Strengths and limitations
We found an optimal classification for evaluating FUAS efficacy with a relatively large sample size and identified a promising subtype of classification for the recurrence of adenomyosis after FUAS, which might be of interest to clinicians to make strategy and researchersto explore the association between standard classification based on MRI and FUAS efficacy. Nonetheless, there are some limitations in our study. First, patients in our study were all under FUAS treatment, which limited our interpretation of the results for the general patient population although our analysis confirmed that classification 2 was related to the rates of dysmenorrhea relief and recurrence after FUAS. Besides, our study informed that FUAS combined with GnRH-a/mirena might be a good choice to decrease the recurrence rates after FUAS. Future validation studies are needed for patients with adenomyosis under other treatments. Second, the retrospective study only focused on clinical symptom improvement, rather than the volume changes of uterus and lesions after FUAS for patients lived far away from our center, and chose to visit local hospital for follow up. Third, dysmenorrhea and menorrhagia were not recorded by the pain severity level of Numerical Rating Scale, but collected from patients-self-reported recurrence of dysmenorrhea and/or menorrhagia as yes or no, which may lead to recall bias. However, as commonly used in the routine clinic visit in the follow-up of adenomyosis patients, the patients-self-reported outcome can also reflect the efficacy of FUAS to some extent. A multicenter study is needed to provide more evidence.