Obstructed Müllerian duct anomalies (MDAs) are a well-recognised risk factor for adolescent endometriosis, yet little is known about the anatomical distribution of lesions in patients with unilateral and complete outlet obstructions 1, 2, 3, 4, 5, 6, 7

During embryogenesis, the uterus and cervix develop through fusion, canalisation and resorption of the paired Müllerian ducts. Vaginal formation is complex and is thought to arise from the Wolffian ducts and Müllerian tubercle8. MDAs arise from aberrations of this process and are classified as obstructive and non-obstructive types. Obstructive MDAs are defined by a non-patent outflow tract. The obstruction may occur at the level of the uterine body, cervix or vagina.

Obstructive MDAs can be further divided into unilateral or complete. Unilateral obstruction involves one blocked and one patent outflow tract. Uterus didelphys with an obstructed right or left hemi-vagina is an example of this type of obstruction. In such cases, menstrual efflux occurs from the unobstructed side but is blocked on the other. Complete obstruction, by contrast, involves no patent outflow tract and results in concealed menses which presents as primary amenorrhoea. Cervical agenesis, vaginal agenesis, and transverse vaginal septum are all examples of this type of obstruction.

The true prevalence of obstructive MDAs in the general population is unknown. A systematic review estimated the overall prevalence of all MDAs (obstructed and non-obstructed) to be 5.5% among females, with higher rates among those with a history of miscarriage, and infertility associated with miscarriage 9. Obstructive MDAs are linked to an increased incidence of retrograde menstruation at laparoscopy, pelvic pain and endometriosis 3, 4, 5.Unlike non-obstructive variants, which are often asymptomatic, complete obstructive MDAs are typically recognised earlier due to severe pain, primary amenorrhoea despite normal secondary sexual characteristics and evidence of haematocolpos or haematometra on evaluation10, 11, 12, 13, 14. In unilateral obstruction (e.g. OHVIRA Triad), visible menses from the unobstructed side may delay diagnosis 6.

Endometriosis, defined as the presence of endometrial-like tissue outside the uterine cavity, is associated with infertility, pelvic pain, dysmenorrhoea, and possibly miscarriage 5,13,15, 16, 17. Its estimated prevalence has risen from 11.4% to approximately 14% over the last two decades 18,19. In adolescents, the true prevalence remains uncertain due to diagnostic challenges and symptom under-recognition.

Multiple theories have been proposed to explain the pathogenesis of endometriosis, including retrograde menstruation, embryonic rest theory, coelomic metaplasia, stem cell recruitment, and metastatic spread 20. Among these, retrograde menstruation remains the most widely supported. This theory posits that endometrial cells are transported through the fallopian tubes into the peritoneal cavity during menstruation, where they implant on extra-uterine tissues 21. Obstructive MDAs offer a unique ‘natural experiment’ to explore this mechanism, as obstruction could lead to predictable alterations in menstrual flow. In unilateral outlet obstruction, reflux is likely asymmetrically distributed. If retrograde menstruation is the primary driver of disease in patients with MDAs, lesions should localise predominantly on the obstructed side. In contrast, embryonic rest theory would predict a more generalised distribution of lesions, independent of obstruction laterality.

While several studies demonstrate an association between MDAs and endometriosis, few have examined the spatial relationship between endometriotic lesions and the site of obstruction. Isolated case reports have described ipsilateral localisation, and one larger case series of 94 patients with the uterus didelphys and obstructed hemi-vagina reported a 75% prevalence of endometriosis, with all lesions confined to the obstructed side 6,22,23. The present study aims to extend this evidence by systematically evaluating the presence and laterality of endometriosis across a broader range of obstructive MDAs, including both unilateral and complete outlet obstructions.

We hypothesise that lesion distribution is determined by the type of anomaly and site of obstruction, reflecting expected patterns of retrograde menstrual flow. In anomalies with complete outlet obstruction (e.g. cervical and vaginal agenesis with a single uterine body), retrograde menstruation would be expected to occur through both fallopian tubes, resulting in bilateral pelvic endometriosis. In contrast, in anomalies with unilateral obstruction (e.g. uterus didelphys and obstructed right/left hemi-vagina), reflux would predominantly be directed through the fallopian tube on the obstructed side, resulting in ipsilateral lesion localisation.

Certain anomalies represent unique configurations. For example, Müllerian agenesis with a unilateral atrophic uterine remnant containing functional endometrium is characterised by complete outlet obstruction, yet lesion development would be expected to be unilateral. Unlike other forms of complete outlet obstruction, the obstruction in this anomaly is confined to a single unilateral remnant, thereby directing retrograde flow along one pathway. To our knowledge, this is the first study to comprehensively document localisation patterns across such a diverse cohort of obstructive MDAs.