MRI for Diagnosing Intrauterine Adhesions

Abstract

Objectives: To prospectively assess the diagnostic accuracy of MRI and transvaginal ultrasound (TVS) as well as the prognostic value of MRI for intrauterine adhesions (IUAs), using hysteroscopy as the reference standard.

Design: Prospective observational study.

Setting: Tertiary medical centre.

Patient(s): Ninety-two women with amenorrhea, hypomenorrhea, subfertility, or recurrent pregnancy loss who underwent MRI and in whom Asherman's syndrome was suspected upon TVS.

Intervention(s): MRI and TVS were conducted approximately 1 week before hysteroscopy.

Methods: Ninety-two patients suspected of having Asherman's syndrome were examined by MRI and TVS within 7 days of an upcoming hysteroscopy. All hysteroscopy procedures were performed during the early proliferative phase of the menstrual cycle. All hysteroscopic diagnoses were performed by an experienced expert. All MRIs were read by 2 experienced, blinded radiologists.

Results: MRI was highly accurate (94.57%), sensitive (98.8%), and specific (42.9%) for diagnosing IUAs with a positive predictive value of 95.5% and a negative predictive value of 75%. The diagnostic values of MRI and TVS were significantly different according to McNemar tests. Junctional zone signal and junctional zone alterations correlated with the stage of IUAs.

Conclusion: MRI is markedly superior to TVS in terms of diagnostic accuracy for IUAs, with total agreement with hysteroscopic findings. However, the main advantage of MRI is that, unlike TVS and hysterosalpingography, it can be used to assess the risk of hysteroscopy and to predict postoperative recovery and future pregnancy based on the uterine junctional zone.

Keywords: intrauterine adhesions; magnetic resonance imaging; transvaginal ultrasound; uterine junctional zone.