Comparison of safety and efficacy of hysteroscopic LNG-IUS fixation surgery and conventional LNG-IUS placement in the treatment of adenomyosis: A retrospective cohort study

Adenomyosis is characterized as a benign gynecological disorder wherein there is a focal or diffuse infiltration of endometrial tissue into the myometrium, resulting in uterine enlargement and the presentation of symptoms related to abnormal uterine bleeding. [ 1 ] Adenomyosis is a heterogeneous disease, and it has been proposed to be classified into 4 types based on magnetic resonance imaging: intrinsic, extrinsic, intramural, and other types. [ 1 ] The etiology of adenomyosis remains inconclusive, yet several hypotheses have gained traction among the medical community. The most widely accepted paradigm posits that adenomyosis arises from the penetration of the endometrial basalis layer into the myometrial compartment. [ 2 ] An alternative hypothesis suggests that adenomyotic lesions may originate from the metaplastic transformation of ectopic Mullerian duct remnants or from the differentiation of endogenous adult stem cells. [ 2 ] These conceptual frameworks provide insights into how parity could be a contributory factor to the pathogenesis of adenomyosis. The physiological changes during pregnancy, including the invasive potential of the trophoblast, may facilitate the extension of the endometrial lining into the myometrial tissue. [ 3 ] Furthermore, the hyperestrogenic environment induced by pregnancy may promote the development of ectopic endometrial foci. [ 3 ] A cross-sectional analysis of a cohort comprising 650,000 individuals estimated the overall incidence of adenomyosis at 1%, translating to 29 cases per 10,000 person-years over a decade. [ 4 ] Adenomyosis was previously confirmed diagnostically only subsequent to hysterectomy, with a predominance assumed in the demographic exceeding 40 years of age. Advances in imaging technologies have elucidated the presence of adenomyosis in a younger patient cohort as well. [ 5 ] An investigative study encompassing 985 symptomatic individuals at a gynecological clinic, employing stringent ultrasound diagnostic parameters, revealed a prevalence of adenomyosis of 20.9% within the cohort, which included both pre- and postmenopausal women, nulliparous as well as multiparous patients. This aligns with histopathological findings post-hysterectomy, ranging from 10% to 35%. [ 6 ] The Levonorgestrel-releasing Intrauterine System (LNG-IUS) is extensively utilized in clinical settings and has been demonstrated to ameliorate dysmenorrhea and hypermenorrhea associated with adenomyosis. [ 7 – 9 ] Additionally, it is employed in the management of endometrial hyperplasia and for the prevention of recurrence of endometrial polyps. Nevertheless, the displacement and expulsion of the LNG-IUS bring significant clinical challenges, particularly in cases where the uterine cavity is dilated due to adenomyosis or uterine myomas. The standard clinical intervention for LNG-IUS displacement involves reinsertion. In cases of increased uterine volume, pre-procedural administration of gonadotropin-releasing hormone agonists (GnRH-a) for 3 to 6 months is also often prescribed. [ 10 , 11 ] In 2021, the publication by Zhu et al [ 12 ] on hysteroscopic LNG-IUS suture fixation surgery garnered considerable interest in the medical community. We followed this method and made some improvements. In this study, we retrospectively analyzed the outcomes of patients with adenomyosis who received the fixation surgery in our hospital, complemented by postoperative surveillance to evaluate its clinical safety and efficacy.

Conceptualization: Jie Liu. Data curation: Shanshan Li, Wenwen Li, Qi Ye, Xiaojiu Xu, An Zhang, Chongyi Yang, Xuan Liu, Fangjing Hu, Jie Liu. Formal analysis: Shanshan Li. Writing – original draft: Shanshan Li. Writing – review & editing: Wenwen Li, Qi Ye, Xiaojiu Xu, An Zhang, Chongyi Yang, Xuan Liu, Fangjing Hu, Jie Liu.

This was a retrospective analysis comparing the efficacy of hysteroscopic LNG-IUS fixation surgery with conventional LNG-IUS placement. Among patients admitted to the Affiliated Jinhua Hospital of Wenzhou Medical University due to adenomyosis from December 2021 to December 2023, 83 patients received treatment of LNG-IUS, with 38 patients in the hysteroscopic LNG-IUS fixation surgery group and 45 in the conventional LNG-IUS implantation group. These patients were followed up till the LNG-IUS was expelled or the censor date (30 th June 2024). Participants were followed at 3, 6, 12, 18, 24 months after surgery through in-person visits to the clinic. The present study was approved by Ethics Committee of Affiliated Jinhua Hospital of Wenzhou Medical University (No. 2021-148-K-123). All patients signed informed consent. The condition of adenomyosis was confirmed through ultrasonography, where a diagnosis was established if at least 3 of the following ultrasound characteristics were evident: heterogeneous myometrial echotexture, asymmetrical thickness of the anteroposterior wall of the myometrium, globular-appearing uterus, sub-endometrial echogenic linear striations, sub-endometrial-myometrial cysts, or poor definition of the endometrial-myometrial junction. [ 13 ] The inclusion criteria for LNG-IUS treatment included: met the diagnostic criteria for adenomyosis; age between 30 and 50 years; uterine enlargement volume <12 weeks of pregnancy size; normal ovarian function; had completed childbearing and had no fertility requirements within the next year of follow-up; no contraindications for the use of LNG-IUS. There were contraindications of LNG-IUS treatment that led to exclusion of patients, which included known or suspected pregnancy; active pelvic inflammatory disease; current or untreated hormone-sensitive cancers; uterine cavity distortion; acute cervicitis or untreated lower genital tract infection; unexplained abnormal uterine bleeding. Patients who had abnormal liver and kidney function were also excluded. For conventional LNG-IUS placement, patients underwent the standard procedure for insertion performed by an experienced gynecologist. Patients were positioned in the bladder lithotomy position, and the operator applied iodophor solution for the sterilization of the vulva and vagina, followed by the routine placement of a disinfecting towel. Hysteroscopic examination of the cervical canal, uterine cavity, endometrium, uterine cornua, and fallopian tube openings were carried out and curettage was performed if necessary. After stabilizing the cervix, patients received either general anesthesia or local anesthesia by injecting lidocaine solution into the cervical area. The operator probed the uterine cavity to determine its position, and then the LNG-IUS was guided through the cervix and placed into the uterine cavity with a specialized insertion device. Hysteroscopic LNG-IUS fixation surgery was conducted at the last days of the menstrual cycle. Briefly, patients were positioned in the bladder lithotomy position with routine disinfection and draping. Hysteroscopic examination and anesthesia were performed as aforementioned. Next, a 3-0 Prolene nonabsorbable suture was used to tie 6 knots at the intersection of the T-arm (horizontal and vertical arms) of the LNG-IUS to secure it, and 6 additional knots were tied 1cm away from the first set before cutting the excess suture. The LNG-IUS ring was then conventionally placed into the uterine cavity as aforementioned. The hysteroscope was reintroduced to locate the position of the Prolene suture knots, and a self-developed hysteroscopic puncture needle was used to fix this knot into the myometrium at the center of the uterine fundus, 1cm deep. Finally, a second hysteroscopic examination was performed to check the position of the LNG-IUS within the uterine cavity and for any significant bleeding before concluding the surgery. The surgical instruments used included the SOPRO-COMEG HEOS hysteroscopy system (SOPRO-COMEG GmbH), 20° PANOVEW 5.0 integrated hysteroscope (Richard Wolf GmbH), 3-0 Prolene nonabsorbable suture (Ethicon, Bridgewater), and a self-developed hysteroscopic puncture needle by our hospital. The self-developed hysteroscopic puncture needle is shown in Figure S1, Supplemental Digital Content, https://links.lww.com/MD/Q271 . Our primary outcome was to compare the displacement or expulsion risk between hysteroscopic LNG-IUS fixation group and conventional placement group. The LNG-IUS position was checked at the clinic at regular follow-up (3-, 6-, 12-, 18- and 24-months post-insertion). Patients were also required to visit the hospital whenever they felt that LNG-IUS is abnormal. Our secondary outcome was the comparison of therapeutic efficacy of the 2 different treatment groups. We measured dysmenorrhea severity, menstrual blood loss, uterine volume, endometrial thickness, and the laboratory marker carbohydrate antigen 125 (CA125) on the day of insertion, and at 3-, 6-, 12-, 18- and 24-months post-insertion. Briefly, dysmenorrhea severity was determined by the visual analog pain scale (VAS), [ 14 ] with scores ranging from 0 to 10. A higher score indicates more severe pain. Menstrual blood loss was recorded using the pictorial blood loss assessment chart (PBAC). [ 15 ] This score assessed blood clots by size: clots smaller than a 1-yuan coin were considered small and scored 1 point, while clots equaled to or larger than a 1-yuan coin were considered large and scored 5 points. A total score >100 in indicated excessive menstrual bleeding. During the observation period, all patients were told to use the same brand and model of sanitary pads. The uterine volume and endometrial thickness were measured using the color Doppler ultrasound. The volume was calculated using the formula: uterine volume = 0.523 × anteroposterior diameter × longitudinal diameter × transverse diameter. [ 16 ] Venous blood samples (5 mL) were collected from the patients in the morning to determine the levels of CA125 which positively correlated with the severity of adenomyosis. [ 17 ] The incidence of LNG-IUS displacement or expulsion was compared between the 2 groups. The differences in clinical characteristics among the 2 different treatment groups were compared using independent 2-sample t -test (normally distributed data) or Kruskal–Wallis test (non-normally distributed data) for continuous variables, and chi-squared test for categorical variables. The differences in the efficacy parameters between day 0 and post-insertion of LNG-IUS were compared using paired-sample t -test. Data are presented as mean ± standard deviation or median (interquartile range, IQR) as appropriate. We assessed crude survival for time to LNG-IUS displacement or expulsion using the Kaplan–Meier survival plot, stratified by the 2 treatment groups with log-rank test. The Cox proportional-hazards model was performed to estimate the associations of hysteroscopic LNG-IUS fixation group with LNG-IUS displacement or expulsion expressed as hazard ratio with 95% confidence interval (CI). We further adjusted for baseline age, body mass index (BMI), uterine volume, endometrial thickness, CA125, VAS score, PBAC score in the Cox regression. All analyses were carried out using IBM Statistical Package for Social Sciences version 26. A 2-tailed P -value <0.05 was deemed statistically significant.

A total of 83 patients received treatment of LNG-IUS, of which 38 patients underwent the hysteroscopic LNG-IUS fixation surgery and 45 received the conventional LNG-IUS placement, with mean follow-up of 16.7 ± 7.5 months. The baseline characteristics before insertion are shown in Table 1 . These patients were 42.1 ± 6.6 years old with mean BMI of 23.1 ± 1.7 kg/m 2 . All patients had given birth at least once, with 30 (36.1%) patients had LNG-IUS displacement or expulsion history. They were diagnosed with adenomyosis and received LNG-IUS treatment due to overlarge uterine volume (299.3 ± 89.9 mL), thick endometria (0.73 ± 0.08), high CA125 level (93.2 ± 7.6 U/mL), severe dysmenorrhea (VAS score 6.49 ± 2.27) and excessive menstrual blood loss (PBAC score 155.9 ± 24.0). There were no differences in baseline characteristics between the hysteroscopic fixation surgery group and the conventional placement group (Table 1 ). Baseline characteristics of patients. BMI = body mass index, CA125 = carbohydrate antigen 125, PBAC = pictorial blood loss assessment chart, VAS = visual analog pain scale. The difference was compared using independent sample t -test. There were only 2 patients experienced LNG-IUS displacement or expulsion (one displacement event in month 2, 1 expulsion event in month 15) in the hysteroscopic LNG-IUS fixation surgery group (incidence 5.3%), as opposed to 14 events in the conventional placement group (incidence 31.1%). The Kaplan–Meier survival plot was shown in Figure 1 , indicating the significant difference in the survival probability (log-rank test P  = .0096). The Cox proportional-hazards model was performed to estimate the risk associations of hysteroscopic LNG-IUS fixation surgery with LNG-IUS displacement or expulsion (Table 2 ). In the crude model, hysteroscopic fixation surgery significantly decreased the risk (HR 0.15, 95% CI 0.03–0.66, P  = .012) of LNG-IUS displacement or expulsion compared with the conventional placement. Such protective effect was remained (HR 0.13, 95% CI 0.03–0.61, P  = .010) after adjustment for baseline age, BMI, uterine volume, endometrial thickness, CA125, VAS score and PBAC score. Association of hysteroscopic LNG-IUS fixation surgery with LNG-IUS displacement or expulsion. CI = confidence interval, HR = hazard ratio, LNG-IUS = levonorgestrel-releasing intrauterine system. Adjusted model was adjusted for baseline age, body mass index, uterine volume, endometrial thickness, carbohydrate antigen 125 level, visual analog pain scale score and pictorial blood loss assessment chart score. Kaplan–Meier survival plots of the hysteroscopic LNG-IUS fixation surgery group and conventional placement group. LNG-IUS = levonorgestrel-releasing intrauterine system. In the current study, we evaluated the comparative efficacy of hysteroscopic fixation surgery and conventional placement techniques in the managementof adenomyosis. The data presented in the Table 3 and Figure 2 illustrate the changes observed in key clinical parameters at various time points following the treatment interventions. Our results indicate that both hysteroscopic fixation and conventional placement methods led to a marked decrease in uterine volume from the 6-month posttreatment assessments to the 24-month visits compared with the baseline values. However, there were no significant differences in uterine volume reduction throughout the follow-up between the 2 treatments. Similarly, endometrial thickness was observed to decrease significantly over time (from month-3) in both groups. Yet, the hysteroscopic fixation group exhibited a more pronounced reduction, although this did not reach statistical significance when compared with the conventional placement group. CA125 levels, which are often used as a biomarker for uterine conditions, decreased significantly in both treatment groups from baseline from month-3. No significant difference was noted between the 2 groups in terms of CA125 level reduction. The reduction of uterine volume, endometrial thickness and CA125 level tended to reach a plateau after 18 months of the LNG-IUS insertion in both groups (Fig. 2 ). Comparisons of treatment efficacy. CA125 = carbohydrate antigen 125, PBAC = pictorial blood loss assessment chart, VAS = visual analog pain scale. denotes P  < .05, compared with baseline (paired samples t -test). denotes P  < .05, compared with conventional placement (independent samples t -test). Efficacy plots of the 2 different treatment groups. (A) uterine volume, (B) endometrial thickness, (C) carbohydrate antigen 125 level, (D) visual analog pain scale score, (E) pictorial blood loss assessment chart score. In terms of patient-reported outcomes, the VAS score, which measures pain levels, was significantly lower in both groups at all posttreatment time points compared to baseline. Patients who received hysteroscopic LNG-IUS fixation surgery reported significant lower VAS score after 6 months treatment till the end compared with the conventional placement group. Dysmenorrhea relief reached a plateau after 12 months of LNG-IUS insertion in both groups. The assessment of menstrual blood loss using the PBAC score also revealed a significant decrease in both groups posttreatment from month-3. The reduction in PBAC scores was more substantial in the hysteroscopic fixation group at month-3 and month-6 visits compared with the conventional placement group, however this did not differ significantly after 12 months treatment. The improvement of menstrual blood loss reached a plateau after 6 months of LNG-IUS insertion in both groups.

Adenomyosis, a prevalent gynecological condition, is defined by the intrusion of endometrial glands and stroma into the myometrial layer, leading to clinical manifestations such as irregular uterine bleeding, pelvic discomfort, and infertility. [ 18 ] The chronic blood loss associated with this disorder can result in anemia and psychological distress due to the persistent menstrual pain. Given the significant impact of adenomyosis on the quality of life, including menstrual disturbances, fertility, and pregnancy outcomes, there is a critical need for durable and effective treatment strategies. [ 5 ] Some experts have also suggested a potential link between adenomyosis and an elevated risk of endometrial cancer development, as both conditions are associated with shared molecular pathway alterations and similar local microenvironments. [ 19 ] Consequently, the demand for sustained and effective adenomyosis treatments is acute. LNG-IUS lifespan ranges from 3 to 7 years, depending on different available LNG-IUS models on the market. Researches indicate that the LNG-IUS is an effective long-term treatment for managing the symptoms of abnormal uterine bleeding and dysmenorrhea associated with adenomyosis. [ 10 , 11 , 20 ] However, the high incidence of LNG-IUS expulsion usually leads to the replacement of LNG-IUS and may discourage patients from continuing treatment for adenomyosis, particularly those with a uterine volume exceeding 150 mL. [ 21 ] Li et al followed up 828 patients with symptomatic adenomyosis for 8 years, and the expulsion rate for LNG-IUS was 21.8% (118/828). [ 11 ] Sheng J et al observed 94 patients with adenomyosis complicated by dysmenorrhea for 3 years, and the ring expulsion rate reached 10.6%. [ 22 ] Youm J et al followed up 481 patients with LNG-IUS for an average of 13.4 months, and 48 (9.6%) cases experienced ring expulsion, with the earliest occurrence 1 week after insertion and the latest at 27.2 months. [ 23 ] Given the significant efficacy of the LNG-IUS, but the issue of displacement/expulsion that prevents patients from continued use, gynecologists have been attempting various surgical procedures to reduce the rate of LNG-IUS displacement/expulsion. In 2021, Zhu L et al first proposed the use of hysteroscopic LNG-IUS fixation in a patient with adenomyosis who had a history of LNG-IUS expulsion [ 12 ] and garnered doctors’ attention. Zhang H [ 24 ] followed up on 12 patients with adenomyosis who underwent this surgical procedure for about 6 months. All 12 patients had symptoms of dysmenorrhea and/or menorrhagia, which were significantly improved after the surgery. In Zhang H’s study, 3 patients experienced ring displacement or expulsion at different times postoperatively. The cause of the ring displacement and expulsion was considered to be the use of absorbable suture material. After switching to nonabsorbable suture material, no cases of LNG-IUS displacement or expulsion were observed during the follow-up period. In our study, we followed up on 38 patients who underwent hysteroscopic LNG-IUS fixation surgery and 45 patients who received conventional placement, with follow-up time ranging from 3 to 24 months. In the hysteroscopic LNG-IUS fixation group, 1 patient experienced LNG-IUS displacement and 1 experienced expulsion postoperatively, which might be due to accidental improper placement during surgery or early strenuous activity. The remaining patients had ultrasound examinations at various times during the follow-up period, showing proper LNG-IUS placement (effective rate 94.7%), although 42.1% of the patients had a history of LNG-IUS displacement or expulsion before the treatment. In contrast, in the conventional placement group, 14 patients experienced LNG-IUS displacement or expulsion after the procedure, with the earliest occurrence 1 month postoperatively. The expulsion rate in the conventional placement group was generally consistent with previous studies, and most of the patients who experienced displacement or expulsion in this study had such event history. Regarding the efficacy of the 2 different treatment regimens, they both effectively reduced uterine volume, endometrial thickness and CA125 level to a comparable degree. Dysmenorrhea severity and excessive menstrual blood loss were also significantly improved following both treatments as indicated by VAS and PBAC scores, with hysteroscopic fixation surgery offering superior outcomes. We acknowledge certain limitations in our study. First, the sample size is relatively small. A larger-scale study is warranted in the future to obtain more robust and valid results. Second, this is an observational study rather than a randomized controlled trial, thus potential bias might exist. Third, for patients with adenomyosis, factors that affect the efficacy of LNG-IUS may include uterine volume and the type of adenomyosis. Adenomyosis is a heterogeneous disease, and it has been proposed to be classified into type I (intrinsic), type II (extrinsic), type III (intramural), and type IV (other types) based on magnetic resonance imaging. [ 1 ] Chen S et al found that LNG-IUS is effective for types I and II, but less effective for type IV. [ 25 ] However, we did not consider the impact of adenomyosis types at the beginning of this study design, and relevant indicators were not included. In future evaluations of whether a patient is suitable for LNG-IUS fixation, a thorough consideration of the patient’s medical history and magnetic resonance imaging should be undertaken. In conclusion, hysteroscopic LNG-IUS fixation surgery can effectively reduce the short-term rate of ring displacement or expulsion in patients with previous displacement/expulsion history and the presence of high-risk factors. The optimal indications for hysteroscopic LNG-IUS fixation should be in patients who have a history of LNG-IUS insertion, experience significant improvement in symptoms such as heavy menstrual bleeding or dysmenorrhea after placement, but have had premature LNG-IUS expulsion. This surgical approach can also be chosen for patients with newly diagnosed adenomyosis who are at high risk for LNG-IUS displacement/expulsion based on careful medical assessment. By selecting the most suitable population for hysteroscopic LNG-IUS fixation surgery, the effectiveness of the treatment can be further improved. However, hysteroscopic LNG-IUS fixation surgery as an innovative surgical method has limited clinical data and lacks long-term follow-up results. More researches are warranted to further clarify its efficacy and value.

Special thanks to all participants who have contributed the valuable data in the study.