AJR AJR Reprints & E-prints Available. Order Today!
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

This Article
Right arrow Abstract Freely available
Right arrow Figures Only
Right arrow Full Text (PDF)
Right arrow CME/SAM Credit
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Dykes, T. M.
Right arrow Articles by Dodson, W.
Right arrow Search for Related Content
PubMed
Right arrow Articles by Dykes, T. M.
Right arrow Articles by Dodson, W.
Social Bookmarking
Add to CiteULike   Add to Complore   Add to Connotea   Add to Del.icio.us   Add to Digg   Add to Reddit   Add to Technorati  
What's this?
DOI:10.2214/AJR.06.0821
AJR 2007; 189:S1-S10
© American Roentgen Ray Society

Imaging of Congenital Uterine Anomalies: Review and Self-Assessment Module

Thomas M. Dykes1, Cary Siegel2 and William Dodson1

1 Department of Radiology, Pennsylvania State University, Milton S. Hershey Medical Center, 500 University Drive, P.O. Box 850, Hershey, PA 17033.
2 Mallinckrodt Institute of Radiology, Washington University, St. Louis, MO.

Received June 22, 2006; accepted after revision August 28, 2006.

 
Address correspondence to T. M. Dykes (tdykes{at}hmc.psu.edu).


Abstract
Top
 Abstract
INTRODUCTION
 EDUCATIONAL OBJECTIVES
 REQUIRED ACTIVITIES
 RECOMMENDED READING
 INSTRUCTIONS
 BACKGROUND
 References
 
Objective

The educational objectives of this continuing medical education activity are for the reader to exercise, self-assess, and improve skills in diagnostic radiology with regard to the interpretation of hysterosalpingograms and magnetic resonance imaging (MRI) of the female pelvis in the evaluation of recurrent reproductive failure caused by congenital uterine anomalies.

Conclusion

This article reviews the common congenital uterine anomalies, characteristic imaging features of each anomaly by hysterosalpingography and MRI, and the clinical importance of diagnosing and properly categorizing each anomaly.

Keywords: congenital malformations • genitourinary tract imaging • hysterosalpingography • MR imaging • müllerian duct anomaly • women's imaging


INTRODUCTION
Top
 Abstract
 INTRODUCTION
EDUCATIONAL OBJECTIVES
 REQUIRED ACTIVITIES
 RECOMMENDED READING
 INSTRUCTIONS
 BACKGROUND
 References
 
This self-assessment module on imaging of congenital uterine anomalies has an educational component and a self-assessment component. The educational component consists of an AJR Teaching File article and an educational exhibit from the 2007 ARRS Annual Meeting, included as supplemental data to this article, that the participant should review. In addition, there are four interactive case scenarios presented in this article. The self-assessment component consists of eight multiple-choice questions with solutions. All of these materials are available on the ARRS Website (www.arrs.org). To claim CME and SAM credit, each participant must register and then enter his or her responses to the questions online.


EDUCATIONAL OBJECTIVES
Top
 Abstract
 INTRODUCTION
 EDUCATIONAL OBJECTIVES
REQUIRED ACTIVITIES
 RECOMMENDED READING
 INSTRUCTIONS
 BACKGROUND
 References
 
Upon completion of this educational activity the participant will be able to:

  1. Recognize the common congenital uterine anomalies;
  2. Appropriately categorize the anomaly based on the American Fertility Society (now known as the American Society for Reproductive Medicine) classification system;
  3. Understand the clinical significance of each anomaly;
  4. Make appropriate recommendations for additional imaging or intervention.


REQUIRED ACTIVITIES
Top
 Abstract
 INTRODUCTION
 EDUCATIONAL OBJECTIVES
 REQUIRED ACTIVITIES
RECOMMENDED READING
 INSTRUCTIONS
 BACKGROUND
 References
 
(available at www.arrs.org or www.ajronline.org)

  1. Chaudry S. Infertility in a young woman. AJR 2007; 189: S11–S12.
  2. Data supplement included with [1]. http://www.ajronline.org/cgi/content/full/189/3_Supplement/S11/DC1
  3. Four interactive case scenarios presented in this article.


RECOMMENDED READING
Top
 Abstract
 INTRODUCTION
 EDUCATIONAL OBJECTIVES
 REQUIRED ACTIVITIES
 RECOMMENDED READING
INSTRUCTIONS
 BACKGROUND
 References
 
(available at www.arrs.org)

  1. Lindheim SR, Sprague C, Winter TC III. Hysterosalpingography and sonohysterography: lessons in technique. AJR 2006; 186: 24–29
  2. Dalfó AR, Úbeda B, Úbeda A, Monzón M, Rotger R, Ramos R, Palacio A. Diagnostic value of hysterosalpingography in the detection of intrauterine abnormalities: a comparison with hysteroscopy. AJR 2004; 183: 1405–1409
  3. Troiano RN, McCarthy SM. Müllerian duct anomalies: imaging and clinical issues. Radiology 2004; 233: 19–34
  4. O'Neill MJ, Yoder IC, Connolly SA, Mueller PR. Imaging evaluation and classification of developmental anomalies of the female reproductive system with an emphasis on MR imaging. AJR 1999; 173: 407–416


INSTRUCTIONS
Top
 Abstract
 INTRODUCTION
 EDUCATIONAL OBJECTIVES
 REQUIRED ACTIVITIES
 RECOMMENDED READING
 INSTRUCTIONS
BACKGROUND
 References
 

  1. Complete the required activities and review the accompanying electronic exhibit presented as supplemental data to [1].
  2. Visit www.arrs.org and go to the left-hand menu bar under Publications/Journals/SAM articles.
  3. Using your member login, order the online SAM as directed.
  4. Follow the online instructions for entering your responses to the self-assessment questions and complete the test by answering the questions online.


BACKGROUND
Top
 Abstract
 INTRODUCTION
 EDUCATIONAL OBJECTIVES
 REQUIRED ACTIVITIES
 RECOMMENDED READING
 INSTRUCTIONS
 BACKGROUND
References
 
The true incidence of congenital uterine anomalies, also referred to as müllerian duct anomalies, is not accurately known, with reports ranging from a frequency of one of 10 patients to one of 1,600 patients [1]. Summation of data from multiple studies suggests the prevalence of these anomalies in women with a history of multiple pregnancy loss is approximately 3% [2]. Approximately 25% of women with müllerian duct anomalies have associated reproductive problems, compared with 10% of the general population [2].


Figure 1
View larger version (18K):
[in this window]
[in a new window]
[as a PowerPoint slide]
  		Fig. 1 —Classification system of müllerian duct anomalies used by the American Society for Reproductive Medicine [3] (Reprinted with permission). DES=diethylstilbestrol.

 


Figure 2
View larger version (158K):
[in this window]
[in a new window]
[as a PowerPoint slide]
  		Fig. 2A —Sagittal localizer sequence in healthy patient shows imaging planes required for T2 imaging sequences. T2-weighted sagittal MR localizer sequence.

 


Figure 3
View larger version (148K):
[in this window]
[in a new window]
[as a PowerPoint slide]
  		Fig. 2B —Sagittal localizer sequence in healthy patient shows imaging planes required for T2 imaging sequences. Imaging plane prescription (white lines) for T2-weighted MR sequence parallel to the uterine long-axis.

 


Figure 4
View larger version (194K):
[in this window]
[in a new window]
[as a PowerPoint slide]
  		Fig. 2C —Sagittal localizer sequence in healthy patient shows imaging planes required for T2 imaging sequences. Imaging plane prescription (white lines) for T2-weighted MR sequence orthogonal to the uterine long-axis.

 
After recognition of a müllerian duct anomaly, it is important to properly classify the anomaly because the associated risks of poor pregnancy outcome and treatment can vary widely between the anomalies. The most common classification system for congenital uterine anomalies is that used by the American Society for Reproductive Medicine [3]. This continuing medical education activity focuses on types II–VI anomalies that are illustrated in Figure 1. It is important to be mindful that this classification system is only a framework for understanding and reporting these anomalies and that all anomalies will not necessarily fit completely into one of the categories. In those situations in which the identified anomaly does not fit into a category, it is more important to accurately describe the component parts of the anomaly rather than force it into a category that does not fully describe it [2].

A basic understanding of the embryologic model of uterovaginal development assists in understanding the anomalies that this activity addresses. Early in gestation, the distal segments of the paired müllerian ducts migrate caudally and medially, where they fuse inferiorly forming the uterovaginal canal [3, 4]. There is a residual, midline dividing septum that regresses in a caudal to cephalad direction resulting in a single-lumen uterovaginal canal. Incomplete development of one of the müllerian ducts, failure of fusion of the müllerian ducts, or failure of septal resorption can result in the types of anomalies presented [3, 4].

Type II anomalies (unicornuate) represent partial or complete unilateral hypoplasia of one of the müllerian ducts. Type III anomaly (uterus didelphus) results from near complete failure of müllerian duct fusion. Type IV anomaly (bicornuate uterus) is caused by incomplete fusion of the müllerian ducts at the level of the fundus. Type V anomaly (septate) is the result of partial or complete failure of septal resorption after fusion of the müllerian ducts. Last, type VI anomaly (arcuate) is a result of nearly complete septal resorption with a morphology that most closely approximates normal, and it is uncertain whether the arcuate uterus is a normal variant or a true anomaly [3].

Müllerian duct anomalies are often first identified on hysterosalpingography. Additional imaging studies such as MRI are typically required to further evaluate and characterize these anomalies as will be described in the scenarios of this educational activity. MRI has a reported accuracy up to 100% in evaluation of these anomalies [2].

Specific details concerning MRI sequences used in the evaluation of müllerian duct anomalies can be found in the literature [3]. In general terms, the basic imaging sequences needed are:


QUESTION 1

What type of congenital uterine anomaly is shown in Figure 3A, 3B, 3C?

  1. Bicornuate uterus.
  2. Unicornuate uterus without a rudimentary horn.
  3. Unicornuate uterus with a communicating, cavitary, rudimentary horn.
  4. Unicornuate uterus with a noncommunicating, cavitary, rudimentary horn.
  5. Partial septate uterus.

QUESTION 2

Concerning a unicornuate uterus, which of the following is correct?

  1. It is the most common of the müllerian duct anomalies.
  2. The minority of cases of unicornuate uterus have a rudimentary horn.
  3. It is associated with poor pregnancy outcomes.
  4. Unicornuate uterus with a rudimentary horn does not require surgical correction.
  5. It is least commonly associated with renal anomalies.

 


Figure 5
View larger version (109K):
[in this window]
[in a new window]
[as a PowerPoint slide]
  		Fig. 3A —Hysterosalpingogram for recurrent pregnancy loss. MRI obtained subsequent to hysterosalpingography. Hysterosalpingogram shows one uterine horn filled during injection (arrow).

 

Figure 6
View larger version (125K):
[in this window]
[in a new window]
[as a PowerPoint slide]
  		Fig. 3B —Hysterosalpingogram for recurrent pregnancy loss. MRI obtained subsequent to hysterosalpingography. Axial T2 MRI shows fusiform uterine cavity with typical trilaminar appearance of high-signal endometrium (star), low-signal junctional zone (long arrow), and intermediate-signal myometrium (short arrow) of uterus seen in right side of pelvis. This corresponds to cavity opacified on hysterosalpingogram.

 

Figure 7
View larger version (146K):
[in this window]
[in a new window]
[as a PowerPoint slide]
  		Fig. 3C —Hysterosalpingogram for recurrent pregnancy loss. MRI obtained subsequent to hysterosalpingography. Axial T2 MRI slightly lower in pelvis shows second endometrial cavity in left side of pelvis with high-signal endometrium (arrow). There is clear separation of this second cavity from more superior cavity by intermediate-signal myometrium.

 
Scenario 1
Clinical History
Hysterosalpingography was performed for recurrent pregnancy loss. MRI was obtained subsequent to hysterosalpingography.

Solution to Question 1
In this case only one uterine horn is filled on the hysterosalpingogram (Fig. 3A). Bicornuate uterus is characterized by two separate uterine horns filling during hysterosalpingography, with each horn having a fusiform appearance [4]. Septate and bicornuate uteri can have a similar appearance on a hysterosalpingogram and often cannot be differentiated by hysterosalpingography [4]. Option A is not the best response.

A unicornuate uterus is diagnosed on hysterosalpingography by filling of a small, fusiform uterine cavity that tapers at its apex and is often shifted to one side of the pelvis. It resembles one horn of a bicornuate uterus [4]. The presence of a rudimentary horn, if it is noncommunicating, cannot be determined by hysterosalpingography. Based on the hysterosalpingogram alone, this patient could have several possible anomalies. The patient could have a unicornuate uterus with a noncommunicating cavitary rudimentary horn, with a noncavitary rudimentary horn, or with no rudimentary horn [2]. Another possibility would include a uterus didelphus with only one cervix recognized and cannulated by the physician or a uterus didelphus with a blind hemivagina (not possible to see the second cervix). Last, a complete septate uterus with opacification of only one horn is a possibility, although this is much less likely than a unicornuate uterus or unrecognized uterus didelphus. The MRI (Fig. 3C) shows two separate uterine horns, clearly separated from one another by myometrial tissue. The smaller, left-sided uterine horn is a rudimentary horn. Option B is not the best response.

A communicating rudimentary horn of a unicornuate uterus should be identified as filling with contrast material on the hysterosalpingogram [4] or shown to communicate with the unicornuate horn on MRI. In this case, both hysterosalpingography and MRI clearly show that this rudimentary horn is noncommunicating. Option C is not the best response.

A rudimentary horn in a unicornuate uterus can be cavitary (containing endometrium) or noncavitary (containing fibrous tissue). When endometrium is present in the rudimentary horn (cavitary), high-signal zonal anatomy can be seen on MRI [2] (Fig. 3C). When endometrium is absent (noncavitary), generally the tissue is fibrous and low signal on T2-weighted MR images (Figs. 4A, 4B, 4C). As discussed previously, a rudimentary horn in a unicornuate uterus can also be communicating or noncommunicating with the unicornuate horn. The combination of the hysterosalpingography and MRI confirms that this patient has a noncommunicating rudimentary horn that has high signal intensity of endometrium (cavitary). Option D is the best response.


Figure 8
View larger version (151K):
[in this window]
[in a new window]
[as a PowerPoint slide]
  		Fig. 4A —Hysterosalpingogram and subsequent MRI. Hysterosalpingogram demonstrating filling of a left-sided unicornuate uterus (arrow).

 

Figure 9
View larger version (168K):
[in this window]
[in a new window]
[as a PowerPoint slide]
  		Fig. 4B —Hysterosalpingogram and subsequent MRI. Axial T2 MRI shows the unicornuate horn corresponding to that seen on the hysterosalpingogram (arrow). A low-signal-intensity fibrous rudimentary horn is seen in the right pelvis (arrowhead).

 

Figure 10
View larger version (168K):
[in this window]
[in a new window]
[as a PowerPoint slide]
  		Fig. 4C —Hysterosalpingogram and subsequent MRI. Coronal T2 MRI better demonstrates the noncavitary (no high-signal endometrium) right-sided rudimentary horn attached to the left-sided unicornuate horn (arrow).

 

A partial septate uterus would have two cavities filling on hysterosalpingogram, rather than one cavity as seen on this patient's examination. Option E is not the best response.

Solution to Question 2
Unicornuate uterus accounts for about 20% of müllerian duct anomalies [2]. It is not the most common anomaly. Option A is not the best response.

An isolated unicornuate horn (no rudimentary horn) is found in 35% of unicornuate uteri. Sixty-five percent of unicornuate uteri have a rudimentary horn [2]. Option B is not the best response.


Figure 11
View larger version (127K):
[in this window]
[in a new window]
[as a PowerPoint slide]
  		Fig. 5A —Two separate cervices visualized. Two separate injections on hysterosalpingography. Two cervices were visualized and cannulated separately. Opacified uterine horns do not communicate and are widely divergent (arrows).

 


Figure 12
View larger version (126K):
[in this window]
[in a new window]
[as a PowerPoint slide]
  		Fig. 5B —Two separate cervices visualized. Two separate injections on hysterosalpingography. Two cervices were visualized and cannulated separately. Opacified uterine horns do not communicate and are widely divergent (arrows).

 


Figure 13
View larger version (137K):
[in this window]
[in a new window]
[as a PowerPoint slide]
  		Fig. 6A —Two separate cervices visualized. T2 axial MRI show two, widely separate uterine horns in a patient with uterus didelphus (arrows).

 


Figure 14
View larger version (135K):
[in this window]
[in a new window]
[as a PowerPoint slide]
  		Fig. 6B —Two separate cervices visualized. T2 coronal MRI demonstrates separate, noncommunicating uterine horns (black arrows) and cervices (arrowheads) and duplicated vagina (white arrows).

 


Figure 15
View larger version (131K):
[in this window]
[in a new window]
[as a PowerPoint slide]
  		Fig. 7A —Woman with vaginal bleeding and abdominal pain. Uterus didelphus. Coronal T2 image demonstrates a dilated right-sided uterine horn (arrow) and a nondilated left-sided uterine horn (arrowhead).

 


Figure 16
View larger version (152K):
[in this window]
[in a new window]
[as a PowerPoint slide]
  		Fig. 7B —Woman with vaginal bleeding and abdominal pain. Uterus didelphus. Axial T2 image demonstrates the dilated right-sided uterine horn (arrow), nondilated left-sided uterine horn (arrowhead), and a massively dilated, obstructed right hemivagina (star).

 
Although a patient with a unicornuate uterus can have a normal pregnancy, there is an increased risk of pregnancy loss and premature labor [1, 2, 4]. Spontaneous abortion rates reportedly range between 41% and 62% and premature birth rates between 10% and 20%. Other complications include abnormal fetal lie and intrauterine growth restriction [2]. Option C is the best response.

A noncommunicating, cavitary (contains endometrium), rudimentary horn associated with a unicornuate uterus is important and once identified is usually surgically resected. Obstetric and nonobstetric complications associated with this anomaly include dysmenorrhea and hematometra at menarche, endometriosis, and ectopic pregnancy implanting in the rudimentary horn [2, 6]. Pregnancy implanting in the rudimentary horn usually has a disastrous outcome, with most resulting in uterine rupture [2, 6]. Communicating, cavitary rudimentary horns are also considered for surgical removal because a pregnancy that implants in the rudimentary horn rarely yields viable offspring [2]. Noncavitary (no endometrium), rudimentary horns would not have complications of hematometra, dysmenorrhea, and endometriosis. Although a patient potentially could still have an ectopic pregnancy implant in a Fallopian tube connected to a noncavitary, rudimentary horn, the current recommendation for this subset of patients is that they do not require surgical resection [2]. Most patients with rudimentary horns require surgical correction. Option D is not the best response.

Renal anomalies are more commonly associated with unicornuate uterus than with the other müllerian duct anomalies and are reported in up to 40% of cases. The associated renal anomaly is typically ipsilateral to the rudimentary horn [2]. Option E is not the best response.

Conclusion
The diagnosis in this case is unicornuate uterus with a noncommunicating, cavitary, rudimentary horn. The correct diagnosis hinges on proper integration of hysterosalpingography and pelvic MRI. The hysterosalpingogram identifies the typical appearance of a unicornuate uterus with opacification of a small, single uterine horn. The two other possible diagnoses, complete septate and unrecognized uterus didelphus, can be effectively excluded by combining the information available from hysterosalpingography and pelvic MRI. Both horns of a complete septate uterus will typically opacify on hysterosalpingography. On MRI, a complete septate uterus will have two separate and relatively symmetric uterine horns rather than a hypoplastic or absent uterine horn seen with unicornuate uterus. Only one uterine horn may fill on hysterosalpingography in an unrecognized uterus didelphus simulating a unicornuate uterus. However, MRI can definitively distinguish between these two diagnoses, with the uterus didelphus having two nonfused uterine horns and two cervices with only minor degrees of fusion at the cervices.

It is important to identify whether there is a rudimentary horn associated with the unicornuate uterus. An excellent imaging modality to accomplish this task is a pelvic MRI [2]. If MRI confirms that the patient has a unicornuate uterus with a rudimentary horn, it is important to determine whether the rudimentary horn has endometrial signal within it on T2 images (cavitary) or is fibrous tissue (noncavitary). Cavitary rudimentary horns generally require surgical removal for the reasons cited previously whereas noncavitary horns typically do not [2, 7]. It is also important to remember that müllerian duct anomalies are frequently associated with renal anomalies. Renal abnormalities are more commonly associated with unicornuate uterus than any other müllerian anomaly, and the most common abnormality is ipsilateral renal agenesis [2].

Scenario 2
Clinical History
No clinical history was available. During hysterosalpingography two separate cervices were visualized (Fig. 5A, 5B).

Solution to Question 3
A bicornuate uterus has two symmetric uterine cavities that fuse caudally and have some degree of communication between the two cavities, usually at the uterine isthmus. A complete bicornuate uterus has a separating cleft of tissue that extends to the internal cervical os, while lesser degrees of separation of the two uterine horns constitute a partial bicornuate uterus [2, 4]. This patient has two completely separate, noncommunicating uterine horns. An uncommon anomaly, the bicornuate, bicollis (two cervices) uterus will also have some degree of communication between the uterine horns in most cases [2]. Options A and B are not the best responses.


QUESTION 3

What type of congenital uterine anomaly does the patient in Figure 5A, 5B have?

  1. Complete bicornuate uterus.
  2. Partial bicornuate uterus.
  3. Uterus didelphus.
  4. Arcuate uterus.
  5. Partial septate uterus.

QUESTION 4

Concerning uterus didelphus, which of the following is correct?

  1. It is the most common congenital uterine anomaly.
  2. It can be confused with a unicornuate uterus on hysterosalpingograms.
  3. The minority of cases have an associated longitudinal vaginal septum.
  4. It is caused by complete failure of resorption of the uterovaginal septum.
  5. Symptoms are common in nonobstructive forms of uterus didelphus.

 

Uterus didelphys is caused by complete or near complete failure of fusion of the paired müllerian ducts during embryologic development. The result is two separate, symmetric uterine cavities with two cervices and no communication between the uterine cavities. Injection of contrast material into each separate cervix independently opacifies a uterine horn that resembles unicornuate uterus morphology [4]. Another uncommon anomaly that could have a similar appearance on hysterosalpingography is a complete septate uterus with a duplicated cervix [6]. MRI distinguishes between the complete septate which has a fused, external uterine fundal contour, and the uterus didelphys, with its widely separate uterine horns and duplicated cervices as shown in a different patient in Figures 6A and 6B. Option C is the best response.

Arcuate uterus is characterized on hysterosalpingography as having a mild, broad indentation at the superior aspect of a single uterine cavity. Although originally classified as a mild form of bicornuate uterus, newer classification systems place arcuate uterus into a separate category [1, 3, 8]. Arcuate uterus has little to no adverse impact on reproduction [1]. Option D is not the best response.

A partial septate uterus would have two cavities filling on hysterosalpingography injecting a single cervix. This case required injection of two separate cervices to opacify separate uterine cavities on hysterosalpingography. Option E is not the best response.

Solution to Question 4
Uterus didelphus constitutes only 5% of congenital uterine anomalies [2] and is not the most common müllerian anomaly. Option A is not the best response.

Uterus didelphus is associated with a longitudinal vaginal septum in nearly 75% of cases [2, 6]. These longitudinal vaginal septae may be further complicated by a transverse septum obstructing one hemivagina. This combination of factors can result in nonvisualization of one of the hemivagina and the associated cervix. The resultant injection of the only "visualized" cervix appears as a unicornuate uterus on hysterosalpingography [4]. MRI can help identify an obstructed uterus didelphus [7] (Figs. 7A and 7B). Option B is the best response.

Most (75%) of cases of uterus didelphys have an associated longitudinal vaginal septum [2]. So, Option C is not the best response.

Uterus didelphys is caused by near complete failure of fusion of the müllerian ducts, not failure of resorption once fusion has occurred [2, 3]. Option D is not the best response.

Patients with nonobstructive forms of uterus didelphus are usually asymptomatic, whereas patients who have obstructive forms develop symptoms at menarche and have complications related to retrograde menstrual flow, such as pelvic adhesions and endometriosis [2]. Option E is not the best response.

Conclusion
Uterus didelphus is caused by complete or near-complete failure of fusion of the paired müllerian ducts during embryologic development. The hysterosalpingography and images obtained after injection of each cervix are characteristic. MRI demonstrates two separate uteri with divergent horns, as expected with failure of fusion of the müllerian ducts. Two separate cervices are also seen with no or only minor degrees of fusion (Figs. 6A and 6B) [2].

As mentioned previously, obstructive symptoms can occur when the longitudinal vaginal septum is complicated by a transverse septum. These symptoms include hematometrocolpos and endometriosis [2]. Uterus didelphus does have an adverse impact on reproduction, similar to that of a unicornuate uterus [1, 9]. It has the highest possibility for successful pregnancy of the müllerian anomalies except for arcuate uterus [9].

Scenario 3
Clinical History
Two uterine cavities were seen on a previous hysterosalpingogram after cannulating a single cervix.

Solution to Question 5
Uterus didelphus is characterized on MRI as having two separate uterine cavities that are widely separate and two cervices with little to no fusion. There is also commonly a vaginal septum. There is no communication between the uterine cavities [2]. Figure 8A, 8B shows two uterine cavities, but there is clearly communication between the cavities in the lower uterine segment and a single cervix. Option A is not the best response.


Figure 17
View larger version (167K):
[in this window]
[in a new window]
[as a PowerPoint slide]
  		Fig. 8A —Two uterine cavities seen on previous hysterosalpingogram after cannulating single cervix. T2 axial MRI shows two separate uterine cavities (white arrows) separated by deep fundal cleft (black arrow) at uterine fundus. Single cervix is present.

 

Figure 18
View larger version (149K):
[in this window]
[in a new window]
[as a PowerPoint slide]
  		Fig. 8B —Two uterine cavities seen on previous hysterosalpingogram after cannulating single cervix. T2 coronal MRI shows communication of two uterine cavities in lower uterine segment (arrow).

 

QUESTION 5

What type of congenital uterine anomaly is shown in Figure 8A, 8B?

  1. Uterus didelphus.
  2. Partial septate uterus.
  3. Complete septate uterus.
  4. Partial bicornuate uterus.
  5. Complete bicornuate uterus.

 

Septate uterus, either partial or complete, is characterized on MRI as having two separate uterine cavities, but the external fundal contour is unified. This is seen on MRI has having either a convex or flat uterine fundal contour, or a shallow concavity (< 1 cm depth) [2]. This case clearly has a deep fundal cleft with two separate uterine cavities. Neither option B nor C is the best response.

Partial bicornuate uterus is defined as having a tissue cleft partially separating the uterine horns with communication of the two horns generally in the lower uterine segment. Additionally, a bicornuate uterus will have a nonunified, external fundal contour seen as a cleft at least 1 cm deep in the external fundal contour [2]. This case shows both of these imaging features. Option D is the best response.


QUESTION 6

Concerning a bicornuate uterus, which of the following is correct?

  1. It is more common than a septate uterus.
  2. It can reliably be differentiated from a septate uterus on routine hysterosalpingography.
  3. It is differentiated from a septate uterus on MRI by identifying myometrium between the two uterine horns.
  4. It is often complicated by cervical insufficiency.
  5. It is caused by failure of resorption of the uterovaginal septum during fetal development.

 

Complete bicornuate uterus is defined as having an external uterine fundal contour characteristic of bicornuate morphology (described above), with separation of the two uterine horns by a tissue cleft extending to the internal cervical os [6]. This case shows separation of the two uterine horns only to the lower uterine segment, where communication occurs. This tissue cleft does not extend to the internal cervical os. Option E is not the best response.

Solution to Question 6
Bicornuate uterus accounts for approximately 10% of müllerian duct anomalies, whereas septate uterus accounts for about 55% of cases [2]. Option A is not the best response.

Two separate uterine horns are seen in both the septate uterus and the bicornuate uterus on hysterosalpingography [4]. Typically, the uterine horns are separated by a wide angle (105° or more) in a bicornuate uterus on hysterosalpingography, whereas with a septate uterus the uterine horns are separated by a narrow angle (75° or less) [4]. However, there is a large amount of overlap in the findings of a septate and bicornuate uterus on routine hysterosalpingograms, and the reported diagnostic accuracy of hysterosalpingography alone in distinguishing between these two anomalies is 55% [2]. Thurmond et al. [10] reported a technical innovation in hysterosalpingography, the "uterine push–pull technique." Gentle cycles of manual traction-relaxation were performed using a cervical vacuum cup after free-spill of contrast material from at least one Fallopian tube in an attempt to move the intraperitoneal contrast around the uterine fundus, allowing visualization of the fundal contour. This technique was not used with either a balloon catheter or a tenaculum and also was not used when patients had tubal disease or other technical problems that limited the flow of contrast into the peritoneal cavity. Forty-four percent of the patients in their series could not have this technique applied (for technical and patient-related reasons), and 58% of the patients imaged using this technique did not show adequate visualization of the fundal contour. One percent of the patients in their seriesexperienced moderate to severe pain and 4% had self-limited vasovagal symptoms when this technique was attempted. Whether this technique is used or not, when a double uterine horn is identified on hysterosalpingography, and if intraperitoneal contrast material is seen outlining the uterine fundus, close inspection of the fundal contour may allow distinction between septate and bicornuate uterus (Fig. 9). Option B is not the best response.


Figure 19
View larger version (135K):
[in this window]
[in a new window]
[as a PowerPoint slide]
  		Fig. 9 —Double uterine horn is present on hysterosalpingogram. Free spill of contrast into peritoneal cavity allows direct visualization of convex fundal contour (arrowheads) confirming that this is septate uterus, not bicornuate.

 

MRI is an excellent imaging tool that can reliably differentiate between bicornuate and septate uterine anomalies. The characteristic diagnostic feature of a bicornuate uterus is an external uterine fundal cleft at least 1 cm deep seen on images that are acquired parallel to the long axis of the uterus. A septate uterus will have either a shallow cleft (< 1 cm depth), a flat contour, or a convex external uterine contour [2]. The presence or absence of myometrial tissue between the two uterine horns is not helpful in distinguishing between septate and bicornuate uteri because it may be absent or present in a septate uterus [2]. Option C is not the best response.

Bicornuate uterus is reported to have the highest association with cervical insufficiency of the müllerian duct anomalies, reaching 38%, and prophylactic cerclage may be indicated to improve fetal survival rates [2]. Option D is the best response.

Bicornuate uterus is caused embryologically by a defect in müllerian duct fusion not resorption [1, 3]. Option E is not the best response.

Conclusion
Bicornuate uterus is caused by a defect in müllerian duct fusion during embryologic development. Although the diagnosis may be suggested on hysterosalpingography by showing opacification of two uterine horns that are divergent from each other, there is significant overlap between this appearance and a septate uterus. MRI is an excellent tool in both diagnosing bicornuate uterus and differentiating it from a septate uterus. Differentiation of bicornuate from septate uterus has important implications for both prognosis and therapy. Septate uterus has higher rates of spontaneous abortions, worse obstetric outcomes, and lower fetal survival rates compared with bicornuate uterus [1, 2, 6]. Septate uterus is often treated with hysteroscopic resection of the septum [1, 2, 4]. Bicornuate uterus generally does not require surgical intervention [2].

The key imaging feature to assess is the external contour of the uterine fundus. A deep fundal cleft (> 1 cm) is consistent with a bicornuate uterus, whereas a shallow cleft, a flat contour, or a convex fundal contour is diagnostic of a septate uterus. It is also important to determine whether a bicornuate uterus is complete or partial because spontaneous abortion and preterm labor are reportedly higher with a complete bicornuate compared to partial [2].

Scenario 4
Clinical History
Recurrent pregnancy loss.

Solution to Question 7
On hysterosalpingography, a bicornuate or a septate uterine anomaly is seen (Fig. 10A). The MRI (Fig. 10B) shows a flat to mildly convex external uterine fundal contour that is characteristic of a septate uterus. Complete septate uterus is characterized by a septum that separates the two uterine cavities extending to the external cervical os [2]. T2 MR images transverse to the cervix may help in determining whether a septation is partial or complete [5]. This case shows a septum, but it extends only to the lower uterine segment on the hysterosalpingogram and the MRI. Also, there is clearly no septation in the cervix on the MRI. Option A is the best response.


Figure 20
View larger version (103K):
[in this window]
[in a new window]
[as a PowerPoint slide]
  		Fig. 10A —Woman with recurrent pregnancy loss. Hysterosalpingogram with injection of one cervix demonstrates two separate uterine cavities with communication in lower uterine segment.

 

Figure 21
View larger version (154K):
[in this window]
[in a new window]
[as a PowerPoint slide]
  		Fig. 10B —Woman with recurrent pregnancy loss. Axial T2 MRI shows flat external uterine fundal contour (arrowheads).

 
The uterine septum in this case extends only to the lower uterine segment, not to the external cervical os as would be characteristic of a complete septate uterus. Option B is not the best response.

Although the possibility of bicornuate uterus could be considered on the hysterosalpingogram, the MRI clearly demonstrates a uterine fundal contour consistent with a septate uterus, not bicornuate morphology. Neither Option C nor D are the best responses.

Uterus didelphus requires cannulation of two separate cervices to opacify the separate uterine cavities, unlike this patient's hysterosalpingogram. Also, the MRI appearance of a uterus didelphus shows complete lack of unification of the uterine horns unlike this patient's MRI (Fig. 10B) showing uterine fundal configuration consistent with unification. Option E is not the best response.


QUESTION 7

What type of congenital uterine anomaly is shown in Figure 10A, 10B, 10C?

  1. Partial septate uterus.
  2. Complete septate uterus.
  3. Partial bicornuate uterus.
  4. Complete bicornuate uterus.
  5. Uterus didelphus.

 


Figure 22
View larger version (122K):
[in this window]
[in a new window]
[as a PowerPoint slide]
  		Fig. 10C —Woman with recurrent pregnancy loss. Axial T2 MRI lower in pelvis demonstrates single cervix with no septation (arrowhead).

 
Solution to Question 8
The septate uterus is characterized by uterine fundal unification, typically seen on MRI as a flat or convex fundal contour. Bicornuate uterus is characterized on MRI as having uterine fundal concavity greater than 1 cm depth. It is, however, important to not interpret slight degrees of fundal concavity (< 1 cm) as indicative of partial bicornuate morphology. Septate uterus can have slight degrees of uterine fundal concavity [2]. Option A is not the best response.

Septate uterus constitutes about 55% of congenital uterine anomalies [2]. It is the most common anomaly. Option B is not the best response.

Septate uterus is associated with high rates of spontaneous abortions, ranging in the literature from 26% to 94% [1, 2, 4]. The most common congenital uterine anomaly identified in patients presenting with recurrent, spontaneous abortions is a septate uterus [4]. Septate uterus also has the worst obstetric outcomes of the congenital uterine anomalies, with increased premature birth rates and lower fetal survival rates [2, 4]. Option C is the best response.

Repair of a septate uterus is best done by hysteroscopic resection of the septum [1, 2, 4]. Transabdominal metroplasty is used for correction of bicornuate anomalies [2, 6] when clinically indicated. Bicornuate uterus, however, generally does not require surgical intervention [2]. Option D is not the best response.

Resection of the septum does improve reproductive outcome for patients with a septate uterus experiencing repeated, spontaneous abortions [2]. Option E is not the best response.


QUESTION 8

Concerning a septate uterus, which of the following is correct?

  1. Diagnosis is excluded on MRI by any concavity in the external uterine fundal contour.
  2. It is the least common of the congenital uterine anomalies.
  3. It is associated with the worst obstetric outcomes of the müllerian duct anomalies.
  4. Surgical correction requires a transabdominal metroplasty.
  5. Resection of the septum does not improve reproductive outcome.

 


Figure 23
View larger version (151K):
[in this window]
[in a new window]
[as a PowerPoint slide]
  		Fig. 11A —No clinical history available. T2 axial MRI: Septate uterus, complete. Arrowhead demonstrates shallow (< 1 cm depth) fundal concavity, still consistent with septate morphology.

 


Figure 24
View larger version (152K):
[in this window]
[in a new window]
[as a PowerPoint slide]
  		Fig. 11B —No clinical history available. T2 axial MRI slightly lower in pelvis. White arrow shows long fibrous (low signal) septum extending to external cervical os consistent with complete septate uterus. Arrowhead shows shallow cleft in fundus, still consistent with septate uterus.

 
Conclusion
Hysterosalpingograms cannot reliably differentiate septate from bicornuate uterine anomalies. A cross-sectional imaging study (sonography or MRI) is generally required to distinguish between these two anomalies that together comprise nearly two thirds of congenital uterine anomalies. It is critical to correctly distinguish between septate and bicornuate anomalies because bicornuate uterus uncommonly needs surgical intervention, whereas septate uterus commonly does. Also, the type of surgical intervention is different; transabdominal metroplasty for bicornuate uterus versus hysteroscopic resection of the septum for a septate uterus. In particular, it is important to not misinterpret slight concavity of the uterine fundal contour as bicornuate morphology. Figures 11A and 11B are of a patient with a septate uterus with a shallow uterine fundal cleft (< 1 cm depth) separating two uterine cavities. This is still consistent with septate morphology.


References
Top
 Abstract
 INTRODUCTION
 EDUCATIONAL OBJECTIVES
 REQUIRED ACTIVITIES
 RECOMMENDED READING
 INSTRUCTIONS
 BACKGROUND
 References
 

  1. Raga F, Bauset C, Remohi J, et al. Reproductive impact of congenital Müllerian anomalies. Hum Reprod1997; 12:2277 –2281[Abstract/Free Full Text]
  2. Troiano RN, McCarthy SM. Müllerian duct anomalies: imaging and clinical issues. Radiology 2004;233 : 19–34[Abstract/Free Full Text]
  3. The American Fertility Society classifications of adnexal adhesions, distal tubal obstruction, tubal occlusion secondary to tubal ligation, tubal pregnancies, Müllerian anomalies and intrauterine adhesions. Fertil Steril 1988;49 : 944–955[Medline]
  4. Ott DJ, Fayez JA, Zagoria RJ. Hysterosalpingography: a text and atlas, 2nd ed. Baltimore, MD: Williams & Wilkins, 1998: 67
  5. Saleem SN. MR imaging diagnosis of uterovaginal anomalies: current state of the art. RadioGraphics 2003;23 : e13[Abstract/Free Full Text]
  6. Cunningham FG, Leveno KJ. Williams obstetrics, 22nd ed. New York, NY. McGraw-Hill, 2005:953 –956
  7. Minto CL, Hollings N, Hall-Craggs M, Creighton S. Magnetic resonance imaging in the assessment of complex Müllerian anomalies. Br J Obstet Gynaecol 2001;108 : 791–797[CrossRef]
  8. Ubeda B, Paraira M, Alert E, Abuin RA. Hysterosalpingography: spectrum of normal variants and nonpathologic findings. AJR 2001; 177:131 –135[Free Full Text]
  9. Imaoka I, Wada A, Matsuo M, et al. MR imaging of disorders associated with female infertility: use in diagnosis, treatment, and management. RadioGraphics 2003;23 :1401 –1421[Abstract/Free Full Text]
  10. Thurmond AS, Jones MK, Matteri R. Using the uterine push–pull technique to outline the fundal contour on hysterosalpingography. AJR; 175:359 –361

Add to CiteULike CiteULike   Add to Complore Complore   Add to Connotea Connotea   Add to Del.icio.us Del.icio.us   Add to Digg Digg   Add to Reddit Reddit   Add to Technorati Technorati    What's this?



This Article
Right arrow Abstract Freely available
Right arrow Figures Only
Right arrow Full Text (PDF)
Right arrow CME/SAM Credit
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Dykes, T. M.
Right arrow Articles by Dodson, W.
Right arrow Search for Related Content
PubMed
Right arrow Articles by Dykes, T. M.
Right arrow Articles by Dodson, W.
Social Bookmarking
Add to CiteULike