PURPOSE:

We aimed to identify predisposing factors that could help predict the therapeutic response of adenomyosis after uterine artery embolization (UAE).

METHODS:

This was a retrospective, single-center study of patients admitted to the hospital for adenomyosis between 2013 and 2015. Sixty-eight patients with adenomyosis who underwent UAE with tris-acryl gelatin microspheres were divided into two groups based on their therapeutic response (complete or incomplete necrosis of lesions), and pre- and postprocedural pelvic magnetic resonance imaging (MRI) data. Patients were followed up for 12 months after UAE. Improvements in dysmenorrhea and menorrhagia were evaluated based on the symptom relief criteria. Improvement rates in both groups were analyzed and compared. Multivariate logistic regression analysis was used to identify the predisposing factors from retrospectively gathered baseline data that might affect the therapeutic response, including MRI features, clinical symptoms, biochemical index, and accompanying diseases of adenomyosis. Then, a prognostic model was established, and the receiver operating characteristic (ROC) curve of identified factors was drawn to determine their predictive value.

RESULTS:

Following UAE, 46 patients (67.6%) showed complete necrosis, while 22 patients (32.4%) showed incomplete necrosis. At 12-month follow-up, dysmenorrhea symptom improvement was seen in 94.7% of complete necrosis and 50% of incomplete necrosis group (P < 0.001); menorrhagia symptom improvement was seen in 96.2% of complete necrosis and 57.1% of incomplete necrosis groups (P = 0.004). Multivariate logistic regression analysis determined serum cancer antigen 125 (CA125) levels (odds ratio [OR], 1.006; 95% confidence interval [CI], 1.002–1.010; P = 0.005) and accompanying endometriosis (OR, 6.869; 95% CI, 1.881–25.016; P = 0.004) as predisposing factors. The areas under the ROC curve of CA125, endometriosis, and these two indicators combined were 0.785, 0.708, and 0.845, which corresponded to sensitivities of 95.5%, 66.7%, and 68.2% and specificities of 52.2%, 80.0%, and 87.0% at optimal cutoff values, respectively.

CONCLUSION:

Symptom relief of dysmenorrhea and menorrhagia for patients with complete necrosis was significantly better than that for patients with incomplete necrosis. Serum CA125 levels and accompanying endometriosis can effectively distinguish complete necrosis from incomplete necrosis.