Vaginal natural orifice transluminal endoscopic surgery versus laparoscopy for ovarian cystectomy: a prospective, open-label, randomized, non-inferiority pilot trial.

This study confirms that vNOTES is non-inferior to laparoscopy for ovarian cystectomy, with comparable surgical outcomes, while offering additional benefits such as improved cosmetic results and decreased postoperative pain. These findings support vNOTES as a viable alternative to traditional laparoscopy for selected patients undergoing ovarian cystectomy.

Between January 2021 and June 2023, 453 patients with adnexal masses were assessed for eligibility. Among them, 75 satisfied the eligibility criteria and received invitations to participate in the study. Of these, 11 declined to participate. Eventually, 64 patients were randomly allocated to either laparoscopic cystectomy or vNOTES cystectomy and were included for the Intent-To-Treat (ITT) analysis. One case in vNOTES group, which was converted to laparoscopy, was excluded from the per-protocol analysis. The flow chart of the study is presented in Fig.  1 . Baseline characteristics are presented in Table  1 . No suspicious signs of malignancy were identified intraoperatively; therefore, no frozen section examinations were performed in any of the cases. There were no statistically significant differences between the two groups in terms of age, nulliparous status, tumor size, and pathological diagnoses, except for body mass index (BMI). Additional procedures were comparable between the two groups (Table  2 ). Fig. 1 Flow chart. vNOTES Vaginal natural orifice transluminal endoscopic surgery Table 1 Patient baseline demographic and clinical characteristic Variables Laparoscopy (n = 32) vNOTES (n = 32) p Age (years) 28.8 ± 5.0 29.9 ± 5.5 0.395 Body mass index (kg/m 2 ) 22.8 ± 3.1 21.1 ± 2.5 0.020 Tumor size (cm) 5.3 ± 1.8 5.1 ± 1.3 0.727 Nulliparae Yes 17 (53.1%) 11 (34.4%) 0.207 No 15 (46.9%) 21 (65.6%) Pathological diagnoses Dermoid cyst 26 (81.2%) 28 (87.5%) 0.661 Endometriotic cyst 0 (0.0%) 1 (3.1%) Mucinous cystadenoma 3 (9.4%) 1 (3.1%) Serous cystadenoma 1 (3.1%) 1 (3.1%) Simple cyst 2 (6.3%) 1 (3.1%) vNOTES Vaginal natural orifice transluminal endoscopic surgery Table 2 Additional procedures during the allocated surgery Variables Laparoscopy (n = 32) vNOTES (n = 32) p Effect size (95% CI) Additional procedures 6 (18.8%) 7 (21.9%) 0.756 RD 3.1 (− 16.6 – 22.8) Diagnostic hysteroscopy 2 (6.3%) 0 (0.0%) Hysteroscopic adhesiolysis 0 (0.0%) 1 (3.1%) Contralateral ovarian cystectomy † 2 (6.3%) 0 (0.0%) Salpingostomy 0 (0.0%) 1 (3.1%) Ligation of Fallopian tube 0 (0.0%) 1 (3.1%) Pelvic adhesiolysis 0 (0.0%) 4 (12.5%) Myomectomy # 2 (6.3%) 0 (0.0%) † A cyst measuring approximately 1.5–2.0 cm was detected in the contralateral ovary in two cases from the laparoscopy group # Two cases were diagnosed with a single subserous myoma measuring 0.5 cm intraoperatively in the laparoscopy group vNOTES vaginal natural orifice transluminal endoscopic surgery, CI confidence interval, RD risk difference Flow chart. vNOTES Vaginal natural orifice transluminal endoscopic surgery Patient baseline demographic and clinical characteristic vNOTES Vaginal natural orifice transluminal endoscopic surgery Additional procedures during the allocated surgery † A cyst measuring approximately 1.5–2.0 cm was detected in the contralateral ovary in two cases from the laparoscopy group # Two cases were diagnosed with a single subserous myoma measuring 0.5 cm intraoperatively in the laparoscopy group vNOTES vaginal natural orifice transluminal endoscopic surgery, CI confidence interval, RD risk difference After adjusting for the randomization stratification factor (cyst size), the combined operative success rate was 96.9% (31/32) in the VNOTES group and 100% (32/32) in the conventional laparoscopy group, yielding a stratified risk difference (RD) of –0.031 (one-sided 95% CI lower limit: –0.127). Since the lower confidence limit (–0.127) was greater than the pre-specified non-inferiority margin of –0.15, the non-inferiority of VNOTES was statistically established (P for non-inferiority < 0.05). In the prespecified subgroup analysis, the success rates were 100% vs. 100% in the 0–5 cm stratum (n = 16 per group) and 93.8% (15/16) vs. 100% (16/16) in the 6–10 cm stratum for the VNOTES and conventional groups, respectively. Postoperative pain scores in the vNOTES group were significantly lower than in the laparoscopy group (median difference [MED] − 1, 95% CI − 1 – − 1; p = 6.1E-5). The median (IQR) pain scores were 1 (0, 1) in the vNOTES group and 2 (1, 3) in the laparoscopy group, respectively. Time of retrieving mass was also significantly shorter in the vNOTES group compared to the laparoscopy group (MED − 4, 95% CI − 7 – –2; p = 5.73E-7), with a median (IQR) of 3 (2, 3) mins and 7 (4, 12) mins respectively. However, estimated blood loss in the vNOTES group was significantly larger than in the laparoscopy group (MED 20, 95% CI 15 to 30; p = 4.23E-7), with a median (IQR) of 30 (20, 50) and 10 (5, 15) respectively (Table  3 ). There were no statistically significant differences in operative times, time to first postoperative flatus, and usage of analgesics between the laparoscopy group and the vNOTES group. (Table  3 ). Only one case in the vNOTES group required oral analgesics (Tramadol) once. Table 3 Surgical outcomes Variables Laparoscopy (n = 32) vNOTES (n = 32) p Effect size (95% CI) Primary outcomes Successful rate † 100% 96.9% (31/32) N/A RD − 3‧1% (95% CI L  − 12.7%) Secondary outcomes Operative time (min) 89.8 ± 31.6 93.0 ± 33.5 0.689 MD 3.3 (− 12.9 – 19.4) Adjust* 0.709 Adj. MD 3.2 (− 13.8 – 20.2) Estimated Blood loss, median, (IQR), ml 10 (5, 15) 30 (20,50) 4.23E-7 MED 20 (15 – 30) Time for specimen retrieval, median, (IQR), min 7 (4, 12) 3 (2, 3) 5.73E-7 MED − 4 (− 7 – − 2) Time to first flatus (h) 22.3 ± 9.2 18.3 ± 6.6 0.054 MD − 3.9 (− 7.9 – 0.1) Adjust* 0.023 Adj. MD − 4.8 (− 8.9 – − 0.7) Postoperative pain score, median, (IQR) 2 (1, 3) 1 (0, 1) 6.1E-5 MED − 1 (− 1 – − 1) Usage of analgesics Yes 0 (0.0%) 1 (3.1%) 1.000 RD 3.1% (− 2.2 – 8.4%) No 32 (100.0%) 31 (96.9%) † The primary outcome (successful rate) was compared using a stratified Cochran-Mantel–Haenszel analysis to account for randomization strata (cyst size). The non-inferiority margin was –15% *adjusted for baseline BMI imbalance vNOTES Vaginal natural orifice transluminal endoscopic surgery, CI confidence interval, CI L lower confidence interval limit, IQR interquartile range, MD mean difference, MED median difference, RD risk difference, Adj . MD adjusted mean difference Surgical outcomes † The primary outcome (successful rate) was compared using a stratified Cochran-Mantel–Haenszel analysis to account for randomization strata (cyst size). The non-inferiority margin was –15% *adjusted for baseline BMI imbalance vNOTES Vaginal natural orifice transluminal endoscopic surgery, CI confidence interval, CI L lower confidence interval limit, IQR interquartile range, MD mean difference, MED median difference, RD risk difference, Adj . MD adjusted mean difference To address the potential confounding effect of baseline BMI imbalance, sensitivity analyses were performed. The results were consistent: operative time difference remained non-significant (adjusted MD = 3.19 min, 95% CI − 13.81 – 20.19, p = 0.709), while the shorter time to first flatus with vNOTES became significant (adjusted MD = − 4.82 h, 95% CI − 8.96 – − 0.67, p = 0.023). The significant differences in estimated blood loss, postoperative pain score, and time for specimen retrieval were also confirmed (p < 0.001). The incidence rates of complications were comparable between these two groups (3.1% in the vNOTES group versus 6.3% in the laparoscopy group, RD − 3.1%, 95% CI − 13.5% – 7.2%, p = 1.000) (Table  4 ). Results from the per-protocol analysis were mostly consistent with those from the ITT analysis (Tables 5–7 in Supplementary materials). Table 4 Complications No. (%) Variables Laparoscopy (n = 32) vNOTES (n = 32) p Effect size (95% CI) Perioperative complications 2 (6.3%) 1 (3.1%) 1.000 RD − 3.1% (− 13.5 – 7.2%) Grade I* Uroschesis 1 (3.1%) 1 (3.1%) Grade II* Pelvic infected disease 1 (3.1%) 0 (0.0%) Grade III-V* 0 (0.0%) 0 (0.0%) *Complications were defined according to the Clavien-Dindo classification grade vNOTES Vaginal natural orifice transluminal endoscopic surgery, CI confidence interval, RD risk difference Complications *Complications were defined according to the Clavien-Dindo classification grade vNOTES Vaginal natural orifice transluminal endoscopic surgery, CI confidence interval, RD risk difference

This was a single-center, prospective, open-label, non-inferiority, randomized trial conducted at the Maternal and Child Health Hospital of Guangxi Zhuang Autonomous Region, China. The study was approved by the local Ethics Committee (Reference number: (2020–3) 4) on 8 June, 2020. The study was registered with the Chinese Clinical Trials Registry (identifier: ChiCTR2300070890). The work was reported in line with the Consolidated Standards of Reporting Trials (CONSORT) guidelines [ 18 ]. The inclusion criteria for the participants were as follows: (1) Women aged between 18 and 45 years; (2) Diagnosis of a benign ovarian tumor based on ultrasound examination or nuclear magnetic resonance (MRI); (3) Tumor size no greater than 10 cm. Patients who met any of the followings were excluded: virginity; pregnancy; history of pelvic inflammatory disease; history of pelvic surgery; genital tract malformation; active lower genital tract infection; bilateral ovarian tumor; suspected severe endometriosis, including rectovaginal endometriosis or Douglas endometriosis; serum cancer antigen 125 (CA-125) ≥ 40 IU/L; suspected malignancy; patients with significant internal diseases or psychiatric conditions who were not eligible for surgery; patients who decided to proceed with oophorectomy; patients who declined to undergo surgery or participate in the trial. All patients gave written informed consent before the intervention began. Eligible patients were randomized (1:1) to undergo cystectomy by either conventional laparoscopy or vNOTES. The computer-generated random allocation sequence was created by an independent investigator using Excel (Microsoft office 365). Randomization was stratified based on the ultrasound size of the ovarian cyst into category I (cyst < 6 cm) or category II( 6 ≤ cyst < 10 cm). This trial was open-label; therefore, neither participants nor investigators were masked during the treatment. Laparoscopic cystectomies were performed by multiple senior gynecological endoscopic surgeons, whereas all vNOTES procedures were performed by a single designated senior surgeon. The steps of vNOTES cystectomy were performed according to our previous report [ 19 ]. The patient was positioned in dorsal lithotomy and Trendelenburg position under general endotracheal anesthesia. Vaginal retractors were inserted to expose the cervix and vagina, which was then sterilized with povidone-iodine. A 2.5–3.0 cm posterior colpotomy was performed using cold scissors. A laparoscopic single-site Platform was used as the vNOTES port. A cortical incision was made using a monopolar, and a cleavage plane was identified. Ovarian cystectomy was then performed by using grasping forceps and cold scissors. The steps of laparoscopic cystectomy are described as follows: The patient was positioned in dorsal lithotomy and Trendelenburg position under general endotracheal anesthesia. The surgical site was disinfected and draped. A 1-cm incision was made in the umbilicus, and a Veress needle was inserted into the peritoneal cavity. Pneumoperitoneum was created with a pressure of 10 mmHg. After withdrawing the Veress needle, a 10 mm trocar was placed through the umbilical incision. A 10 mm laparoscope was used to examine the pelvis and abdominal cavity. Two or three additional trocars were placed in the bilateral lower abdomen. The capsule of the ovarian tumor was then enucleated and stripped from the healthy ovarian tissue, as described by Wang [ 16 ]. Any slight hemorrhage was coagulated using electrosurgical bipolar forceps. For larger bleeders, the ovarian remnants were sutured for hemostasis using a 2–0 polyglycolic acid suture. The specimens were placed in a disposable endobag and retrieved from the abdomen. Frozen section examinations were performed when suspicious signs of malignancy were identified intraoperatively. The primary outcome was the proportion of patients successfully treated by the allocated procedure without conversion to other surgeries. The secondary outcomes were operative time, estimated blood loss, postoperative pain score, usage of analgesics, time to first flatus, time for specimen retrieval, perioperative complications. Postoperative pain score was measured using a visual analogue scale (VAS) [ 20 ], which was monitored every 12 h after surgery or any time the patient complained about pain until discharge. The maximum pain score recorded was used for each patient in the analysis. Patients were followed up on days 7 and 30 postoperative. Perioperative complications include pelvic infection, hematoma formation, injury to other organs (such as the ureter, bladder, or bowel), wound infection, wound dehiscence (wound separation), and incisional hernia, etc. Complications were defined according to the Clavien-Dindo classification system [ 21 ]. A non-inferiority study was conducted to assess the primary outcome. According to previous report, 97.6% of patients were successfully treated using laparoscopy [ 22 ]. We assumed that 95% of vNOTES procedures would be completed successfully without the need for conversion, deriving an expected event rate (π) of 5%. A non-inferiority margin (δ) of 15% for the difference in conversion rates was considered clinically acceptable [ 13 ]. Using a one-sided test with a significance level (α) of 5% and a power (1-β) of 80%, the required sample size per group was calculated via the formula [ 23 ]: 2{π(1-π)δ −2 }(z 1-α  + z 1-β ) 2 . This calculation yielded a requirement of 26 participants per group to demonstrate the non-inferiority of vNOTES compared to conventional laparoscopy. To account for a potential drop-out rate of 20%, the final sample size was set at 64 participants (32 per group). The primary outcome, operative success rate, was compared between groups using a stratified Cochran-Mantel–Haenszel (CMH) method to account for the cyst size stratification (cyst < 6 cm and 6 ≤ cyst < 10 cm) used in randomization. The combined risk difference (RD) and its one-sided 95% confidence interval (CI) were derived from this analysis. Non-inferiority of VNOTES to conventional laparoscopy would be concluded if the lower limit of the one-sided 95% CI for the RD was greater than the non-inferiority margin (Δ) of −0.15. Secondary outcomes were compared between the two groups using two-sided tests. All analyses were conducted according to the intention-to-treat (ITT) principle. Additionally, a per-protocol analysis was performed in cases where conversion occurred. Normally distributed data were described using mean ± standard deviation (SD), while skewed data were described using median (interquartile range, IQR). We compared the two treatment arms using either the Chi-squared test or Fisher’s exact test for categorical variables, and the independent samples t test or Mann–Whitney U test for normally and skewed distributed variables respectively. Sensitivity analyses would be performed to address confounding if substantial baseline imbalances occurred. The level of statistical significance was set at p < 0.05. Statistical analyses were performed using SPSS version 26.0 software.

vNOTES is non-inferior to laparoscopy for ovarian cystectomy. vNOTES demonstrates superior cosmetic outcomes, lower postoperative VAS scores, and shorter specimen retrieval times. Further multicenter randomized trials with larger series are required to confirm these findings.

To the best of our knowledge, the present study is the first-ever report of a prospective randomized trial comparing ovarian cystectomy by vNOTES and laparoscopy. The findings of this trial reveal that ovarian cystectomy by vNOTES is non-inferior to laparoscopy, based on the successful rate of ovarian cystectomy performed by the allocated procedure without conversion. In the present study, no significant difference in the conversion rate was observed between vNOTES and laparoscopy, although one case of vNOTES (3.1%) was converted to laparoscopy due to severe endometriosis. Since endometriosis is closely associated with severe pelvic adhesion, it may not be suitable for vNOTES. Based on the current evidence, endometriomas are rarely included in vNOTES procedures [ 4 , 6 ]. Therefore, patients should be carefully assessed before being involved in vNOTES. In accordance with previous studies on vNOTES (Baekelandt et al., 2021; Huang et al., 2020; Wang, 2016) [ 4 , 10 , 16 ], patients who were virgins, suspected of having malignancy, had a history of pelvic inflammatory disease, a history of pelvic surgery, or an active lower genital tract infection were excluded from the current trial. However, a previous cesarean section and nulliparity may not be considered contraindications [ 16 , 24 ]. There is no consensus on the appropriate size of an ovarian mass for performing ovarian cystectomy by vNOTES [ 10 , 16 ]. Based on our experience, patients with ovarian masses no greater than 10 cm are eligible for vNOTES cystectomy. Besides its minimally invasive nature, the cosmetic benefits and reduced postoperative pain associated with vNOTES have been documented in in the studies by Baekelandt et al. (2018, 2021) [ 4 , 11 ]. The present study confirms the finding of Baekelandt et al. [ 4 ], from their randomized trial that postoperative pain scores are significantly lower with vNOTES than with conventional laparoscopy. This finding may be related to the characteristics of the pelvic nerve afferents that innervate the vagina, which tend to produce diffuse sensations that are difficult to localize [ 25 , 26 ]. Additionally, the flexibility of the vagina can facilitate specimen retrieval through a posterior colpotomy incision, especially for dermoid cysts containing bone, skin, and hair [ 22 ]. In the present study, vNOTES was associated with shorter specimen retrieval times compared to laparoscopy. The shorter time to first flatus observed in the vNOTES group further supports its role in promoting faster recovery, which is consistent with the findings from the Per-Protocol Population analysis. The incidence of complications was low in both the vNOTES and laparoscopy groups, and the difference was not statistically significant (3.1% vs 6.3%, p = 1.000). Given the low event rate, the study lacked sufficient statistical power to reliably detect a difference in this outcome. A randomized study conducted by Baekelandt et al. (2021), indicated a trend toward a higher incidence of severe events in the vNOTES group compared to the laparoscopy group (15% vs 3%, p = 0.08), although there were no statistically significant differences. Most of these events in the vNOTES group were postoperative bleedings (4/5) [ 4 ]. In the present trial, the estimated blood loss in the vNOTES group was greater than that in the laparoscopy group (median [IQR], 30 [20–50] ml vs 10 [5–15] ml, p < 0.001), which is consistent with the findings of a case-matched study by Wang et al. (2016) comparing natural orifice transluminal endoscopic surgery (NOTES)-assisted versus laparoscopic ovarian cystectomy [ 16 ]. Thus, vNOTES may be associated with greater blood loss, and extra attention should be paid to achieve complete hemostasis. Infection has been a concern for vNOTES. However, current evidence reveals that the infection rate remains relatively low in vNOTES (0.0–0.9%) [ 4 , 5 , 16 ]. This trial observed no cases of pelvic inflammation in the vNOTES group, thereby confirming the earlier finding by Baekelandt et al. (2021) that vNOTES does not increase the risk of pelvic infection [ 4 ]. However, since the low event rate of complications, the present study was not powered to detect differences in this outcome. A key strength of this study is that it constitutes the first prospective randomized trial directly comparing vNOTES with laparoscopy for ovarian cystectomy. However, several limitations warrant acknowledgment. First, the trial was non-blinded, which may introduce performance or assessment bias, though blinding was impractical in this surgical context. Second, being a single-center study, our findings require validation through future multicenter trials to ensure generalizability. Third, the involvement of multiple surgeons in performing the laparoscopic procedures and the performance of varied ancillary procedures could introduce technical heterogeneity, potentially limiting intergroup comparability. Additionally, the higher baseline Body Mass Index (BMI) observed in the laparoscopy group compared to the vNOTES group might have confounded the surgical outcomes. However, sensitivity analyses indicated that this difference did not substantially affect most surgical outcomes, with the exception of time to first flatus. Finally, the follow-up period was limited to short-term endpoints; long-term outcomes such as cyst recurrence and reproductive function remain to be investigated.

Benign ovarian cysts are one of the most common disorders in females [ 1 ]. Up to 10% of women in the United States will undergo surgical removal of an adnexal mass sometime in their lifetime [ 2 ]. Over the past decades, laparoscopic surgery has been the dominant minimally invasive approach for treating benign ovarian tumors [ 3 ]. However, vaginal natural orifice transluminal endoscopic surgery (vNOTES) is emerging as laparoscopy evolves and the demand for cosmetic outcomes increases [ 4 – 8 ]. vNOTES was first reported as a new treatment for benign ovarian tumors in 2012 [ 9 ]. Since then, vNOTES has garnered significant attention due to its advantages of minimal pain, scarless outcomes, and rapid recovery [ 4 , 10 – 13 ]. Accumulating evidence has shown that adnexectomy via vNOTES is both feasible and safe [ 4 , 14 , 15 ]. A prospective randomized controlled trial reported that vNOTES was non-inferior to laparoscopy for a successful adnexectomy without conversion, and it also demonstrated shorter operating times and reduced postoperative pain [ 4 ]. However, a significant number of patients are diagnosed with benign ovarian tumors during their reproductive years [ 1 ]. Thus, ovarian cystectomy is preferred over adnexectomy for patients in this group who desire to preserve fertility. In previous studies, ovarian cystectomy by vNOTES has been shown to yield superior outcomes. However, most of these studies have been limited by small sample sizes or retrospective designs [ 6 , 9 , 11 , 16 , 17 ]. To further establish the feasibility, safety, and advantages of performing ovarian cystectomy via vNOTES, a more comprehensive investigation is warranted. The present study aims to compare the surgical outcomes of ovarian cystectomy between vNOTES and laparoscopy, thereby providing a thorough evaluation of the clinical results and benefits of both procedures. This research will contribute valuable insights into the potential advantages and limitations of vNOTES, ultimately guiding clinical decision-making and improving patient care.

Below is the link to the electronic supplementary material. Supplementary file1 (DOCX 19 KB) Supplementary file2 (DOCX 21 KB) Supplementary file3 (DOCX 17 KB) Supplementary file1 (DOCX 19 KB) Supplementary file2 (DOCX 21 KB) Supplementary file3 (DOCX 17 KB)