The present study aimed to develop a nomogram to predict long-term outcomes of uterine artery embolisation (UAE) for treating adenomyosis.
We reviewed data of 221 patients with adenomyosis who underwent UAE between May 2016 and January 2018. Predictive factors were identified using multivariate logistic regression analysis. A nomogram to predict the outcome of UAE was created for the training set. The performance of the predictive model was assessed by discrimination (quantified using the area under the curve, AUC) and calibration (evaluated by calibration curves and Hosmer-Lemeshow test) internally within the training set. Finally, an independent external validation was conducted using the validation set.
In total, 201 patients were included. In the training set (n = 137), 96 (70.1%) exhibited a good response (GR), and 41 (39.9%) showed a poor response (PR). In the validation set (n = 64), 44 (68.7%) showed GR and 20 (31.3%) showed PR. The dysmenorrhoea score, T2 signal type, CA125, apparent diffusion coefficient, accompanying endometriosis, and accompanying fibroids were identified as associated factors and used in the nomogram. The AUC of the nomogram was 0.800 (95% confidence interval [CI] 0.724-0.877) and 0.798 (95% CI 0.686-0.909) in the training and validation sets, respectively. The calibration curves and Hosmer-Lemeshow test showed optimal agreement between predicted and actual probabilities (training set: P = 0.754; validation set: P = 0.453).
We developed a nomogram that could predict the outcome of UAE in patients with adenomyosis. This model has the potential to select patients for UAE.

Copyright © 2022. Published by Elsevier B.V.

Author