Efficacy of balloon stent or oral estrogen for adhesion prevention in septate uterus: A randomized clinical trial.

To the Editor: Septate uterus is the most common congenital uterine malformation associated with an increased risk of miscarriage, preterm birth, and fetal malpresentation and reduced clinical pregnancy rates. The European classification defines septate uterus as having an internal fundal indentation exceeding 50% of the uterine wall thickness and an external indentation less than 50% of the uterine wall thickness.[1] Transcervical resection of septum (TCRS) may reduce miscarriage rates and improve live birth rates in patients with adverse reproductive histories, although this remains controversial.[2, 3] TCRS is widely performed in China, with 5484 and 5472 patients undergoing the procedure in 2020 and 2021 respectively, according to the National Center for Quality Control of Medical Records. However, the risk of de novo intrauterine adhesion (IUA) after TCRS is high. To prevent IUAs, various antiadhesive strategies, including absorbable barriers, physical barriers, and hormonal therapy, have been developed. Estrogen and intrauterine balloons have shown efficacy in preventing IUAs by promoting endometrial proliferation and separating the endometrial walls. However, the effectiveness of these methods in patients performed with TCRS has been poorly studied.[2] We designed a multicenter, randomized controlled trial (RCT) to assess the efficacy of intrauterine balloon stents and oral estrogen compared with no intervention in preventing de novo IUAs after TCRS (ChiCTR2000032061). The research protocol was approved by the Ethics Committee of Peking Union Medical College Hospital (PUMCH; No. JS-2268).[4] All the participants provided informed consent. Patients were recruited from 10 grade-A tertiary hospitals in China [Supplementary file, https://links.lww.com/CM9/C304] between May 2020 and April 2022. Patients were eligible if they were 20–35 years old (considering the legal marriage age), nulliparous, willing to be pregnant, had regular menstrual cycles, diagnosed with septate uterus according to the European Society of Human Reproduction and Embryology (ESHRE)/European Society of Gynecological Endoscopy (ESGE) classification criteria, and agreed to receive TCRS monitored by concomitant laparoscopy.[1] Exclusion criteria included the presence of double cervix, vaginal septum, IUAs, a history of metroplasty, severe endometriosis (The revised American Society for Reproductive Medicine [rASRM] stage 3–4), adenomyosis, submucosal tumors, endometrial malignancies, or breast cancer. Eligible participants were randomly assigned 1:1:1 to one of the three groups: (1) surgery-only (Group A), (2) balloon stent (Group B), and (3) estrogen (Group C). The randomization and implementation of the study were detailed in the protocol.[4] The patients did not know the allocation before the surgery. The investigators were informed of the group assignment after randomization, while the outcome assessors and data analysts were blinded. The surgeon who performed the second-look hysteroscopy were unaware of the group assignments. Surgical procedures involved resecting the uterine septum using bipolar energy instruments under general anesthesia, with laparoscopy to evaluate the uterine fundus and the integrity of the myometrium. For the balloon group (Group B), the intrauterine balloon stent (COOK Medical®, Bloomington, USA, 5 mL) was inserted at the end of the operation and removed in an outpatient setting one week later. The participants in the estrogen group (Group C) received 2 mg of oral estradiol valerate twice daily for two periods. The severity of IUAs was evaluated using the American Fertility Society (AFS) classification during second-look hysteroscopy. The trial was designed as a superiority study. Power and sample size calculations were described in the protocol.[4] The aim was to recruit 243 patients. The primary outcome was the occurrence of de novo IUAs confirmed by second-look hysteroscopy, and secondary outcomes included live birth rate (>28 weeks), pregnancy rate, ongoing pregnancy rate, pregnancy loss rate, and operation-related complications, such as uterine perforation, bleeding, pelvic infection, and other adverse events. Data were analyzed using Stata 17.0 software (StataCorp LLC, Texas, USA). A chi-squared test was used to compare rates. Continuous variables were presented as mean ± Standard Deviation (SD) or median (interquartile range [IQR]), and were compared using ANOVA or Kruskal-Wallis test. Discrete variables were presented as number (%). For the primary outcome, primary analysis was conducted on an intention-to-treat basis, with sensitivity analysis using per protocol analysis. A two-sided P <0.05 was considered statistically significant. A total of 874 patients were assessed for eligibility, with 253 randomly assigned to the surgery-only group (78 patients), balloon stent group (91 patients), and estrogen group (84 patients) [Supplementary Figure 1, https://links.lww.com/CM9/C304]. Demographic characteristics were balanced across groups [Supplementary Table 1, https://links.lww.com/CM9/C304]. The mean age was 28.6 ± 3.4 (range, 20–35) years. Supplementary Table 2, https://links.lww.com/CM9/C304, shows the representativeness of the participants to the general Chinese population. The interval between surgery and second-look hysteroscopy was comparable across groups: Group A (70 [IQR: 62–87] days), Group B (63 [IQR: 56–79] days), and Group C (76 [IQR: 62–87] days). In total, 184 patients (72.7%) had an incomplete septum, and 69 patients (27.3%) had a complete septum. Four patients (three from the surgery-only group and one from the estrogen group) did not attend the second-look hysteroscopy, with two patients from the surgery-only group becoming pregnant without contraception. No crossover occurred between groups. There was no significant difference in perioperative characteristics among the groups [Supplementary Table 3, https://links.lww.com/CM9/C304]. No balloons fell out in the balloon stent group after the operation. All the patients in the estrogen group adhered to the prescribed regimen. There was no statistically significant difference in the incidence of IUAs among the surgery-only, balloon, and estrogen group (7 of 78 [9.0%], 7 of 91 [7.7%], 11 of 84 [13.1%], respectively; P = 0.464). Additionally, no significant difference was found in the severity of adhesions between the groups. The incidence of the residual fundal notch also showed no significant difference across groups (7 of 78 [9.0%], 11 of 91 [12.1%], 5 of 84 [5.9%], respectively; P = 0.369) [Supplementary Table 4, https://links.lww.com/CM9/C304]. Comparing the balloon group with the surgery-only group, there was no statistically significant difference in the incidence of IUAs (7.7% vs. 9.0%; relative risk: 0.86; 95% CI, 0.31–2.34; P = 0.764). Similarly, there was no difference in the incidence of a residual fundal notch (12.1% vs. 9.0%; P = 0.517). No significant difference was found in the incidence of postoperative IUAs between the estrogen and surgery-only groups (13.1% vs. 9.0%; relative risk: 1.46; 95% CI, 0.59–3.58; P = 0.409). There was no statistically significant difference between the two groups regarding residual fundal notch (5.9% vs. 9.0%; P = 0.468). [Supplementary Table 4, https://links.lww.com/CM9/C304]. In the per-protocol analyses, the results were generally consistent with those of the primary analysis. Details are provided in Supplementary Table 5, https://links.lww.com/CM9/C304. None of the patients experienced complications during anesthesia, hysteroscopy, or laparoscopy. No adverse events related to the antiadhesion interventions were reported. In the 12-month follow-up after the second-look hysteroscopy, there were no significant differences in cumulative pregnancy rates among the three groups [Supplementary Figure 2, https://links.lww.com/CM9/C304]. Among patients attempting conception (n = 215), intrauterine pregnancy rates (64.7%, 65.3%, 75.0%) and live birth rates (20.6%, 21.3%, 15.3%) were similar across the surgery-only, balloon, and estrogen groups respectively [Supplementary Table 6, https://links.lww.com/CM9/C304]. All live births were full-term. Infertility and pregnancy loss rates also showed no significant difference among the three groups. At the one-year follow-up after second-look hysteroscopy, 87 patients were in a state of pregnancy, with unknown pregnancy outcomes. In this multicenter RCT involving patients with septate uteri, we found no significant difference in the incidence of de novo IUAs after TCRS when using balloon stents, oral estrogen, or no preventive intervention. The IUA guidelines proposed by the American Association of Gynecologic Laparoscopists (AAGL) and the ESGE supported the value of balloon stents (Level A) and estrogen (Level B) in preventing IUA reformation after adhesiolysis, but these findings do not extend to the primary prevention after TCRS. According to the guidelines of the American Society for Reproductive Medicine (ASRM) in 2016, there is no sufficient evidence to support or oppose any method for IUA prevention after TCRS (grade C).[2] Previous prospective studies on using estrogen or intrauterine balloon stents to prevent IUAs after TCRS have shown similar results with our findings, with no significant difference in IUA occurrence [Supplementary Table 7, https://links.lww.com/CM9/C304]. However, the sample size of these studies was small (n = 46/28/20, respectively) with limited power of test. A large-scale retrospective study was conducted on 238 Chinese patients also found no significant differences between patients allocated estrogen (2–4 mg of estradiol valerate), intrauterine devices, intrauterine balloons, or no preventive intervention.[5] Additionally, we observed no significant difference in pregnancy or pregnancy loss between the three groups one year following second-look hysteroscopy, which was consistent with previous studies, suggesting that these antiadhesion interventions would not improve short-term fertility outcomes [Supplementary references, https://links.lww.com/CM9/C304]. However, a considerable number of patients were still ongoing pregnant, necessitating continued follow-up. This study is a large multicenter RCT investigating the efficacy of postoperative adhesion prevention for TCRS, with 98.4% of the patients completing the second-look hysteroscopy even during the Coronavirus Disease 2019 (COVID-19) pandemic. However, the sample size may have been underpowered to detect significant differences due to smaller-than-expected differences between the groups. The use of a single balloon stent type and dose of estrogen may also limit the generalizability of the results. Further research with larger sample sizes and varying treatment regimens is needed. The autocrosslinked hyaluronic acid gel, which has shown efficacy in preventing IUAs after TCRS according to the literature, was not applied in this study because it had not been introduced into Chinese hospitals when this trial started. But from the perspective of clinical significance, the current results have suggested that using estrogen to prevent adhesion may be unnecessary. Considering the large population in China, our findings may change the current clinical situation of widespread use of antiadhesion agents in patients receiving TCRS, which is of great significance in reducing unnecessary medical intervention and expenses. The results of this multicenter RCT do not support the use of oral estrogen or intrauterine balloons to prevent IUA formation after TCRS or improve fertility outcomes. Acknowledgments We thank all participants in this study, all research staff in study sites, staff of the Resman central randomization platform, and members of the data safety monitoring board for their oversight of this trial. Funding This work was supported by grants from the National Key Research and Development Program of China (No. 2021YFC2701405), the National Natural Science Foundation of China (Nos. 82171614 and 82271656), Beijing Natural Science Foundation (No. 7232125). Conflicts of interest None.