Müllerian Anomalies: Classification and Imaging

Müllerian anomalies (congenital uterine and vaginal anomalies) result from abnormal formation, fusion, or resorption of the paramesonephric (Müllerian) ducts during embryogenesis. They are classified using the 2021 ASRM Müllerian Anomalies Classification (MAC2021), which divides anomalies into nine categories (agenesis/hypoplasia, unicornuate, didelphys, bicornuate, septate, arcuate, DES-related, and complex/unclassified), or the ESHRE/ESGE (CONUTA) system, which uses main classes U0–U6 with separate cervical and vaginal codes. MRI is the reference-standard imaging modality because it depicts external fundal contour and internal cavity morphology simultaneously, which is what separates a septate uterus (normal convex or minimally indented fundus) from a bicornuate uterus (fundal cleft >1 cm) — a distinction that determines whether a patient is offered surgery. Canalization defects (septate, bicornuate) carry the highest risk of miscarriage and preterm birth, while unification defects (unicornuate, didelphys) largely spare fertility but still raise obstetric risk.

Quick Reference

• Reference imaging modality: MRI (T2-weighted, small-FOV, true axial-oblique to the endometrial canal) for classification; 3D transvaginal ultrasound is an accepted alternative in expert hands.
• Key discriminator — septate vs. bicornuate: external fundal contour. Concave fundal cleft >1 cm = bicornuate (two horns, external anomaly). Convex, flat, or indentation <1 cm externally with an internal septum = septate.
• ASRM 2021 septate uterus criteria: septal length >1 cm from the interostial (bicornual) line, with the leading edge of the septum forming an angle <90°.
• ASRM 2021 arcuate uterus criteria: fundal endometrial indentation ≥1.0 cm with an obtuse (>90°) angle and normal external fundal contour; considered a normal anatomic variant, not a true anomaly.
• ASRM 2021 categories (9): normal, agenesis/hypoplasia, unicornuate, didelphys, bicornuate, septate, arcuate, DES-related, complex/atypical.
• ESHRE/ESGE (CONUTA) main classes: U0 normal, U1 dysmorphic (incl. T-shaped), U2 septate, U3 bicorporeal, U4 hemi-uterus, U5 aplastic, U6 unclassifiable — plus independent cervical (C0–C4) and vaginal (V0–V4) codes.
• Highest fertility impact: canalization defects (septate/bicornuate) reduce fertility and raise miscarriage/preterm birth rates; unification defects (unicornuate/didelphys) do not reduce fertility but still raise miscarriage, preterm birth, and malpresentation risk.
• Always assess concurrently: renal anomalies (unicornuate uterus: ~40% ipsilateral renal agenesis) and vaginal septa (didelphys: obstructed hemivagina + ipsilateral renal anomaly = OHVIRA/Herlyn-Werner-Wunderlich syndrome).
• Report the actionable finding: state fundal contour, septal/cleft length, angle, cavity number, cervical number, vaginal septum, and renal status — this is what determines surgical candidacy, not the class label alone.

Background and Embryology

The paired Müllerian (paramesonephric) ducts form the fallopian tubes, uterus, cervix, and upper vagina. Normal development requires three sequential steps: (1) organogenesis — formation of the ducts; (2) fusion — the two ducts fuse in the midline to form a single uterus and cervix; and (3) septal resorption — the median septum created by fusion is resorbed to produce a single endometrial cavity. A defect at any step produces a characteristic anomaly: failure of duct formation causes agenesis or unicornuate uterus, failure of fusion causes didelphys or bicornuate uterus, and failure of septal resorption causes septate or arcuate uterus. Because the Müllerian ducts and the metanephric (renal) system develop in close proximity during the same embryologic window, renal anomalies are common companions to Müllerian anomalies and should be actively sought on every study.

Classification Systems

Two systems dominate current practice. The American Society for Reproductive Medicine’s 2021 classification (MAC2021) is the most widely used in North America and was developed with radiologist input specifically to be MRI-friendly, replacing the older 1988 AFS system². The ESHRE/ESGE (CONUTA) consensus classification, more common in Europe, categorizes anomalies by degree of anatomic deviation from normal and separately codes the cervix and vagina³. The two systems agree on the extremes (normal, didelphys, unicornuate, agenesis) but diverge most on borderline cavity indentations: ESHRE/ESGE’s septal-length threshold for diagnosing a septate uterus is lower than ASRM’s, so it classifies a substantially higher proportion of uteri as septate — in head-to-head comparison, ESHRE/ESGE labeled roughly four times as many uteri as septate than the older AFS system did, raising concern about overtreatment of anatomic variants that may be clinically insignificant^4,5. This is a critical point when reporting: a description of the actual anatomy (contour, angle, length) is more useful to the referring clinician than the classification label alone, since the label a patient receives can change depending on which system is applied.

ASRM 2021 category	Approximate ESHRE/ESGE equivalent
Normal uterus	U0
Agenesis/hypoplasia	U5 (aplastic)
Unicornuate	U4 (hemi-uterus)
Didelphys	U3 with C2/V1 (two cervices, longitudinal vaginal septum)
Bicornuate	U3 (bicorporeal)
Septate	U2
Arcuate	U1 or normal variant (system-dependent)
Not separately coded	U1 dysmorphic / T-shaped uterus

Imaging Technique

MRI is recommended as the primary diagnostic tool in MAC2021 because a single study can characterize external fundal contour, internal cavity configuration, cervical number, vaginal anatomy, and the kidneys^2,6. Protocol should include small field-of-view, high-resolution T2-weighted sequences acquired in the sagittal plane and in a true axial-oblique plane angled perpendicular to the endometrial canal (not simply axial to the body), which is the plane that best displays the fundal contour and any septum/cleft in cross-section. 3D transvaginal ultrasound, which similarly permits coronal reconstruction of the fundal contour, is an accepted first-line or adjunct modality where available and operator expertise is high, though MRI remains preferred for complex or atypical anomalies and when a rudimentary horn or obstructed segment must be characterized^1,6. 2D ultrasound and hysterosalpingography (HSG) can suggest an anomaly (e.g., a fundal cleft or two cervical canals) but cannot reliably distinguish septate from bicornuate uteri because neither depicts the external fundal contour.

Imaging Findings by Subtype

Agenesis / Hypoplasia (including MRKH syndrome)

MRI shows absent or rudimentary uterine tissue, typically as bilateral fibromuscular remnants (“Müllerian buds”) without an identifiable endometrial cavity, and a blind-ending or absent upper vagina. Ovaries are typically normal since they arise from a separate embryologic origin, an important distinguishing feature from gonadal dysgenesis. Renal agenesis or ectopia is present in a substantial minority and must be documented. Cervical, fundal, and combined subtypes are described depending on which segment is affected.

Unicornuate Uterus

A single elongated, banana-shaped endometrial cavity is displaced laterally, with a single fallopian tube on that side. Roughly two-thirds have an associated rudimentary contralateral horn, which must be characterized as communicating, non-communicating (with functioning endometrium — a cause of cyclic pain and hematometra requiring excision), or with no cavity at all. Ipsilateral renal agenesis occurs in approximately 40% of cases and should be actively assessed on every study^1,2.

Uterus Didelphys

Complete failure of fusion produces two separate, fully divergent uterine horns, each with its own cervix (didelphic cervix) and often its own endometrial cavity with no communication between the two. A longitudinal or oblique vaginal septum is present in the majority of cases. When the vaginal septum is obstructive, it produces OHVIRA/Herlyn-Werner-Wunderlich syndrome — didelphys uterus, obstructed hemivagina, and ipsilateral renal agenesis — which presents with cyclic pain, a hematocolpos, and a pelvic mass, and is frequently missed until adolescence.

Bicornuate Uterus

Partial failure of fusion leaves two symmetric endometrial cavities connected by fibromuscular tissue, but critically, the external fundal contour shows a deep cleft (>1 cm) between the two horns — this fundal cleft is what separates bicornuate from septate on MRI or 3D ultrasound. A single or double cervix may be present (bicornuate unicollis vs. bicollis). The intercornual distance is typically widened compared with a septate uterus.

Septate Uterus

The most common clinically significant Müllerian anomaly and the one most linked to reproductive loss. The external fundal contour is normal (convex, flat, or minimally indented, <1 cm), while a fibrous or fibromuscular septum divides the endometrial cavity internally. MAC2021 defines it quantitatively: septal length >1 cm measured from the interostial (bicornual) line, with the leading edge of the septum forming an angle <90°². On T2-weighted MRI the septum is typically low signal (fibrous) but may contain some intermediate-signal myometrial tissue; this composition does not change the diagnosis but can be noted. Complete septa extend to the internal os or beyond (sometimes with a duplicated cervix mimicking didelphys); partial septa do not reach the os.

Arcuate Uterus

A mild, smooth fundal indentation of the endometrial cavity with an obtuse angle and a normal external fundal contour. MAC2021 sets the threshold for the indentation at ≥1.0 cm, but distinguishes it from septate by the obtuse leading-edge angle and specifies that a normal uterus has no indentation at all². The arcuate uterus is treated as a normal anatomic variant of no independent surgical significance, though some data associate it with second-trimester loss¹.

Dysmorphic / T-shaped Uterus (ESHRE/ESGE U1)

Not separately coded in MAC2021 but recognized in ESHRE/ESGE as U1: a normal external contour with an abnormally narrow, T-shaped or constricted endometrial cavity, classically associated with in-utero diethylstilbestrol (DES) exposure but also seen without a DES history. MAC2021 retains a distinct DES-related category for anomalies attributable to that specific exposure.

Complex / Atypical Anomalies

Anomalies that combine features of more than one category (e.g., unicornuate uterus with a contralateral septate horn) or that do not fit any defined category are placed here (MAC2021 “complex/atypical”; ESHRE/ESGE U6). Precise descriptive reporting of the actual anatomy is more valuable than forcing a label in these cases.

Differential Diagnosis and Imaging Pitfalls

• Septate vs. bicornuate uterus: the single most consequential distinction, since septate uteri are treated with hysteroscopic septoplasty while bicornuate uteri are generally managed medically/obstetrically, not surgically. The differentiator is the external fundal contour, which requires a sequence that shows the fundus in cross-section (axial-oblique T2 MRI or coronal 3D ultrasound) — a standard axial pelvic MRI or 2D ultrasound can miss this and lead to misclassification.
• Septate vs. arcuate uterus: both have a normal external contour; the distinction rests on indentation length and the angle of the leading edge (acute/<90° for septate vs. obtuse/>90° for arcuate). This distinction matters because septate uteri are offered treatment and arcuate uteri generally are not.
• Didelphys vs. complete septate uterus with duplicated cervix: both can show two cervices and a longitudinal vaginal septum; look for complete divergence of the uterine horns (didelphys) versus two horns joined by a common fundal myometrial mantle with a normal, non-cleft external contour (septate).
• Unicornuate uterus vs. small didelphic/bicornuate horn: a rudimentary horn can be subtle on ultrasound; MRI is more sensitive for confirming absence or presence of endometrial tissue within it, which changes management (a functioning non-communicating horn requires surgical excision).
• Always screen the kidneys in any suspected Müllerian anomaly — renal agenesis/ectopia is a frequent companion finding (especially with unicornuate and didelphys/OHVIRA anomalies) and its absence or presence should be explicitly stated in the report.

Clinical Impact and Fertility

The clinical significance of a Müllerian anomaly depends on its category. A large systematic review by Chan et al. found that canalization defects (septate and bicornuate uteri) are associated with reduced fertility and increased rates of first-trimester miscarriage and preterm delivery, while unification defects (unicornuate and didelphys uteri) do not reduce fertility but are still associated with increased miscarriage, preterm delivery, and fetal malpresentation at delivery; arcuate uteri were specifically associated with second-trimester miscarriage⁷. This is the clinical basis for radiology’s central role: because septate uterus is both the most common anomaly with reproductive consequences and the one for which surgical correction (hysteroscopic septoplasty) is an option, an accurate, reproducible description of fundal contour, septal length, and angle directly determines whether a patient is referred for treatment. A precise MRI or 3D ultrasound report — rather than a bare classification label — is what allows the reproductive endocrinologist or gynecologic surgeon to counsel the patient and plan (or avoid) intervention.

Frequently Asked Questions

What is the difference between a septate and a bicornuate uterus?

A septate uterus has a normal, non-cleft external fundal contour with an internal fibrous septum dividing the cavity, while a bicornuate uterus has an external fundal cleft greater than 1 cm separating two distinct uterine horns. This distinction is made on MRI or 3D ultrasound and determines management, since septate uteri can be corrected hysteroscopically while bicornuate uteri generally are not treated surgically.

Is MRI or ultrasound better for diagnosing Müllerian anomalies?

MRI is the reference-standard modality because it reliably depicts both the external fundal contour and the internal cavity in a single study, which is what most classification systems require. 3D transvaginal ultrasound can achieve comparable accuracy in experienced hands and is an accepted first-line alternative, but standard 2D ultrasound and hysterosalpingography cannot reliably show the external contour and should not be used alone to distinguish septate from bicornuate anomalies.

Does a septate uterus cause infertility?

Septate uterus is associated with reduced fertility and increased rates of first-trimester miscarriage and preterm birth, more so than any other Müllerian anomaly category, which is why it is the anomaly most often treated with hysteroscopic septoplasty.

What does an arcuate uterus mean for pregnancy?

The arcuate uterus is generally considered a normal anatomic variant rather than a true anomaly, with a mild, obtuse-angled fundal indentation and normal external contour. It is not treated surgically, though some data associate it with an increased risk of second-trimester miscarriage.

Why do Müllerian anomalies often come with kidney problems?

The Müllerian (paramesonephric) ducts and the metanephric (renal) system develop adjacent to one another at the same embryologic stage, so a disruption that affects duct formation frequently affects renal development as well. Unicornuate uterus has ipsilateral renal agenesis in roughly 40% of cases, and obstructed hemivagina with didelphys uterus (OHVIRA syndrome) classically includes ipsilateral renal agenesis — the kidneys should be evaluated whenever a Müllerian anomaly is found.

Which classification system should radiologists use, ASRM or ESHRE/ESGE?

Both are in active clinical use, and reports increasingly note both or simply describe the anatomy directly (fundal contour, cavity number, cervical number, vaginal septum). ASRM 2021 was developed with radiologist input and is more MRI-oriented, while ESHRE/ESGE classifies a broader range of cavity indentations as septate and additionally codes the cervix and vagina independently; the choice often follows regional/institutional convention.

References

1. Dixit R, Duggireddy CS, Pradhan GS. Mullerian anomalies: revisiting imaging and classification. Insights Imaging. 2025;16:40. PMID: 39962011
2. Pfeifer SM, Attaran M, Goldstein J, et al. ASRM müllerian anomalies classification 2021. Fertil Steril. 2021;116(5):1238-1252. PMID: 34756327
3. Grimbizis GF, Gordts S, Di Spiezio Sardo A, et al. The ESHRE/ESGE consensus on the classification of female genital tract congenital anomalies. Hum Reprod. 2013;28(8):2032-2044. PMID: 23771171
4. Ludwin A, Ludwin I. Comparison of the ESHRE-ESGE and ASRM classifications of Müllerian duct anomalies in everyday practice. Hum Reprod. 2015;30(3):569-580. PMID: 25534461
5. Ludwin A, Tudorache S, Martins WP. ASRM Müllerian Anomalies Classification 2021: a critical review. Ultrasound Obstet Gynecol. 2022;60(1):7-21. PMID: 35678250
6. Al Najar MS, Al Ryalat NT, Sadaqah JS, Husami RY, Alzoubi KH. MRI Evaluation of Mullerian Duct Anomalies: Practical Classification by the New ASRM System. J Multidiscip Healthc. 2022;15:2579-2589. PMID: 36388626
7. Chan YY, Jayaprakasan K, Tan A, Thornton JG, Coomarasamy A, Raine-Fenning NJ. Reproductive outcomes in women with congenital uterine anomalies: a systematic review. Ultrasound Obstet Gynecol. 2011;38(4):371-382. PMID: 21830244