Magnetic resonance imaging (MRI), due to its optimal delineation of anatomy, has become the mainstay in imaging for diagnosing Müllerian duct anomalies (MDA). Pelvic MRI is requested for various conditions such as primary amenorrhoea, infertility or poor obstetric history with regard to MDA, as identifying the exact aetiology for these conditions is vital. Knowledge regarding the classification of MDA is important, as the treatment varies with respect to the different classes. As all the lesions do not fit within the classification of the American Society for Reproductive Medicine, a new anatomy-based classification was established by the European Society of Human Reproduction and Embryology and the European Society for Gynecological Endoscopy, to fulfil the needs of experts. We aim to discuss various classes of classified and unclassified MDA with regard to both the above-mentioned classifications and illustrate some of them using various cases based on pelvic MRI studies.
Embryologically, the two-paired Müllerian ducts fuse in the midline in the lower part by the 6th to the 11th week of gestation. Lateral fusion forms the uterus and upper two-thirds of the vagina and the cephalad parts of the separated ducts form the fallopian tubes. Resorption of the midline septum results in a single uterine cavity. The sinovaginal bulb forms the lower third of the vagina, which fuses with the lower part of the fused Müllerian ducts.
Any deviation from the above process leads to Müllerian duct anomalies (MDA). Three-dimensional transvaginal ultrasound is almost as sensitive as magnetic resonance imaging (MRI) in diagnosing congenital uterine anomalies.
Clinical presentation of the MDA is variable, where patients with obstructive anomalies present earlier. In these young women, the goals of therapy are to relieve the obstruction immediately, restore normal menstruation and sexual function and preserve reproductive potential.
T2-weighted sequences in all three planes (axial, coronal and sagittal) are adequate. Firstly, a sagittal T2-weighted image is obtained (TR/TE-4860/86, slice thickness [ST] 5mm and NEX-2), following which oblique coronal (
Oblique coronal image (a) obtained by planning along the long axis of uterus (white line) on the sagittal T2 sequence (b).
(a) Sagittal T2 sequence demonstrates a rudimentary uterus (white arrow). (b) Planning of the T2 CUBE curved maximum intensity projection (MIP). (c) T2 cube curved MIP demonstrates a bicornuate rudimentary uterus with an intercornual distance of more than 4 cm and widely separated (yellow arrows). (d) Coronal T2 sequence shows both ovaries (white arrows).
T1-weighted sequences are used to define any haemorrhagic component in the obstructed anomaly such as haematometra, adenomyosis, ovarian chocolate cyst and benign teratoma. It is also useful for uterine cysts with mucinous components and paravaginal Gartner’s duct cysts (
(a) Coronal T2 sequence demonstrates a complete septate uterus with a myoma (star) in the septum. (b) Sagittal T2 sequence shows a Gartner’s duct cyst with focal sacculations (double-headed white arrow) from the fornix extending along the anterior vaginal wall, well delineated against the instilled gel in the vagina. (c and d) Axial T1 sequence demonstrates a left ureterocoele with a blind-ending atretic ureter (white arrows) in the broad ligament. (e) Coronal T2 sequence shows a normal right kidney; left kidney not visualised. (f) Incidental left adrenal adenoma (fat component in out-of-phase images) marked by a white arrow.
Coronal screening of the upper abdomen is mandatory in all cases of MDA to evaluate for associated renal abnormalities (
American Society for Reproductive Medicine (
The American Society for Reproductive Medicine classification.
American Society for Reproductive Medicine classification.
Class | Classification |
---|---|
Class I | Agenesis or hypoplasia – (a–e) (vaginal or cervical or fundal or tubal or combined) |
Class II | Unicornuate – (a–d) (communicating horn or non-communicating horn or no cavity or no horn) |
Class III | Uterine didelphys |
Class IV | Bicornuate uterus (a and b – Complete or partial) |
Class V | Septate uterus (a and b – Complete or partial) |
Class VI | Arcuate uterus |
Class VII | Diethylstilboestrol related |
Uterine anatomical deviations are classified as normal uterus U0, dysmorphic uterus U1, septate uterus U2, bicorporeal uterus U3, hemi-uterus U4, aplastic uterus U5 and still unclassified cases U6.
The European Society of Human Reproduction and Embryology and European Society for Gynecological Endoscopy classification.
The European Society of Human Reproduction and Embryology and the European Society for Gynecological Endoscopy classification.
An external fundal contour with dipping of more than 10% or 1 cm,
To differentiate bicornuate and septate uterus: an intercornual distance of more than 4 cm, an intercornual angle of more than 105° and external fundal dipping favours a bicornuate uterus.
Flat or subtle fundal dipping (less than 1 cm), an intercornual distance of less than 4 cm and an acute intercornual angle of less than 75° suggests a septate uterus.
In an arcuate uterus, myometrial dipping in the inner aspect of less than 10% is seen with a convex or flat outer fundal contour. The inner dipping edge is not sharp and saddle shaped,
The unicornuate uterus does not have the usual rounded fundal contour and oblong shape. Zonal anatomy and endometrial–myometrial differentiation is maintained.
Absence of myometrial–endometrial differentiation or fundus, body and cervix differentiation represent a hypoplastic uterus.
Bicornuate and didelphys uterus of the previous ASRM classification incorporate under bicorporeal uterus U3, subcategories 3a and 3b, respectively. A dysfused septate uterus is 3c.
Unicornuate uterus is classified under hemi-uterus U4.
A clear description of the septate uterus is provided as a fundal midline indentation of >50% of the uterine wall thickness. Anything less is included in the subcategory of U1, a dysmorphic uterus.
Arcuate uterus (type VI of ASRM) is not included in the ESHRE–ESGE classification system, avoiding unnecessary apprehension created by this almost normal variant as an anomaly. In the ESHRE–ESGE classification, these patients are classified as normal or septate, depending on the degree of midline indentation.
The ESHRE and ESGE classification classifies 38 of the 39 congenital anomalies of the female genital tract, whereas the ASRM classification is not as comprehensive.
Independent classification of uterine, cervical and vaginal anomalies is made. It is very useful, especially in cases of obstructive cervical or vaginal malformations with a normal uterus, which constitutes about 22 out of the 39 types of anomalies.
The ESHRE and ESGE classification uses common terms for describing anomalies and the exact anatomical status of the female genital tract, rather than the liberal non-specific terminologies used in the ASRM classification.
The ESHRE and ESGE classification includes associated anomalies of non-Müllerian origin as part of its system to accommodate the renal tract malformations which are common with MDA.
The embryological origin of the anomalies can be easily identified on the basis of the ESHRE and ESGE classification.
The ESHRE and ESGE classification places less severe anomalies at the beginning with the more deformed anomalies placed as the later classes and sub-classes.
The ESHRE and ESGE classification defines uterine deformities based on the proportions of uterine anatomical landmarks, such as uterine wall, rather than using definite objective numerical values, keeping in consideration the variability of uterine dimensions in different patients.
Written informed consent was obtained from all the patients for publication of this review article, including all the images.
A uterus having an interostial straight or curved line with less than 50% of the uterine wall thickness, internal indentation in the midline fundus and a flat external contour on coronal MRI section is considered a normal uterus U0 (
Coronal oblique T2 sequence shows a normal uterus. No fundal external dipping, intercornual distance of less than 4 cm and internal fundal muscular indentation of less than 50% of the uterine wall thickness. Associated right lateral wall intramural fibroid and left benign teratoma seen.
A dysmorphic uterus is sub-classified by the shape of the uterine cavity as T-shaped, infantile and others. The other category is used to differentiate minor deformities.
This is due to an absorption defect. A uterus having a flat external contour and more than 50% internal fundal midline indentation is classified as a septate uterus. It is divided into subtypes a and b (partial [
Coronal oblique T2 sequence shows no fundal external dipping, an intercornual distance of less than 4 cm and an internal fundal muscular indentation of more than 50% of the uterine wall thickness. Associated left pelvic kidney is seen.
Coronal oblique T2 sequence (a) shows no fundal external dipping. There is a partial septum with a muscular and fibrous component separating the endometrial cavity in the fundal region. A single cervix is seen. T2 CUBE maximum intensity projection (MIP) image (b) demonstrates well the inferior sharp end of the septum. Partial septate uterus. Associated left ovarian cyst.
Coronal oblique T2 sequence demonstrates a complete septate uterus. The outer fundal contour is flat. The intercornual distance is 3.5 cm. Internal indentation of the muscular and fibrous component is seen in fundal region (a) and there is only a fibrous component in the lower part (white arrow) (b), extending down to the cervix and two vaginal cavities (c – yellow double-headed arrows).
An arcuate uterus may be classified as a partial septate uterus in the ESHRE–ESGE classification (
Ludwin et al.
Patients with a complete septate uterus may have multiple associated findings. Myomas were seen in the septum and left broad ligament in one of our patients (not shown). There was an associated Gartner’s duct cyst that appeared hyperintense on the T1 sequence, extending from the fornix to the anterior vaginal wall. This patient also had a left ureterocoele with a short ureter ending blindly. She had a normal right kidney and an absent left kidney with an incidental left adrenal adenoma (
A bicorporeal uterus is caused by fusion defects, characterised by external indentation of more than 50% of the uterine wall thickness. This is subdivided into partial and complete (
(a, b) Axial T2 sequence demonstrates a fundal external indentation exceeding 50% of the uterine wall thickness, dividing the uterine corpus and cervix completely. (c, d) Axial T2 sequence demonstrates two vaginal cavities (white arrows).
Increased risk of placental abnormalities has been reported with uterine anomalies,
Sagittal T2 sequence (a) demonstrates two fundi and two uterine bodies (white arrows) with two endometrial cavities and two cervices (black arrows). Post-partum changes seen in the anterior uterus. (b) Axial T2 sequence shows post-partum changes with the placenta in fundal and left lateral wall (white arrow) – placenta increta.
Coronal short tau inversion recovery (STIR
A unilaterally formed uterus, which occurs as a result of a formation defect with an absent or incompletely formed contralateral part, is usually laterally deviated. It is sub-classified into subtypes a and b by the presence or absence (
Sagittal T2 sequence (a) demonstrates non-extension of the endometrial cavity up to the fundal region. Coronal oblique T2 sequence (b) shows a unicornuate uterus. Banana-shaped endometrial cavity displaced to left side with right ovarian benign teratoma.
An aplastic uterus is a formation defect with bilateral (
Sagittal T2 sequence (a) demonstrating a rudimentary uterus. Short tau inversion recovery (STIR) coronal (b) shows the rudimentary uterus with a cavity and absent ovaries.
Fundal, body and cervix differentiation as well as endometrial–myometrial differentiation are not seen (
Failure of the Müllerian duct to develop leads to Müllerian agenesis. Parikh et al. state that the most common cause of primary amenorrhea is Müllerian agenesis.
Axial T2 sequence (a) demonstrates a band of vagina (double-headed white arrow) in between the rectum (asterisk) and urethra (yellow arrow). Axial T2 sequence (b) identifies both ovaries (yellow arrows) and a left pelvic kidney. Mayer–Rokintasky–Küster–Hauser syndrome.
Rare anomalies or combined pathologies that do not comply correctly with one of the above six groups are categorised into unclassified (
Coronal T2 sequence (a) demonstrates a uterus with endo-myometrial differentiation on the left (white arrow) and an additional uterus-like structure (double-headed arrow) on the right side attached to the left-sided uterus by a fibrous band. (b) Coronal T2 thick maximum intensity projection (MIP) delineates well both uteri (white arrows) and connecting fibrous band (long white arrow). (c) Oblique reformation of the left-sided uterus shows a unicornuate uterus with a rudimentary non-communicating horn (white arrow). (d) Sagittal T2 sequence shows cervical aplasia. (e) Sagittal T2 sequence demonstrates the blind-ended vagina (white arrow). (f) Sagittal T2 sequence with vaginal gel instillation shows a transverse septum superiorly (yellow arrow) and a longitudinal septum (white arrows) in the lower part.
Sagittal T2 sequence demonstrates the uterus and cervix distended with haematometra with a thick non-canalised inferior margin of the cervix.
The ESHRE and ESGE classification is an anatomical and embryological-based classification and includes cervical and vaginal anomalies. A volumetric T2-weighted MRI sequence such as T2 CUBE is the sequence of choice. Although the ESHRE classification is more accommodative than the ASRM classification, there are still a few unclassified anomalies. Magnetic resonance imaging acquired after the instillation of vaginal gel helps to clearly delineate the transverse and longitudinal vaginal septum. Knowledge of the various classifications, differentiating points, imaging characteristics, associated anomalies, syndromes and various available treatment options is vital for the radiologist to guide the treating experts to achieve the best patient outcome.
We thank Dr Prabu Dhanasingh and Dr Sunil Kumar for their contribution in figure and artwork preparation.
The authors declare that they have no financial or personal relationship(s) that may have inappropriately influenced them in writing this article.
D.J. was the project leader. D.J. and V.I. were responsible for experimental and project design. Both authors collected and reported the cases and wrote the manuscript.