Robot-assisted versus conventional laparoscopic surgery for endometrial cancer: an updated systematic review and meta-analysis Authors: Kanwal Majeed, Muhammad Ahmed, Hassam Ali, Aamna Kashif, Muzna Murtaza, Zainab Muhammad Hanif, Fnu Sidra, Atiya Batool, Jones Onesime, Aymar Akilimali Published: 2026 ABSTRACT AbstractBackgroundEndometrial cancer is commonly treated with minimally invasive staging procedures. Whether robotic assistance improves outcomes compared with conventional laparoscopy remains debatable. Robot-assisted surgery (RAS) and conventional laparoscopic surgery (CLS) have been used to manage endometrial cancer.ObjectivesTo compare the perioperative and oncologic outcomes of robot-assisted and conventional laparoscopy for endometrial cancer.MethodsWe conducted a meta-analysis of studies comparing robot-assisted and conventional laparoscopic surgeries for endometrial cancer. We searched PubMed, the Cochrane Library, and Google Scholar from inception to April 15, 2025, for comparative studies. Random effects meta-analysis was used to estimate risk ratios or mean differences, and heterogeneity was assessed via the I² statistic. Relevant randomized controlled trials (RCTs) and observational studies were identified. The perioperative outcomes of interest included intraoperative visceral injuries, operation time, estimated blood loss (EBL), blood transfusion, total number of lymph nodes harvested (TLNH), conversion to laparotomy, length of hospital stay, recurrence and overall 3-year and 5-year survival rates. Data were synthesized via random effects models, and heterogeneity was evaluated via I² statistics.ResultsThirty studies (2 RCTs and 28 observational studies) including 10,673 patients were analyzed. RAS was associated with reduced estimated blood loss (MD − 78.36) and a lower rate of conversion to laparotomy (RR 0.35). There were no statistically significant differences between groups in operative time, blood transfusion, total lymph node harvested, length of hospital stay, or recurrence. Three-year and five-year overall survival were not significantly different; however, these estimates were based on a limited number of studies and should be interpreted cautiously. Substantial heterogeneity was observed across several outcomes.ConclusionRALS is associated with reduced blood loss and fewer conversions to laparotomy compared with CLS, while most perioperative and oncologic outcomes remain comparable. Given the predominance of observational evidence, persistent heterogeneity, and limited survival data, these findings should be interpreted with caution, and high-quality randomized trials are needed to better define the comparative effectiveness of these approaches.Supplementary InformationThe online version contains supplementary material available at 10.1186/s12893-026-03802-6.Keywords: Conventional laparoscopic surgery, Endometrial cancer, Meta-analysis, Perioperative outcomes, Robot-assisted surgery Introduction Endometrial cancer is the most common gynecologic malignancy in high-income countries, and its incidence is increasing [1]. Its incidence is increasing, driven by obesity, aging, and hormonal factors, making it a growing public health concern. The disease primarily presents with postmenopausal bleeding [2]. Surgical staging is essential for assessing the extent of the disease and guiding treatment. While traditionally performed via open laparotomy, minimally invasive approaches such as laparoscopy and robotic surgery are now preferred for treating endometrial cancer, offering lower morbidity without compromising oncologic outcomes [3]. In this context, comparing robotic-assisted and conventional laparoscopic approaches has become increasingly relevant for optimizing the surgical management of endometrial cancer. Conventional laparoscopy is a minimally invasive surgical technique in which a laparoscope is inserted into the abdomen to guide hysterectomy and bilateral salpingo-oophorectomy [4]. As laparoscopy is less invasive, it is associated with less pain, faster recovery, and a shorter duration of hospital stay [5]. Despite its many advantages over laparotomy, it may result in major complications, including injury to major vessels or nerves, lymphedema, and associated cellulitis [6]. The risk of these complications necessitates the procedure to be performed by a surgeon experienced in laparoscopic hysterectomy and lymph node sampling [5]. While laparoscopic surgery remains a widely adopted minimally invasive technique, robot-assisted laparoscopy (RALS) has emerged as an advanced alternative. RALS, also known as robotic surgery, has rapidly gained acceptance among gynecologic oncologists since 2005, when it was approved by the US Food and Drug Administration (FDA) [7]. Compared with conventional laparoscopy, RALS may provide greater dexterity, tremor filtration, reduced reliance on assistance, and a shorter learning curve [8]. Several studies comparing robotic-assisted hysterectomy with lymphadenectomy to traditional laparoscopy suggest advantages to the robotic approach, such as reduced blood loss and a shorter operative time [9–13]. However, Cardenas-Goicoechea et al. reported that the mean number of lymph nodes retrieved was similar between the groups, with no significant differences in the time to discharge, readmission, or reoperation rates, indicating broadly comparable surgical outcomes between robotic-assisted and conventional laparoscopy [9]. It has also been shown to be a feasible, safe, and reproducible alternative to laparoscopic surgery, even in obese or elderly patients with endometrial cancer [14]. Despite these theoretical advantages, evidence specifically addressing patients with severe obesity remains particularly relevant, as this population presents unique surgical challenges. A retrospective study evaluating robotic-assisted, laparoscopic, and vaginal hysterectomy in morbidly obese women with endometrial cancer demonstrated that minimally invasive approaches, particularly robotic-assisted surgery, are feasible and safe in this high-risk group, with acceptable perioperative outcomes. These findings further support the potential role of robotic platforms in overcoming technical limitations associated with conventional laparoscopy in patients with elevated body mass index [15]. Nonetheless, drawbacks such as longer operative times, higher costs, technical limitations of the robotic system, and the need for specialized staff and training remain important considerations [16]. Inconsistent findings from studies comparing conventional and robot-assisted laparoscopy for endometrial cancer underscore the need for clearer evidence through comprehensive analysis. Our review was prospectively limited to evaluating clinical effectiveness, including operative, perioperative, and oncologic adequacy outcomes, whereas economic evaluations, such as cost-effectiveness, were outside the scope of this analysis. This updated meta-analysis includes additional recent studies comparing robotic-assisted and conventional laparoscopic surgeries for endometrial cancer. It evaluates perioperative and oncologic outcomes, including recurrence and overall survival, and incorporates newly published randomized controlled trials. Subgroup analyses and meta-regression were performed to explore sources of heterogeneity, providing a more detailed assessment of variability across studies. Methods This meta-analysis was reported in accordance with the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines [17]. Search strategy This systematic review and meta-analysis was prospectively registered with PROSPERO (CRD420251120306). A systematic literature search was conducted in PubMed, the Cochrane Library, and Google Scholar from inception to April 15, 2025, and predefined terms for robotic surgery, laparoscopy, and endometrial cancer were used to identify studies comparing robot-assisted and conventional laparoscopic surgery for endometrial cancer. Boolean operators were applied to combine synonyms (ORs) and different concepts (ANDs). MeSH terms (for PubMed) and free-text keywords were included for every concept. Filters for human and English-language publications were applied to the search. Google Scholar was searched for supplementary or gray literature, including trial registries and conference abstracts, although limitations regarding the reproducibility of Google Scholar searches are acknowledged. The search strategy followed the Cochrane Handbook recommendations, using Boolean operators, MeSH terms, and filters to ensure comprehensive and reproducible results. The full search strategies for each database, including Boolean operators, MeSH terms, and keywords, are provided in Supplementary Table S1. Study selection Two reviewers screened the titles and abstracts, assessed the full texts, and resolved disagreements by consensus with a third reviewer. The study selection process followed the PRISMA guidelines and is illustrated in the PRISMA flow diagram (Figure S1). Eligibility criteria Studies were included if they were randomized controlled trials or observational studies comparing robot-assisted laparoscopic surgery with conventional laparoscopic surgery in patients with endometrial cancer and reported at least one predefined outcome measure. For multiple publications from the same cohort, only the most comprehensive or recent report was included. The exclusion criteria were conference abstracts, reviews, editorials, letters, case reports, non-English publications, studies on recurrent endometrial cancer, and studies without direct comparisons between the RALS and CLS. Data extraction and outcomes Two reviewers independently extracted the data via a standardized collection form. The extracted information included study characteristics (first author, year, country), study design, sample size, patient characteristics (age, body mass index, disease stage, tumor grade), and surgical characteristics (uterine weight and adjuvant therapy). The primary outcome was intraoperative visceral injury (ureteral, bladder, bowel, or vascular injury). The secondary outcomes included the operative time, estimated blood loss, blood transfusion rate, total number of lymph nodes harvested, conversion to laparotomy, length of hospital stay, recurrence rate, and overall, three- and five-year survival. Quality assessment Two reviewers independently assessed the methodological quality. Randomized controlled trials were evaluated via the Cochrane risk of bias tool (RoB 2.0), and observational studies were assessed via the Newcastle–Ottawa Scale (NOS). Disagreements were resolved by consensus with a third reviewer’s input. The certainty of evidence for the reported outcomes was not formally assessed using GRADE; therefore, conclusions regarding comparative effectiveness should be interpreted with caution [18, 19]. Statistical analysis Random effects meta-analysis was performed via Review Manager (RevMan) version 5.4.1. Survival outcomes (e.g., 3-year and 5-year survival) were analyzed via risk ratios (RRs), as the included studies reported data primarily in the form of total events and the total number of patients within each group. Hazard ratios (HRs) are not consistently available and cannot be derived because of the lack of time-to-event data. Additionally, Kaplan–Meier curves are not uniformly reported in a manner suitable for reliable digitization; therefore, the reconstruction of HRs from survival curves was not performed. Consequently, survival outcomes were pooled using RRs, which is consistent with the approach applied to other dichotomous outcomes in this study. For continuous outcomes (e.g., operative time and estimated blood loss), mean differences (MDs) with 95% confidence intervals (Cls) were used, depending on measurement consistency across studies. Higgins I² was used to measure heterogeneity [20]. Heterogeneity among studies was assessed via the I² statistic, with values of 25–50%, 50–75%, and > 75% considered mild, moderate, and severe heterogeneity, respectively. Statistical significance was set at p < 0.05 for all analyses. Pooled effect estimates were calculated via a random-effects model to account for the anticipated clinical and methodological variability across the studies. Publication bias was evaluated for outcomes including ≥ 10 studies through visual inspection of funnel plots in accordance with the Cochrane Handbook recommendations. In cases of funnel plot asymmetry, Egger’s regression test was performed via Comprehensive Meta-Analysis software to further assess small-study effects. Outcomes with fewer than 10 studies were not formally assessed for publication bias because of the limited reliability of such analyses. To explore the potential sources of between-study heterogeneity, a series of a priori subgroup analyses were conducted on the basis of clinically and methodologically relevant characteristics. The studies were stratified according to study design (randomized controlled trials versus observational studies), disease stage (early stage versus mixed or advanced-stage disease), body mass index (BMI) (obese [≥ 30 kg/m²] versus nonobese [< 30 kg/m²]), data reporting format (studies reporting outcomes as the mean ± standard deviation versus those requiring statistical conversion from median and dispersion measures), geographic region (United States, Europe, and Asia), and study size (large studies [≥ 200 patients] versus small studies [< 200 patients]). Subgroup analyses were performed for all continuous outcomes, including estimated blood loss (EBL), length of hospital stay (LOS), operative time, and total lymph node harvest (TLNH). Differences between subgroups were assessed via the chi-square test for subgroup differences, as implemented in Review Manager (RevMan), with a P value < 0.05 indicating statistical significance. For studies reporting continuous data as medians, values were converted to means and standard deviations via established statistical methods (e.g., Wan et al.), and additional subgroup analyses were conducted to evaluate the potential impact of such transformations on pooled estimates. Meta-regression analyses were performed to explore the potential sources of heterogeneity. Two prespecified moderators, age and body mass index (BMI), were evaluated across four continuous outcomes that demonstrated substantial heterogeneity. Random-effects meta-regression models using the maximum likelihood estimator were applied. Results Study selection and characteristics A systematic search of the literature was conducted, yielding 5524 articles. After screening, duplicates and ineligible articles were excluded. A total of 30 studies were included in this meta-analysis [7, 11, 12, 21–47]. Two of them were RCTs, and 28 were cohorts with 10,673 patients [3553 in the RALS group and 7120 in the CLS]). The mean age of the patients in the RALS group was 56.34 years, whereas that in the CLS group was 56.94 years. The baseline characteristics of the included studies are summarized in Table 1, and the baseline characteristics of the patients are presented in Table S2. Most of the included studies were observational in design (28 of 30), with only two randomized controlled trials. In addition, baseline oncologic characteristics, including disease stage, tumor grade, uterine weight, and adjuvant therapy, were inconsistently reported across studies, with several variables not reported in a substantial proportion of studies. Table 1Baseline characteristics of included articlesAuthor, YearStudy designCountryPatient populationTotal populationRALSCLSArgenta et al. 2022 [21]RetrospectiveUSA5664611027Bell et al., 2008 [11]RetrospectiveUSA403070Boggess et al., 2008 [12]RetrospectiveUSA10381184Cardenas Goicoechea et al., 2014 [48]RetrospectiveUSA187245432Chiou et al., 2015 [23]RetrospectiveTaiwan86150236Coronado et al., 2012 [24]RetrospectiveSpain7184155Corrado et al., 2015 [25]RetrospectiveItaly72277349Escobar et al., 2012 [49]RetrospectiveKorea303060Eoh et al. 2021 [50]RetrospectiveKorea31532483563Eoh et al., 2023 [27]RetrospectiveKorea5514521003Gehrig et al., 2008 [29]RetrospectiveUSA493281Hiratsuka et al., 2023 [30]RetrospectiveJapan343165Hoekstra et al. 2009 [31]ProspectiveUSA32739Johanssons et al., 2023RetrospectiveAustralia394180Jung et al., 2010 [33]ProspectiveKorea282553Kivekas et al., 2025 [51]RCTFinland494897Lim et al. 2011 [34]RetrospectiveUSA122122244Maenpaa et al., 2016 [35]RCTFinland504999Magrina et al., 2011 [52]RetrospectiveUSA6737104Manchana et al., 2015 [37]RetrospectiveThailand284775Martino et al., 2011 [38]RetrospectiveUSA101114215Nezhat et al., 2019 [39]RetrospectiveUSA5748105Perroni et al., 2021 [53]RetrospectiveItaly4557661221Seamon et al. 2009 [41]RetrospectiveUSA10576181Seror et al., 2014 [42]RetrospectiveFrance40106146Shah et al., 2011 [43]RetrospectiveUSA43118161Silva et al., 2018 [44]RetrospectiveBrazil444589Turunen et al. 2013 [45]RetrospectiveFinland67150217Venkat et al., 2012 [46]RetrospectiveUSA272754Yoon et al., 2024 [54]RetrospectiveKorea95173268NOS Newcastle-Ottawa Scale, RALS Robot assisted laparoscopic surgery, CLS Conventional laparoscopic surgery Baseline characteristics of included articles NOS Newcastle-Ottawa Scale, RALS Robot assisted laparoscopic surgery, CLS Conventional laparoscopic surgery Quality assessment and publication bias The quality assessment of 28 cohort studies and two RCTs was performed via the Newcastle‒Ottawa quality assessment scale and the RoB-2 tool, respectively. All the included RCTs had a low risk of bias (Figure S2) (Table S3). Outcomes consisting of 10 or more studies were assessed for publication bias via funnel plots. (Fig S3-S5). Funnel plots that revealed asymmetry were further assessed for confirmation via Begg’s and Egger’s tests. The p values for Begg’s and Egger’s tests are provided in Table S4. Primary outcome Intraoperative visceral injuries This meta-analysis compared robotic-assisted laparoscopic surgery (RALS) and conventional laparoscopic surgery (CLS) for endometrial cancer in terms of intraoperative visceral injuries. Fourteen studies were included, encompassing 5,774 patients (1,274 in the RALS group and 4,500 in the CLS group). The pooled analysis assessed the incidence of intraoperative visceral injuries and revealed no statistically significant reduction in the RALS group compared with the CLS group. A random-effects model using the risk ratio (RR) was applied to account for the variability across the studies. The risk of intraoperative visceral injury was similar between the RALS and CLS groups [RR = 0.72; 95% CI: 0.43–1.19; p = 0.20]. Heterogeneity among the studies was minimal (I² = 0%), indicating consistent findings across the included studies. Overall, these results suggest that the RALS does not significantly reduce the risk of intraoperative visceral injury compared with the CLS. (Fig. 1A). A Intra Operative visceral injury (B) Blood transfusion. Squares indicate risk ratios; horizontal lines represent 95% confidence intervals; diamonds indicate pooled estimates Secondary outcomes Blood transfusion Blood transfusion rates were reported in ten studies, including 1,805 patients (768 in the RALS group and 1,037 in the CLS group). There was no appreciable difference in blood transfusion rates between the two groups according to the pooled analysis [RR, 0.73; 95% CI [0.40, 1.33]; p = 0.30, I²=27%] (Fig. 1B). Conversion to laparotomy The conversion rate was examined in 17 studies encompassing 2,791 patients: 1,247 in the RALS group and 1,544 in the CLS group. The meta-analysis revealed that patients in the RALS group had significantly fewer conversions to laparotomy than those in the CLS group did [RR 0.35, 95% CI [0.23–0.53]; p < 0.00001, I²=0%] (Fig. 2A). A Conversion to Laparotomy (B) Recurrence. Squares indicate risk ratios; horizontal lines represent 95% confidence intervals; diamonds indicate pooled estimates Recurrence Six studies reported recurrence rates among 2,921 patients (1,265 and 1,656 in the RALS and CLS groups, respectively). There was no conclusive evidence that RALS reduced the recurrence risk compared with CLS [RR, 0.85; 95% CI [0.59, 1.23]; p = 0.39, I² = 34%] (Fig. 2B). Estimated blood loss The estimated blood loss was analyzed in 20 studies involving 2,989 patients (1,302 RALS and 1,687 CLS). Pooled analysis showed a significant reduction in blood loss with RALS compared with CLS [MD: −78.36, 95% CI [− 99.07, − 57.65]; p = 0.0001] however, heterogeneity was high (I² = 81%), sensitivity and subgroup analyses did not fully explain the high heterogeneity. Therefore, these results should be interpreted as indicative of a trend rather than definitive (Fig. 3A). A Estimated blood loss (B) Total lymph node harvested. Squares indicate mean difference; horizontal lines represent 95% confidence intervals; diamonds indicate pooled estimates. Arrow indicates that the confidence interval of [44] extended beyond the axis limits Total number of lymph nodes harvested Twelve studies involving 1,980 patients (908 RALS and 1,072 CLS) reported total lymph node harvest. Pooled analysis showed no statistically significant difference between RALS and CLS [MD 0.98; 95% CI − 2.36 to 4.31; p = 0.57]; however, heterogeneity was very high (I² = 90%), and sensitivity and subgroup analyses did not fully explain this variability. Therefore, these results should be interpreted as indicative of a directional trend rather than a definitive effect (Fig. 3B). Length of hospital stay Twenty-one studies involving 6,444 patients (1,555 RALS and 4,889 CLS) reported the length of hospital stay. Pooled analysis showed no statistically significant difference between RALS and CLS [MD − 0.35 days; 95% CI − 0.72 to 0.02; p = 0.07]; however, heterogeneity was very high (I² = 94%), and sensitivity and subgroup analyses did not fully explain this variability. Therefore, these results should be interpreted as indicative of a directional trend rather than a definitive effect (Fig. 4A). A Length of hospital stay (B) Operation time minutes. Squares indicate mean difference; horizontal lines represent 95% confidence intervals; diamonds indicate pooled estimates Operative time Twenty-two studies involving 4,295 patients (1,799 RALS and 2,496 CLS) reported operative time. Pooled analysis showed no statistically significant difference between RALS and CLS [MD 11.49 min; 95% CI − 7.12 to 30.10; p = 0.23]; however, heterogeneity was very high (I² = 97%), and sensitivity and subgroup analyses did not fully explain this variability. Therefore, these results should be interpreted as indicative of a directional trend rather than a definitive effect (Fig. 4B). 3- and 5-year survival Overall survival outcomes were reported by four studies for 3-year survival (n = 1,796) and four studies for 5-year survival (n = 6,028). No statistically significant differences were observed between RALS and CLS for either 3-year survival [RR: 1.01, 95% CI 0.96–1.06; P = 0.82, I² = 67%] or 5-year survival [RR: 1.01, 95% CI 0.98–1.05; p = 0.38, I² = 67%] (Fig. 5A–B). However, these findings are based on a limited number of studies and were derived using pooled risk ratios at fixed time points rather than time-to-event analyses, which may limit their robustness. A Overall 3-year survival (B) Overall 5-year survival. Squares indicate risk ratios; horizontal lines represent 95% confidence intervals; diamonds indicate pooled estimates Sensitivity analysis Sensitivity analysis for the outcome of total number of lymph nodes harvested, in which we removed one study at a time, revealed that after exclusion of the study by Lim et al., heterogeneity decreased from 90% to 66%, suggesting that heterogeneity was a significant source of variability [RR = 2.03; P = 0.03, I2 = 66%; 95% CI (0.15–3.91)] (Fig S6B). This study reported a substantially greater negative mean difference than the other studies did, potentially due to its case-matched design using the first 122 patients in each surgical group, thereby capturing early-phase learning curves. Such methodological factors, including potential selection bias and differences in surgical experience, may have contributed to the more extreme outcomes, which in turn inflated the overall heterogeneity in the meta-analysis. Leave-one-out sensitivity analysis was performed for the outcome estimated blood loss, which revealed a notable reduction in heterogeneity (I² from 81% to 65%) upon exclusion of the study by Maenpaa et al. [35] [MD = -84.60; P < 0.00001, I2 = 65%; 95% CI (-101.12– -68.09)] (Fig S6A). This study differed methodologically from the others, as it was an RCT, whereas the remaining studies were cohort designs. The distinct design and potentially stricter control of confounding variables in the RCT may have contributed to the variability in estimated blood loss outcomes, thus accounting for its disproportionate impact on overall heterogeneity. For operative time, heterogeneity was substantial (I² = 97%), and sensitivity analysis by removing one study at a time did not result in any reduction in heterogeneity, indicating that variability was not driven by a single outlier but likely reflected true interstudy differences in surgical settings, case complexity, and institutional protocols. For hospital stay, heterogeneity was also substantial (I² = 94%), and sensitivity analysis by excluding one study at a time did not result in any reduction in heterogeneity, indicating that variability was not driven by a single outlier. High heterogeneity persisted for several outcomes despite leave-one-out and subgroup analyses, suggesting genuine between-study differences in the case-mix and perioperative pathways. Continuous outcomes with subgroup analyses Study design The differences in the study designs were notable. For EBL, the effect was significant in observational studies (MD: −84.60 mL, P < 0.00001; I² = 65%) but not in RCTs (MD: −4.00 mL, P = 0.71), with a significant subgroup interaction (P < 0.00001). Similarly, the operative time favored RALS in the RCT (MD: −41.00 min, P < 0.00001) but not in the observational studies (MD: 14.10 min, P = 0.14), again with a significant subgroup difference (P < 0.00001). In contrast, LOS and TLNH did not significantly differ between the study designs (Figure S7). Disease stage Across disease stages, the reduction in EBL remained consistent in both early-stage (MD: −161.60 mL, P = 0.01; I² = 48%) and mixed/advanced disease (MD: −70.76 mL, P < 0.00001; I² = 84%), without a significant subgroup difference (P = 0.16). The LOS remained comparable across stages. The operative time was longer in early-stage disease patients (MD: 26.45 min, P < 0.0001) than in advanced-stage patients. In contrast, TLNH was greater in mixed/advanced disease patients (MD: 3.61 nodes, P = 0.009), although the subgroup interactions were not statistically significant (Figure S8). Body Mass Index (BMI) RALS consistently reduced EBL in both obese (MD: −79.45 mL, P < 0.00001; I² = 86%) and nonobese patients (MD: −75.37 mL, P < 0.00001; I² = 30%), with no subgroup difference (P = 0.83). The LOS was modestly reduced in nonobese patients (MD: −0.66 days, P = 0.02) but not in obese patients. Operative time and TLNH remained similar across BMI groups (Figure S9). Data transformation method The effect of RALS on EBL remained significant regardless of reporting format—both in directly reported data (MD: −94.20 mL, P < 0.00001; I² = 51%) and converted data (MD: −61.76 mL, P = 0.0001; I² = 84%), with no subgroup difference (P = 0.09). For LOS, only directly reported data showed a significant reduction (MD: −0.52 days, P = 0.01), whereas no differences were observed for operative time or TLNH across transformation methods (Figure S10). Geographic region The reduction in EBL was consistent across regions, including the USA (MD: −86.80 mL, P < 0.00001), Europe (MD: −63.22 mL, P = 0.01), and Asia (MD: −70.24 mL, P = 0.0001), with no subgroup interaction (P = 0.63). LOS, operative time, and TLNH showed no consistent regional differences, although LOS demonstrated a modest subgroup interaction (P = 0.03) (Figure S11). Study size RALS reduced EBL in both large (MD: −92.44 mL, P < 0.00001) and small studies (MD: −72.65 mL, P < 0.00001), with no subgroup difference (P = 0.27). The LOS was shorter in large studies (MD: −1.19 days, P = 0.02) but not in small studies, whereas the TLNH was greater in small studies (MD: 3.79 nodes, P = 0.03). The operative time remained comparable regardless of the study size (Figure S12). Meta-regression analysis Meta-regression analyses were performed to assess the impact of age and body mass index (BMI), including operation time, estimated blood loss, total number of lymph nodes harvested (TLNH), and length of hospital stay (LOH), on outcomes with high heterogeneity. Neither age nor BMI was significantly associated with effect size across any of the outcomes (all p > 0.05). The proportion of between-study variance explained by these moderators was minimal (R² analog range: 0.00–0.06), indicating that these variables did not significantly account for heterogeneity. The scatter plots are provided in the supplementary figures (Figure S13-S16). Discussion Surgical care is being reimagined with previously unattainable precision and control as we advance into an era characterized by robotic-assisted techniques [55]. Our meta-analysis attempts to evaluate the relative effectiveness, safety, and potential for improved patient outcomes in this quickly developing field of minimally invasive methods by contrasting robot-assisted and conventional laparoscopic procedures for endometrial cancer. Our findings suggest that RALS is associated with reduced estimated blood loss and a lower rate of conversion to laparotomy compared with CLS. However, most other perioperative and oncologic outcomes were broadly comparable between the two approaches, indicating no clear overall superiority. Overall, these findings indicate that while RALS may offer specific technical advantages, it does not demonstrate clear overall superiority over CLS. Our updated meta-analysis included 30 studies with 10,673 patients and had almost three times the sample size of the 2016 meta-analysis by Xie et al. [47], which included 19 studies with 3,056 patients. The larger sample size strengthens the validity of our findings and supports the robustness of our estimates. Furthermore, our analysis broadens the focus by incorporating a wider variety of outcomes that were not thoroughly covered in previous studies, such as recurrence and overall survival. We improve the statistical power and offer more precise information about the long-term safety and effectiveness of both surgical techniques by integrating more recent research. Given the lack of a significant difference in the incidence of intraoperative visceral injuries, our meta-analysis indicates that the RALS and CLS are similarly effective in this regard. Robotic-assisted surgery includes features such as 3D visualization, wristed instruments, tremor filtration, and ergonomic design, which have been described in previous studies [56–58]. Moreover, the evolution of robotic platforms continues to expand the scope of minimally invasive surgery in endometrial cancer. Recent evidence from pilot studies evaluating robotic single-port hysterectomy using the da Vinci SP system suggests that further technological refinements may enhance surgical feasibility while maintaining safety and efficacy in selected patients. These advancements highlight that robotic surgery is not a static modality but an evolving field, and current findings should be interpreted within the context of ongoing technological innovation [59]. While these characteristics may theoretically enhance surgical precision and dexterity, their impact on clinical outcomes remains to be confirmed. Our pooled data did not reveal a statistically significant reduction in intraoperative visceral injuries with the RALS, suggesting that any potential mechanistic advantages may not translate into measurable differences in this outcome. The improved accuracy and flexibility of robotic instruments, which enable more controlled dissection and less trauma to surrounding tissues during surgical staging procedures, may be responsible for reduction in estimated blood loss observed with RALS in our study [23]. In robotic surgery, regulated tissue handling and precise dissection are made possible by increased precision, enhanced dexterity, and superior three-dimensional visualization, all of which probably help minimize blood loss [55]. However, while the reduction in estimated blood loss with RALS was statistically significant (− 78 mL), the absolute difference was modest in the context of minimally invasive hysterectomy, where average blood loss is typically < 200 mL and may have limited clinical impact for most patients. Although the robotic arm experienced somewhat lower blood loss of approximately 50 mL, this did not always result in the decrease of hemoglobin levels or the need for blood transfusions. Although some studies reported that robotic surgery resulted in decreased transfusion rates, our study revealed that there was no statistically significant difference in blood transfusion rates between the two groups [60]. In the treatment of endometrial cancer, robotic-assisted laparoscopic surgery has been reported in some studies to be quicker than traditional laparoscopy, with a shorter overall operating room duration. This study by Mäenpää et al. highlights the improved efficiency of the robotic approach [35]. Since robotic devices require additional setup and preparation time, Ran et al.‘s study revealed that conventional laparoscopic surgery is typically faster than robotic surgery. The overall surgery and operating room times were longer with the robotic approach, even though the main surgical time was similar for both methods [61]. Although there are conflicting reports in the literature concerning which procedure is more time-efficient, our research revealed no significant differences in the operating time between robotic and laparoscopic surgery, indicating similar procedure times in clinical settings. Our meta-analysis revealed that RALS was associated with a lower rate of conversion to laparotomy compared with CLS in terms of surgical outcomes. According to several studies, patients who underwent RALS had a lower conversion rate to laparotomy than those who underwent CLS, despite having greater BMIs. This may reflect the technical advantages of the robotic platform; however, patient selection and surgeon experience likely also contribute. The evidence indicates that RALS is still practical and successful in obese patients with endometrial cancer, even though a greater BMI is known to make laparoscopic procedures more difficult and increase the chance of conversion [47, 62]. According to the article by Mäenpää et al., due to complications such as adhesions, disseminated cancer, trocar site bleeding, and a uterus that is too large to be removed vaginally, conversion to laparotomy was more common in the group that underwent conventional laparoscopy. This suggests that the robotic approach provides better handling of intraoperative challenges [35]. While RALS demonstrates lower intraoperative blood loss and reduced conversion rates, it is important to recognize that these outcomes may not be solely attributable to robotic technology. Surgeon expertise, patient selection, and institutional protocols likely contribute substantially to the observed benefits. The technological advantages of robotic systems may therefore be synergistic with surgical proficiency rather than independently causative, representing a key limitation of the current evidence. Although this difference was not statistically significant, our research revealed that the RALS group tended to stay in the hospital for a shorter period than did the CLS group. Compared with conventional surgery, robotic-assisted surgery was associated with a trend toward shorter length of hospital stay, most likely because it is minimally invasive, with smaller incisions, less trauma, and quicker recovery. While robotic methods might provide additional reductions in the length of hospital stay compared with laparoscopic surgery, this result can still be affected by patient comorbidities, surgeon experience, institutional volume, and treatment complexity [63]. The total number of lymph nodes harvested in the RALS and CLS groups did not differ significantly, suggesting that the two surgical techniques were equally effective in terms of oncologic adequacy. Within the Mäenpää trial, the groups that underwent robotic-assisted and traditional laparoscopic surgery had similar median yields of 25 and 23 lymph nodes, respectively. Given the hospital’s extensive experience and ability to perform laparoscopic surgery for endometrial cancer, the comparable yield is probably the result of the consistent efficacy of lymphadenectomy across both surgical procedures [35]. No statistically significant differences were observed between the RALS and CLS groups in 3-year and 5-year overall survival or recurrence rates. However, these findings should be interpreted cautiously, as only four studies reported survival data, and patient staging, tumor characteristics, and adjuvant therapy were inconsistently reported or not controlled across studies. In addition, survival outcomes were analyzed using pooled risk ratios at fixed time points rather than time-to-event measures such as hazard ratios, which is methodologically less robust and may not fully capture differences over time. Consequently, the strength of evidence for long-term oncologic equivalence between the RALS and CLS is limited. Vanesa Delso et al. conducted a retrospective analysis and discovered no statistically significant differences between RALS and CLS in terms of disease-specific survival, overall survival, or disease-free survival in women with early-stage endometrial cancer. The authors concluded that survival outcomes were unaffected by the minimally invasive surgical technique selected, whether CLS or RALS. This was explained by the comparable oncologic safety of the two methods, which both demonstrated improved quality of life and fewer perioperative complications while preserving similar long-term survival outcomes [64]. The RALS and CLS groups in Yoon et al.’s study had recurrence rates of 6.3% and 10.4%, respectively. The CLS group had a slightly greater recurrence rate than the other groups did, but this difference was not statistically significant. The greater recurrence rate in the CLS group might be explained by identified risk factors for endometrial cancer, such as tumor size and depth of myometrial invasion [54]. Importantly, the certainty of evidence remains limited. The predominance of observational studies, substantial heterogeneity, and inconsistent reporting of key oncologic variables reduce confidence in the pooled estimates. As such, the observed advantages of RALS should be interpreted cautiously and should not be considered definitive evidence of superiority. Although our study focused primarily on clinical outcomes, cost-effectiveness is an important consideration in the adoption of robotic surgery. Reported costs across studies are highly heterogeneous, reflecting differences in healthcare systems, currencies, and reporting years. The costs were not standardized or adjusted for inflation, limiting the reliability of direct comparisons between the RALS and CLS. Consequently, conclusions regarding cost-effectiveness should be interpreted with caution. Future studies employing standardized economic evaluations are warranted to provide clearer insights into the financial implications of robotic-assisted surgery. Limitations The shortcomings of this meta-analysis are numerous. Most included studies were observational, limiting causal inference and introducing potential bias. Potential unmeasured confounders such as surgeon experience, case selection, and institutional expertise likely contributed to the observed differences. Therefore, the advantages of RALS may reflect synergistic effects of technology and surgical proficiency rather than purely technological causation. Comparability was also impacted by significant variation among studies regarding surgical procedures, patient groups, outcome definitions, and the language bias introduced by limiting inclusion to English-language publications. Additionally, our search was limited to PubMed, the Cochrane Library, and Google Scholar, so relevant studies indexed in other databases may have been missed. Another important limitation is the incomplete reporting of baseline oncologic and patient factors such as tumor stage and (neo)adjuvant therapy. Despite the sensitivity analyses, substantial heterogeneity persisted (I² > 90% for several outcomes), including key outcomes such as operative time, blood loss, and length of hospital stay. Additionally, the use of risk ratios instead of hazard ratios for survival outcomes further limits the strength of conclusions regarding long-term oncologic effectiveness. The variability in discharge criteria across studies may have influenced LOS reporting. Operative time and other perioperative outcomes are known to improve with increasing robotic case volume; however, primary studies rarely report surgeon experience or institutional volume in a standardized manner, preventing formal adjustment. Future multicenter trials should stratify outcomes by surgeon experience to better capture the effect of the learning curve. Findings on quality of life, long-term oncologic outcomes, and cost-effectiveness were constrained by the paucity of available data. This limitation is particularly relevant given emerging evidence from alternative minimally invasive approaches, such as transvaginal natural orifice transluminal endoscopic surgery (vNOTES), which have demonstrated differences in health-related quality of life compared with conventional laparoscopy. Such findings emphasize that perioperative outcomes alone may not fully capture patient-centered benefits, underscoring the need for future studies to incorporate standardized quality-of-life assessments when comparing surgical techniques [65]. Although formal tests for publication bias were largely nonsignificant, the possibility of publication bias cannot be completely ruled out. Moreover, oncologic outcomes are limited by the small number of studies reporting survival, as well as variability in patient staging and adjuvant therapy, restricting confidence in conclusions about long-term oncologic equivalence. These constraints highlight the necessity of carefully planned RCTs with standardized reporting. Future directions Although our meta-analysis identified potential advantages of the RALS over CLS for certain intraoperative and postoperative outcomes, more large-scale, high-quality randomized controlled trials are needed to confirm these results and more thoroughly evaluate long-term oncologic outcomes. Future studies should also concentrate on quality-of-life metrics, patient-reported outcomes, and cost-effectiveness evaluations to enhance clinical decision-making. The wider application and durability of the RALS in standard gynecologic oncology treatment will also require research into the learning curve, surgeon expertise, and access to robotic technology across various healthcare settings. Future research should also address the challenges faced in resource-limited settings where access to robotic platforms is scarce. The high acquisition and maintenance costs, coupled with the need for specialized training, may hinder widespread adoption despite potential perioperative advantages. In such environments, conventional laparoscopy is likely to remain the more practical option. Accordingly, incorporating cost-effectiveness and health equity considerations will be essential to informing policy and investment decisions across diverse healthcare systems. Conclusion Our meta-analysis revealed that, compared with CLS, RALS was associated with reduced intraoperative blood loss and a lower conversion rate to laparotomy. RALS also showed nonsignificant trends toward fewer intraoperative visceral injuries and shorter hospital stays. In contrast, no significant differences were observed between the two approaches in terms of operative time, transfusion rate, total number of lymph nodes harvested, recurrence, or 3- and 5-year survival outcomes. Overall, while RALS may offer certain perioperative advantages, it does not demonstrate clear overall superiority over CLS, and findings are limited by heterogeneity and the predominance of observational evidence. The financial implications of robotic surgery, including acquisition, maintenance, and procedural costs, remain uncertain; dedicated studies are needed to provide evidence on cost-effectiveness and inform resource allocation decisions. Supplementary Information Supplementary Material 1. Acknowledgements The authors thank the leadership of the Medical Research Circle (MedReC), Democratic Republic of the Congo, for institutional support. Use of artificial intelligence tools No 0.35 intelligence was used in generating the manuscript. Provenance and peer review Not commissioned, externally peer reviewed. Author declaration All the authors have read and approved the final version of the manuscript. Kanwal Majeed and Aymar Akilimali have full access to all of the data in this study and take complete responsibility for the integrity of the data and the accuracy of the data analysis. Transparency statement Kanwal Majeed affirms that this manuscript is an honest, accurate, and transparent account of the study being reported; that no important aspects of the study have been omitted; and that any discrepancies from the study as planned have been explained. Abbreviations Body mass index Confidence interval Conventional laparoscopic surgery Estimated blood loss Food and Drug Administration Higgins’ heterogeneity statistic Mean difference Newcastle–Ottawa Scale Preferred Reporting Items for Systematic Reviews and Meta-Analyses Quality-adjusted life year Randomized controlled trial Robotic-assisted laparoscopic surgery Review Manager Risk of Bias Risk ratio Total number of lymph nodes harvested Authors’ contributions Conceptualization and study design: Kanwal Majeed, Muhammad Ahmed, Zainab Muhammad Hanif and Aamna Kashif, Project administration: Kanwal Majeed, Zainab Muhammad Hanif and Aamna Kashif, Supervisor: Kanwal Majeed and Aymar Akilimali, Investigation: Kanwal Majeed, Muhammad Ahmed and Zainab Muhammad Hanif, Literature search and data synthesis: Muhammad Ahmed, Hassam Ali, Muzna Murtaza, Zainab Muhammad Hanif and Aymar Akilimali, Collection and assembly of data: Muhammad Ahmed, Hassam Ali, Muzna Murtaza and Jones Onesime, Manuscript drafting: Kanwal Majeed, Hassam Ali, Muzna Murtaza, Zainab Muhammad Hanif, Aamna Kashif, FNU Sidra, Atiya Batool, Jones Onesime and Aymar Akilimali *, Manuscript review: * All authors *, Final approval of manuscript* : All authors. Funding The authors did not receive any financial support for this work. No funding has been received for this study. Data availability The authors confirm that the data supporting the findings of this study are available within the supplementary materials. Declarations Ethics approval and consent to participate Not applicable. Consent for publication Not applicable. Competing interests The authors declare no competing interests. Footnotes Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. References Associated Data Supplementary Materials Supplementary Material 1. Data Availability Statement The authors confirm that the data supporting the findings of this study are available within the supplementary materials.