Clinical efficacy analysis of LNG-IUS tail anchoring to the superficial myometrium under hysteroscopic direct vision.

There were 75 patients in the common placement group and 131 patients in the anchoring group. Differences in the baseline data between the two groups were compared. There were no significant differences in age, gravidity, parity, uterine volume, hemoglobin concentration, uterine cavity length, or delivery history ( P >0.05). However, there were significant differences in the proportions of patients with adenomyosis at the surgeon level between the two groups ( P <0.05), as shown in Table  1 (Table  1 Comparison of general data between the two groups). Table 1 Comparison of general data between the two groups Variables Common placement group ( n = 75) Anchoring group ( n = 131) p Age(y) 37.52 ± 6.24 39.06 ± 5.06 0.071 Gravidity( n ) 3.00 (2.00, 4.00) 3.00 (2.00, 4.00) 0.920 Parity( n ) 1.00 (1.00, 2.00) 1.00 (1.00, 2.00) 0.617 Uterine volume(cm 3 ) 115.93 ± 65.74 123.10 ± 67.06 0.458 Hemoglobin concentration(g/L) 114.74 ± 22.03 115.69 ± 19.43 0.749 Uterine cavity length(cm) 8.41 ± 1.00 8.59 ± 1.15 0.247 Delivery history ( n ,%)  No 11 (14.67) 10 (7.63) 0.108  Yes 64 (85.33) 121 (92.37) Adenomyosis( n ,%)  No 39 (52.00) 28 (21.37)  < 0.001  Yes 36 (48.00) 103 (78.63) Surgeon level( n ,%)  Experienced group 22 (29.33) 80 (61.07)  < 0.001  Inexperienced group 53 (70.67) 51 (38.93) For statistical methods and significance, please consult section"1.5 Statistical analysis".The difference was statistically significant at the p < 0.05 level Comparison of general data between the two groups For statistical methods and significance, please consult section"1.5 Statistical analysis".The difference was statistically significant at the p < 0.05 level Fifty-three patients in the anchoring group and 53 patients in the common placement group were ultimately included after matching. The reanalysis of the data revealed no significant differences between the anchoring group and the common placement group in terms of age, gravidity, parity, uterine volume, hemoglobin concentration, uterine cavity length, delivery history, proportion of patients with adenomyosis or surgeon level ( P >0.05), as shown in Table  2  (Table  2 Comparison of general data between the two groups of matched patients). Table 2 Comparison of general data between the two groups of matched patients Variables Common placement group ( n = 53) Anchoring group ( n = 53) p Age 38.00 ± 6.15 39.90 ± 5.06 0.179 Gravidity 3.00 (2.00, 4.00) 3.00 (2.00, 4.00) 0.341 Parity 1.00 (1.00, 2.00) 1.00 (1.00, 2.00) 0.645 Uterine volume 105.37 ± 61.67 115.75 ± 59.74 0.803 Hemoglobin concentration 116.2 ± 21.80 114.00 ± 21.10 0.949 Uterine cavity length 8.35 ± 1.01 8.40 ± 0.91 0.911 Delivery history, n (%) 0.713  No 5 (9.43) 3 (5.66)  Yes 48 (90.57) 50 (94.34) Adenomyosis, n (%) 0.843  No 22 (41.5) 21 (39.6)  Yes 31 (58.5) 32 (60.4) Surgeon level, n (%) 0.689  Experienced group 21 (39.6) 19 (35.8)  Inexperienced group 32 (60.4) 34 (64.2) For statistical methods and significance, please consult section "1.5 Statistical analysis". The difference was statistically significant at the p < 0.05 level Comparison of general data between the two groups of matched patients For statistical methods and significance, please consult section "1.5 Statistical analysis". The difference was statistically significant at the p < 0.05 level After PSM, the re-expulsion rate was 18.87% for the anchoring group and 43.40% for the common placement group. The re-expulsion rate was significantly lower in the anchoring group than in the common placement group, as shown in Table  3 (Table  3 Comparison of the drop-out rates of the two groups of matched patients). The re-expulsion time was significantly lower in the common placement group than in the anchoring group ( P <0.05). The re-expulsion times of the two groups of patients who were successfully matched are shown in Table  4 (Table  4 Comparison of re-expulsion time between two groups of matched patients). Table 3 Comparison of the drop-out rates of the two groups of matched patients Group Retention, n (%) Expulsion, n (%) χ2 p Common placement group ( n = 53) 30 (56.60) 23 (43.40) 7.436 0.006 Anchoring group ( n = 53) 43 (81.13) 10 (18.87) The chi − square test was employed to compare the rates between the two groups.The difference was statistically significant at the p < 0.05 level Table 4 Comparison of re-expulsion time between two groups of matched patients Group Re-expulsion time (months) z p Common placement group 9.00(3.00 ~ 12.00) −3.005 0.002 Anchoring group 18.00(12.50 ~ 36.00) The chi − square test was employed to compare the rates between the two groups. The difference was statistically significant at the p < 0.05 level Comparison of the drop-out rates of the two groups of matched patients The chi − square test was employed to compare the rates between the two groups.The difference was statistically significant at the p < 0.05 level Comparison of re-expulsion time between two groups of matched patients The chi − square test was employed to compare the rates between the two groups. The difference was statistically significant at the p < 0.05 level No cases of uterine perforation, water poisoning, increased bleeding, or postoperative infection occurred. Patients with preoperative dysmenorrhea reported experiencing dysmenorrhea relief after LNG-IUS placement. There was no increase in pain in the anchoring group. Recurrent LNG-IUS expulsion was observed in 36 of 75 patients in the common placement group and in 25 of 131 patients in the anchoring group after surgery, as shown in Table  5 (Table  5 Statistics of the postoperative LNG-IUS in the two groups). Table 5 Statistics of the postoperative LNG-IUS in the two groups Group Expulsion, n (%) Retention, n (%) Common placement group ( n = 75) 36 (48.00) 39 (52.00) Anchoring group ( n = 131) 25 (19.08) 106 (80.92) The chi − square test was employed to compare the rates between the two groups. The difference was statistically significant at the p < 0.05 level Statistics of the postoperative LNG-IUS in the two groups The chi − square test was employed to compare the rates between the two groups. The difference was statistically significant at the p < 0.05 level In the anchoring group, 19.08% of patients experienced re-expulsion of the LNG-IUS. The patients in the anchoring group were divided into short-tail and long-tail groups on the basis of the length of the tail. There were no significant differences in age, gravidity, parity, uterine volume, hemoglobin concentration, uterine cavity length, delivery history, proportion of patients with adenomyosis or surgeon level between the two groups ( P >0.05), as shown in Table  6 (Table  6 Comparison of the general data of patients between the short-tail group and the long-tail group). The re-expulsion rate in the long-tail group (6.90%) was significantly lower than that in the short-tail group (28.77%), and the difference was significant ( P <0.05). The odds ratio (95% confidence interval) was 5.452 (95% CI: 1.752–16.962), which passed the z test (z = 2.929, p = 0.003), as shown in Table  7 (Table  7 Comparison of reshedding rates between the short-tail and long-tail groups). Table 6 Comparison of the general data of patients between the short-tail group and the long-tail group Variables Short-tail group ( n = 73) Long-tail group ( n = 58) p Age 39.21 ± 5.48 38.88 ± 4.58 0.365 Gravidity 2.00 (2.00, 4.00) 3.00 (2.00, 4.00) 0.570 Parity 1.00 (1.00, 2.00) 1.00 (1.00, 2.00) 0.721 Uterine volume 121.20 ± 68.78 125.50 ± 65.35 0.717 Hemoglobin concentration 117.85 ± 17.94 112.97 ± 21.00 0.154 Uterine cavity length 8.59 ± 1.09 8.60 ± 1.22 0.977 Delivery history, n (%) 0.344  No 7 (9.59) 3 (5.17)  Yes 66 (90.41) 55 (94.83) Adenomyosis, n (%) 0.549  No 17 (23.29) 11 (18.97)  Yes 56 (76.71) 47 (81.03) Surgeon level, n (%) 0.609  Experienced group 46 (63.01) 34 (58.62)  Inexperienced group 27 (36.99) 24 (41.38) For statistical methods and significance, please consult section "1.5 Statistical analysis". The difference was statistically significant at the p < 0.05 level Table 7 Comparison of reshedding rates between the short-tail and long-tail groups Group Expulsion, n (%) Retention, n (%) χ 2 p OR OR(95% CI) Short-tail group ( n = 73) 21 (28.77) 52 (71.23) 10.01 0.002 5.452 1.752–6.962 Long-tail group ( n = 58) 4 (6.90) 54 (93.10) The chi − square test was employed to compare the rates between the two groups.  The difference was statistically significant at the p < 0.05 level Comparison of the general data of patients between the short-tail group and the long-tail group For statistical methods and significance, please consult section "1.5 Statistical analysis". The difference was statistically significant at the p < 0.05 level Comparison of reshedding rates between the short-tail and long-tail groups The chi − square test was employed to compare the rates between the two groups.  The difference was statistically significant at the p < 0.05 level

This was a retrospective case‒control study conducted at the Third Xiangya Hospital, Central South University. Patients who had undergone LNG-IUS insertion between March 1, 2017, and March 1, 2022, were included in the study. The screening criteria were focused primarily on patients with at least one previous history of LNG-IUS expulsion who were requiring conservative therapy with LNG-IUS and who were reluctant to undergo endometrial or uterine removal. Patients with known or suspected malignancies of the reproductive system were excluded from the study. A total of 206 patients met these criteria. This research received approval from the Ethics Committee of the Third Xiangya Hospital of Central South University, and all patients were informed about the study and signed an informed consent form (Ethics Approval number: 24301). The patients were divided into a common placement group (LNG-IUS placement under hysteroscopy) and an anchoring group (LNG-IUS tail anchoring to the superficial myometrium under hysteroscopic direct vision). Comparisons of the re-expulsion rate of the LNG-IUS and the incidence of adverse reactions were performed between the two groups. The anchoring group was divided into short-tail and long-tail groups on the basis of tail length, and the re-expulsion rate of LNG-IUS was compared between the two groups. Upon completion of the preoperative assessment, the patient voided her bladder and assumed the lithotomy position. The patient was then administered intravenous general anesthesia, and basic vulvar and vaginal disinfection was performed. The hysteroscope is performed with a 5 mm diameter STORZ hysteroscope system (Germany) or an Olympus hysteroscope system (Japan). The cervical canal, uterine cavity, endometrium, bilateral uterine angles, and fallopian tube orifice were meticulously visualized, and specimens were procured for histopathological analysis. Curettage or resection of any intrauterine lesions was performed as necessary. The length of the patient's uterine cavity was measured, cervical dilation was used, the LNG-IUS was inserted into the uterine cavity through the cervix, the arms were extended and placed to the fundus, and the cannula and putter were withdrawn. Under direct hysteroscopic visualization, the distal end of the tail (approximately 5–10 cm from the external cervical opening) was grasped with 5 Fr single-jointed forceps, entered the uterine cavity through the cervical canal and submerged through the superficial muscularis propria of the lower and middle portions of the left or right wall of the uterine cavity, with an upward and downward submerged distance of 1.5–2.5 cm, and then passed through the upper end of the lateral wall to reveal the tail. The tail was then pulled to the external cervical opening via miniature forceps to moderately pull and adjust the tail wire through the muscularis propria. The tail filament was pulled out to the outside of the cervical vertebra, and the other tail filament was cut at a distance of 2 cm from the outside of the cervix see Figs. 1 (Surgical illustration of anchoring surgery) and 2 (Surgical video illustration of anchoring surgery). The link of Surgical video was placed at the end of the article as Supplementary Information). Fig. 1 Surgical illustration of anchoring surgery. A  LNG-IUS in normal position; B  The tail of LNG-IUS submerged from the lateral wall of the uterus from the bottom to the top; C  Tying and fixing of the tail that comes out of the upper segment of the uterine sidewall Fig. 2 Surgical video illustration of anchoring surgery. ( A  The distal end of the tail (approximately 5–10 cm from the external cervical opening) was grasped with a 5 Fr single-jointed forceps, entered the uterine cavity. B  The starting position is the middle and lower part of the left or right wall of uterus. C  The tail of LNG-IUS submerged through the superficial muscularis propria from the middle and lower part of the left or right wall of uterus from bottom to top, and the sneak distance is about 1.5–2.5 cm. D  The tail of LNG-IUS passed through the upper end of the lateral wall. E  Pull down and adjust the position of the tail of LNG-IUS. F  The tail of LNG-IUS wire is knotted and fixed) Surgical illustration of anchoring surgery. A  LNG-IUS in normal position; B  The tail of LNG-IUS submerged from the lateral wall of the uterus from the bottom to the top; C  Tying and fixing of the tail that comes out of the upper segment of the uterine sidewall Surgical video illustration of anchoring surgery. ( A  The distal end of the tail (approximately 5–10 cm from the external cervical opening) was grasped with a 5 Fr single-jointed forceps, entered the uterine cavity. B  The starting position is the middle and lower part of the left or right wall of uterus. C  The tail of LNG-IUS submerged through the superficial muscularis propria from the middle and lower part of the left or right wall of uterus from bottom to top, and the sneak distance is about 1.5–2.5 cm. D  The tail of LNG-IUS passed through the upper end of the lateral wall. E  Pull down and adjust the position of the tail of LNG-IUS. F  The tail of LNG-IUS wire is knotted and fixed) We obtained general patient information from the hospital's electronic medical records, including age, gravidity, parity, uterine volume, hemoglobin concentration, disease type, and surgical conditions (such as uterine cavity length, operator expertise level, and LNG-IUS tail length). All patients were followed up to document the rate of re-expulsion and the incidence of adverse reactions. General information included preoperative uterine volume measured by B-ultrasound, which was calculated via the following formula: uterine volume (mL) = length (cm) × height (cm) × width (cm) × 0.5236. The surgical evaluation indicators included uterine cavity length and operator experience level. The latter were categorized into junior and senior groups. The junior group comprised physicians with less than 5 years of work experience, whereas the senior group consisted of physicians with more than 5 years of work experience and those holding intermediate, associate senior, or full senior professional titles. The length of the LNG-IUS tail was classified as short (the length of the retained tail measured approximately 5–7 cm) or long (the length of the retained tail measured approximately 7–10 cm). The position of the LNG-IUS was assessed by B-ultrasound, with normal positioning defined as the lower pole located above the internal cervical orifice. No LNG-IUS was found in the uterine cavity, and an ectopic LNG-IUS was excluded from the pelvic and abdominal cavities. If the LNG-IUS protruded from the vaginal opening or the lower extremity of the longitudinal arm of the LNG-IUS below the internal cervical opening during the operation, it was considered detachment. The main adverse reactions included uterine perforation, fluid overload, massive bleeding, and infection. Data analysis was performed via SPSS 26.0 software. Patients were divided into two groups on the basis of the surgical method used, and propensity score matching (PSM) with a 1:1 matching ratio and a caliper value of 0.02 was used. Propensity score matching (PSM): The test group and the control group are screened via specific statistical methods to ensure that the selected subjects have comparable clinical characteristics, thus minimizing potential confounders. The re-expulsion rate of the matched patients in the two groups was analyzed via the chi-square test to assess the clinical efficacy of LNG-IUS tail wire uterine superficial myometrial anchoring under hysteroscopy ( P <0.05). Normally distributed data were compared via independent sample t tests, whereas nonnormally distributed data were compared via nonparametric Wilcoxon rank-sum tests. Between-group comparisons of count data were analyzed via the chi-square test.

The factors related to LNG-IUS expulsion found in current studies include adenomyosis, heavy menstruation, a large uterine volume, and surgical experience. A higher expulsion rate is related not only to abnormal uterine contractility caused by diffuse or limited muscle fiber connective tissue hyperplasia in adenomyosis but also to the fact that adenomyosis often has a large uterine volume or uterine cavity volume and a large menstrual volume. A larger uterine cavity or uterine cavity malformation weakens the effect of intrauterine immobilization of the LNG-IUS, coupled with a series of inflammatory reactions around the lesion. Menstrual blood flow scouring, foreign bodies, and exclusive aspects of joint action can all lead to LNG-IUS in large uterine adenomyosis patients being likely to fall off. YOUM J et al. reported that the incidence of shedding in patients with adenomyosis was significantly greater than that in patients with a normal uterus [ 8 ]. The multivariate logistic regression analysis of factors related to LNG-IUS re-expulsion revealed that increased uterine volume may increase the rate of LNG-IUS re-expulsion. Moreover, when the LNG-IUS is placed again in patients with a history of LNG-IUS loss, the operation method is also an independent factor affecting the re-expulsion of the LNG-IUS. The efficacy of LNG-IUS anchoring was analyzed by controlling for confounding factors via PSM or by analyzing influencing factors affecting the re-expulsion of LNG-IUS. These results indicated that anchoring surgery could effectively prevent the redischarge of LNG-IUS (Table  8 Univariate and multivariate logistic regression analysis of LNG-IUS re-expulsion related factors). Table 8 Univariate and multivariate logistic regression analysis of LNG-IUS re-expulsion related factors Variables Univariatelogistic regression analysis Multivariate logistic regression analysis P OR (95%CI) P OR (95%CI) Age 0.083 1.05 (0.99 ~ 1.11) 0.151 1.05 (0.98 ~ 1.12) Gravidity 0.060 1.14 (0.99 ~ 1.30) 0.642 1.05 (0.86 ~ 1.27) Parity 0.241 1.26 (0.85 ~ 1.87) 0.711 1.12 (0.60 ~ 2.10) Uterine volume 0.004 1.01 (1.01 ~ 1.01) 0.009 1.01 (1.01 ~ 1.01) Hemoglobin concentration 0.244 0.99 (0.98 ~ 1.01) 0.954 1.00 (0.98 ~ 1.02) Uterine cavity length 0.765 1.04 (0.80 ~ 1.37) 0.881 1.02 (0.75 ~ 1.41) Delivery history  0 1.00 (Reference) 1.00 (Reference)  1 0.270 1.89 (0.61 ~ 5.88) 0.869 1.14 (0.25 ~ 5.17) Disease (Non-adenomyosis or Adenomyosis)  0 1.00 (Reference) 1.00 (Reference)  1 0.482 0.80 (0.42 ~ 1.50) 0.797 1.11 (0.51 ~ 2.39) Surgeon level  0 1.00 (Reference) 1.00 (Reference)  1 0.013 0.46 (0.25 ~ 0.85) 0.351 0.72 (0.35 ~ 1.45) Operation method  0 1.00 (Reference) 1.00 (Reference)  1  < 0.001 3.91 (2.09 ~ 7.34)  < 0.001 4.62 (2.14 ~ 9.99) OR Odds Ratio, CI Confidence Interval Univariate and multivariate logistic regression analysis of LNG-IUS re-expulsion related factors Patients who previously experienced IUD detachment reportedly have a greater risk of re-expulsion after reinsertion, with net cumulative re-expulsion rates of 21.7% and 31.4% at 6 and 12 months, respectively [ 9 , 10 ]. Currently, the surgical techniques used to reduce the failure rate of LNG-IUS include the "prefixed LNG-IUS method" and the "new intrauterine knot method to fix LNG-IUS", both of which utilize nonabsorbable sutures to secure the LNG-IUS in the uterine cavity [ 11 ]. There are differences in the procedural methods used. In the former approach, a single knot is not tied within the uterine cavity; instead, the end of the single knot is inserted into the perfusion hole of the hysteroscope or pushed into the uterine cavity 2–3 times using a knot pusher before the suture is cut off. However, this repetitive entry and exit from the uterine cavity may increase the risk of infection and prolong the operation time, potentially increasing the risk of fluid overload. In contrast, in the latter approach, a slip tie is first performed outside of the uterine cavity before the needle inside is sutured via a laparoscopic needle holder. The pull suture technique is then used to tighten one end of the suture to secure the LNG-IUS while releasing it. The slip-knot technique simplifies the knotting technique, but both methods require strict aseptic conditions, which increases costs. It is unclear whether the use of nonabsorbable sutures increases the risk of intrauterine foreign body reactions, inflammation, or irregular bleeding. Our method has been validated in previous studies (the sample size in a previous study included 20 participants), demonstrating the effectiveness of anchoring the superficial myometrium of the LNG-IUS tail under hysteroscopy to prevent expulsion. We increased the sample size and used rigorous statistical analysis, which revealed a significant difference in re-expulsion rates between the two groups ( P < 0.05). No adverse events were reported, indicating that this approach is safe and feasible. There was a recurrence rate of 19.08% (25/131) in the anchoring group. Subgroup comparisons were made according to the length of the tail used in the anchoring procedure. The re-expulsion rate of the long-tail group (6.90%) was significantly lower than that of the short-tail filament group (28.77%), and there was a significant difference between the two groups ( p <0.05). Compared with the other methods, the short-tail anchoring method was more likely to fall off, and the length of the modified long tail filaments fixed was approximately 7–10 cm, which was less likely to fall off. The long tail wire ensures that the stealth distance in the muscle layer is between 1.5 cm and 2.5 cm. At the same time, leaving enough tail wire length can ensure that the tail penetrating from the muscle layer can be knotted and fixed, avoid ineffective anchoring, strengthen the effect of anchoring, and effectively reduce the shedding rate. This study fully confirmed that in patients with a history of LNG-IUS expulsion, anchoring can effectively prevent the occurrence of re-expulsion and that the duration of re-expulsion is longer (Table  4 ), the duration of LNG-IUS use is longer, and the utilization rate is higher than that of patients with conventional placement. The method we adopted on the equipment only needs ordinary hysteroscopy equipment, does not require the addition of additional surgical materials or human resources, and has good economic benefits. This method is effective and safe and can extend the life of LNG-IUS, avoid repeated surgery and waste medical resources. In addition, this study selected only patients from our hospital, resulting in a certain selectivity bias in the study results, and it needs to be conducted in more regions with a larger sample size and promoted in more hospitals. In conclusion, the anchoring of the superficial muscularis of the tail filament of the LNG-IUS under direct hysteroscopy provides a new option for patients with a history of LNG-IUS loss, which can maximally prevent the recurrence of loss, ensure the treatment effect, and have obvious clinical application and promotion value. This method has the advantages of simple operation, no need for special equipment or instruments, and minimal surgical trauma and is suitable for all levels of hospitals.

The LNG-IUS is widely utilized for the management of menorrhagia and dysmenorrhea-related disorders. Research has demonstrated that, in addition to its contraceptive properties, the LNG-IUS is efficacious in treating abnormal urine bleeding from ovulatory dysfunction, adenomyosis, and endometriosis. It plays a crucial role in the prevention and treatment of various gynecological conditions [ 1 ]. However, a single design may increase the susceptibility of the LNG-IUS to fall off because of factors such as uterine cavity enlargement, abnormal uterine contractility and menorrhagia. In their retrospective investigation of 1100 patients with adenomyosis who were treated with LNG-IUS, Li L et al. reported that the cumulative expulsion rate after 12 months of LNG-IUS placement in patients with adenomyosis was as high as 11% [ 2 ]. The reported expulsion rate of LNG-IUS ranges from 9.1% to 37.5% [ 3 , 4 ]. Patients experiencing LNG-IUS expulsion may experience rapid recurrence of dysmenorrhea and menorrhagia, potentially diminishing the effectiveness of treatment and leading to treatment interruption or failure. Furthermore, the cost of the LNG-IUS is high, and substantial medical resources are needed for both the insertion and removal of the LNG-IUS. Therefore, mitigating LNG-IUS expulsion and extending the retention period have emerged as pressing concerns that require attention. Both high-intensity focused ultrasound (HIFU) and gonadotropin-releasing hormone agonist (GnRH-a) have been shown to significantly reduce the uterine volume and prolong the use rate of the LNG-IUS [ 5 ]. However, the outcome of HIFU treatment is typically incomplete, and patients can easily relapse. After 3 months of GnRH-a treatment, there may be a partial rebound in the uterine volume and an increased likelihood of LNG-IUS expulsion [ 6 ]. Therefore, we took a different approach and conducted an innovative study on surgical methods involving anchoring the LNG-IUS tail to the superficial myometrium under hysteroscopic direct vision (referred to as anchoring surgery). Initial studies have indicated that this procedure effectively prevents LNG-IUS expulsion [ 7 ]. To further investigate the clinical efficacy of this technique, we increased the sample size of our study.

Supplementary Material 1. Surgical video. Supplementary Material 1. Surgical video.