Comparison of conventional multiport laparoscopy, single-port laparoscopy, and vNOTES for the management of large adnexal masses (≥ 10 cm): a retrospective cohort study.

This was a multicenter, retrospective cohort study comparing postoperative pain and recovery after salpingo-oophorectomy performed using three different MIS methods between January 2020 and December 2025. This study was performed in line with the principles of the declaration of Helsinki. It was approved by the institutional review board (Protocol: 26003; Decision 1: 44) in February 2026. Written informed consent was obtained from all participants for both the surgical procedure and the use of their anonymized data for research publication. The inclusion criteria were as follows: age ≥ 18 years, CML, SPL, or vNOTES approach for salpingo-oophorectomy, preoperative imaging suggesting benign adnexal disease ≥ 10 cm, and complete perioperative and follow-up data were available. The exclusion criteria were as follows: deep infiltrating endometriosis, suspected or confirmed invasive ovarian malignancy preoperatively, concomitant major gynecologic or non-gynecologic procedures (e.g., hysterectomy, extensive adhesiolysis, oncologic debulking), previous pelvic radiation, American Society of Anesthesiologists (ASA) physical status ≥ IV, and refusal or inability to complete postoperative questionnaires. Before surgery, all eligible patients received comprehensive counseling regarding the CML, SPL, and vNOTES approaches. This counseling included detailed information on incision types, expected postoperative pain, cosmetic outcomes, and potential risks, after which patients were allowed to choose their preferred method, provided that there were no contraindications to vNOTES, such as a very narrow vaginal introitus, extensive suspected intra-abdominal adhesions, or technical limitations identified by the surgeon. This patient-preference-based allocation reflects real-world clinical practice; however, it introduces a significant source of selection bias. Patients who self-selected into the vNOTES group may differ systematically from those who chose CML or SPL in ways not fully captured by the recorded covariates—including motivation, health literacy, pain tolerance, and preference for scar-free surgery—potentially inflating the apparent benefits of vNOTES on patient-reported outcomes. This limitation is important and is addressed in detail in the discussion. All patients underwent a standardized preoperative evaluation, including medical history, physical and bimanual pelvic examinations, and transvaginal and/or transabdominal ultrasonography. For adnexal masses with suspicious ultrasonographic features, magnetic resonance imaging was performed to further characterize their morphology and size. The International Ovarian Tumor Analysis simple rules were applied to distinguish between benign and malignant adnexal masses, and patients with features suspicious for malignancy were excluded or managed according to oncologic protocols. The recorded demographic and clinical variables included age, body mass index (BMI), parity, menopausal status, comorbidities, smoking status, history of previous abdominal or pelvic surgery, BRCA mutation status, serum CA-125 level, and detailed imaging characteristics of the adnexal mass (size, laterality, and presumed histological type). Two experienced gynecologic oncologists (KG and VG) performed all surgeries, each with extensive prior experience in minimally invasive gynecologic procedures, including CML, SPL, and vNOTES. Enhanced Recovery After Surgery (ERAS) protocols were institutionalized for all patients (Supplementary Table). Local anesthesia was not applied to the incision sites. Pneumoperitoneum was successfully established in all three groups, with pressure of 8–10 mmHg for VNOTES and 10–12 mmHg for CML and SPL. Under general anesthesia, patient positioning was as follows: For the CML and LESS groups, after gaining abdominal access and creating pneumoperitoneum, patients were placed in a steep Trendelenburg position. In the vNOTES group, the Trendelenburg position was adopted prior to making the initial vaginal incision to allow for cephalad bowel displacement and enhance pelvic visibility. The surgical procedures for CML, SPL, and vNOTES salpingo-oophorectomy followed the standardized techniques previously established and reported by our group [ 10 , 11 ]. Port setup time was defined as the interval from the initial skin or vaginal incision to the completion of port placement and the first adequate intra-abdominal visualization with a laparoscope. Operation time was defined as the interval from the completion of port setup to the final closure of the skin or vaginal mucosa; it therefore excluded port setup and was measured separately from specimen removal time, which began at the opening of the retrieval bag or initiation of extraction and ended with complete removal of the adnexa from the abdominal cavity. Total procedure time (from initial incision to final closure) can be approximated as the sum of port setup, operation, and specimen removal times. Estimated blood loss (EBL) was recorded from the anesthesia records and the operative reports. Intraoperative complications were defined as any adverse events occurring from incision to closure, including organ injury, major bleeding requiring transfusion or conversion to laparotomy, and equipment-related problems. Postoperative complications were defined as events occurring within 30 days after surgery, including but not limited to febrile episodes, wound infection, vaginal cuff dehiscence, and rehospitalization. Complications were scored using the Clavien-Dindo classification system [ 12 ]. Postoperative pain was evaluated using a patient-reported visual analog scale (VAS) and the Faces Pain Scale revised (FPS-R). The VAS ranged from 0 (no pain) to 10 (worst imaginable pain), and the patients rated their pain intensity at 6, 12, and 24 h after surgery. The revised FPS was used as an observational tool to assess facial expressions corresponding to pain intensity at the same time points as the self-report scale. Assessments were performed by independent clinicians (BSU, EB) who were familiar with the scale and blinded to the patient’s surgical group allocation. The requirement for additional analgesic medication within the first 24 h postoperatively was also noted. At 24 h after surgery, overall postoperative recovery was assessed using the validated TUQuality of Recovery-40 (QoR-40) questionnaire. The QoR-40 includes 40 items across five domains (physical comfort, emotional state, physical independence, psychological support, and pain), each scored on a 5-point Likert scale, yielding total scores ranging from 40 (very poor recovery) to 200 (excellent recovery). A validated Turkish version of the QoR-40 [ 13 ] was used, and the patients completed the questionnaire under supervision to ensure its completeness. For some analyses, a high QoR-40 score was defined a priori as a total score above the cohort median (150). At the 3-month postoperative follow-up, cosmetic outcomes at the port/incision sites were evaluated using the Patient and Observer Scar Assessment Scale (POSAS), which consists of the Observer Scar Assessment Scale (OSAS) and Patient Scar Assessment Scale (PSAS) [ 14 ]. Because vNOTES is performed via posterior colpotomy and produces no visible abdominal scar, the POSAS instrument is not applicable to this approach; cosmetic outcomes were therefore formally compared only between the CML and SPL groups, with vNOTES reported descriptively as a scar-free approach. Direct numerical comparison of POSAS scores between vNOTES and the abdominal approaches is not meaningful and was deliberately avoided to prevent misleading interpretation. The OSAS (completed by the surgeon) and PSAS (completed by the patient) each included six items scored from 1 (normal skin/best) to 10 (worst possible), with total scores ranging from 6 to 60; higher scores indicated worse scar appearance and symptoms. Sexual function was assessed using the Female Sexual Function Index(FSFI), in which five domains (desire, arousal, lubrication, orgasm, and satisfaction) were scored from 0 to 10 and summed to a total score between 0 and 50, with higher scores indicating better sexual function. All patient-reported outcome measures used in this study (VAS, FPS-R, QoR-40, POSAS, and FSFI) are previously validated and widely used instruments in clinical research. Categorical factors were described using frequencies and percentages. The χ2 test was used to analyze categorical throughput. Normal distributions were evaluated based on the skewness and kurtosis values. For parametric data, a one-way ANOVA was conducted, while the Kruskal–Wallis test was used to assess nonparametric data. Results for parametric data are reported as mean ± SD, whereas non-parametric data are presented as medians (ranges). Logistic regression was used to identify independent predictors of an excellent quality of recovery, defined a priori as a QoR-40 score ≥ 150 (cohort median). Variables entered into the univariate model were selected on the basis of clinical and biological plausibility as potential confounders of postoperative recovery in MIS for adnexal surgery: age (≥ 50 vs. < 50 years), BMI (≥ 30 vs. < 30 kg/m²), multiparity, postmenopausal status, previous pelvic surgery, prior caesarean section, and surgical approach (vNOTES vs. CML/SPL). Variables with P  < 0.20 in univariate analysis, together with surgical approach (the primary exposure of interest), were entered into a multivariate logistic regression model using backward elimination; the final model retained surgical approach and all baseline variables listed above to adjust for residual imbalance between groups. Outcomes were expressed as odds ratios (OR) with 95% confidence intervals (CIs). Propensity score matching (PSM) was not performed. We opted against PSM for three reasons: (i) the overall sample size ( n  = 87) and the smallest group size (vNOTES, n  = 23) are too limited to generate a stable propensity score model without over-fitting; (ii) 1:1 matching would have required discarding a substantial proportion of unmatched patients, further reducing statistical power for secondary outcomes and complications; and (iii) baseline characteristics were already well balanced across groups (Table  1 ), so the incremental benefit of PSM over standard covariate adjustment was expected to be limited. Instead, homogeneity was confirmed by comparing baseline characteristics, and a multivariate logistic regression model adjusting for age, BMI, parity, menopausal status, previous pelvic surgery, and prior caesarean section was used to mitigate residual confounding. Data entry and analysis were performed using SPSS version 24 (SPSS Inc., Chicago, IL, USA). Statistical significance was set at P  < 0.05. Table 1 Demographic and clinical characteristics of the patients CML ( n  = 38) SPL ( n  = 26) vNOTES ( n  = 23) P Age (years) a 47.9 ± 8.5 49.3 ± 8.8 47.9 ± 8.9 0.794 Postmenopausal status b 15 (39.5) 12 (46.3) 9 (39.1) 0.840 Previous pelvic surgery b 0.941 Myomectomy 2 (5.3) 1 (3.8) Ovarian Cystectomy 3 (7.9) 3 (11.5) Ectopic Pregnancy Surgery 1 (2.6) 1 (3.8) 1 (4.3) Other ovarian oophorectomy 1 (2.6) - - C-Section b 0.907 − 0 20 (52.6) 14 (53.8) 13 (56.5) − 1 6 (15.8) 3 (11.5) 3 (13.0) - Multi (≥ 2) 12 (31.6) 9 (34.6) 7 (30.4) BMI (kg/cm 2 ) a 29.7 ± 5.0 30.4 ± 4.3 32.0 ± 5.7 0.258 < 30 21 (55.3) 13 (50.0) 7 (30.4) 0.194 ≥ 35 9 (23.7) 7 (26.9) 5 (21.7) ≥ 40 8 (21.1) 6 (23.1) 11 (47.8) Parity a 1.9 ± 0.9 1.8 ± 0.8 2.0 ± 0.9 0.759 Nullipar b 2 (5.3) 1 (3.8) 1 (4.3) 0.990 Unipar 10 (26.3) 7 (26.9) 5 (21.7) Multipar 26 (68.4) 18 (69.2) 17 (73.9) Mass size (cm) a 14.5 ± 4.7 15.6 ± 4.6 14.4 ± 4.2 0.589 10–14 cm 23 (60.5) 12 (46.2) 15 (65.2) 0.505 15–19 cm 11 (28.9) 10 (38.5) 4 (17.4) ≥ 20 cm 4 (10.5) 4 (15.4) 4 (17.4) CA-125 (U/ml) a 19.9 ± 7.9 19.8 ± 7.7 18.5 ± 9.6 0.808 Pathology b 0.963 Benign 36 (94.7) 25 (96.2) 22 (95.7) Borderline 2 (5.3) 1 (3.8) 1 (4.3) SPL  Single-port laparoscopic surgery, CML  Conventional multiport laparoscopic surgery, BMI  Body mass index, OSAS  Observer Scar Assessment Scale, PSAS  Patient Observer Scar Assessment Scale a mean ± SD, b n (%) Demographic and clinical characteristics of the patients SPL  Single-port laparoscopic surgery, CML  Conventional multiport laparoscopic surgery, BMI  Body mass index, OSAS  Observer Scar Assessment Scale, PSAS  Patient Observer Scar Assessment Scale a mean ± SD, b n (%)

A total of 87 patients met the inclusion criteria and were categorized into three groups: CML ( n  = 38), SPL ( n  = 26), and vNOTES ( n  = 23). All patients underwent unilateral salpingo-oophorectomy in our department In four cases involving borderline ovarian tumors, additional peritoneal and omental biopsies were performed. The baseline demographic and clinical characteristics revealed no significant differences among the three groups (Table  1 ). Significant differences were observed in the surgical timing. The port setup time was shortest for CML (2.5 ± 0.5 min) and longest for vNOTES (7.8 ± 1.5 min) ( P  < 0.001). The total operation time differed significantly among the groups, with CML (61.0 ± 14.7 min) being faster than vNOTES (71.5 ± 13.1 min) ( P  = 0.021). However, vNOTES and SPL demonstrated significantly shorter specimen removal times (6.1 ± 1.8 min and 6.6 ± 1.3 min, respectively) than CML (9.3 ± 2.2 min) ( P  < 0.001). No significant differences were found in estimated blood loss ( P  = 0.375), and there were no conversions to laparotomy or intraoperative complications in any of the groups. .(Table  2 ). Table 2 Primary and secondary outcome measures CML ( n  = 38) SPL ( n  = 26) vNOTES ( n  = 23) P Port set-up time (min) a 2.5 ± 0.5 4.4 ± 1.1 7.8 ± 1.5 < 0.001 Operation time (min) a 61.0 ± 14.7 65.1 ± 13.1 71.5 ± 13.1 0.021 Specimen removal time (min) a 9.3 ± 2.2 6.6 ± 1.3 6.1 ± 1.8 < 0.001 Estimated blood loss (mL) a 77 ± 23 83 ± 16 84 ± 24 0.375 Return to laparotomy b - - - Intraoperative complications b - - - Postoperative complications b Grade 2 Moderate Complications - Ileus - Urinary infection 1 (2.6) - - 1 (3.8) - 2 (8.7) 0.521 0.195 Additional analgesic requirement b 9 (23.7) 3 (15.4) 1 (4.3) 0.137 Length of stay hospital (day) c 1 (1–3) 1 (1–3) 1 (1–3) 0.323 VAS a , 6 h 5.0 ± 1.1 4.8 ± 1.1 4.2 ± 1.0 0.041 VAS a , 12 h 2.9 ± 0.8 2.6 ± 0.6 2.2 ± 0.6 0.002 VAS a , 24 h 2.0 ± 0.6 1.6 ± 0.6 1.5 ± 0.5 0.009 FPS a , 6 h 5.8 ± 1.4 5.0 ± 1.1 4.9 ± 1.2 0.010 FPS a , 12 h 3.7 ± 1.0 2.8 ± 0.6 2.6 ± 0.6 < 0.001 FPS a , 24 h 2.9 ± 1.0 1.8 ± 0.7 1.6 ± 0.4 < 0.001 Quality of recovery-40 score a 145.0 ± 8.7 147.9 ± 9.6 153.5 ± 9.7 0.004 OSAS score a 17.8 ± 2.6 23.1 ± 2.7 - < 0.001 PSAS score a 18.4 ± 4.5 23.6 ± 2.8 - < 0.001 Sexual function index a , 3 months 31.7 ± 2.3 32.0 ± 2.5 33.0 ± 3.4 0.197 Sexual function index a , 6 months 32.1 ± 1.9 32.7 ± 1.9 33.3 ± 2.4 0.129 SPL  Single-port laparoscopic surgery, CML  Conventional multiport laparoscopic surgery, VAS  Visual analog scale, FPS  Faces pain scale, OSAS  Observer Scar Assessment Scale, PSAS  Patient Observer Scar Assessment Scale a Mean ± SD, b n (%), c Median (Min-Max) Primary and secondary outcome measures Postoperative complications b Grade 2 Moderate Complications - Ileus - Urinary infection 1 (2.6) - - 1 (3.8) - 2 (8.7) 0.521 0.195 SPL  Single-port laparoscopic surgery, CML  Conventional multiport laparoscopic surgery, VAS  Visual analog scale, FPS  Faces pain scale, OSAS  Observer Scar Assessment Scale, PSAS  Patient Observer Scar Assessment Scale a Mean ± SD, b n (%), c Median (Min-Max) vNOTES reported significantly lower VAS scores at 6, 12, and 24 h post-surgery compared to the other groups ( P  = 0.041, P  = 0.002, and P  = 0.009, respectively). Similar to the VAS scores, the FPS-R scores were the lowest in the vNOTES cohort across all postoperative time points. The overall Quality of Recovery-40 (QoR-40) score was significantly higher in the vNOTES group (153.5 ± 9.7) than in the SPL (147.9 ± 9.6) and CML (145.0 ± 8.7) groups ( P  = 0.004). The additional analgesic requirement was the lowest in the vNOTES group (4.3%), although this difference was not statistically significant ( P  = 0.137). Cosmetic outcomes were compared only between CML and SPL, because vNOTES produces no visible abdominal scar and is not amenable to POSAS scoring. Patients undergoing CML reported significantly better cosmetic outcomes, indicated by lower OSAS (17.8 ± 2.6) and PSAS (18.4 ± 4.5) scores, compared with SPL (23.1 ± 2.7 and 23.6 ± 2.8, respectively; P  < 0.001 for both). vNOTES was scar-free by design and is therefore reported separately and descriptively rather than scored on POSAS. Long-term functional outcomes, specifically the Female Sexual Function Index scores, showed no significant differences among the three surgical approaches at either the 3-month or 6-month follow-up ( P  = 0.197 and P  = 0.129, respectively). Logistic regression analysis was performed to identify factors predicting a high QoR-40 score (≥ 150). In both univariate and multivariate OR analyses, the vNOTES approach was identified as the only significant predictor of achieving superior recovery quality (OR, 1.8; 95% CI, 1.1–3.1; P  = 0.012). Other variables, including age, BMI, parity, and previous surgical history, were not significant (Table  3 ). Table 3 Results of univariate and multivariate analyses performed to identify risk factors for a QoR-40 score ≥ 150 Univariate Multivariate OR 95% CI P OR 95% CI P Age (≥ 50 years) 1.0 0.9–1.1 0.557 1.0 0.9–1.1 0.393 BMI (≥ 30 kg/cm 2 ) 1.0 0.9–1.1 0.965 0.9 0.8–1.1 0.478 Multiparity 0.7 0.4–1.2 0.298 0.7 0.4–1.2 0.258 Postmenopausal status 1.3 0.5–3.2 0.457 0.8 0.1–4.7 0.806 Previous pelvic surgery 0.8 0.5–1.4 0.519 0.6 0.3–1.2 0.220 C-Section 0.6 0.4–1.1 0.077 0.6 0.3–1.1 0.059 vNOTES 1.9 1.2–3.3 0.015 1.8 1.1–3.1 0.012 QoR  Quality of Recovery, BMI  Body mass index, OR  Odds ratio, CI  Confidence interval Results of univariate and multivariate analyses performed to identify risk factors for a QoR-40 score ≥ 150 QoR  Quality of Recovery, BMI  Body mass index, OR  Odds ratio, CI  Confidence interval

K.G : Conceptualization, Methodology, Writing original draf. E.B : Data curation, Writing and Editing. B.S.U : Data curation, Writing and Editing. B.N.E: Visualization, Validation. S.U : Visualization, Validation. F.K : Visualization, Resorces. V.G : Formal analysis Writing-review and editing. Statistical support. All authors read and approved the final manuscript before submission.

This multicenter retrospective cohort study provides novel comparative data on CML, SPL, and vNOTES for the management of large adnexal masses (≥ 10 cm). All three minimally invasive approaches demonstrated acceptable safety profiles in experienced hands. While CML maintained an advantage in overall operative speed, vNOTES was associated with more efficient specimen extraction, lower early postoperative pain scores, and enhanced quality of recovery. However, given the retrospective design, patient-preference-based allocation, inherent selection bias, and limited sample size, these findings should be interpreted with appropriate caution. The favorable outcomes observed with vNOTES may partially reflect systematic differences between patient groups rather than the surgical approach alone. Prospective randomized trials with adequate power are needed to definitively establish the comparative effectiveness of vNOTES in this challenging clinical scenario, with BMI included as a key stratification variable.

To the best of our knowledge, this is the first study to directly compare conventional CML, SPL, and vNOTES, specifically for managing adnexal masses measuring 10 cm or larger. While MIS is the preferred standard for benign adnexal disease, large masses have historically been considered relative contraindications for single-port or transvaginal approaches because of technical constraints. Our key findings revealed that while CML achieves the shortest port setup and overall operative times, vNOTES offers notable benefits, including quicker specimen retrieval, reduced postoperative pain, and better overall recovery outcomes, effectively establishing itself as a viable and patient-centered alternative for large-volume adnexal surgery. These findings should, however, be interpreted in the context of the methodological limitations discussed below. Our analysis revealed significant timing variations across the surgical phases. CML was associated with the shortest port setup and total operation times, likely due to the straightforward nature of multiport triangulation and the surgeon’s familiarity with the technique. In contrast, multiple randomized controlled trials and meta-analyses comparing single-port laparoscopy to conventional laparoscopy have reported longer operative times for single-port approaches owing to restricted instrument maneuverability and collision [ 8 , 14 , 15 ]. However, our data showed that vNOTES and SPL had a significant advantage over CML in terms of the specimen removal time. This ‘extraction advantage’ results from the larger incisions used in vNOTES (posterior colpotomy) and SPL (umbilical incision), which are more substantial compared to the smaller standard port sites typically utilized for specimen extraction in CML, allowing for the removal of larger masses. Recent technical reports corroborate this, demonstrating that vNOTES provides a highly effective route for safely bagging and extracting very large adnexal masses (up to 20 cm) transvaginally, thereby avoiding the need to enlarge small abdominal ports or perform extensive intra-abdominal tumor morcellation. Transvaginal extraction has been highlighted as a superior method for large solid tumors, bypassing the limitations of abdominal incision. Thus, for large masses, the initial time lost during the complex setup in vNOTES was partially recovered during the efficient extraction phase. The most significant clinical outcome of this study was the superior quality of recovery and reduced early postoperative pain observed in the vNOTES group. Patients who underwent vNOTES reported significantly lower VAS and FPS-R pain scores at 6, 12, and 24 h postoperatively. This finding is consistent with the landmark NOTABLE trial, which demonstrated that vNOTES adnexectomy results in significantly lower postoperative pain scores and reduced analgesic requirements compared with conventional laparoscopy [ 9 ]. This pain reduction stems from differences in somatic and visceral innervation. Abdominal wall incisions used in CML and SPL trigger sharp somatic nociceptive pain, whereas the posterior vaginal fornix accessed in vNOTES is innervated by visceral autonomous fibers, resulting in a more diffuse and less severe pain. Furthermore, vNOTES allows surgeons to operate at lower pneumoperitoneal pressures, thereby reducing peritoneal irritation and early postoperative discomfort. Conversely, previous meta-analyses comparing SPL and CML found no significant overall advantage of single-port laparoscopy in terms of postoperative pain at 24 h, suggesting that the complete avoidance of abdominal wall trauma via vNOTES is the primary driver of enhanced recovery [ 8 , 14 ]. Crucially, our multivariate logistic regression confirmed that the vNOTES approach was the only independent predictor of achieving a high-quality recovery (QoR-40 ≥ 150), outweighing other variables such as age, BMI, and history of surgery. Despite the technical difficulties associated with masses larger than 10 cm, our study demonstrated the safety of all three MIS techniques. No conversions to laparotomy occurred, and there were no significant intraoperative complications, such as organ damage or severe hemorrhage. This indicates that when performed by skilled surgeons, the limited range of motion in vNOTES or SPL does not jeopardize patient safety, even in cases involving large benign masses. Given the small sample size, however, the study was underpowered to detect rare intraoperative or postoperative complications; the absence of events should therefore not be interpreted as evidence of equivalence between techniques for safety outcomes. Regarding long-term outcomes, our study found no significant differences in sexual function among the three surgical approaches at either the 3-month or 6-month follow-up. This directly addresses a common concern regarding transvaginal access and is strongly supported by recent secondary analyses of the NOTABLE and HALON trials, which concluded that the colpotomy involved in vNOTES does not negatively affect women’s sexual functioning or cause new-onset dyspareunia [ 16 ]. Cosmetic outcomes were favourable for CML compared with SPL in the abdominal-approach analysis, most likely because of the larger single umbilical incision required for SPL specimen extraction. vNOTES inherently avoids any visible abdominal scarring but, for this reason, cannot be directly compared with CML or SPL on the POSAS instrument; the cosmetic advantage of vNOTES is categorical rather than quantitative, and comparisons should be interpreted on that basis. An additional clinically important observation in our cohort is the BMI distribution across surgical groups. Patients with a BMI ≥ 40 kg/m² were almost twice as frequent in the vNOTES arm (47.8%) as in the CML (21.1%) and SPL (23.1%) arms, whereas patients with a BMI < 30 kg/m² were more often managed with CML (55.3%) or SPL (50.0%) than with vNOTES (30.4%). Although the overall between-group BMI comparison did not reach statistical significance, given the sample size the distribution is clinically relevant and, importantly, is unfavourable to vNOTES: obesity is a well-established risk factor for prolonged operative time, wound complications, and delayed recovery after abdominal MIS. Despite this less favourable baseline profile, patients managed by vNOTES still reported lower pain scores, lower rescue analgesic use, and the highest QoR-40 scores. This pattern is consistent with a growing body of literature demonstrating that vNOTES is feasible and effective for hysterectomy in obese and morbidly obese patients, with favourable peri- and postoperative outcomes compared with conventional laparoscopy [ 17 , 18 ]. Taken together, these data suggest that BMI—and in particular morbid obesity—may be an especially important variable to consider when selecting the surgical approach in women undergoing MIS for large benign adnexal masses, and that vNOTES may be particularly advantageous in this subgroup. However, because our study was not specifically powered for BMI-stratified analyses, and patients were not randomly allocated, this hypothesis should be tested in dedicated prospective studies. Our study has several important limitations that warrant careful consideration. First, the retrospective, non-randomized design and the allocation of patients to surgical approach on the basis of patient preference, together with explicit procedure-specific contraindications for vNOTES (narrow vaginal introitus, suspected dense intra-abdominal adhesions, or technical limitations identified by the surgeon), introduce a significant risk of selection bias. By design, patients with anatomical or surgical-history features favourable to a transvaginal approach were more likely to be allocated to vNOTES, whereas patients with features that complicate vaginal access were channelled to CML or SPL. This structural imbalance may have independently favoured outcomes in the vNOTES arm, particularly pain and quality-of-recovery endpoints, and the reader should interpret the observed between-group differences with this limitation clearly in mind. Second, we did not perform propensity score matching. We opted against PSM because, given our cohort size and the markedly unequal group sizes, PSM would have required discarding a substantial proportion of unmatched patients and generated unstable propensity scores, severely compromising statistical power for our secondary outcomes. Instead, we addressed potential confounding using two complementary strategies. Baseline testing demonstrated homogeneity across all recorded variables and, despite vNOTES being contraindicated in patients with suspected adhesions, the rates of previous pelvic surgery and prior caesarean section were statistically comparable across all three groups, suggesting similar baseline surgical-history risk. Rather than discarding data via PSM, we used a multivariate logistic regression model adjusted for age, BMI, multiparity, postmenopausal status, previous pelvic surgery, and prior caesarean section. Nevertheless, as with all observational studies, unmeasured confounding (e.g., surgeon preference, intra-operative adhesion severity not captured in preoperative imaging, pre-existing pelvic pain, and socio-educational determinants of self-reported recovery) cannot be excluded. Third, the overall sample size ( n  = 87) and the relatively small vNOTES group ( n  = 23) limit statistical power, particularly for complications, rescue analgesic use, and subgroup analyses such as BMI-stratified comparisons; non-significant between-group differences should therefore not be interpreted as evidence of equivalence. Fourth, although the BMI distribution favoured heavier patients in the vNOTES group, the study was not specifically designed to analyse outcomes by BMI category, and larger studies are required to confirm the suggested benefit of vNOTES in morbidly obese patients. Fifth, POSAS-based cosmetic outcomes could not be formally compared across all three groups because vNOTES leaves no visible abdominal scar; the cosmetic comparison was therefore restricted to CML vs. SPL, and the scar-free nature of vNOTES was reported descriptively. Finally, operator experience was limited to two high-volume gynecologic oncologists, which may limit the generalizability of our findings to centres with less vNOTES expertise.

Adnexal masses (AM) are common in women, with up to 10% requiring surgery during their lifetime [ 1 ]. Laparoscopy has become popular for the surgical management of benign AM, although it is challenging for large cysts [ 2 ]. Minimally invasive surgery (MIS) is preferred because patients experience more rapid recovery, less adhesion, a faster return to social activities, and better cosmetic outcomes than with other surgical approaches [ 2 ]. However, preoperative assessment of the nature of an adnexal mass is crucial. For masses with an intermediate or high risk of malignancy [ 3 , 4 ], priority must be given to preventing cyst rupture and intra-abdominal spilling. Accordingly, laparotomy may be required. The comprehensive literature base supports that size alone is not a contraindication to MIS; critical factors include preoperative assessment for benign disease, surgeon experience, and adherence to oncologic principles [ 5 – 7 ]. Conventional multiport laparoscopy (CML) remains the most common method for adnexal surgery and oophorectomy. This technique is technically straightforward, is more versatile, and is considered safer in complex cases than single-port laparoscopy. Building on the principles of MIS, Single-port laparoscopy (SPL) oophorectomy is a safe and feasible alternative to CML. It provides better cosmetic results and modest reductions in early postoperative discomfort in selected patients. However, a recent systematic review and meta-analysis found no significant differences in the overall outcomes between SPL and CML. Therefore, the findings do not demonstrate a clear preference for either technique in adnexal surgery [ 8 ]. Over the past decade, vaginal natural orifice transluminal endoscopic surgery (vNOTES) has broadened the spectrum of vaginal surgical options by integrating endoscopic visualization with a transvaginal approach, allowing adnexal procedures to be performed via posterior colpotomy without visible abdominal scarring. A recent randomized study reported shorter operative times and reduced postoperative pain after vNOTES adnexectomy compared with conventional laparoscopic adnexectomy [ 9 ]. Despite these advances, the management of large adnexal masses remains challenging. However, for lesions ≥ 10 cm, SPL and vNOTES are technically more challenging and potentially riskier than CML, mainly due to limited instrument maneuverability and suboptimal triangulation. The restricted range of motion and constrained working space when manipulating large masses have hindered the wider adoption of these techniques in clinical practices. To our knowledge, no prior study has directly compared CML, SPL, and vNOTES in the management of adnexal masses ≥ 10 cm, which poses a substantial clinical challenge. Accordingly, the present study was designed to fill this gap by comparing the surgical outcomes of CML, SPL, and vNOTES oophorectomies in patients with large adnexal masses (≥ 10 cm). Based on the emerging evidence suggesting patient-centered benefits of vNOTES, we hypothesized that vNOTES would demonstrate superior outcomes in terms of postoperative pain and quality of recovery, despite potential initial technical challenges for large masses.

The datasets generated and analysed during our study are not publicly available due to patient confidentiality anddata protection regulations. However, they can be obtained from the corresponding author upon reasonable request.

Supplementary Material 1. Supplementary Material 1.