciencedirect.com

Imaging of the uterovaginal anomalies

Doaa Ibraheem Hasan a,1, Hazim Ibrahim Aly Tantawy a,*, Sherin Attia Shazly b

a Radiodiagnosis Department, Faculty of Medicine, Zagazig University, Egypt b Gynecology and Obstetric Department, Faculty of Medicine, Zagazig University, Egypt

Received 8 September 2010; accepted 6 October 2010 Available online 3 December 2010

KEYWORDS Abstract Objective: To assess the role of different imaging modalities including ultrasonography, Uterovaginal; hysterosalpingogram, and magnetic resonance imaging in detection of variable Mu¨llerian anoma- MRI; lies. Preoperative proper diagnosis data about Mu¨llerian anomalies necessary for clear indications HSG (hysterosalpingogra- of how and when to operate. phy); Patients and methods: A retrospective MRI study of 34 patients was done in the period from Feb- MDA (Mu¨llerian duct ruary 2008 to February 2010, their age ranging from 3 months to 38 years (mean 24 years), with anomaly); uterovaginal anomalies. Ultrasonography was performed for all cases; HSG was performed in 16 TAUS (trans-abdominal cases before MRI imaging. ultrasound); Results: MRI is the most reliable method for evaluating uterovaginal anomalies, particularly in TVUS (trans-vaginal ultrasound) pediatrics and virgins. MRI is an accurate examination for identification and categorization of MDAs and should be carried out prior to any surgery, in this study MRI allowed correct diagnosis of 34 uterine anomalies (accuracy 100%) whereas US was correct in 30 out of 34 cases (accuracy 88%). HSG had a limited role as cannot be preformed for virgins, and cannot identify non-commu- nicating horns in unicornuate cases. Conclusion: MRI is the examination of choice in uterovaginal anomalies. Endovaginal ultrasound cannot be preformed for children or females who have never had sexual intercourse. TAUS have not proved completely reliable in Mu¨llerian duct anomalies. Ó 2010 Egyptian Society of Radiology and Nuclear Medicine. Production and hosting by Elsevier B.V. All rights reserved.

* Corresponding author. Mobile: +20 101534149. E-mail addresses: [email protected] (D.I. Hasan), hazimtantawy@ 1. Introduction yahoo.com (H.I.A. Tantawy). 1 Mobile: +20 113694905. The overall published data suggest a prevalence range of uter- ovaginal anomalies of around 1–6% (1,2). Congenital anoma- 0378-603X Ó 2010 Egyptian Society of Radiology and Nuclear lies of the Mu¨llerian ducts (MDA), the most common of which Medicine. Production and hosting by Elsevier B.V. Open access under CC BY-NC-ND license. are uterine malformations, are associated with fertility prob- Peer review under responsibility of Egyptian Society of Radiology and lems (3). Multifactorial polygenic and familial factors are Nuclear Medicine. involved in their formation. The result may be deficient devel- doi:10.1016/j.ejrnm.2010.10.008 opment (agenesis, rudimentary horn, unicornuate ), non-fusion (didelphys or bicornuate uterus), or defective canalization of the Mu¨llerian ducts (septate uterus) (4). Production and hosting by Elsevier (DES) is a synthetic estrogen crosses the placenta and exerts a direct effect on daughters shows genital 518 D.I. Hasan et al. tract anomalies that are completely unrelated to fusion anom- evaluation of tube permeability and can detect the presence alies (5,6). DES exposure in utero causes uterine abnormalities of uterine septa, intrauterine synechiae, submucous fibroids that were well shown at HSG and consist of irregular constric- and endometrial polyps. However, it cannot be performed tions and a T-shaped uterine cavity, with irregularity of the on patients who are virgins and it does not allow the external endometrial surface, uterine hypoplasia, and diverticula of uterine anatomy to be viewed, which hampers the differential the fallopian tubes that cause poor pregnancy outcome (7). diagnosis between uterus didelphys and septate uterus (14,16). MDAs may become clinically evident at different ages Moreover, HSG exposes the ovaries to ionizing radiation; depending on their specific characteristics and associated dis- the injection of contrast may also cause allergies and discom- orders. In the newborn/infant age, an initial presentation is a fort; and there is also a risk of uterine infection (16). palpable abdominal or pelvic mass due to a utero or/and vag- inal obstruction causing intraluminal fluid retention which can 2. Embryology be discovered in the adolescent age group as delayed menarche or primary with/without a fluid retention in the The female reproductive system develops from the two paired uterus (hydro/) and/or vagina (hydro/hematocol- Mu¨llerian ducts (synonym: paramesonephric duct) that start pos) which may present as a painful intra-abdominal swelling off in the embryonal mesoderm lateral to each Wolffian duct (8,9). In the childbearing age, MDAs can be present with var- (synonym: mesonephric duct). The paired Mu¨llerian ducts de- ious problems including infertility, repeated spontaneous abor- velop in medial and caudal directions; the cranial part remains tions, premature delivery, and fetal intrauterine growth non-fused and forms the fallopian tubes. The caudal part fuses retardation. There can be some associated renal, skeletal or to a single canal forming the uterus and the upper two thirds of spine malformations (10,11). the vagina. This is called lateral fusion. In a process called ver- As the complexity of presentations diagnosing of Mu¨llerian tical fusion, the intervening midline septum of both ducts malformations requires the use of more than one imaging undergoes regression. The caudal part of the vagina arises modality (12). from the sinovaginal bulb and fuses with the lower fused TVS allows a more detailed analysis of the , Mu¨llerian ducts (Fig. 1). uterine cavity and , its specificity ranges from 85% to Hence, associated malformations of the kidney, but not of 92%, while the specificity of MRI ranges from 96% to 100% the ovaries are frequently observed together with MDAs (17). for diagnosing Mu¨llerian malformations (13). Classification system of the Mu¨llerian duct anomalies used by HSG is the method traditionally used to evaluate the cervi- the American Society for Reproductive Medicine (18) (Fig. 2). cal canal, uterine cavity and fallopian tubes. Its efficacy in diagnosing anomalies is debatable and varies according to 3. Patients and methods the specific type of malformation (14). The specificity values range from 6% to 60%, depending on the malformation inves- Patients with Mu¨llerian abnormalities were referred to radiol- tigated and the technician’s skill (15). HSG enables accurate ogy department in Zagazig university hospital between Febru-

Fig. 1 The female reproductive system embryology. Imaging of the uterovaginal anomalies 519

Table 1 Correlation between the age range and different clinical presentations. Age range Clinical presentation Number of cases 3 months to 4 years Pelvic mass (n =3) 3 10–16 years Chronic pain (n =5) 5 1ry amenorrhea (n =6) 6 19–38 years 1ry infertility (n = 16) 16 2ry infertility (n =2) 2 Abortions (n =2) 2

and cervix performed for the patients were given initial idea as inspection of the low vaginal septum. HSG was done for small number of cases (16 cases refereed for infertility investigation). Fig. 2 The classification system of MDAs developed by the The arcuate uterus (class VII) is the most common anomaly American Fertility Society. in general (Table 2). In contrast, the septate uterus is the com- monest anomaly in the infertile population. US and MRI findings were reported in Tables 3 and 4, respectively, the findings will be described. Class VII was not ary 2008 and February 2010 after clinical examination, US or found in our study. Associated anomalies like HSG revealed uterine, cervical, and/or vaginal congenital were found in four cases, renal agenesis was found in three abnormalities. The majority of the cases were presented by cases. infertility or amenorrhea. Laparotomy and laparoscopic con- MRI allowed correct diagnosis of 34 uterine anomalies firmation of these radiological impressions were made in 13 (accuracy 100%) whereas US was correct in 30 of 34 cases and 12 patients, respectively, (hysteroscopy were made for nine (accuracy 88%). non-virgin cases). Thirty-four female patients were investi- Class I, the most common form is Mayer–Rokitansky–Kus- gated by US, and MRI. HSG was done for 16 non-virgin cases. ter–Hauser syndrome (MRKH), which consists of vaginal US examination was performed using the trans-abdominal agenesis and a rudimentary or absent uterus (Fig. 3). Also approach only (for infants and virgin patients) or combined for class I, we had correctly diagnosed one case of cervical trans-abdominal and transvaginal approaches for the remain- agenesis in the presence of maldifferentiation of the uterine ing cases. body. The MRI study was performed for all cases using a 0.5 T Class II results from the failure of one of both of the paired imager (Signa, General Electric Medical Systems) with body- Mu¨llerian ducts. If no rudimentary horn is present, usually no coil. Intravenous contrast was used in two cases. We obtained further treatment is required. Similarly, no intervention is SE T1 weighted, FSE T2-weighted, and fast STIR with fat sup- needed if there is a rudimentary horn that does not contain pression sequences. Sagittal, coronal and paraxial images were endometrium. However, if a rudimentary horn contains endo- acquired. The paraxial images were obtained to produce true metrium, regardless of whether it communicates with the main coronal images of the uterus. In evaluating MDAs, imaging uterine cavity, surgical excision is recommended to prevent ret- the uterus in its true coronal plane is essential to assess the rograde menses or ectopic pregnancy. US examination was external fundal contour. non-specific for unicornuate uterus (two of three cases were HSG was performed in 16 cases according to the standard diagnosed), defining only a single cavity (Fig. 4), but without technique in the pre-ovulatory phase of the menstrual cycle sufficient details of adnexal regions or the rudimentary horn. and without the use of anesthesia, only antispasmodic drug MRI was excellent in depicting the uterine morphology (three was used. A vaginal speculum was used to visualize the cervix, which was cleaned with antiseptic before being grasped with a single tooth tenaculum. A cervical cannula or a catheter was Table 2 Percentage of different classes in our study. used, approximately 10–15 mL of a water-soluble contrast medium was injected manually. Patients were lying on their Class (n)% backs during the procedure. Radiograms were done in supine (I) Complete agenesis (n =2) position following opacification of the uterine cavity, fallopian Cx* and UV** agenesis (n = 1) 14.7 tubes, and peritoneal spill. Delayed film was taken after Hypoplastic ut*** (n =2) 10 min. (II) Unicornuate (n = 3) 8.8 US, HSG and MRI findings and diagnosis were correlated (III) Ut*** didelphys (n = 4) 11.7 with the results of the laparotomy (in 13 cases), laparoscopy (IV) Bicornuate (n = 6) 17.6 (V) Septate ut*** (n = 5) 14.7 (in 12 cases), and hystoscopy (in 9 cases). (VI) Arcuate ut*** (n = 9) 26.4 (VII) DES drug related (n =0) 4. Results Transverse VS+ (n = 2) 5.8 n = number, \ = cervix, \\ = upper vagina, \\\ = ut, The majority of the MDAs presented by 1ry infertility or + = transverse vaginal septum. amenorrhea (Table 1). Clinical examinations of the vagina 520 D.I. Hasan et al.

Table 3 US findings of the patients. Class (n) US findings (I) Complete agenesis (n = 2) Absent uterine body, cervix, and upper vagina (2/2) Cx* and UV** agenesis (n = 1) Pelvic mass (no uterine body or cervix or vagina identified) (0/1) Hypoplastic uterus (n = 2) Small sized uterus (2/2) (II) Unicornuate (n = 3) Reduced uterine volume ellipsoidal configuration (2/3) (III) Ut*** didelphys (n = 4) Two separated uterine bodies one case with bilateral multiple fibroids (2/4). Two uterine bodies with one with unilateral obstructed (1/4) (IV) Bicornuate (n = 6) Divergent small ut*** horns, fundal contour concavity (<4 cm) (5/6) (V) Septate ut*** (n = 5) Fundal indentation, echogenic mass dividing the ut*** cavity (5/5) (VI) Arcuate ut*** (n = 9) Smooth indentation of the fundus, normal outer fundal contour (9/9) Transverse VS+ (n = 2) Midline anechoic pelvic cystic mass (1/2) n = number, \ = cervix, \\ = upper vagina, \\\ = ut, + = transverse vaginal septum.

Table 4 MRI findings of the patients. Class (n) MRI findings (I) Complete agenesis (n = 2) Absent uterus and upper vagina Cx* and UV** agenesis (n = 1) Absent CX* and UV**. Uterus was heterogeneous mass, lost zonal anatomy Hypoplastic uterus (n = 2) Small endometrial cavity and shorter distance between the uterine horns (II) Unicornuate (n = 3) One uterine horn-IIA in one case and IIB in one case (III) Ut*** didelphys (n = 4) Two separate complete chambers (IV) Bicornuate (n = 6) Divergent small ut*** horns, fundal contour concavity (<4 cm) (V) Septate ut*** (n = 5) Normal external contour, the septum was a low-intensity signals on T2WIs (VI) Arcuate ut*** (n = 9) Outer fundal contour remains convex, a thin, fibrous septum Transverse VS+ (n = 2) (one case) (one case) n = number, \ = cervix, \\ = upper vagina, \\\ = ut, + = transverse vaginal septum.

Fig. 3 Mayer-Rokitansky-Kuster-Hauser syndrome (class 1) in 14 years old female patient presented by 1ry amenorrhea. (A) The longitudinal abdominal US of the pelvis demonstrates that there is no uterus between the UB and the rectum. (B) Sagittal T2WI documents the absence of uterine tissues. (C & D) Axial T2WI shows absence of vaginal tissue between the bladder and the rectum, normal ovaries in higher cuts. Imaging of the uterovaginal anomalies 521

Fig. 4 Unicornuate uterus in 32-years-old female patient. (A) TVUS shows banana shape single uterine horn (B) HSG shows the left uterine horn and missing RT. horn of the uterus. (C & D) axial T2 & T1WIs show left sided solitary banana shaped uterine horn, normal zonal anatomy identified in T2 image and non communicating rudimentary horn containing blood with minimal amount of free fluid in the pelvis. out of three cases diagnosed). In one case of unicornuate Class IV is due to incomplete fusion of the Mu¨llerian ducts uterus with rudimentary horn non-communicating with the resulting in bicornuate uterus. A fundal cleft of more than main cavity HSG can suggest the rudimentary horn in commu- 1 cm distinguishes a bicornuate uterus from a septate uterus, nicating type one (class IIA, Fig. 2). bicornuate uteri rarely require surgery. However, if a patient Class III is the failure of Mu¨llerian duct fusion leading to is symptomatic, a metroplasty can be performed. Moreover, uterus didelphys, in the majority of cases there is a transverse the MRI allowed differentiating definitively between bicornuate vaginal septum with subsequently obstructive, hydro/hemato- uterus (IV) and septate uterus (class V), furthermore evaluating colpometra results. Resection of the septum establishes normal composition, thickness and extension of the uterine septum and menses via the vagina. US image demonstrated widely separate the aspect of the fundal contour. This is a very important distinc- endometrial canals and two complete cervical canals (three out tion to do because it significantly affects patient treatment: a of four cases were diagnosed), one false negative case missed as septate uterus requires hysteroscopic septectomy, while a bicor- a large displaced one uterine body), while nuate uterus does not require surgical treatment. T2WIs MR demonstrated two separate cervices and complete Class V is the septate uterus, which results from a failure of duplication of widely separated uterine. One case with ob- resorbed central intervening tissue, septum, between the two structed hemivagina was found (Fig. 5). endometrial canals.

Fig. 5 Uterine didelphys in a 13-year-old female patient (A) abdominal US, sagittal view shows distended vagina by fluid. (B & C) axial & sagittal T2-WIs show complete separation of the uterine horns, a dilated right-sided uterine horn and a non dilated left-sided uterine horn with 2 endocervical canals, which is indicative of uterine didelphys. Also noted obstructed right hemivagina indicate presence of blood in various stages of evolution and left pelvic simple identified. 522 D.I. Hasan et al.

Fig. 6 Complete septate uterus in 23-years-old female patient. (A) Abdominal US in transverse view shows normal fundal contour, two separated endometrial line (E) by a septum. (B) axis T2-WI shows a septate uterus with a complete septum extending to the external os. The septum is thin and has isointense signal intensity to that indicates it is muscular. (C) Coronal T2-WI shows lower extension of the septum and left sided pelvic endometrioma of low signal intensity of retained blood products.

Complete and incomplete forms were existing in our study. one (class IIA). This is in agreement with O’Neill et al. (24), In the former, the septum extends into the cervix (Fig. 6). they stated that sonography and HSG cannot reliably detect- the rudimentary Mu¨llerian remnants and preferred MRI in evaluation of unicornuate uterus. 5. Discussion Renal malformations are common with unicornuate uterus and occur mostly on the same side as the rudimentary horn Once an MDA is suggested based on the patient’s history and (25), in this study two cases present with renal aplasia, one case physical examination, the next diagnostic step includes an that had unilateral agenesis and one ectopic kidney. imaging workup, which often detects the underlying anomaly These anomalies are explained by Oppelt et al. (22), as the and guides further surgical interventions. fundamental embryonic structure involved in the process; the MRI is a very accurate imaging modality in uterine evalu- mesoderm establishes the connection between the differentia- ation and contributes significantly to treatment planning, tion of the uterus from the Wolffian and Mu¨llerian ducts, with although ultrasonography remains the modality of choice as the development of the urogenital system. an initial study of patients who are suspected of having MDAs. In the present study regarding class III, TAUS demon- We propose, in accordance with many authors in the liter- strated widely separate endometrial canals and two complete ature, to reserve MRI imaging for patients with a technically cervical canals in three out of four cases (75% accuracy), while inadequate or indeterminate ultrasound examination. T2WIs MR demonstrated two separate cervices and complete In the present study MRI allowed correct diagnosis of 34 duplication of widely separated uterine (100% accuracy). One uterine anomalies (accuracy 100%) whereas US was correct case with obstructed hemivagina was found. in 30 of 34 cases (accuracy 88%). In agreement with Acie´n Other researchers (14,15) stated that the key to the MR (19), who had reported similar percentage of the different imaging examination is the documentation of a deep midline MDAs, as the common types were septated uteri and bicornu- fundal cleft in the external contour of the uterus, two widely ate, while bicornuate, didelphys and unicornuate comprised divergent uterine horns, and two cervices. the remaining lesser amount. This was of clinical interest, as Class IV due to incomplete fusion of the Mu¨llerian ducts re- the former anomaly can be easily managed by hysteroscopy, sult in bicornuate uterus. In our study bicornuate uterus cases while the others need more complicated procedures or have were diagnosed by US in five out of six cases, one false diag- no surgical approach (20). nosed as septate caused by inadequate visualization of the fun- Class I, the most common form is Mayer–Rokitansky– dal cleft. This abnormality was most commonly misdiagnosed Kuster–Hauser syndrome (MRKH). In the present study US as a normal or bicornuate uterine cavity. and MRI diagnosed all MRKH cases; one of them had unilat- While MR in the present study was correctly diagnosed 6 eral ovarian aplasia and renal aplasia. As established, data out of 6 cases (100%) compared to US (83%). Accurate were reported by other authors (21,22), for the association of diagnoses were made for bicornuate uteri using either mullerian hypoplasia with ovarian and renal agenesis. modality owing mainly to the presence of a characteristic In our study we had also correctly diagnosed one case of fundal notch. cervical agenesis in the presence of maldifferentiation of the On the basis of external fundal outline appearance. In such uterine body. Diagnosis of absence of the uterine cervix and anomalies, the muscular or fibrotic nature of any intra-cavi- endocervical canal necessitates hysterectomy in young patient tary septum was assessed for the septal thickness and its signal population; an artificial cervix cannot be created (23). intensity at MRI (26). Class II results from the failure of one of both of the paired Class V is the septate uterus, which results from a failure of Mu¨llerian ducts. In our study US was non-specific for unicor- resorbed central intervening tissue, septum, between the two nuate uterus (two out of three cases were diagnosed). MRI was endometrial canals. Both US and MRI were of the same accu- excellent in depicting the uterine morphology (three out of racy in diagnosis of class V. But the nature of the septum is three cases diagnosed) in one case of unicornuate uterus with based on the signal intensity on MR series. rudimentary horn non-communicating with the main cavity. Pellerito et al. (26) stated also that the assessment can be HSG can suggest the rudimentary horn in communicating type made of the angle of divergence of the cavities at HSG; limited Imaging of the uterovaginal anomalies 523 role of the HSG in some instances of vaginal or cervical septa, (10) Buttram VC, Gibbons WE. Mu¨llerian anomalies: a proposed or both. So MRI gives the all needed data not only for the type classification: an analysis of 144 cases. Fertil Steril 1979;32:40–6. of anomaly but also for the nature of the tissue as fibrous or (11) Oppelt P, Stefan PR, Anja K, Sara B, Georges A, Kurt S, et al. muscular septum types. Clinical aspects of Mayer–Rokitansky–Kuster–Hauser syndrome: Anomalies of the vaginal septum should be resected at the recommendations for clinical diagnosis and staging. Hum Reprod 2006;21(3):792–7. time of diagnosis, thereby resolving problems of (12) Se´rgio R, Renata A, Thais V, Marina G, Priscila S, Jose´de A. and permitting adequate drainage of menstrual flow. Mu¨llerian duct anomalies: review of current management. Sao Two cases were found, in which MRI has consistently dem- Paulo Med J 2009;127(2):92–6. onstrated high sensitivity and specificity for evaluation of vag- (13) Pui MH. Imaging diagnosis of congenital uterine malformation. inal anomalies. In agreement with Pellerito et al. (26) who Comput Med Imaging Graph 2004;28(7):425–33. found that MRI correctly diagnosed 24 of 24 anomalies (14) Scarsbrook AF, Moore NR. MRI appearances of Mu¨llerian duct (100% accuracy), compared to (92% accuracy) detected on abnormalities. Clin Radiol 2003;58(10):747–54. endovaginal sonography (EVS). (15) Braun P, Grau FV, Pons RM, Enguix DP. Is hysterosalpingog- The patients were analyzed for the presence of a dermoid raphy able to diagnose all uterine malformations correctly? A and submucosal leiomyoma endometriosis. The endometriosis retrospective study. Eur J Radiol 2005;53(2):274–9. (16) Sorensen SS. Hysteroscopic evaluation and endocrinological was found by MRI in four cases and one case with ovarian aspects of women with Mu¨llerian anomalies and . cyst. Evaluating patients with Mu¨llerian anomalies contributes Int J Fertil 1987;32(6):445–52. proof in favor of the theories of retrograde menstruation and (17) Larsen WJ. Development of the urogenital system. In: Sherman celomic metaplasia (27). The sensitivity for detection of focal LS, Potter SS, Scott WH, editors. Human embryology. New implants with US has been reported as low as 11%, compared York, NY: Churchill Livingstone; 2001. p. 277–9. with sensitivity and a specificity of 100% detection of endome- (18) AFS. The American Fertility Society classifications of adnexal triomas by MRI (28). adhesions, distal tubal obstruction, tubal occlusion secondary to tubal ligation, tubal pregnancies, Mu¨llerian anomalies and 6. Conclusion intrauterine adhesions. Fertil Steril 1988;49:944–55. (19) Acie´n P. Embryological observations on the female genital tract. Hum Reprod 1992;7(4):437–45. TVUS has not proved completely reliable in Mu¨llerian duct (20) Pelosi III MA, Pelosi MA. Laparoscopic-assisted transvaginal anomalies. MRI is an accurate examination for identification metroplasty for the treatment of bicornuate uterus: a case study. and categorization of the MDAs and should be carried out Fertil Steril 1996;65:886–90. prior to any surgery (laparotomy/laparoscopy or/and hystero- (21) Creatsas G, Deligeoroglou E, Makrakis E, Kontoravdis A, scopic resections, metroplasty or vaginoplasty). Papadimitriou L. Creation of a neovagina following Williams vaginoplasty and the Creatsas modification in 111 patients with References Mayer–Rokitansky–Kuester–Hauser syndrome. Fertil Steril 2001;76:1036–40. (22) Oppelt P, Strissel PL, Kellermann A, Seeber S, Humeny A, (1) Sorensen SS. Estimated prevalence of Mu¨llerian anomalies. Acta Beckmann MW. Correlation of DNA sequence variations of the Obstet Gynecol Scand 1988;67:441–5. entire anti-Mu¨llerian hormone (AMH) gene promoter and AMH (2) Zanetti BE, Ferrari LR, Rossi C. Classification and radiographic protein expression with the Mayer–Rokitansky–Kuester–Hauser features of uterine malformations: hysterosalpingographic study. syndrome. Hum Reprod 2005;20:149–57. Br Radiol 1978;51:161–70. (23) Mueller GC, Hero KH, Yolanda RS, Elisabeth Q, Ruth CC, (3) Acie´n P, Susarte F, Romero J, Galan J, Mayol MJ, Quereda FJ, Timothy D, et al. Mu¨llerian duct anomalies: comparison of MRI et al. Complex genital malformation: ectopic ureter ending in a diagnosis and clinical diagnosis. AJR 2007;189:1294–302. supposed mesonephric duct in a woman with renal agenesis and (24) O’Neill MJ, Yoder Isabel C, Connolly Susan A, Muller Peter R. ipsilateral blind hemivagina. Eur J Obstet Gynecol Reprod Biol Imaging evaluation and classification of development anomalies 2004;117(1):105–8. of the female reproductive system with an emphasis on MR (4) Haddad B, Barranger E, Paniel BJ. Blind hemivagina: long term imaging. AJR 1999;173:407–16. follow-up and reproductive performance in 42 cases. Hum (25) Prada Arias M, Muguerza VR, Montero SM, Vazquez Castelo Reprod 1999;14(8):1962–4. JL, Arias GM, Rodriguez CA. Uterus didelphys with obstructed (5) Kaufman R, Adam E, Binder GL, Gerthoffer E. Upper genital tract hemivagina and multicystic dysplastic kidney. Eur J Pediatr Surg changes and pregnancy outcome in offspring exposed in utero to 2005;15(6):441–5. diethylstilbestrol. Am J Obstet Gynecol 1980;137:299–308. (26) Pellerito JS, McCarthy SM, Doyle MB, Glickman MG, DeCher- (6) Buttrani VC. Mu¨llerian anomalies and their management. Fertil ney AH. Diagnosis of uterine anomalies: relative accuracy of MR Steril 1983;40:159–63. imaging, endovaginal sonography, and hysterosalpingography. (7) Julia RF. MR imaging of Mullerian anomalies: impact on ¨ Radiology 1992;183(3):795–800. therapy. AJR 1996;167:1491–5. (27) Mazouni C, Girard G, Deter R, Haumonte JB, Blanc B, Bretelle (8) Pittock ST, Babovic VD, Lteif A. Mayer–Rokitansky–Kuster– F. Diagnosis of Mullerian anomalies in adults: evaluation of Hauser anomaly and its associated malformations. Am J Med practice. Fertil Steril 2008;89(1):219–22. Genet 2008;135:314–6. (28) Gougoutas CA, Siegelman ES, Hunt J, Outwater E. Pelvic (9) Acie´n P, Acie´n M. Malformations of the female genital tract endometriosis: various manifestations and MR imaging findings. and embryological bases. Curr Women Health Rev 2007;3(4): AJR 2000;175:353–8. 248–88.