Clinical Outcomes of Uterine Artery Embolization in Adenomyosis With and Without Coexisting Uterine Fibroids or Ovarian Endometrioma: A Retrospective Clinical Analysis

Uterine artery embolization (UAE) is a minimally invasive therapeutic option for adenomyosis (AM). However, its clinical effectiveness in patients with comorbid gynecologic disorders remains unclear. This study aimed to evaluate the clinical efficacy of UAE in AM patients with or without concomitant uterine fibroids (UFs) or ovarian endometrioma (OE). A total of 243 women with AM who underwent UAE were retrospectively enrolled, including 141 patients with AM alone, 71 with concomitant UFs, and 31 with concomitant OE. Based on clinical response, 193 patients were classified as having effective treatment outcomes and 50 as ineffective. Baseline demographic and clinical characteristics were recorded. Symptom severity scale (SSS), health-related quality of life (HRQOL), and numerical rating scale (NRS) scores were assessed preoperatively, at 3 months after UAE, and at the last follow-up. Significant differences in baseline age and cancer antigen 125 (CA125) levels were observed among the three groups (p < 0.05). Postoperatively, all groups showed significant reductions in SSS and NRS scores, as well as significant improvements in HRQOL scores, at both 3 months and the end of follow-up compared with baseline (all p < 0.001). At the final follow-up, patients in the AM + OE group exhibited significantly higher SSS and NRS scores as well as significantly lower HRQOL scores compared with the other two groups (p < 0.05). In contrast, no significant differences were observed between the simple AM and AM + UFs groups. Patients in the effective group were significantly older, had higher parity, lower CA125 levels, and showed a higher incidence of concomitant UFs but a lower incidence of OE compared with the ineffective group (all p < 0.05). HRQOL scores were significantly higher, and NRS scores were significantly lower in the effective group than in the ineffective group at both postoperative time points (all p < 0.001). The overall clinical efficacy rate was 79.42%, with no significant difference between the simple AM and AM + UFs groups (80.85% vs 88.73%, pa = 0.15). However, the clinical efficacy rate was significantly lower in the AM + OE group (51.61%) than in both the simple AM (80.85%, p = 0.001) and AM + UFs (88.73%, p < 0.001) groups. Kaplan-Meier analysis indicated that concomitant OE was associated with an increased risk of treatment failure following UAE. UAE effectively alleviates symptoms and improves quality of life in patients with AM, particularly in those with isolated AM or concomitant UFs. However, clinical outcomes are less favorable in patients with concomitant OE, suggesting the need for individualized therapeutic strategies in this subgroup.

- adenomyosis - uterine artery embolization - leiomyoma - endometrioma - quality of life Adenomyosis (AM) is generally defined as a benign gynecological disorder characterized by ectopic endometrial tissue invasion into the myometrium [1]. The clinical manifestations of AM include abnormal uterine bleeding (irregular, heavy, or intermenstrual bleeding), pelvic pain, dysmenorrhea, and dyspareunia [2]. Although AM was historically considered to predominantly affect multiparous individuals presenting with dysmenorrhea and menorrhagia, it is now increasingly diagnosed in younger patients, including those experiencing infertility or subfertility [3]. Moreover, AM commonly coexists with other gynecological conditions, particularly endometriosis and uterine fibroids (UFs), which complicates the attribution of clinical symptoms solely to the adenomyotic process [4]. Additional diagnostic challenges emerge from the lack of universally accepted diagnostic criteria for AM [5]. Therefore, the management of AM remains complex and encompasses surgical, medical, and radiological interventions [6]. However, effective treatment strategies with durable symptom control and minimal side effects remain limited. Ovarian endometrioma (OE), commonly referred to as “chocolate cysts”, is an ovarian cystic lesion resulting from endometriosis, a prevalent gynecological disorder characterized by the growth of endometrial tissue outside the uterus. This condition is associated with clinical symptoms, including pelvic pain, dysmenorrhea, and infertility [7]. The presence of OE in AM patients may be associated with more severe uterine dysfunction and symptom burden [8]. UFs, the most prevalent benign tumors affecting women globally, represent a significant health burden, with an estimated prevalence of up to 80% among premenopausal women, and are frequently associated with pain, heavy menstrual bleeding, and infertility [9]. Imaging studies by Neal M. Lonky et al. [10] have demonstrated that two-dimensional ultrasound exhibits limited sensitivity for detecting AM when the uterus is concurrently affected by UFs. However, the impact of the different comorbidities on clinical outcomes in patients with AM has been insufficiently explored. Uterine artery embolization (UAE) is a minimally invasive procedure that has been widely recognized for the management of symptomatic UFs and has also demonstrated efficacy in the treatment of symptomatic AM [11]. Growing evidence suggests that UAE does not adversely affect fertility potential, or menstrual resumption [12,13]. In recent years, its favorable safety profile and therapeutic effectiveness have contributed to increasing clinical acceptance of UAE [14]. The clinical benefits of UAE in AM have been well documented, particularly in alleviating dysmenorrhea and improving quality of life (QOL) [15]. However, endometriosis has been identified as a potential negative prognostic factor for UAE, as patients with coexisting AM and endometriosis are more likely to experience incomplete lesion necrosis following embolization [16]. Despite these observations, comparative data on UAE outcomes among AM patients with different gynecological comorbidities remain limited. Therefore, this study aimed to evaluate the therapeutic efficacy of UAE and to compare clinical outcomes in AM patients with varying comorbid conditions, thereby providing evidence to support more individualized and optimized treatment strategies. From May 2017 to March 2022, 273 patients with adenomyosis (AM) who underwent uterine artery embolization (UAE) were retrospectively screened. According to the inclusion and exclusion criteria, 30 patients were excluded, and 243 eligible patients were finally enrolled in this study. The enrolled patients were divided into three groups: 141 with AM alone (simple adenomyosis), 71 with AM combined with uterine fibroids (UFs), and 31 with AM combined with ovarian endometrioma (OE) (Fig. 1). Complete clinical and follow-up data were available for all included patients. This retrospective study was performed at The Affiliated Guangdong Second Provincial General Hospital of Jinan University. Inclusion criteria: (1) age Exclusion criteria: (1) concomitant uterine fibroids and ovarian endometrioma; (2) recent use of hormonal agents, including oral contraceptives or progestins, within the past three months; (3) incomplete clinical data. The diagnostic criteria for AM were defined as follows [17]: all patients were diagnosed with adenomyosis based on clinical history, symptomatology, and magnetic resonance imaging (MRI) results. Clinical history and symptoms included dysmenorrhea, menorrhagia, pelvic pain, and dyspareunia, with dysmenorrhea and menorrhagia representing the predominant complaints in this cohort. MRI diagnostic criteria for adenomyosis [18] included a maximum junctional zone thickness The diagnostic criteria for OE were as follows [19]: (i) diagnosis of OE based on MRI findings, with definitive diagnosis established when lesion consisted of multiple cysts showing uniformly high signal intensity on T1-weighted imaging (T1WI), regardless of signal intensity on T2-weighted imaging (T2WI), or when a cyst with high signal intensity on T1WI demonstrated low signal intensity on T2WI, often interspersed with areas of high signal; (ii) a maximum diameter of ovarian endometrioma The diagnostic criteria for UFs were defined as follows [21]: (1) essential criterion: transvaginal ultrasound confirmation of Patients were placed in the supine position, and routine disinfection was performed at the right femoral artery puncture site. Local anesthesia was administered using 2% lidocaine (0.1 g: 5 mL, Hualu Pharmaceutical Co., Ltd., Liaocheng, China). Under digital subtraction angiography guidance, arterial puncture was performed using the Seldinger technique, followed by insertion of a 5-F vascular sheath. A catheter was then selectively advanced into the left uterine artery. Embolization microspheres (300–500 µm; Hengrui Medical Technology Co., Ltd., Suzhou, China) were used as embolic agents and delivered through the catheter until the distal branches of the artery were no longer visualized. The same procedure was subsequently performed for embolization of the contralateral uterine artery. After catheter removal, hemostasis was achieved by manual compression of the puncture site for 10–15 minutes, followed by application of a pressure dressing with an elastic bandage. Patients were then transferred back to the ward for postoperative observation. Baseline clinical indicators were extracted from electronic medical records, including age, gravidity, parity, uterine volume, duration of dysmenorrhea, hemoglobin levels, cancer antigen 125 (CA125) levels, and follow-up duration. Numerical rating scale (NRS), symptom severity scale (SSS), and health-related quality of life (HRQOL) scores were recorded preoperatively, at 3 months postoperatively, and at the final follow-up. Uterine volume was measured and calculated using the ellipsoid formula [22]: uterine volume = D1 The NRS score is recognized by the American Pain Society as a standard tool for pain assessment [23]. It consists of 11 numerical levels ranging from 0 to 10, with 0 representing no pain and 10 indicating the most severe, intolerable pain. In this study, the NRS score was used to assess the subjective severity of menstrual pain experienced by patients. The Uterine Fibroid Symptom and Quality of Life (UFS-QOL) questionnaire comprises two domains: the SSS scale and HRQOL scale [24], with a total of 37 items, including 8 symptom items and 29 HRQOL items. The questionnaire was self-administered. Symptom items evaluate the frequency and severity of disease-related symptoms, while the HRQOL items assess multiple domains of well-being, including fatigue, sleep, self-image, emotional distress/psychological burden, fear of embarrassment, interference with daily activities, social relationships, and sexual function. All items are rated on a five-point Likert scale. Symptom frequency and HRQOL items range from “never” to “always”, whereas symptom severity ranges from “not at all” to “a large extent”. The HRQOL total score is calculated by summing the subscale scores, excluding the symptom subscale. For HRQOL and its subscales, higher scores indicate better QOL, whereas higher SSS scores reflect greater symptom severity. Both SSS and HRQOL scores are standardized on a 0–100 scale. All questionnaire assessments (NRS and UFS-QOL) were conducted face-to-face by trained gynecologic nurses with more than three years of clinical experience. All evaluators received standardized training in questionnaire administration and scoring procedures before study initiation. The same evaluation team conducted follow-up assessments throughout the study to ensure inter-rater consistency and data reliability. When assistance was required (e.g., for comprehension of questionnaire items), standardized and neutral explanations were provided without influencing patient responses. Follow-up data were collected through review of electronic medical records, outpatient and inpatient records, and supplemented by telephone interviews. Starting from 3 months post-surgery, SSS scores were obtained at 3-month intervals. Follow-up was terminated once the treatment was deemed ineffective. Patients who achieved effective treatment outcomes were followed for up to 36 months post-surgery. Clinical efficacy was evaluated according to predefined criteria. Treatment was classified as effective when the SSS remission rate exceeded 50%; otherwise, treatment was classified as ineffective. The SSS remission rate was calculated using the following formula: (SSS score before treatment – SSS score after treatment) / SSS score before treatment Statistical analyses and graphical presentations were performed using SPSS version 21.0 (IBM Corp., Armonk, NY, USA) and GraphPad Prism 6.0 software (GraphPad Software Inc., San Diego, CA, USA). The Kolmogorov-Smirnov test was used to evaluate the normality of data distributions. Homogeneity of variances was assessed using Levene’s test or the Brown-Forsythe test, as appropriate. Normally distributed continuous variables were expressed as mean For within-group longitudinal comparisons across time points (preoperative, 3 months postoperatively, and final follow-up), the Friedman test ( To identify independent factors associated with treatment efficacy, logistic regression analyses were performed. Univariate logistic regression was first performed to assess associations between preoperative baseline variables and clinical outcomes (effective vs ineffective). Variables with p The Kaplan-Meier method was used to evaluate the effect of UAE on clinical efficacy in patients with AM, and differences between groups were assessed using the log-rank test. Post-hoc power analyses for subgroup comparisons were performed using Cohen’s h and the normal approximation (two-sided Comparative analyses of baseline clinical characteristics among the three groups are presented in Table 1. Significant differences were observed in age (F = 15.06, p | Clinical indicators | Simple AM group (n = 141) | AM + UFs group (n = 71) | AM + OE group (n = 31) | Statistic | p | pa | pb | pc | | Age (years) | 39.39 | 43.08 | 39.42 | F = 15.06 | 0.975 | ||| | Gravidity | 3.00 (2.00, 4.00) | 3.00 (1.00, 4.00) | 3.00 (2.00, 3.50) | H = 1.34 | 0.51 | 0.34 | 0.38 | 0.92 | | Parity | 2.00 (1.00, 2.00) | 1.00 (1.00, 2.00) | 1.00 (1.00, 2.00) | H = 3.83 | 0.15 | 0.08 | 0.21 | 0.90 | | Uterine volume (cm3) | 250.40 (183.10, 365.10) | 276.60 (201.75, 361.90) | 307.60 (231.40, 381.00) | H = 2.98 | 0.23 | 0.27 | 0.13 | 0.43 | | Duration of dysmenorrhea (years) | 3.00 (2.00, 6.00) | 4.00 (1.50, 6.00) | 5.00 (2.00, 7.50) | H = 2.44 | 0.30 | 0.89 | 0.12 | 0.20 | | Hemoglobin (g/L) | 101.00 (80.00, 121.00) | 104.00 (85.00, 125.00) | 108.00 (87.00, 127.50) | H = 2.19 | 0.34 | 0.22 | 0.27 | 0.79 | | CA125 (U/mL) | 72.10 (41.45, 132.42) | 65.35 (33.51, 103.79) | 94.38 (49.86, 199.87) | H = 7.04 | 0.03 | 0.11 | 0.11 | 0.003 | | Follow-up duration | 29.80 (23.03, 33.93) | 27.03 (23.77, 32.38) | 35.77 (25.17, 36.00) | H = 6.20 | 0.05 | 0.61 | 0.03 | 0.02 | Note: CA125, cancer antigen 125; UFs, uterine fibroids; OE, ovarian endometrioma. Measurement data with normal distribution were expressed as mean The median follow-up duration was 29.80 months (IQR: 23.03, 33.93) in the simple AM group, 27.03 months (IQR: 23.77, 32.38) in the AM + UFs group, and 35.77 months (IQR: 25.17, 36.00) in the AM + OE group (H = 6.20, p = 0.05). Follow-up duration in the AM + OE group was significantly longer than that of the simple AM (pb = 0.03) and AM + UFs groups (pc = 0.02). In contrast, no significant differences were identified among the three groups in terms of gravidity, parity, uterine volume, duration of dysmenorrhea, or hemoglobin levels (all p Changes in SSS scores among the three AM groups were evaluated at baseline, 3 months postoperatively, and at the final follow-up (Table 2). Preoperatively, no significant differences in SSS scores were observed among the three groups (H = 3.25, p = 0.197; all pa, pb, pc | Clinical indicators | Simple AM group (n = 141) | AM + UFs group (n = 71) | AM + OE group (n = 31) | Statistic | p | pa | pb | pc | | Pre-operation | 53.13 (37.50, 62.50) | 50.00 (40.63, 70.31) | 43.75 (28.13, 59.38) | H = 3.25 | 0.197 | 0.70 | 0.11 | 0.08 | | 3 months postoperatively | 12.50 (3.13, 18.75) | 9.38 (6.25, 18.75) | 15.63 (9.38, 25.00) | H = 3.43 | 0.18 | 0.93 | 0.08 | 0.09 | | At the end of follow-up | 9.38 (0.00, 15.63) | 6.25 (0.00, 15.63) | 15.63 (6.25, 31.26) | H = 8.84 | 0.012 | 0.30 | 0.01 | 0.003 | | Within-group comparisons | |||||||| | 3 months vs pre-operation | |||||||| | End vs pre-operation | |||||||| | End vs 3 months | 0.153 | 0.020 | 0.760 | Note: SSS, symptom severity scale. Non-normally distributed continuous variables were expressed as median (Q1, Q3) [interquartile range (IQR)]. Kruskal-Wallis H test was used for comparisons among multiple groups, followed by the Dunn’s post hoc test. For within-group longitudinal comparisons across different time points (pre-operation, 3 months postoperatively, and at the end of follow-up), the Friedman test ( At the final follow-up, SSS scores further reduced in all groups compared with preoperative values (all p Changes in HRQOL scores among the three groups are summarized in Table 3. Preoperatively, no significant differences were observed among the groups (H = 0.69, p = 0.708; pa = 0.94, pb = 0.46, pc = 0.40). At 3 months postoperatively, HRQOL scores increased markedly in all groups compared with preoperative values (all p | Clinical indicators | Simple AM group (n = 141) | AM + UFs group (n = 71) | AM + OE group (n = 31) | Statistic | p | pa | pb | pc | | Pre-operation | 43.97 (28.45, 61.21) | 46.55 (27.16, 59.48) | 40.52 (28.45, 48.71) | H = 0.69 | 0.708 | 0.94 | 0.46 | 0.40 | | 3 months postoperatively | 80.17 (69.83, 93.97) | 84.48 (67.25, 93.97) | 71.55 (59.48, 87.93) | H = 4.01 | 0.135 | 0.61 | 0.07 | 0.06 | | At the end of follow-up | 91.38 (81.03, 97.41) | 91.38 (82.33, 96.98) | 74.14 (53.88, 90.09) | H = 14.34 | 0.52 | ||| | Within-group comparisons | |||||||| | 3 months vs pre-operation | |||||||| | End vs pre-operation | |||||||| | End vs 3 months | 0.799 | Note: HRQOL, health-related quality of life. Non-normally distributed continuous variables were expressed as median (Q1, Q3) [interquartile range (IQR)]. Kruskal-Wallis H test was used for comparisons among multiple groups, followed by Dunn’s post hoc test. For within-group longitudinal comparisons across different time points (pre-operation, 3 months postoperatively, and at the end of follow-up), the Friedman test ( Changes in NRS scores among the three groups were evaluated at baseline, 3 months postoperatively, and at the final follow-up (Table 4). No significant differences in NRS scores were observed among the three groups at baseline (H = 4.13, p = 0.127; pa = 0.05, pb = 0.79, pc = 0.26) or at 3 months postoperatively (H = 1.37, p = 0.504; pa = 0.95, pb = 0.24, pc = 0.34). At the final follow-up, NRS scores were significantly higher in the AM + OE group than in both the simple AM group (pb = 0.02) and the AM + UFs group (pc = 0.003), while no significant difference was observed between the latter two groups (pa = 0.26) (H = 7.68, p = 0.021). Within-group comparisons revealed that NRS scores decreased significantly in all groups at both postoperative time points compared with baseline (all p | Clinical indicators | Simple AM group (n = 141) | AM + UFs group (n = 71) | AM + OE group (n = 31) | Statistic | p | pa | pb | pc | | Pre-operation | 6.00 (4.00, 9.00) | 5.00 (3.00, 8.50) | 6.00 (4.00, 10.00) | H = 4.13 | 0.127 | 0.05 | 0.79 | 0.26 | | 3 months postoperatively | 1.00 (0.00, 2.00) | 1.00 (0.00, 2.00) | 1.00 (0.00, 3.00) | H = 1.37 | 0.504 | 0.95 | 0.24 | 0.34 | | At the end of follow-up | 0.00 (0.00, 2.00) | 0.00 (0.00, 1.00) | 1.00 (0.00, 3.50) | H = 7.68 | 0.021 | 0.26 | 0.02 | 0.003 | | Within-group comparisons | |||||||| | 3 months vs pre-operation | |||||||| | End vs pre-operation | |||||||| | End vs 3 months | 0.821 | 0.068 | 0.432 | Note: NRS, numerical rating scale. Non-normally distributed continuous variables were expressed as median (Q1, Q3) [interquartile range (IQR)]. Kruskal-Wallis H test was used for comparisons among multiple groups, followed by Dunn’s post hoc test. For within-group longitudinal comparisons across different time points (pre-operation, 3 months postoperatively, and at the end of follow-up), the Friedman test ( Based on predefined clinical efficacy criteria, patients were stratified into an effective group (n = 193) and an ineffective group (n = 50). As shown in Table 5, significant differences were observed between the two groups in age, parity, CA125 levels, and the presence of concomitant UFs and OE (all p | Clinical indicators | Effective group (n = 193) | Ineffective group (n = 50) | Statistic | p | | Age (years) | 41.13 | 37.92 | t = –4.16 | | | Gravidity | 3.00 (2.00, 4.00) | 2.50 (1.00, 4.00) | Z = –1.28 | 0.200 | | Parity | 2.00 (1.00, 2.00) | 1.00 (1.00, 2.00) | Z = –2.01 | 0.045 | | Uterine volume (cm3) | 259.00 (191.60, 381.30) | 266.55 (191.25, 336.38) | Z = –0.25 | 0.803 | | Duration of dysmenorrhea (years) | 3.00 (2.00, 6.00) | 4.00 (2.00, 7.00) | Z = –0.88 | 0.379 | | Hemoglobin (g/L) | 101.00 (82.00, 121.00) | 113.50 (84.75, 125.50) | Z = –0.91 | 0.360 | | CA125 (U/mL) | 72.10 (36.20, 121.92) | 82.62 (47.45, 171.40) | Z = –1.99 | 0.046 | | Comorbid UFs [n, %] | 63 (32.64%) | 8 (16.00%) | 0.021 | | | Comorbid OE [n, %] | 16 (8.29%) | 15 (30.00%) | Note: CA125, cancer antigen 125. Measurement data with normal distribution were expressed as mean To identify predictors of treatment response, univariate logistic regression analyses were conducted with clinical efficacy (effective = 1; ineffective = 0) as the dependent variable (Table 6). Age, CA125 levels, coexistence of OE, and preoperative SSS were significantly associated with treatment efficacy (all p | Variables | SE | Z | p | OR (95% CI) | || | Group classification | |||||| | AM | — | — | — | — | 1.00 (Reference) | | | AM + UFs | 0.580 | 0.420 | 1.381 | 0.167 | 1.786 (0.786–4.060) | | | AM + OE | –1.320 | 0.430 | –3.070 | 0.002 | 0.267 (0.118–0.620) | | | Age (years) | 0.110 | 0.030 | 3.733 | 1.116 (1.061–1.187) | || | Uterine volume (cm3) | 0.001 | 0.001 | 0.924 | 0.356 | 1.001 (0.999–1.002) | | | Gravidity | 0.120 | 0.098 | 1.224 | 0.221 | 1.128 (0.930–1.368) | | | Parity | 0.308 | 0.169 | 1.822 | 0.068 | 1.361 (0.968–1.909) | | | Duration of dysmenorrhea (years) | –0.029 | 0.041 | –0.707 | 0.480 | 0.972 (0.902–1.048) | | | Hemoglobin (g/L) | –0.009 | 0.009 | –1.000 | 0.317 | 0.991 (0.974–1.009) | | | CA125 (U/mL) | –0.005 | 0.002 | –2.500 | 0.012 | 0.995 (0.991–0.999) | | | Pre-SSS | 0.030 | 0.010 | 3.000 | 0.003 | 1.030 (1.010–1.050) | | | Pre-HRQOL | 0.009 | 0.009 | 1.000 | 0.317 | 1.009 (0.991–1.027) | | | Pre-NRS | –0.096 | 0.054 | –1.778 | 0.075 | 0.908 (0.817–1.010) | | | Follow-up duration | 0.028 | 0.022 | 1.273 | 0.203 | 1.028 (0.981–1.079) | Note: OR, odds ratio; CI, confidence interval; SE, standard error. | Variables | SE | Z | p | OR (95% CI) | VIF | || | Intercept | –5.120 | 1.550 | –3.299 | 0.001 | 0.006 (0.0003–0.124) | || | Group classification | ||||||| | AM | — | — | — | — | 1.00 (Reference) | || | AM + UFs | 0.180 | 0.460 | 0.391 | 0.696 | 1.197 (0.486–2.951) | 1.422 | | | AM + OE | –1.250 | 0.430 | –2.907 | 0.004 | 0.287 (0.133–0.619) | 1.381 | | | Age (years) | 0.130 | 0.040 | 3.250 | 0.001 | 1.139 (1.051–1.234) | 1.213 | | | CA125 (U/mL) | –0.010 | 0.003 | –3.333 | 0.001 | 0.990 (0.984–0.996) | 1.294 | | | Pre-SSS | 0.020 | 0.010 | 2.000 | 0.046 | 1.020 (1.000–1.041) | 1.452 | Note: OR, odds ratio; CI, confidence interval; VIF, variance inflation factor. Changes in HRQOL and NRS scores were compared between the effective and ineffective groups at baseline, 3 months postoperatively, and at the final follow-up (Table 8). Preoperatively, no significant differences were observed between the two groups in either HRQOL or NRS scores (both p | Clinical indicators | Time | Effective group (n = 193) | Ineffective group (n = 50) | Statistic | pa | | HRQOL score | Pre-operation | 43.97 (27.59, 60.34) | 43.97 (31.46, 59.27) | Z = –0.29 | 0.77 | | 3 months postoperatively | 83.62 (70.69, 94.83) | 68.97 (53.45, 76.72) | Z = –4.30 | || | At the end of follow-up | 93.10 (83.62, 98.28) | 70.26 (50.22, 85.56) | Z = –6.88 | || | Within-group comparisons | ||||| | 3 months vs pre-operation | ||||| | End vs pre-operation | ||||| | End vs 3 months | 0.750 | |||| | NRS score | Pre-operation | 6.00 (4.00, 9.00) | 7.50 (4.00, 10.00) | Z = –1.91 | 0.056 | | 3 months postoperatively | 0.00 (0.00, 2.00) | 2.00 (1.00, 4.00) | Z = –4.99 | || | At the end of follow-up | 0.00 (0.00, 1.00) | 3.00 (1.00, 4.00) | Z = –7.25 | || | Within-group comparisons | ||||| | 3 months vs pre-operation | ||||| | End vs pre-operation | ||||| | End vs 3 months | 0.233 | 0.199 | Note: HRQOL, health-related quality of life; NRS, numerical rating scale; Non-normally distributed continuous variables were expressed as median (Q1, Q3) [interquartile range (IQR)]. Mann-Whitney U test was used for comparison between the two groups. For within-group longitudinal comparisons across different time points (pre-operation, 3 months postoperatively, and at the end of follow-up), the Friedman test ( The therapeutic efficacy of UAE differed significantly among AM subtypes (Table 9). The clinical efficacy rate was comparable between the simple AM group and the AM + UFs group (80.85% vs 88.73%, pa = 0.15). In contrast, the AM + OE group exhibited a markedly lower clinical efficacy rate (51.61%) compared with both the simple AM group (pb = 0.001) and the AM + UFs group (pc | Clinical indicators | Simple AM group (n = 141) | AM + UFs group (n = 71) | AM + OE group (n = 31) | p | pa | pb | pc | | | Clinical efficacy | 0.001 | 0.15 | 0.001 | ||||| | Effective [n, %] | 114 (80.85) | 63 (88.73) | 16 (51.61) | ||||| | Ineffective [n, %] | 27 (19.15) | 8 (11.27) | 15 (48.39) | ||||| | Post-hoc power | 0.89 (AM + OE vs Simple AM) | 0.98 (AM + OE vs AM + UFs) | Note: the measurement data were expressed by the number of cases and percentage, and the chi-square test was used for comparisons between groups. The pa represented AM + UFs group vs Simple AM group; pb represented AM + OE group vs Simple AM group; pc represented AM + OE group vs AM + UFs group; p Kaplan-Meier survival analysis (Fig. 2) demonstrated no significant difference in the Kaplan-Meier curve between the simple AM group and the AM + UFs group (p This study demonstrates that UAE results in significant and sustained improvements in symptom severity, pain and HRQOL in patients with AM. Notably, therapeutic responses varied by comorbid conditions: outcomes after UAE were comparable between patients with isolated AM and those with concomitant UFs, whereas patients with concomitant OE experienced substantially less symptom relief and a significantly higher risk of treatment failure. Baseline characteristics revealed significant age differences among the three patient groups. Patients with AM complicated by UFs were significantly older, which is consistent with previous reports indicating that the prevalence of UFs elevates with age and reaches a peak incidence at approximately 40 years [25]. UAE has been well established as a mature and widely accepted treatment modality for uterine fibroids and has recently emerged as an effective and cost-efficient option for the management of AM [26]. Previous studies have demonstrated that UAE yields substantial clinical improvement in patients with isolated AM as well as those with coexisting uterine conditions [27]. Consistent with these findings, the present study observed that older patients and those with AM coexisting with UFs experienced more favorable therapeutic responses, in line with evidence supporting the robust efficacy of UAE in AM patients with concomitant fibroids [11]. The UFS-QOL English questionnaire, first released in 2002, remains the sole validated instrument designed to evaluate UFs-related symptoms and their impact on QOL [24]. This instrument incorporates the SSS, in which higher scores indicate greater symptom burden, and the HRQOL scale, in which higher scores indicate improved QOL [28]. Previous studies have reported that UAE can reduce SSS cores by approximately 21–39 points within 6 months in patients with UFs [29]. Moreover, significant improvements in HRQOL and SSS scores have been documented at three months post-UAE in patients with AM [30]. Aligned with earlier evidence [15,31], our findings demonstrate that UAE significantly reduces symptom severity and improves quality of life across all AM subtypes. Notably, during follow-up, patients with AM complicated by ovarian endometrioma (AM + OE) exhibited relatively poorer long-term outcomes in symptom severity and quality of life compared with patients with simple AM or AM combined with uterine fibroids (AM + UFs). Quality of life continued to improve beyond 3 months postoperatively in the simple AM and AM + UFs groups, whereas improvements in the AM + OE group were comparatively limited. Taken together, these findings indicate that UAE effectively alleviates symptoms and improves quality of life in patients with AM. However, patients with simple AM or AM + UFs generally achieve more sustained benefits, whereas those with AM + OE, although still deriving clinical benefit, demonstrate relatively weaker treatment responses and less favorable long-term outcomes. The NRS is widely employed to quantify pain score intensity in clinical and research settings [32]. A previous study has shown that NRS scores in patients with AM drop substantially following treatment [33]. Gailė Maldutytė et al. [34] reported a significant positive interrelation between the number of sonographic features of AM and NRS scores, indicating a greater pain burden with increasing disease severity. Consistent with these observations, improvements in pelvic pain and QOL have been reported in women with AM who undergo UAE [35]. Approximately three months following UAE treatment, patients with AM typically exhibit notable symptom relief, marked improvement in overall health status, and a substantial decrease in pain scores [36]. This study also demonstrated a marked postoperative reduction in NRS scores across all groups, with the AM + UFs subgroup experiencing the most pronounced improvement in pain. In contrast, patients with AM + OE showed a more modest decline in pain scores. Univariate analyses revealed that age, CA125 levels, coexistence of OE, and preoperative SSS were significantly associated with treatment response. In the multivariate logistic regression model, concomitant OE remained an independent predictor of suboptimal response to UAE, indicating reduced treatment sensitivity in the AM + OE subtype. Conversely, older age and higher baseline SSS independently predicted favorable outcomes, while elevated CA125 levels were negatively associated with treatment efficacy. These findings are clinically plausible, as patients with a greater symptom burden often experience more substantial symptom relief following targeted vascular occlusion, whereas elevated inflammatory or disease-activity markers, such as CA125, may reflect a more complex pathological state that attenuates therapeutic benefit [16]. Kaplan-Meier analysis further corroborated that patients with AM + OE had a higher probability of treatment failure following UAE. Moreover, response-stratified analyses demonstrated that patients in the effective group achieved sustained and significant improvements in HRQOL and NRS scores postoperatively. Although improvements were less pronounced in the ineffective group, both pain and HRQOL remained significantly improved relative to baseline values, suggesting that UAE confers a degree of symptom relief even among suboptimal responders. Notably, the attenuated therapeutic response observed in the AM + OE subgroup is unlikely to be solely attributable to the presence of OE. In this cohort, patients did not present with typical OE-associated symptoms, such as non-cyclic pelvic pain or dyspareunia, implying that the observed treatment pattern may instead reflect the intrinsic biological characteristics of this adenomyosis subtype. Previous studies have reported that certain AM lesions are characterized by more diffuse infiltration, heightened local inflammatory and pro-angiogenic activity, extensive collateral pelvic vascularization, and pronounced neural remodeling, all of which may limit post-embolization ischemic necrosis and increase susceptibility to reperfusion, thereby blunting pain control and lesion regression [37,38,39]. Additionally, increased myometrial fibrosis may confer greater tolerance to ischemia, further diminishing the therapeutic impact of embolization [40]. Collectively, these mechanisms support the notion that AM + OE represents a biologically more treatment-resistant phenotype. Future studies integrating preoperative imaging biomarkers, inflammatory indices, and measures of neural sensitization may improve patient stratification and help optimize therapeutic decision-making and outcome prediction. Taken together, our findings demonstrate that UAE effectively alleviates symptoms and QOL in patients with AM, with superior efficacy observed in those with pure AM or AM complicated by UFs, and reduced effectiveness in cases with concomitant OE, which was also associated with an increased risk of treatment failure. However, several limitations of this study should be acknowledged. First, the retrospective design of this study may have introduced selection and information biases, and it lacks the methodological rigor of a randomized controlled trial. Second, the sample size, particularly that of patients with AM complicated by OE, was relatively small, which may compromise the stability of subgroup analyses. Although post-hoc power analysis indicated adequate power for the observed between-group differences, the limited number of AM + OE cases may still introduce selection bias and limit the precision and generalizability of subgroup estimates. Therefore, these subgroup findings should be validated in larger, well-designed prospective studies. Additionally, variability in follow-up durations at study endpoints, including early withdrawal and completion at the 36-month follow-up, limits the interpretation of long-term outcomes, such that Kaplan-Meier analyses can only descriptively indicate poorer long-term efficacy in the AM + OE group. Future studies should aim to elucidate the biological mechanisms through which concomitant OE compromises the efficacy of UAE in AM, potentially involving alterations in the local microenvironment, vascular architecture, or variations in hormone sensitivity. Moreover, prospective randomized controlled trials comparing UAE with medical and surgical treatment interventions across different AM subtypes are warranted. Finally, a large-scale multicenter study is planned to further validate and refine the present findings. UAE provides significant and sustained improvements in pain, symptom severity, and quality of life in patients with adenomyosis. Therapeutic efficacy was comparable between patients with simple AM and those with AM combined with uterine fibroids, whereas individuals with concomitant ovarian endometrioma experienced substantially poorer outcomes and a higher risk of treatment failure. Older age and lower CA125 levels were associated with a favorable treatment response, while ovarian endometrioma emerged as an independent adverse predictor of UAE efficacy. Collectively, these findings support UAE as an effective uterus-preserving therapeutic option for AM and underscore the value of preoperative patient stratification, particularly regarding ovarian endometrioma, to optimize therapeutic decision-making and clinical outcomes. • UAE significantly improved SSS, HRQOL, and NRS scores in patients with AM, with benefits evident at 3 months and sustained throughout follow-up. • Clinical efficacy was comparable between the simple AM and AM + UFs groups, while patients with concomitant OE experienced significantly poorer symptom relief and lower response rates. • Multivariate logistic regression identified OE as an independent predictor of poor response to UAE, whereas older age and higher baseline symptom severity predicted favorable outcomes and elevated CA125 levels were associated with reduced treatment efficacy. • Responder demonstrated continued improvement in HRQOL and stable pain control over time, while non-responders showed deterioration after 3 months despite initial symptom improvement. • Survival analysis further demonstrated that patients with AM + OE had a markedly higher risk of treatment failure following UAE than the other AM subtypes. The datasets analyzed during the current study are available from the corresponding authors upon reasonable request. ZT: Conceptualization, Methodology, Data curation, Formal analysis, Writing—original draft. HL: Methodology, Validation, Statistical analysis. CX: Clinical data collection, Patient follow-up, Resources. ZO: Technical support, Imaging interpretation, Supervision of interventional procedures. YW: Conceptualization, Writing – review & editing, Project administration, Supervision. WW: Conceptualization, Study design, Writing – review & editing, Funding acquisition, Supervision. All authors contributed to the important editorial changes in the manuscript. All authors read and approved the final manuscript. All authors have participated sufficiently in the work and agreed to be accountable for all aspects of the work. This retrospective study was approved by the Ethics Committee of The Affiliated Guangdong Second Provincial General Hospital of Jinan University (Approval No. 2024-KY-KZ-062-02). All procedures were conducted following the principles outlined in the Declaration of Helsinki and the measures for the Ethical Review of Biomedical Research Involving Humans issued by the National Health and Family Planning Commission of the People’s Republic of China in 2016. As this study involved a secondary analysis of medical records obtained during routine clinical care, with a large sample size, complete removal of patient identifiers, no disclosure of personal privacy, no involvement of sensitive information, family genetic data, or genetic testing, and no inclusion of vulnerable populations, the requirement for individual informed consent was waived by The Affiliated Guangdong Second Provincial General Hospital of Jinan University. Not applicable. This study was funded by grants from the basic and applied basic research project of high-level university/Dengfeng Hospital of the municipal university (college) joint funding project of Guangzhou Basic Research Plan (Grant No. 2023A03J0277) and the Science and Technology Planning Project of Guangzhou (Grant No.2025A03J4451). The authors declare no conflicts of interest.

Publisher’s Note: IMR Press stays neutral with regard to jurisdictional claims in published maps and institutional affiliations.