Factors associated with hysterectomy after failure of levonorgestrel-releasing intrauterine system treatment in women with abnormal uterine bleeding

This descriptive study included 233 women diagnosed with AUB who received an LNG-IUS between November 2019 and December 2022 at Antalya Training and Research Hospital. Of these, 60 patients (25.7%) underwent hysterectomy due to treatment-resistant bleeding, while 173 patients (74.2%) continued follow-up without surgical intervention. Baseline demographic and clinical characteristics are summarized in Table  1 . Women in the hysterectomy group were significantly older compared with those who did not undergo hysterectomy (45.35 vs. 41.76 years, p  = 0.001). No significant differences were observed in BMI and parity between the groups. Among the 233 patients, 41.2% had normal hemoglobin levels, 18.4% had mild anemia, 31.8% had moderate anemia, and 8.6% had severe anemia. In the non-hysterectomy group ( n  = 173), the rates were 42.8% normal hemoglobin, 18.5% mild anemia, 30.0% moderate anemia, and 8.7% severe anemia. In the hysterectomy group ( n  = 60), the corresponding rates were 36.7%, 18.3%, 36.7%, and 8.3%, respectively (Table  2 ). Table 1 Demographic and clinical characteristics of women according to hysterectomy status Variable Without Hysterectomy ( n  = 173) Hysterectomy ( n  = 60) p -value Age (years, median (IQR) 41.76 (38–46) 45.35 (44–48) 0.001* BMI (kg/m²), median (IQR) 27.20 (25–29.3) 27.91 (25.70–29.20) 0.99 Parity, median (IQR) 2 (2–3) 2 (2–3) 0.93 Hemoglobin level (g/dL) 11.29 ± 2.05 11.06 ± 1.91 0.45 * p -values were calculated using the Mann–Whitney U test Demographic and clinical characteristics of women according to hysterectomy status Hemoglobin level (g/dL) * p -values were calculated using the Mann–Whitney U test Table 2 Demonstrated level of anemia between the groups Level of anemia Without hysterectomy Hysterectomy  Overall study n  %  n % n  % Normal (≥ 12 g/dL) 74 42.8 22  36.7 96  41.2 Mild (11–11.9 g/dL) 32  18.5 11  18.3 43  18.4 Moderate (8–10.9 g/dL) 52 30 22  36.7 74  31.8 Severe (< 8 g/dL) 15 8.7 5  8.3 20 8.6 Total 173  100 60  100 233 100 Demonstrated level of anemia between the groups Normal (≥ 12 g/dL) Mild (11–11.9 g/dL) Moderate (8–10.9 g/dL) Severe (< 8 g/dL) The clinical diagnoses of patients according to the PALM-COEIN classification in the groups are shown in Table  3 . The distribution of PALM–COEIN diagnoses differed significantly between women who underwent hysterectomy and those who did not (χ² = 43.58, df = 5, p  < 0.001). After Bonferroni correction (α = 0.008), only myoma remained significantly more frequent in the hysterectomy group (adjusted p  < 0.001), while cases without structural pathology were significantly more common in the non-hysterectomy group (adjusted p  < 0.001). Histopathological findings of patients who underwent hysterectomy are presented in Table  4 . Among the 60 hysterectomy cases, 92% ( n  = 55) had a structural uterine pathology. The most common diagnosis was leiomyoma (38.3%), followed by adenomyosis (21.7%). Combination pathologies, such as adenomyosis + leiomyoma (11.7%) and adenomyosis + endometrial polyp + leiomyoma (5.0%), were observed. Five patients (8.3%) had no gross uterine pathology; however, endometrial biopsy revealed simple hyperplasia without atypia ( n  = 3), endometrial polyp ( n  = 1), and proliferative endometrium ( n  = 1). Overall, more than 90% of patients with treatment-resistant AUB had an underlying structural uterine pathology. Leiomyoma was the most frequent cause, followed by adenomyosis. The presence of multiple or complex lesions was strongly associated with LNG-IUS failure and the need for surgical management. Table 3 Comparison of PALM-COEIN diagnoses between hysterectomy and non-hysterectomy groups Diagnosis Without Hysterectomy ( n  = 173) % Hysterectomy ( n  = 60) % Adjusted p value† Adenomyosis 16 9.2 8 13.3 1.000 Adenomyosis + Myoma 0 0 4 6.7 0.024 Endometrial Polyp 4 2.3 2 3.3 1.000 Myoma 32 18.5 27 45.0 < 0.001* Myoma + Endometrial Polyp 0 0 2 3.3 0.393 Normal/No structural lesion 121 70 17 28.3 < 0.001* Total 173 100.0 60 100.0 * Statistically significant after Bonferroni correction (α = 0.008) † Adjusted p values were calculated using Fisher’s exact test with Bonferroni correction for six pairwise comparisons Comparison of PALM-COEIN diagnoses between hysterectomy and non-hysterectomy groups * Statistically significant after Bonferroni correction (α = 0.008) † Adjusted p values were calculated using Fisher’s exact test with Bonferroni correction for six pairwise comparisons Table 4 Histopathological diagnosis of women who underwent hysterectomy ( n  = 60) Adenomyosis Frequency Percent (%) 13 21.7 Adenomyosis + Endometrial polyp + Myoma 3 5.0 Adenomyosis + Myoma 7 11.7 Endometrial polyp 3 5.0 Endometrial polyp + Myoma 6 10.0 Myoma 23 38.3 None 5 8.3 Total 60 100 Note. Frequencies and percentages are presented Histopathological diagnosis of women who underwent hysterectomy ( n  = 60) Note. Frequencies and percentages are presented

This descriptive retrospective study was conducted at Antalya Training and Research Hospital. It included 233 women who were diagnosed with AUB and treated with an LNG-IUS (Mirena ® , Schering AG, Germany) between November 2019 and December 2022. All procedures were carried out in accordance with the ethical standards of the institutional and national committees on human experimentation and with the 1975 Declaration of Helsinki, as revised in 2008. Ethical approval for this study was obtained from the Antalya Training and Research Hospital Ethics Committee in 2023 (protocol number: 2023 − 119). Patient data, including age, body mass index (BMI), underlying etiology of AUB according to the PALM-COEIN classification, ultrasound findings, and endometrial biopsy results, were obtained from hospital medical records. During follow-up, patients who experienced treatment-resistant bleeding and subsequently underwent hysterectomy were identified, and their histopathological findings were analyzed (Fig.  1 ). Fig. 1 Flowchart showing patient selection, inclusion and exclusion criteria, and follow-up outcomes Flowchart showing patient selection, inclusion and exclusion criteria, and follow-up outcomes Statistical analyses were performed using IBM SPSS Statistics 25.0 software (IBM Corp., Armonk, NY, USA). The distribution of continuous variables was assessed using the Kolmogorov-Smirnov test. Continuous variables with normal distribution was shown mean and standard distribution; continuous variables that did not show a normal distribution were summarized as the median and interquartile range (IQR), while categorical variables were expressed as counts and percentages. The Mann-Whitney U test or independent t-test was utilized to evaluate the difference between the hysterectomy and non-hysterectomy groups for continuous variables, and the p -values were calculated using this test. Statistical significance was set at a two-sided p -value < 0.05. Categorical variables were analyzed using Pearson’s chi-square test. When significant differences were observed, post-hoc analyses were performed using Fisher’s exact test with Bonferroni adjustment to control for type I error inflation across six pairwise comparisons (adjusted significance level α = 0.008).

Abnormal uterine bleeding (AUB) is defined as uterine bleeding episodes outside the normal range of menstrual cycle frequency, regularity, duration, or volume [ 1 ]. It may be classified as acute, when bleeding is severe enough to threaten hemodynamic stability [ 2 ], or chronic, when abnormal bleeding recurs for at least six months [ 3 ]. The lifetime prevalence of AUB is estimated to be 10%–35%, with heavy menstrual bleeding (HMB) affecting 14%–25% of women [ 4 ]. AUB is particularly common during adolescence and the perimenopausal period and may cause fatigue, reduced daily functioning, emotional stress, and complications such as iron deficiency anemia [ 5 ]. Thus, AUB is not only an individual gynecological problem but also a significant public health issue. Furthermore, direct and indirect costs, including lost workdays, reduced productivity, and the associated economic burden, underscore the overall public health significance of AUB [ 6 ]. Organic uterine tumors, including endometrial polyps, submucosal fibroids, endometrial hyperplasia, and endometrial cancer, constitute an important proportion of the etiological factors of abnormal uterine bleeding, particularly in the perimenopausal and postmenopausal periods. Accurate and timely identification of these structural causes is essential for determining appropriate management strategies and for avoiding missed diagnoses of malignancy [ 7 ]. Transvaginal ultrasonography is widely accepted as the first-line imaging modality for evaluating abnormal uterine bleeding because of its non-invasive nature, wide availability, cost-effectiveness, and ability to provide detailed assessment of the uterine cavity and endometrial structures, particularly in low-resource settings [ 8 ]. The International Federation of Gynecology and Obstetrics (FIGO) classification system is used to determine the AUB etiopathology. The FIGO classification system categorizes AUB into structural (PALM: Polyp, Adenomyosis, Leiomyoma, Malignancy/Hyperplasia) and non-structural (COEIN: Coagulopathy, Ovulatory dysfunction, Endometrial, Iatrogenic, and Not otherwise classified) causes, providing a standardized framework for diagnosis and management [ 9 ]. AUB management depends on the bleeding severity, patient age, comorbidities, and fertility preferences. Medical therapy is considered the first-line approach, with both hormonal and non-hormonal options available. The levonorgestrel-releasing intrauterine system (LNG-IUS) has been shown to be the most effective initial treatment in reducing menstrual blood loss [ 10 ]. Other options include antifibrinolytics such as tranexamic acid and GnRH antagonists [ 11 ]. The LNG-IUS contains 52 mg of levonorgestrel, releasing approximately 20 µg/day initially, which gradually decreases to 14 µg/day after five years. Most of the hormone remains localized in the uterus, suppressing endometrial growth while exerting minimal systemic effects. In addition to its contraceptive action, it is also effective in the management of endometrial hyperplasia, adenomyosis, and early-stage endometrial cancer [ 12 ]. For patients with refractory symptoms or severe bleeding unresponsive to medical treatment, surgical interventions such as endometrial ablation, uterine artery embolization, or hysterectomy may be considered. Hysterectomy remains the definitive treatment, with higher patient satisfaction and improved overall health outcomes reported compared with other modalities [ 13 ]. This study’s primary aim is to compare the demographic, clinical, ultrasound, and histopathological characteristics of women requiring a hysterectomy due to AUB with those of women who were successfully treated without a hysterectomy following LNG-IUS implantation. The secondary aim is to emphasize the importance of early diagnosis and personalized treatment strategies in managing AUB by identifying clinical and pathological factors associated with treatment failure and subsequent hysterectomy.

This study’s results suggest that the effectiveness of the LNG-IUS for AUB is strongly influenced by the presence of underlying structural uterine pathology. More than 90% of women who required a hysterectomy despite LNG-IUS therapy had at least one structural lesion.

The study’s results indicate that the main reason for LNG-IUS treatment failure is underlying structural pathologies. Among the 60 patients who underwent hysterectomy due to LNG-IUS failure, 55 (92%) had at least one structural uterine abnormality. Specifically, myoma was the most frequent finding (45%), followed by adenomyosis (13.3%) and combinations such as adenomyosis + myoma (6.7%) or myoma + endometrial polyp (3.3%). Only 17 patients (28.3%) had “normal” uterine findings, in contrast to 69.9% normal findings in the non-hysterectomy group. These results indicate that structural pathologies are strongly associated with treatment-resistant AUB. Beyond its contraceptive effect, the LNG-IUS has been reported to be effective in the treatment of dysmenorrhea, leiomyoma, endometriosis, adenomyosis, and endometrial hyperplasia [ 14 ]. Wheeler et al. showed that the LNG-IUS is a viable alternative to hysterectomy, particularly for ovulatory or endometrial causes of AUB [ 15 ]. Li et al. demonstrated that the LNG-IUS effectively shortened menstrual duration and induced amenorrhea during long-term follow-up in patients with adenomyosis [ 16 ]. Uterine leiomyomas have also been successfully treated with LNG-IUS in several studies [ 17 , 18 ]. Furthermore, Banu and Manyonda reported that the LNG-IUS provided quality-of-life (QoL) improvements comparable to hysterectomy in women with leiomyoma [ 19 ]. Senol et al. reported that LNG-IUS is an effective method to increase hemoglobin levels in patients with severe menorrhagia due to myoma [ 20 ]. Although supportive evidence indicates a reduction in bleeding associated with leiomyoma, two systematic reviews concluded that the available data remain insufficient to firmly recommend the LNG-IUS for the treatment of AUB with leiomyoma [ 21 , 22 ]. Consistently, Atak et al. observed the lowest continuation rates and the highest surgical intervention rates in patients with leiomyoma, identifying this group as the least responsive to the LNG-IUS in terms of bleeding reduction [ 23 ]. Compared with previous studies, our findings suggest that LNG-IUS failure is primarily associated with underlying structural uterine pathology. Most patients requiring a hysterectomy had an underlying structural cause of AUB, most commonly leiomyoma, followed by adenomyosis. The presence of coexisting lesions appeared to further reduce the treatment’s effectiveness. Historically, hysterectomy was regarded as the definitive treatment for HMB. However, advances in medical management have substantially reduced the need for this surgical approach. Hurskainen et al. [ 24 ] conducted a randomized controlled trial comparing the LNG-IUS and hysterectomy, reporting similar improvements in QoL but lower costs in the LNG-IUS group. Lähteenmäki et al. [ 25 ] found that offering the LNG-IUS during the surgical waiting period led to surgery cancellation in 64.3% of women after six months, compared with only 14.3% in the control group. Likewise, Spencer et al. [ 26 ] demonstrated superior QoL and lower costs with the LNG-IUS compared to hysterectomy in HMB management. In our clinic, LNG-IUS insertion is the first approach for AUB treatment. If the LNG-IUS was insufficient for AUB treatment, these patients underwent hysterectomy. AUB is also increasingly being acknowledged as a major public health problem because of its high prevalence and profound influence on women’s physical and social health [ 27 ]. Improving access to prompt diagnosis and appropriate treatment, alongside promoting health education and awareness campaigns, would help decrease the burden of anemia and fatigue related AUB and its impact on productivity and QoL [ 28 ]. In this context, our findings indicating a predominance of structural uterine pathology among women requiring a hysterectomy suggest that public health strategies focusing on early diagnosis and raising awareness could reduce the morbidity and economic burden associated with AUB [ 29 ]. The study results highlight the importance of comprehensive pre-procedural evaluation for structural uterine lesions in patients being considered for LNG-IUS therapy. Women with leiomyoma, adenomyosis, or combined pathologies should be counseled about the possibility of limited treatment response and the potential need for surgical intervention. By contrast, patients without structural abnormalities are more likely to achieve substantial benefit in terms of symptom control reinforcing the LNG-IUS as a first-line medical therapy for AUB. The strengths of this study are the real clinical data, all pathologies of the uterine entity histopathologically confirmed, and a standardized, etiological analysis following the PALM-COEIN classification, thus making it possible to detect factors associated with the failure of LNG-IUS treatment and resulting in hysterectomy. This study has several limitations. Its retrospective design and reliance on hospital records, together with the relatively small number of patients who ultimately underwent hysterectomy, may limit our findings’ generalizability. Larger prospective studies (generalizable data, if possible) and relative risk calculations are needed to better identify predictors of LNG-IUS failure and to further refine patient selection for this therapy.