Cervical cancer ranks fourth in the incidence of cancer with an estimated 661,021 new cases emerging in 2022, and is also the fourth leading cause of cancer-related mortality among women globally (1). Radical hysterectomy and pelvic lymph node dissection (PLND) is recommended as the standard treatment for patients with stage IB1, IB2 and IIA1 using the International Federation of Gynecology and Obstetrics (FIGO) staging system (Version 2018) (2). Over the past decades, minimally invasive surgery (MIS) has been widely adopted, offering benefits such as reduced blood loss, shorter hospital stays, and faster recovery compared to open laparotomy.

The publication of the Laparoscopic Approach to Cervical Cancer (LACC) trial in 2018, which reported inferior survival outcomes for MIS compared to laparotomy, raised controversies and technical refinement within this field (3). The results might ascribe to squeezing and damage to the cervical lesion by uterine manipulator, intraperitoneal CO2 circulation, and unsealed colpotomy leading to tumor cell growth and dissemination. In response, researchers have tried to find improvements against these factors rather than directly abandon it. Studies have proved that the survival outcomes of patients with cervical lesions smaller than 2 cm after MIS were not inferior to those undergoing laparotomy (4, 5). In addition, researchers also explored technical innovations like no-manipulation techniques or gasless laparoscopy, and strictly followed the tumor-free principle including sealed colpotomy, contained tissue extraction, and repeated pelvic irrigation to take full advantages of MIS without compromising oncological outcomes (6–8). Moreover, surgeons’ experience and skills determine the excision adequacy of radical hysterectomy, which is also one of the most important prognostic factors.

Laparoendoscopic single-site surgery (LESS), which enters into the peritoneal cavity through a single umbilical incision or vagina, has been greatly appreciated in gynecology because it shows advantages in minor invasion, quick and safe specimen extraction, alleviated pain, fast recovery, as well as satisfying patients’ expectation in cosmesis (9). However, the omission of a uterine manipulator introduces significant technical challenges, particularly in achieving adequate exposure for precise parametrial dissection and systematic lymphadenectomy. These challenges are markedly exacerbated in the LESS approach due to its inherent ergonomic constraints, such as loss of triangulation, instrumental collision, and a limited ability for surgical assistance.

The emerging robotic surgical system, characterized by magnified three-dimensional (3D) camera, articulated instruments, and ergonomic surgeon console, presents a potential solution to these amplified technical difficulties. These features enhance visual clarity, operative dexterity, precision and reduce surgeons’ fatigue to facilitate complex dissection in the confined single-site workspace, making the complementation of a no-manipulation radical hysterectomy more feasible.

While several studies have described the feasibility of both traditional LESS (10) and robotic laparoendoscopic single-site surgery (R-LESS) in managing early-stage cervical cancer (11–13), a direct comparison between these two approaches with no-manipulation technique is lacking. Therefore, the primary objective of this study was to determine whether R-LESS radical hysterectomy with no-manipulation technique was superior to traditional LESS in the perioperative and survival outcomes for early-stage cervical cancer.

A total of 124 patients were included in this study, comprising 73 cases in the R-LESS group and 51 cases in the LESS group. After the introduction of da Vinci Xi system, radical hysterectomies for cervical cancer were mostly performed by R-LESS.

Baseline characteristics were summarized in Table 1 which were all balanced with no statistical significance between the two groups. The clinical stage distributions of patients were comparable between the two groups (p=0.479). Regarding no-manipulation techniques adopted in the R-LESS group, 17 of 73 cases (23.3%) were performed with Zheng’s 4C suspension method, and the remaining 56 cases (76.7%) utilized the banding method. The width of resected parametrium and length of vagina reached at least 3 cm. As for the extent of lymphadenectomy in the R-LESS group, 32 cases (43.8%) underwent SLND, and the others received systematic PLND, in which 6 patients (8.2%) only had comprehensive lymphadenectomy on one side due to negative mapping or previous surgery. The LESS group only had 4 cases (7.8%) undergoing SLND, with the majority, 47 cases (92.2%), receiving systematic PLND.

The R-LESS group experienced a significant reduction in both OT (236.5 ± 52.8 vs 390.8 ± 73.5 min, p<0.001) and EBL (median, 50.0 vs 150.0 mL, p<0.001) compared to the LESS group. Even for those who had systematic PLND bilaterally, R-LESS consistently enabled a substantial reduction in OT (259.3 ± 47.3 vs 397.0 ± 70.3 min, p<0.001). Regarding postoperative rehabilitation indexes, both groups showed comparable and favorable outcomes except for a shorter drainage time (3.7 ± 1.2 vs 4.7 ± 1.9 days, p=0.001) after R-LESS.

Four cases (7.8%) converted to multi-port laparoscopy in the LESS group due to difficult operation, whereas the R-LESS group achieved a complete success with all cases completed as planned without conversion. Three patients (4.1%) undergoing R-LESS encountered intraoperative complications including 1 bowel injury and 2 vascular injuries. In the LESS group, 5 patients (9.8%) experienced a total of 6 intraoperative complications including 2 bowel injuries and 1 bladder injury addressed with suturing repair, 2 ureteral stenoses requiring placement of ureteral stents under cystoscopy, and 1 vascular injury converting to multi-port laparoscopy because of excessive bleeding. The LESS group exhibited higher rates of intraoperative complication and conversion compared to the R-LESS group, though these results didn’t reach statistical significance (p>0.05).

Postoperative complication rates were also comparable (28.8% for R-LESS and 35.3% for LESS, p=0.441). Among those who underwent R-LESS, 1 patient required reoperation to evacuate an infected pelvic hematoma 30 days after surgery because of persistent fever, and 4 patients diagnosed with urogenital fistula by CT urography required ureteral reimplantation. In the LESS group, reoperations were needed for 4 patients, one of whom experienced intestinal perforation and 3 suffered from urogenital fistula (Table 2).

No significant differences were observed between the two groups regarding histologic types, with squamous carcinoma (61.6% in R-LESS and 62.7% in LESS) being the predominant type. Besides, patients with poor differentiation (54.8% in R-LESS and 62.7% in LESS), deep stromal invasion (35.6% vs 31.4%), parametrial involvement (5.5% vs 2.0%), vaginal fornix involvement (17.8% vs 5.9%), positive margin (4.1% vs 0%), lymph node metastasis (9.6% vs 11.8%), and length of resected vagina after being fixed with 10% formalin (2.0 ± 0.6 vs 2.1 ± 0.5 cm) were also comparable (p≥0.05). However, the LESS group had significantly more patients with LVSI (49.0% vs 31.5%, p=0.049). For those who received bilaterally systematic PLND, there was no statistical difference in the mean number of retrieved nodes (28.8 ± 5.4 vs 30.4 ± 6.5, p=0.233). After integrating pathological information, the distributions of patients’ postoperative disease status were found to be similar between the two groups (p=0.544).

The determinations of adjuvant therapies were made by the surgeon and chemo-radiational oncologists as recommended by the National Comprehensive Cancer Network (NCCN) guidelines. No difference was found regarding the postoperative interventions (p=0.136). The median follow-up duration was 30.2 (95%CI, 25.4-35.0) months for all 124 enrolled patients, 18.9 (95%CI, 15.1-22.7) months for the R-LESS group, and 45.3 (95%CI, 44.4-46.2) months for the LESS group. Among the 5 patients with recurrence in the LESS group, sites included lung metastases (n=3), bone metastasis (n=1), and retroperitoneal lymph node metastasis (n=1). In the R-LESS group, the 3 recurrences consisted of vaginal cuff recurrence (n=1), lung metastasis (n=1), and bone metastasis (n=1). The recurrence rate (9.8% vs 4.1%) and death rate (5.9% vs 1.4%) after LESS were higher than those after R-LESS, but the differences didn’t reach statistical significance (p>0.05). The 4.5-year DFS of all 124 patients was 92.3%, and the 3-year OS was 96.3%. The 3-year DFS (95.8% for R-LESS vs 89.8% for LESS, p=0.399) and OS (96.3% vs 95.8%, p=0.752) were similar between the two groups (Figure 2, Table 3).

Cox proportional hazards regression was performed to analyze factors influencing disease recurrence and survival. On univariate analysis for DFS, histologic type, positive resection margin, lymph node metastasis, and postoperative pathological status were associated with an increased risk of recurrence (p<0.05). In the multivariate model, histologic type (HR 2.0, 95% CI 1.3-3.2; p=0.002) and lymph node metastasis (HR 16.8, 95% CI 3.7-75.1; p<0.001) remained independent risk factors for recurrence (Table 4). The wide confidence interval for lymph node metastasis may due to the small event number, and this result should be interpreted with caution. For OS, multivariate analysis identified histologic type (HR 2.5, 95% CI 1.3-4.9; p=0.005) as the only independent risk factor for death in this cohort (Table 5). The surgical approach (R-LESS vs. LESS) and no-manipulation technique adopted were not independent risk factors for either DFS or OS in the multivariate models.

Given the observed significant difference in OT between the groups, an effect size (Cohen’s d) of 2.484 was calculated. With the actual sample sizes and a two-tailed α level of 0.05, the analysis yielded a statistical power exceeding 90%, indicating that the study was adequately powered for this important perioperative outcome.

We presented our experience in two innovative no-manipulation techniques specifically utilized in transumbilical single-site radical hysterectomy for early-stage cervical cancer, which were designed to improve patients’ survival outcomes while preserving the benefits of MIS. They not only facilitated uterine manipulation but also achieved sufficient exposure of parametrial and LND areas.

In our study, the R-LESS group had significantly less OT and EBL, which was consistent with previous literatures. Gao et al. reported a comparative study of R-LESS and LESS in treating stage IB1 cervical cancer that showed remarkable advantages of R-LESS in shortening OT (223.56 ± 15.43 vs 248.61 ± 20.89 min, p<0.01) and reducing intraoperative bleeding (217.25 ± 16.77 vs 294.74 ± 24.00 mL, p<0.01) (11). Nie et al. reached the same conclusion in the comparison of robotic and laparoscopic multi-port surgery for radical hysterectomy (16). The high-definition 3D vision of da Vinci Xi system facilitated in clearly identifying and distinguishing parametrial structures, ureters and vessels, which cooperated with the flexible and stable instruments to perform dissection and resection quickly and precisely. The unipolar scissors, with the ability to provide sustaining energy stimulation, enhanced the efficiency of LND, offering a distinct advantage over the ultrasonic scalpel typically used in LESS. Moreover, the banding method was simpler and significantly reduced operative time with no need to adjust and replace the suspension sutures. Under the magnified view, any bleeding spot could be promptly identified and addressed with bipolar forceps, thus minimizing the time spent on frequent instrument exchanges that were often required in LESS. Furthermore, surgeons didn’t have to stand beside the operating table for hours of high-intensity work during R-LESS procedures, conserving both energy and physical strength that contributed to efficient and high-quality completion of the surgery.

Our findings indicated that patients experienced a quick recovery after single-site surgery, despite the LESS group had a longer duration of drainage due to a larger proportion of systematic PLND. In a previous study regarding radical hysterectomy for cervical cancer, no significant differences were observed between the robotic and laparoscopic groups in terms of catheter time, drainage time, exhaust time, and hospital stay (17). But the results from Luo et al. supported that the robotic surgery would promote a quicker recovery when compared to traditional laparoscopy (18). The placement of a pelvic drainage tube served a dual purpose of dynamically monitoring potential complications such as internal bleeding, lymphatic leakage, and ureteral fistula, as well as minimizing the risk of infection caused by pelvic effusions. The single-site surgery encouraged patients’ mobility by diminishing postoperative pain, which in return, facilitated drainage and bowel function recovery. Most patients were discharged the day after removing drainage tubes.

The intraoperative complication rate of the LESS group was slightly higher than that observed in the R-LESS group. Urinary injuries only occurred after LESS, and the incidence rate was in accordance with previous reports regarding MIS (19, 20). We speculated that the infiltration of cervical lesion and prior abdominal surgery caused dense adhesions of paravesical fossae and ureters to surrounding tissues, making it hard to distinguish anatomic structures under the two-dimensional vision provided by traditional laparoscopy. Moreover, the technical limitations inherent to LESS made it exceptionally difficult to perform meticulous dissection. Four conversions to multi-port laparoscopy all occurred in traditional LESS before the first case of R-LESS. Except for reduced surgical difficulty brought by the robotic system, the surgeon had accumulated considerable expertise in single-site radical hysterectomy prior to the initiation of R-LESS.

The postoperative complication rate was 28.8% for R-LESS and 35.3% for C-LESS with no statistical difference that were similar to Gallotta’s report regarding robotic and laparoscopic multi-port surgery (21). But there were also studies supported that the robotic surgery could significantly reduce postoperative complications following radical hysterectomy (16, 22). Urogenital fistula was the most severe complication after radical hysterectomy, potentially resulting from direct damage, delayed thermal injury, or accidental suture of the bladder or ureters during vaginal cuff closure. Postoperative radiotherapy could also lead to tissue ischemia, necrosis and fistula formation. The incidence rate of urogenital fistula in our study was similar to that reported in laparotomy (23), suggesting that the LESS approach would not increase the risk.

Numbers of retrieved nodes in both groups were more than 21.0-27.4 reported in laparotomic PLND (23, 24). Gao et al. had reached the same conclusion as ours that LESS had satisfying lymph node retrieval no matter whether it was assisted by the robotic system or not (21.37 ± 4.11 vs 20.71 ± 3.47, p=0.475) (11). However, another study indicated that robotic surgery had significantly more resected nodes in comparison to traditional multi-port laparoscopy (29.45 ± 9.78 vs 21.05 ± 11.22, p<0.001) (25). The enhanced vision and flexible instruments of robotic system helped inexperienced surgeons rapidly acquire proficiency in complex procedures, ensure the thoroughness of LND, and reduce missed diagnosis and recurrence.

The 4.5-year DFS and 3-year OS rates observed in all 124 patients were better than those reported for the laparoscopic group in LACC trial. Besides, previous studies have also provided evidence supporting that the survival outcomes and patterns of recurrence following robotic radical hysterectomy were not different from those of the laparotomic approach (26, 27). Given a relatively shorter follow-up time for the R-LESS group, a comparison was made only for the 3-year DFS and OS rates, which revealed no statistical significance. However, determining whether no-manipulation techniques could alter the controversies regarding MIS for cervical cancer will necessitate a larger cohort with extended follow-up data.

To our knowledge, this represents the first comparative study between R-LESS and LESS radical hysterectomy for early-stage cervical cancer employing two distinct no-manipulation techniques, both of which are easily implemented while ensuring adequate surgical excision. All procedures were performed by the same surgeon, thereby minimizing potential bias related to surgical skill and experience. This study also has several limitations. This study has several limitations. As a single−center retrospective analysis, the generalizability of our findings is inherently constrained by possible selection bias and institution−specific practices. Therefore, extrapolation of these results to broader populations should be made with caution. Although the sample size is substantial for a single−center series, it may still be underpowered to detect small but clinically meaningful differences, particularly in survival outcomes and complication rates. Furthermore, a temporal disparity in surgical dates between the two cohorts led to a significant difference in median follow−up duration, which could introduce bias into survival analyses. Continued follow-up is essential to confirm the durability of these promising early survival outcomes. Cohort studies comparing laparotomy and MIS with no-manipulation technique for radical hysterectomy with balanced follow-ups are expected to further validate its efficacy.

It should be noted that the widespread application of R-LESS must be contextualized within its economic implications and associated learning curve. The substantial capital investment required for the robotic system, alongside ongoing maintenance and instrument costs, presents a significant barrier to its broad accessibility. Patients’ financial capacity and acceptance of emerging technologies should be carefully considered. For those with economic hardship, traditional LESS using Zheng’s 4C suspension method remains a safe and viable alternative. Beyond yielding superior perioperative outcomes, another major advantage of R-LESS lies in its ability to enhance anatomical visualization while improving operational flexibility and precision. It enables assistance without requiring additional ports, thereby reducing the complexity of such complex procedures and shortening the learning curve.

This study demonstrated that R-LESS and LESS are both feasible and safe in radical hysterectomy with no-manipulation technique. R-LESS is superior to LESS with significantly reduced OT and EBL with a tendency of less conversions and complications. Both approaches show comparable short-to-medium survival outcomes when sufficient excision extents are reached assisted by valid no-manipulation techniques. However, definitive conclusions regarding oncologic non-inferiority require future investigation with larger sample sizes and longer follow-up. Moreover, prospective randomized studies comparing LESS no-manipulation technique with laparotomy are needed to further elucidate its safety and advantages.