Magnetic Resonance Imaging-Based Classifications for Symptom of Adenomyosis

Abstract

Objectives: The aim of the study was to identify an optimal magnetic resonance imaging (MRI)-based classification for the severity of adenomyosis and explore the factors associated with disease severity (dysmenorrhea or menorrhagia). Design: and Participants: Several classifications based on MRI have been proposed, and their phenotypes are reported to be associated with the severity of adenomyosis. However, a consensus classification based on MRI findings has not yet been reached. Our study was designed to retrospectively analyze data from a cohort of patients in the Affiliated Nanchong Central Hospital of North Sichuan Medical College from June 2017 to December 2021 before focused ultrasound ablation surgery (FUAS), identify the optimal classification of adenomyosis severity from different classification criteria, and explore factors associated with the presence of symptoms. Methods: The proportions of disease severity among different classification groups were compared to obtain the one generating the most considerable χ2 value, which was identified as the optimal classification for informing disease severity. A logistic regression model was constructed to explore factors associated with disease severity. Results: Classification of Kobayashi H (classification 4) concerning the affected areas and size (volumes of lesions) was recognized as the optimal one, which identified dysmenorrhea (χ2 = 18.550, p value = 0.002) and menorrhagia (χ2 = 15.060, p value = 0.010) secondary to adenomyosis. For volumes of the uterine wall <2/3, the dysmenorrhea rate in subtype 4 was higher than that in subtype 1 (χ2 = 4.114, p value = 0.043), and the dysmenorrhea rate in subtype 5 was higher than that in subtype 2 (χ2 = 4.357, p value = 0.037). Age (odds ratio [OR] = 0.899, 95% confidence interval [CI] = 0.810∼0.997, p value = 0.044) and external phenotype (OR = 3.588, 95% CI = 1.018∼12.643, p value = 0.047) were associated with dysmenorrhea. Concerning volumes of the uterine wall ≥2/3, the menorrhagia rate in subtype 3 remarkably increased compared with that in subtype 6 (χ2 = 9.776, p value = 0.002), and internal phenotype was identified as an independent factor associated with menorrhagia (OR = 1.706, 95% CI = 1.131∼2.573, p value = 0.011). Limitations: Patients in our study were all included before FUAS, which limited our result interpretation for the general patient population. Conclusions: MRI-based classification 4 is identified as an optimal classification for informing the severity of adenomyosis. The phenotype of classification is the main characteristic associated with disease severity.