Vnotes versus multiport laparoscopy for Borderline ovarian tumor staging: a multicenter propensity score–matched cohort study

Abstract

Borderline ovarian tumors (BOTs) require meticulous surgical staging. While conventional multiport laparoscopy (CML) remains the standard minimally invasive approach, vaginal natural orifice transluminal endoscopic surgery (vNOTES) offers potential recovery advantages through scarless access. This study compared perioperative and postoperative recovery outcomes between vNOTES and CML for BOT staging. Methods This multicenter retrospective comparative cohort study was conducted at five tertiary referral centers between January 2022 and March 2026. Consecutive patients with histopathologically confirmed BOTs undergoing minimally invasive management were assigned to vNOTES or CML groups according to surgeon and patient preference, anatomical suitability, and institutional practice. Propensity score matching (1:1 nearest-neighbor, 13 covariates) yielded 24 matched pairs (n = 48). Primary outcomes were the perioperative endpoints of operative time, estimated blood loss, and intraoperative complications. Secondary outcomes were postoperative recovery endpoints, including postoperative pain (VAS, FPS), analgesic requirements, hospital stay, QoR-40, FSFI scores, and 30-day complications. Results Median operative time was modestly longer with vNOTES (105.0 vs. 90.0 min; p = 0.039), attributable to extended port set-up time, while specimen removal was significantly shorter (12.0 vs. 15.0 min; p = 0.031). The vNOTES group demonstrated lower additional analgesic requirements (12.5% vs. 33.3%; p = 0.028), consistently lower pain scores across all time points (VAS at 6h: 2.8 ± 1.4 vs. 4.1 ± 1.6; p = 0.007), shorter hospital stay (1.4 ± 0.7 vs. 2.1 ± 0.9 days; p = 0.004), and higher QoR-40 scores (168.4 ± 12.3 vs. 152.7 ± 14.8; p = 0.002). Postoperative complication rates (8.3% vs. 16.7%; p = 0.384) and restaging rates (4.2% vs. 8.3%; p = 0.550) were comparable, although the small sample limited power for these uncommon events and their confidence intervals were wide. No recurrences were observed over a median follow-up of only 14.3–18.7 months, a period too short to assess oncologic safety in borderline ovarian tumors. Conclusion In this small, retrospectively matched cohort of carefully selected patients treated at experienced centers, vNOTES for BOT staging was associated with improved early recovery—including reduced pain, analgesic use, and hospital stay—without an apparent increase in perioperative morbidity. These findings pertain strictly to perioperative and short-term recovery outcomes and do not permit any inference regarding oncological safety. Given the observational design, substantial residual selection and indication bias that propensity matching cannot fully remove, and limited oncologic follow-up, these findings should be regarded as hypothesis-generating. Prospective studies with longer oncologic follow-up and cost-effectiveness analyses are warranted before broader adoption.