Congenital MUllerian anomalies: diagnostic accuracy of three-dimensional ultrasound

Size: px

Start display at page:

Download "Congenital MUllerian anomalies: diagnostic accuracy of three-dimensional ultrasound"

Frederick Booker
6 years ago
Views:

1 FERTILITY AND STERILITY Copyright ~" 1996 American Society for Reproductive Medicine Printed on acid-free paper in U. s. A. Congenital MUllerian anomalies: diagnostic accuracy of three-dimensional ultrasound Francisco Raga, M.D.* Fernando Bonilla-Musoles, M.D. * Javier Blanes, M.D.* Newton G. Osborne, M.D.t:J: Universidad de Valencia, Valencia, Spain, and Howard University College of Medicine, Washington, D.C. Objective: To determine whether it is possible to identify and diagnose accurately Mullerian anomalies with three-dimensional (3-D) ultrasound (US). Design: Controlled blinded clinical study. Setting: Normal human volunteers undergoing infertility evaluation in a university hospital. Patients: Forty-two patients who underwent laparoscopy and hysterosalpingography as part of their work up for infertility and were found to have either a normal uterus (30 patients) or a Mullerian abnormality (12 patients) consented to be evaluated with 3-D US by sonographers who were unaware oftheir infertility history or oftheir laparoscopy and hysterosalpingography diagnoses. Interventions: Transvaginal 3-D US evaluation over a 10- to 15-minute duration. Main Outcome Measures: Three-dimensional imaging was successful in all cases. Results: Sonographers identified a Mullerian anomaly in all cases and came up with the correct classification in 11 of 12 cases. All patients with a normal uterus were identified correctly. Conclusions: In all patients with Mullerian anomalies, 3-D US examination of the endometrial cavity correlated with hysterosalpingography. In 91.6% of patients, 3-D US correlated with the external uterine configuration observed by laparoscopy. This technique may be used reliably in an office setting to diagnose and classify Mullerian anomalies. Fertil Steril 1996;65:523-8 Key Words: Three-dimensional ultrasound (3-D US), Mullerian anomalies, laparoscopy, HSG There is little doubt that certain congenital uterine anomalies can result in infertility and even sterility in some instances. There also is compelling evidence that a significant fraction of certain obstetric complications, such as second trimester abortion, premature birth, intrauterine growth retardation, fetal malpresentation, and retained placenta can be due to Mullerian tract anomalies 0). The most common uterine anomalies associated Received April 25, 1995; revised and accepted September 29, * Departmento de Obstetricia y Ginecologia, Facultad de Medicina, Universidad de Valencia. t Department of Obstetrics and Gynecology, Howard University College of Medicine. :j: Reprint requests: Newton G. Osborne, M.D., Ph.D., Department of Obstetrics and Gynecology, Howard University College of Medicine, 2041 Georgia Avenue, N.W., Washington, D.C (FAX: ). with obstetric failure are septate, bicornuate, and didelphic uteri. Conventional ultrasound (US) has been suggested as a screening or even diagnostic test for the detection of these uterine malformations (2). However, at present, accurate diagnosis can be made only by an evaluation of the uterine cavity with hysteroscopy or hysterosalpingography (HSG) along with observation of the external profile of the uterus at the time of laparotomy or laparoscopy. Because many women with these anomalies remain undiagnosed in their lifetime, the real incidence and prevalence of Mullerian tract malformations is difficult to estimate. The introduction of three-dimensional (3-D) US as a diagnostic instrument in reproductive medicine has added a new dimension to US examination of the uterus (3). The purpose of this study was to evaluate the accuracy of 3-D US as a detecting instrument for uterine anomalies previously diagnosed by HSG and laparoscopy. Raga et al. Congenital Mullerian anomalies 523

MATERIALS AND METHODS Forty-two patients either with normal uteri or with documented Mullerian tract anomalies who attended the Infertility Clinics of the Department of Obstetrics and Gynecology at

Uterine malformations were classified according to The American Fertility Society classification system (4). The patients were middle-class married Spanish women with a history of primary infertility.

2 MATERIALS AND METHODS Forty-two patients either with normal uteri or with documented Mullerian tract anomalies who attended the Infertility Clinics of the Department of Obstetrics and Gynecology at the University of Valencia School of Medicine gave their informed consent to participate in a transvaginal pelvic 3-D US examination study. Uterine malformations were classified according to The American Fertility Society classification system (4). The patients were middle-class married Spanish women with a history of primary infertility. Although the patients had a mean age of 26 years (range 23 to 34 years), cases and controls were not matched for age. All patients had been evaluated previously by laparoscopy and HSG. However, the sonographers (F.R. and F.B.-M.) were blinded to the patients's histories and to the diagnoses that resulted from the endoscopic and radiologic examinations. Normal patients and patients with Mullerian anomalies who agreed to be screened by 3-D US were advised not to disclose their diagnoses to the sonographers. All patients were evaluated by 3-D US when they came for follow up during the luteal phase of a menstrual cycle that occurred 2: 1 month after evaluation by HSG and laparoscopy. The equipment used for transvaginal 3-D examinations was a Combison 530 US machine (Kretz Technik, Zipf, Austria). Patients were instructed to empty their bladders before the transvaginal 3-D US examination. The US vaginal probe was protected with a sterile condom before the examination. To take advantage of the better contrast resolution of the secretory endometrium during evaluation of the uterine cavity, all patients were examined during the luteal phase. The transvaginal 3-D US diagnoses were then compared with the diagnoses obtained by HSG and endoscopy. The 3-D US image is generated by superimposing the programmed volume box over the two-dimensional (2-D) US scan image of the uterus. After the desired 2-D image is chosen for study, the volumetric rotor is set into operation. The vaginal transducer then performs a sweep of transversal sections that are to be stored in the US computer. The computer integrates the images and enables the sonographer to view three planes simultaneously on a monitor screen by generating a plane (plane C) that is perpendicular to the transducer. Once the perpendicular plane (plane C) to the transducer is obtained, the calculated 3-D image with the complete volume scan is stored in a removable computer disk. This scanning procedure takes between 2 and 10 seconds, depending on the size of the volume box. At this point, the examination ofthe patient is complete. At a later 524 Raga et al. Congenital Mullerian anomalies Figure 1 (A), Three-dimensional sonogram of a normal uterus in three perpendicular planes. Frontal section (upper left), median sagittal section (upper right), horizontal section (lower left). The diagrammatic presentation (lower right) displays the position of the plane in the scanned 3-D volume. (B), Three-dimensional transparent-maximum reconstruction of the normal uterine cavity. time the sonographer can select any of the stored sections for detailed analysis. All sections stored in the computer disk are available for study. Three-dimensional images are generated only when the three planes (frontal section, median sagittal section, and horizontal section) for any section are integrated and displayed on the screen (Fig. 1A). It is possible to rotate and translate any plane of the volume stored so that a multitude of images of the uterine cavity, uterine contour, and cervix can Fertility and Sterility

be generated. The sonographer in this way is then able to form a 3-D plastic image of the uterine cavity (endometrial echo), of the uterine contour, or of the cervix.

The number of slices (1 to 200) and the rotation angle (1 to 360 ) also must be defined.

3 be generated. The sonographer in this way is then able to form a 3-D plastic image of the uterine cavity (endometrial echo), of the uterine contour, or of the cervix. To generate a final 3-D image of the uterine cavity, for example, a threshold has to be defined up to which echogenicity should be taken for reconstruction ofthe uterine cavity. The number of slices (1 to 200) and the rotation angle (1 to 360 ) also must be defined. The time taken for image definition varies between 20 and 25 seconds for each slice selected so that the entire process of scanning and calculating takes between 5 and 10 minutes, depending on the number of slices and rotation angle selected. Depending on the structure to be studied, different 3-D modes can be elaborated. The surface reconstruction mode allows study of the outer contour or profile of the uterus so that it is possible to determine whether there is a sagittal notch on the fundus or to measure the depth of the myometrial spur in the wall that separates the two hemicavities of a didelphic uterus. The transparent maximum-minimum mode reveals objects with high echogenicity in the interior of an organ such as the uterus, giving striking images of the endometrial cavity, for example (Fig. 1B). RESULTS Three-dimensional reconstruction of the endometrial echo allowed accurate reconstruction of the uterine cavity in all cases. The sonographers were able to detect accurately by 3-D US all uterine abnormalities and were able to identify all patients with a normal uterus who were included in the study as controls. It was possible to determine on examination of the fundus whether there was a fundal cleft or whether there was a single or double cervix when an image of the cervical area was generated. Class II The 3-D reconstruction ofthe uterine cavity identified one patient with a unicornuate uterus by showing a single uterine cavity and the presence of a rudimentary horn on one side. The 3-D image revealed that the rudimentary horn communicated with the well developed uterine cavity (subclass a) (Fig. 2A). Class III There was one case of a uterus with two separate endometrial cavities and a duplicate cervix identified with 3-D US. A uterus didelphys results when there is a complete failure of MUllerian duct midline fusion (Fig.2B). Figure 2 (A), Three-dimensional uterine cavity reconstruction of a unicornuate uterus. The presence of an atretic rudimentary horn which communicates with the fully developed cavity is depicted clearly. (B), Transparent-maximum volumetric reconstruction of a didelphic uterus (class III). R, rudimentary horn be partial (subclass b) or may extend to the cervix (subclass a). The 3-D image shown in Figure 2B demonstrates the presence of two uterine cavities. In the four cases identified, the anomaly did not include the cervix, so the partial anomaly was diagnosed as subclass b. However, it was not possible to distinguish these bicornuate uteri from a septate uterus without 3-D US evaluation of the external uterine contour. Class IV Class V The incomplete fusion of the Mullerian ducts produces a bicornuate uterus. The failure to fuse may Failure of resorption of the medial segments of the paramesonephric ducts after fusion results in Raga et al. Congenital Mallerian anomalies 525

3A). We were able to identify correctly all but one of the external uterine contours previously detected during laparoscopy.

4 Class VI The 3-D US reconstruction of the endometrial cavity of an arcuate uterus demonstrates a concavity in the uterine fundus (Fig. 3B). The surface reconstruction mode was used simultaneously with imaging of three perpendicular planes for 3-D US diagnostic evaluation of uterine contour along with an US tomography of the uterus (Fig. 3A). We were able to identify correctly all but one of the external uterine contours previously detected during laparoscopy. The only false-negative external contour 3-D US diagnosis was a partial bicornuate uterus identified as having a normal external uterine contour by 3-D imaging due to the presence of a strategically located small fundal leiomyoma. DISCUSSION Figure 3 (A), Ultrasound tomography in frontal section ofbicornuate uterus. The 3-D reconstruction of the uterine profile demonstrates the presence of a large fundal cleft (arrows) as well as the typical two endometrial cavities (arrowheads) of a bicornuate uterus. (B), Three-dimensional reconstruction of the uterine cavity of the patient with an arcuate uterus. Observe the concavity (arrows) at the uterine fundus. a septate uterus. The septum can project slightly (partial, subclass b) from the top of the uterine fundus or it can extend to the cervical os (complete, subclass a). As was the case with bicornuate uteri, two distinct endometrial cavities with a single cervix were demonstrated by 3-D imaging. In four cases the defect was partial (subclass b) (Fig. 3A). In one case the anomaly reached the cervicoisthmic region (subclass a). Nevertheless, we were unable to differentiate septate uteri from bicornuate uteri without 3-D US evaluation of the uterine contour. 526 Raga et al. Congenital Mullerian anomalies Although several authors report the incidence of uterine malformations to be between 1 and 10 per 1,500 patients (5, 6), the actual incidence of Mullerian duct anomalies is unknown. The discrepancies in incidence reported may be due to the use ofinaccurate diagnostic methods in some cases and in part to the fact that many of these defects remain undiagnosed during the lifetime of many women (7). Many methods are used to identify Mullerian tract defects. Among these are a simple pelvic examination, HSG, hysteroscopy, laparoscopy, 2-D US, hysterosalpingosonography, and even magnetic resonance imaging (MRI) (1, 8-10). There is general agreement that, with the possible exception of class I anomalies in which diagnosis is evident with a speculum examination, a simple pelvic examination is of limited diagnostic value. Hysterosalpingography provides valuable information about the endometrial cavity and about tubal patency. However, the usefulness of HSG is limited as a technique to classify uterine anomalies because it does not provide definitive information about the external contour of the uterus. The radiologic image of the architecture of the endometrial cavity cannot differentiate reliably between didelphys, bicornuate, and septate uteri. Hysteroscopy allows both direct visualization of the uterine cavity and operative intervention when used to remove septae. However, as is the case with HSG, it cannot be used to evaluate the external contour of the uterus reliably. Laparoscopy is an invasive surgical procedure that usually requires general anesthesia, but at present is essential to assess the uterine contour. It is also very helpful for the evaluation of tubal patency and for the assessment of other associated pelvic or abdominal pathology. Fertility and Sterility

By its ability to delineate both internal and external uterine architecture, MRI may provide an alternative diagnostic method for the evaluation ofmullerian tract anomalies.

5 By its ability to delineate both internal and external uterine architecture, MRI may provide an alternative diagnostic method for the evaluation ofmullerian tract anomalies. However, it is an expensive modality that is ill suited for office practice. Ultrasound has the advantages of minimal invasiveness, relatively low cost, and ease of performance. Several investigators have reported that 2-D US is a useful technique for the evaluation of Mullerian tract anomalies (2, 9-12). Transabdominal US was the first echographic technique used for this purpose. However, transvaginal sonography, because of an ability to be closer to the pelvic organs, allows better anatomic delineation of pelvic structures in addition to providing images with better contrast and detail resolution. Hysterosalpingosonography, a new diagnostic US technique, allows visualization of the endometrial cavity in addition to allowing evaluation of fallopian tube patency without exposure to radiation (13). Obstetrics and gynecology is likely to undergo revolutionary changes in diagnostic imaging with the arrival of 3-D US (3). Preliminary reports have focused on the diagnostic potential of this technique in reproductive medicine (14). With the use of abdominal 2-D US, the uterine cavity could not be depicted clearly in up to 35% of cases (2). In the present study transvaginal 3-D volume scanning identified accurately uterine cavity anomalies in all cases. Although transvaginal probe 2-D sonography is an excellent screening examination for the presence of intrauterine pathology in women presenting with abnormal uterine bleeding (15), it does not seem to be as effective as 3-D US for the diagnosis of Mullerian anomalies. Jurkovic et al. (16) reported recently that, with the ability to view three orthogonal planes, they were able to obtain an image equal to that of HSG and superior to transvaginal 2-D US images. Although the authors did not use 3-D reconstruction in their study, they reported a 98% sensitivity and 100% specificity using three orthogonal planes when viewing normal uteri, compared with 88% sensitivity and 94% specificity with transvaginal 2-D US. With arcuate uteri the 3-D sensitivity and specificity were 100% compared with 67% and 94%, respectively, for transvaginal 2-D US. With major Mullerian anomalies, the sensitivity and specificity of 3-D US were both 100% compared with 100% sensitivity and 95% specificity for transvaginal 2-D sonography. However, the positive predictive value reported for major anomalies was 100% for 3- D compared with 50% for 2-D sonography. The report of Jurkovic et al. (16), which appeared after our work was submitted for publication, corresponds with our experience. In our hands, transvaginal 2- D US detected Mullerian anomalies in only 75% of cases. With transvaginal 2-D US, we detected only three of five septate uteri and we missed the arcuate uterus. With abdominal 2-D US, visualization ofthe external contour of the uterus is reported to be adequate in 90% of cases (2). In the present study, 3-D US described accurately the uterine contour in 30 patients with normal uteri and in 11 of 12 patients with an anomaly of uterine contour. The only misdiagnosis was due to the presence of a strategically located uterine myoma. One ofthe disadvantages of 3-D US compared with HSG and laparoscopy for the diagnosis of Mullerian tract anomalies is the lack of information about tubal patency with plain US techniques. However, this disadvantage is likely to be overcome with development, modifications, or improvements of new techniques like hysterosalpingosonography. In summary, we were able to identify all patients with uterine malformations when 3-D US was used to examine both the cavity and external contour of the uterus. In all but one case, we were able to classify patients with Mullerian anomalies according to the American Fertility Society classification system (4), including classification of a minor arcuate malformation of the uterus in one of the patients. The only false-negative contour was that of a bicornuate subclass b uterus classified by 3-D US as having a normal external contour, thus classifying it as a class b septate uterus. Three-dimensional US is a noninvasive imaging technique with the ability to generate accurate images of the endometrial cavity and of the external contour of the uterus. We believe that the advantages of this technique along with the capacity to create 3-D spatial reconstructions of pelvic organs is certain to make 3-D US the procedure of choice for the study of congenital uterine anomalies. REFERENCES 1. Acien P. Reproductive performance of women with uterine malformations. Hum Reprod 1993;8: Nicolini U, Bellotti M, Bonazzi B, Zamberletti D, Candiani GB. Can ultrasound be used to screen uterine malformations? Fertil SteriI1987;47: Steiner H, Staudach A, Spitzer D, Schaffer HI. Three-dimensional ultrasound in obstetrics and gynecology: techniques, possibilities and limitations. Hum Reprod 1994;9: The American Fertility Society. The American Fertility Society classification of adnexal adhesions, distal tubal occlusion, tubal occlusion secondary to tubal ligation, tubal pregnancies, Mullerian anomalies and intrauterine adhesions. Fertil Steril 1988; 49: Green LK, Harris RE. Uterine anomalies. Frequency of diagnosis and associated obstetric complications. Obstet Gynecol 1976;47: Heinonen PK, Saarikoski S, Pystynen P. Reproductive performance of women with uterine anomalies. Acta Obstet Gynecol Scand 1982;61: Raga et al. Congenital Mullerian anomalies 527

6 7. Simon C, Martinez L, Pardo F, TortajadaM, Pellicer A. Mullerian defects in women with normal reproductive outcome. Fertil SterilI991;56:1l Letterie GS, Wilson J, Miyazawa K. Magnetic resonance imaging of Mullerian tract abnormalities. Fertil Steril 1988; 50: Fedele L, Ferrazzi E, Dorta M, Vercellini P, Candiani GB. Ultrasonography in the differential diagnosis of "double" uteri. Fertil Steril 1988;50: Randolph JR Jr, Ying YK, Maier DB, Schmidt CL, Riddick DH. Comparison of real-time ultrasonography, hysterosalpingography, and laparoscopy/hysteroscopy in the evaluation of uterine abnormalities and tubal patency. Fertil Sterill986; 46: Masri MN, Setchell ME, Chard T. Transvaginal ultrasound for diagnosis of uterine malformations. Br J Obstet Gynaecol 1990; 97: Pellerito JS, McCarthy SM, Doyle MB, Glickman MG, De- Cherney AH. Diagnosis of uterine anomalies: relative accuracy ofmri, endovaginal sonography, and hysterosalpingography. Radiology 1992; 183: Bonilla-Musoles F, Simon C, Serra V, Sampaio M, Pellicer A. An assessment of hysterosalpingosonography (HSSG) as a diagnostic tool for uterine cavity defects and tubal patency. J Clin Ultrasound 1992;20: Feichtinger W. Transvaginal three-dimensional imaging. Ultrasound Obstet GynecolI993;3: Indman PD. Abnormal uterine bleeding: accuracy of vaginal probe ultrasound in predicting abnormal hysteroscopic findings. J Reprod Med 1995;40: Jurkovic D, Geipel A, Gruboeck K, Jauniaux E, Natucci M, Campbell S. Three-dimensional ultrasound for the assessment of uterine anatomy and detection of congenital anomalies: a comparison with hysterosalpingography and two-dimensional sonography. Ultrasound Obstet Gynecol 1995; 5: Raga et a!. Congenital Mullerian anomalies Fertility and Sterility

MULLERIAN DUCT ANOMALY: A CASE REPORT

MULLERIAN DUCT ANOMALY: A CASE REPORT MULLERIAN DUCT ANOMALY: A Sunny Goyal 1, Ankur Aggarwal 2, Hemant Kumar Mishra 3, Tushar Prabha 4, Vipin kumar Bakshi 5 HOW TO CITE THIS ARTICLE: Sunny Goyal, Ankur Aggarwal, Hemant Kumar Mishra, Tushar