OC21.04: Does hormone use influence the diagnostic performance of sonographic signs of adenomyosis: a prospective study

Numerous studies have explored the diagnostic accuracy of adenomyosis features, yet the impact of hormone use on these assessments remains unexplored. The aim of our study was to assess if hormone use affects the prevalence and diagnostic value of adenomyosis features on transvaginal ultrasound (TVS). Prospective, observational cohort study performed at four centres. Premenopausal women aged 25–52 years scheduled for hysterectomy between 02/2020 and 06/2022 were consecutively included. All participants underwent a systematic 2D and 3D TVS. Stored images, videos and 3D volumes were assessed by an ultrasound expert in diagnosing adenomyosis. Direct signs (myometrial cysts, hyperechogenic islands, echogenic subendometrial lines and buds) and indirect signs (globular uterus, fan-shaped shadowing, wall asymmetry, irregular junctional zone) were evaluated. Dedicated histopathological examination following hysterectomy served as the diagnostic gold standard. Hormone use included progestins (Prg) or oral combined contraceptives (COC), but not GnRH-a. 464 women were included in the study, 392 had a moderate or good image quality and were included in the final analysis, 216/392 (55.1%) had adenomyosis on histology. 153/392 (39%) used hormones (IUD n = 100, COC n = 19, Prg-only-pill n = 17, Prg injection n = 6, NETA n = 4, Other n = 7). Myometrial cysts were significantly more frequently observed in women not using hormones (22.4% vs 13.7%, p = 0.034). The sensitivity, specificity, and diagnostic accuracy for myometrial cysts was better in women without hormone use than with (36.7 vs 21.1%; 98.0 vs 93.5%; 62 vs 57.5%). The prevalence of the other signs or overall accuracy for diagnosing adenomyosis did not significantly differ between groups. We found that hormone use influences the prevalence and diagnostic performance of myometrial cysts, but not the overall performance of TVS.