A short anogenital distance on MRI is a marker of endometriosis

Study question: Could the anogenital distance (AGD) as assessed by MRI (MRI-AGD) be a diagnostic tool for endometriosis?

Summary answer: A short MRI-AGD is a strong diagnostic marker of endometriosis.

What is known already: A short clinically assessed AGD (C-AGD) is associated with the presence of endometriosis.

Study design size duration: This study is a re-analysis of previously published data from a case-control study.

Participants/materials setting methods: Women undergoing pelvic surgery from January 2018 to June 2019 and who had a preoperative pelvic MRI were included. C-AGD was measured at the beginning of the surgery by a different operator who was unaware of the endometriosis status. MRI-AGD was measured retrospectively by a senior radiologist who was blinded to the final diagnosis. Two measurements were made: from the posterior wall of the clitoris to the anterior edge of the anal canal (MRI-AGD-AC), and from the posterior wall of the vagina to the anterior edge of the anal canal (MRI-AGD-AF).

Main results and the role of chance: The study compared MRI-AGD of 67 women with endometriosis to 31 without endometriosis (controls). Average MRI-AGD-AF measurements were 13.3 mm (±3.9) and 21.2 mm (±5.4) in the endometriosis and non-endometriosis groups, respectively (P < 10^-5). Average MRI-AGD-AC measurements were 40.4 mm (±7.3) and 51.1 mm (±8.6) for the endometriosis and non-endometriosis groups, respectively (P < 10^-5). There was no difference of MRI-AGD in women with and without endometrioma (P = 0.21), or digestive involvement (P = 0.26). Moreover, MRI-AGD values were independent of the revised score of the American Society of Reproductive Medicine and the Enzian score. The diagnosis of endometriosis was negatively associated with both the MRI-AGD-AF (β = -7.79, 95% CI (-9.88; -5.71), P < 0.001) and MRI-AGD-AC (β = -9.51 mm, 95% CI (-12.7; 6.24), P < 0.001) in multivariable analysis. Age (β = +0.31 mm, 95% CI (0.09; 0.53), P = 0.006) and BMI (β = +0.44 mm, 95% CI (0.17; 0.72), P = 0.001) were positively associated with the MRI-AGD-AC measurements in multivariable analysis. MRI-AGD-AF had an AUC of 0.869 (95% CI (0.79; 0.95)) and outperformed C-AGD. Using an optimal cut-off of 20 mm for MRI-AGD-AF, a sensitivity of 97.01% and a specificity of 70.97% were noted.

Limitations reasons for caution: This was a retrospective analysis and no adolescents had been included.

Wider implications of the findings: This study is consistent with previous works associating a short C-AGD with endometriosis and the absence of correlation with the disease phenotype. MRI-AGD is more accurate than C-AGD in this setting and could be evaluated in the MRI examination of patients with suspected endometriosis.

Study funding/competing interests: N/A.

Trial registration number: The protocol was approved by the 'Groupe Nantais d'Ethique dans le Domaine de la Santé' and registered under reference 02651077.