Robotic-assisted versus conventional laparoscopic colorectal resection for colorectal cancer: a systematic review, meta-analysis and GRADE assessment of perioperative and postoperative outcomes

¹Department of Medicine, Sir Seewoosagur Ramgoolam Medical College, Mauritius, ²Department of Pharmacology, Sir Seewoosagur Ramgoolam Medical College, Moka, Mauritius, ³Department of Surgery, Joe Morolong Memorial Hospital, Vryburg, North West, South Africa, ⁴Department of Urology and Robotic Surgery, University of Florida, Florida, United States of America

^&Corresponding author
Indrajit Banerjee, Department of Pharmacology, Sir Seewoosagur Ramgoolam Medical College, Moka, Mauritius

Introduction    

According to WHO, colorectal cancer is the third most common cancer worldwide and is the second leading cause of cancer-related deaths. (World Health Organization, 2026) [1]. This global burden highlights the importance of minimizing perioperative and post-operative complications in a bid to decrease morbidity and mortality. Conventional laparoscopy is a safe and effective option [2]; however, its limitations include: 2-dimensional view, unstable assistant-controlled camera, poor ergonomics, straight-tip instruments, fulcrum effect, and enhanced tremor effect [3]. These limitations are especially relevant where the surgeon must navigate the narrow pelvis to ensure the preservation of pelvic autonomic nerves to maintain postoperative urinary and sexual functions.

Robotic-assisted surgery addresses the above limitations by providing better dexterity, reduced tremors, and better visualization [4]; however, it comes with longer operative time and a higher cost [5]. Previous systematic reviews have reported several advantages of robotic-assisted surgery over conventional laparoscopic surgery, which focus on lower conversion rates as the primary outcome [6], but limited literature focuses on functional and oncological outcomes.

Functional outcomes such as urinary and sexual function post-operatively can substantially affect the patient's quality of life. A previous review conducted in 2018 suggested that robotic surgery may offer faster recovery in urogenital function as compared to laparoscopic surgery [7]; therefore, we have used more recent data to offer a more accurate and up-to-date analysis. Despite the increasing number of systematic reviews investigating robotic-assisted surgery, an important evidence gap persists, as certainty of evidence for key clinical outcomes has not been comprehensively evaluated using the GRADE framework.

This systematic review and meta-analysis were conducted to compare robotic-assisted surgery with conventional laparoscopic surgery in patients undergoing colorectal cancer surgery. We have evaluated functional outcomes at six months, including urinary and sexual function, and operative outcomes such as lymph node harvest yield and intraoperative blood loss. Postoperative outcomes such as length of hospital stay and days to first flatus were also assessed.

Methods     

This systematic review and meta-Analysis were conducted in accordance with the Preferred Reporting Items for Systematic Reviews and Meta-Analyses 2020 (PRISMA) Figure 1 [8].

Registration and protocol: registration of the review was performed on PROSPERO, an international prospective register of systematic reviews, and this research was registered under ID number: CRD420261451189. The full protocol can be accessed in PROSPERO [9].

Search strategy: an electronic search was conducted on PubMed, Cochrane Central Register of Controlled Trials (CENTRAL), and Google Scholar in order to find relevant manuscripts to the study. A combination of keywords and Boolean operators was used for the data extraction: (robotic surgical procedures) AND (conventional Laparoscopic) AND (colorectal cancer))/ ((robotic surgical procedures) AND (conventional laparoscopic) AND (bladder function))/ ((robotic surgical procedures) AND (conventional laparoscopic) AND (sexual function))/ ((robotic surgical procedures) AND (conventional laparoscopy) AND (outcomes)). The strategy employed in the search, along with the total number of articles screened, is presented in the PRISMA flow diagram (Figure 1).

PICOS Framework: the eligibility criteria for this study were defined in accordance with the PICOS framework (Population (P), Intervention (I), Comparator (C), Outcomes (O), Study design (S)). Population: Adult patients with a confirmed diagnosis of colorectal cancer who were scheduled for surgical resection of the cancer. Intervention: patients undergoing robotic-assisted surgery. Comparator: conventional laparoscopic procedure.

Outcomes: the primary outcomes assessed were bladder function, male sexual function, and female sexual function; secondary outcomes included lymph nodes harvested, length of hospital stay, blood loss intraoperatively, and days before first flatus.

Study design: randomized Controlled Trials (RCTs) published in English between 2016 and 2026.

Eligibility criteria: all the titles were initially screened, and appropriate abstracts were reviewed by two independent reviewers (RC and AB) using the inclusion criteria, and the full texts of potentially eligible articles were then obtained and assessed against the predefined inclusion and exclusion criteria by the same reviewers. The date of the last search was 25^th July 2026.

Inclusion criteria: all Randomized Controlled Trials (RCTs) meeting the following inclusion criteria were taken into consideration: adult patients undergoing elective surgery for colorectal cancer, comparing robotic-assisted surgery with conventional laparoscopic surgery. Studies that reported at least one of the predefined outcomes were included. Full-text articles were included and assessed for eligibility in this systematic review. Only the RCTs available in English were included in this study.

Exclusion criteria: paediatric patients or those who underwent open or emergency surgery were excluded. All non-RCTs, cohort studies, case-control studies, cross-sectional studies, case series, case reports, in vitro studies, and animal experiments were not included in this study.

Data extraction: all the titles were screened independently by two researchers (RC and AB), and any disagreements were resolved by the involvement of a senior third reviewer (IB). The full-text RCTs that satisfied the eligibility criteria were identified for inclusion in the final selection. A total of six RCTs were ultimately included for systematic review, and 3 RCTs were included for meta-analysis. Table 1 shows the study authors and year, country of study, type of study, sample size, type of surgery, urinary function, male sexual function, female sexual function, lymph nodes harvested, length of hospital stay, blood loss intraoperatively, and days before first flatus.

Risk of bias assessment: the quality of the studies was assessed independently by two researchers (RC and AB). The Cochrane risk-of-bias tool for randomized trials (RoB2) was used for quality assessment [10]. The RoB2 tool categorizes each domain as having a low risk of bias, some concerns, or high risk of bias. The findings were visually summarized using the Risk-of-bias Visualization (ROBVIS) web application program, which generated traffic light plots and weighted bar charts [11] (Figure 2).

Statistical analysis: meta-Analysis was performed using the RevMan Web Application. All the continuous outcomes (bladder function, male sexual function, female sexual function, lymph nodes harvested, length of hospital stay, blood loss intraoperatively, and days to first flatus) were reported as mean with standard deviations. In papers that did not publish this data (e.g., in studies reporting median and interquartile ranges), the reviewers estimated the mean and standard deviation according to Luo [12] and Wan's methods [13]. Mean differences (MD) with 95% confidence intervals (CI) were calculated using the RevMan Web application. Heterogeneity was assessed using I² statistics, with >50% being considered significant heterogeneity, while studies with <50% I² were considered as low heterogeneity. A random-effects model was used for studies with high heterogeneity (I² >50%), and a fixed-effect model was utilized for those studies with low heterogeneity (I²<50%). P-values < 0.05 were considered statistically significant. In order to explain the high heterogeneity across certain factors, a sensitivity analysis was performed on female sexual function, lymph nodes harvested, and length of hospital stay. The findings were visually summarised by generating forest plots.

GRADE assessment: the certainty of evidence for each outcome was evaluated by using the Grading of Recommendations Assessment, Development and Evaluation (GRADE) approach, and classified as high, moderate, low, or very low certainty. Assessments were based on five domains: Risk of Bias, Inconsistency, Indirectness, Imprecision, and Publication Bias. A summary of findings (SoF) was prepared by using the GRADEpro Guideline Development Tool (GRADEpro GDT), incorporating pooled effect estimates, absolute effects, and the corresponding certainty ratings for all the key outcomes.

Results     

An extensive review of the literature was conducted, which generated a total of 19,507 articles (PubMed: 1423, Cochrane: 84, Google Scholar: 18,000). 1507 of these were identified as duplicates and were excluded thereafter by EndNote 21. Therefore, 18,000 records were screened after de-duplication. Non-randomized controlled trials, cohort studies, case-control studies, case reports, animal experiments, and in vitro studies were excluded (n=17,920). Ultimately, 80 full-text articles were assessed for eligibility. This yielded 6 RCTs that were analysed for comparison of operative and postoperative outcomes between conventional laparoscopic surgery and robotic-assisted surgery in patients suffering from colorectal cancer for systematic review, and 3 RCTs were analysed for data pooling and development of forest plots.

Risk of bias assessment: the methodological quality of the included studies was assessed using the Cochrane Risk of Bias (ROB2) tool. The results indicate risk of bias from randomisation process (low risk: 66.7%, some concerns: 16.7%, high risk: 16.7%), low risk of bias from deviations from intended interventions (low risk: 100%), low risk of bias from missing outcomes (low risk: 100%), bias due to measurement of outcome (low risk: 83.3%, high risk: 16.7%), bias due to selection of reported result (low risk: 83.3%, high risk: 16.7%). Overall, four studies (66.7%) were judged to have a low risk of bias, one study (16.7%) had some concerns, and one study (16.7%) was judged to have a high risk of bias.

Primary outcomes

Bladder function (I-PSS): three studies involving 1620 patients (811 robotic-assisted and 809 laparoscopic) evaluated bladder function 6 months post-operatively in patients having undergone CLS vs RAS using I-PSS (International Prostate Symptom Score) as a primary outcome. The mean difference was found to be -1.37, 95% CI [-1.72, -1.01], and the overall effect was P<0.00001, showing statistical significance that robotic-assisted surgery offers lower I-PSS. No statistical heterogeneity was observed among the included studies (I²=0%, P=0.80) (Figure 3A).

Male sexual function (IIEF): the forest plot in Figure 3B demonstrates male sexual function 6 months post-operatively using IIEF (International Index of Erectile Function) for CLS vs RAS. Three studies including 914 patients (466 RAS and 448 CLS) were analyzed. The MD was found to be 2.83 with 95% CI [2.11, 3.55] and an overall effect of P<0.00001, providing statistical evidence favouring robotic-assisted surgery for the same. No heterogeneity was found among the included studies (I²=0%, P=0.64).

Female sexual function (FSFI): three studies including 581 patients (293 RAS and 288 CLS) were considered for this analysis. FSFI (Female Sexual Function Index) was used to determine female sexual function 6 months post-operatively to compare CLS and RAS. The MD was found to be 2.88 [1.79,3.97] with 95% CI and an overall effect of P<0.00001 favouring RAS. High heterogeneity was found among the included studies. (Tau²=0.61, I²=67%, P=0.07). The result is displayed in Figure 3C.

Secondary outcomes: perioperative outcomes

Lymph nodes harvested: four studies including 937 patients (472 RAS, 465 CLS) evaluated the number of lymph nodes harvested perioperatively in patients undergoing CLS vs RAS. The analysis did not demonstrate any statistically significant difference in lymph nodes harvested perioperatively. The MD= 1.24, 95% CI [-3.78, 6.26], and P=0.63. High heterogeneity was observed among the studies (Tau²=22.93, I²=96%, p<0.00001) (Figure 4A).

Blood loss (ml): two studies including 376 patients (193 RAS, 183 CLS) demonstrated the blood loss in mL in each surgery. The mean difference was found to be -26.70 with 95% CI [-32.61, -20.78] and P<0.00001 in favour of robotic-assisted surgery. No heterogeneity was found between the studies (I²=0%, P=0.93) (Figure 4B).

Post-operative outcomes

Length of hospital stay (days): three studies including 478 patients (244 RAS, 234 CLS) showed that length of hospital stay in days postoperatively in patients having undergone either surgery. The MD=-1.24 with 95% CI [-2.55, 0.06] and overall effect of P=0.06 showing no statistical significance. High heterogeneity was found among the studies (I²=82%, P=0.0010) (Figure 4C).

Days to first flatus: the forest plot in Figure 4D shows the analysis of two studies including 376 patients (193 robotic assisted, 183 laparoscopic) for the number of days taken to pass first flatus post-operatively following either laparoscopic or robotic assisted surgery. The mean difference was found to be -0.99 with 95% CI [-1.14, -0.83] with P<0.00001 favouring robotic assisted surgery. No heterogeneity was found between these two studies (I²=0%, P=0.37)

GRADE assessment: Figure 5 presents the GRADE assessment of all Randomized Controlled trials, which provide a high-certainty starting level of evidence. For the primary outcomes, the certainty of evidence remained high for the outcomes of bladder function, male sexual function, and female sexual function, as no serious concerns were identified regarding risk of bias, inconsistency, indirectness, imprecision, or publication bias. The certainty of the evidence for secondary outcomes, blood loss and days to first flatulence remained moderate, due to imprecision from a small sample size. The certainty of evidence was downgraded by 1 level. For the lymph node harvested, the certainty of evidence was downgraded by 2 levels to low due to considerable statistical heterogeneity (I²> 50%) and wide CI. Overall GRADE assessment indicates high confidence in the evidence supporting that robotic-assisted surgery is superior to conventional laparoscopy. Figure 6 shows sensitivity with heterogeneity> 50%. Figure 6A female sexual function, Figure 6B lymph nodes harvested, and Figure 6C length of hospital stay were the secondary outcomes of Robotic-Assisted versus Conventional Laparoscopic Colorectal Resection for Colorectal Cancer

Discussion     

Several previous reviews such as Mohammed et al. and Wang et al. have primarily compared the perioperative outcomes such as lower conversion rates in RAS compared to conventional laparoscopic colorectal resection [14,15], but increased operative time for robotic-assisted surgery as reported by Huang et al. [16]. Therefore, we have focused on postoperative outcomes such as urinary and sexual function for a different comparison between the two modalities of surgery while also evaluating selected perioperative and oncological outcomes.

Our review has demonstrated that RAS was associated with improved postoperative urinary function compared to CLS. We have used the IPSS (International Prostate Symptom Score) validated by Barry et al. [17]. for comparison across the studies considered. A higher IPSS score is equivalent to worse function. We have considered Feng et al. a (2022), Feng et al. b (2022) and TengTeng et al. for the analysis [18-20]. There were similar findings in the review conducted by Abdelsamad et al. [21]. This advantage may be attributable to enhanced vision and precise manipulation of robotic-assisted surgery in the complex anatomy and narrow pelvic cavity [20]. The absence of heterogeneity (I²=0%) further strengthens the consistency of this finding across included trials.

Similarly, we have observed significantly higher postoperative IIEF scores in robotic-assisted surgery, which is associated with better male sexual function. This may be attributed to better nerve preservation due to three-dimensional imaging, excellent ergonomics, tremor elimination, ambidextrous capability, motion scaling, and instruments with multiple degrees of freedom as reported by Baik [22]. The absence of heterogeneity among included studies supports this outcome. Additionally, significantly higher postoperative FSFI scores were observed in RAS, which is associated with overall better female sexual function. However, it was associated with high heterogeneity. Subsequently, we conducted a sensitivity analysis to reduce the heterogeneity. The latter may have been due to fewer women being sexually active before and after the surgery. It may also be attributed to the mean estimation from IQR.

Robotic-assisted surgery did not provide a higher lymph node yield than CLS. Adequate lymph node retrieval is important for accurate pathological staging and may influence decisions regarding further treatment. According to O'Boyle et al. patients who had more nodes removed tended toward improved survival, independent of stage [23]. However, the high heterogeneity suggests that the effect was not consistent throughout the studies [18,20,24,25]. The heterogeneity may be attributed to tumour location and neoadjuvant therapy. However, after subsequent sensitivity analysis, heterogeneity was significantly reduced.

We have found that RAS showed lower intraoperative blood loss compared to CLS. Simillis et al. concluded that increased operative blood loss may adversely impact short-term and long-term postoperative outcomes and that measures should be taken to minimize operative blood loss during colorectal cancer surgery [26]. However, a reduction of about 26.70 ml may not necessarily translate into better outcomes. No heterogeneity was found, highlighting consistency between the studies [18,27]. RAS was associated with significantly shorter hospital stay. This may be attributed to the more precise nature of RAS as well as all the above perioperative advantages. The study conducted by Lagoe et al. suggested that hospital length of stay reduction at the community level is an important opportunity for improving health care outcomes and efficiency [28]. However, the high heterogeneity demonstrates inconsistencies throughout the included studies. This may be due to differing institutional discharge criteria or the implementation of ERAS. Our analysis has shown that, on average, robotic-assisted surgery demonstrated first flatus in patients postoperatively by almost a day earlier. Time of first postoperative flatus is usually used to assess bowel function during early postoperative recovery [29]. It is a good marker of bowel function and crucial in any postoperative patient.

Conclusion     

This systematic review and meta-analysis demonstrate that robotic-assisted surgery is superior to conventional laparoscopy. By overcoming the technical constraints of conventional laparoscopy, RAS delivers higher precision in the narrow pelvis which is reflected by the improved preservation of autonomic nerves, resulting in better postoperative urinary and sexual functions. Beyond functional outcomes, RAS achieved higher lymph node yield, reduced blood loss, along with accelerated recovery and shortened length of hospital stays. However, the observed heterogeneity across several parameters emphasises the need for in-depth evaluation through large-scale, high-quality trials.

Competing interests     

The authors declare no conflict of interests.

Authors' contributions     

Conceptualization: Ritisha Conjobeeharry, Anshika Bundhoo, Indrajit Banerjee. Methodology: investigation, data curation, formal analysis and resources: Ritisha Conjobeeharry, Anshika Bundhoo, Jared Robinson. Visualization, writing-original draft, writing-review, and editing: Ritisha Conjobeeharry, Anshika Bundhoo, Indraneel Banerjee, Jared Robinson, Indrajit Banerjee. Supervision: Indrajit Banerjee, Indraneel Banerjee. All the authors have read and approved the final version of this manuscript.

Acknowledgments     

The authors are thankful to Professor Dr. A.P. Singh, SSR Medical College, Mauritius, for his constant encouragement to conduct this research.

Table and figures     

Table 1: data extraction table of robotic-assisted versus conventional laparoscopic colorectal resection for colorectal cancer

Figure 1: Preferred Reporting Items for Systematic Reviews and Meta-Analysis 2020 (PRISMA) flow chart of robotic-assisted versus conventional laparoscopic colorectal resection for colorectal cancer

Figure 2: risk of bias for Randomized Controlled Trials (RCTs) - A) traffic plot, B) weighted bar plots of robotic-assisted versus conventional laparoscopic colorectal resection for colorectal cancer

Figure 3: forest plots for outcomes; A) bladder function; B) male sexual function; C) female sexual function for robotic-assisted versus conventional laparoscopic colorectal resection for colorectal cancer

Figure 4: A) meta-analysis of lymph node harvest; B) blood loss; C) length of hospital stay; D) days to first flatus for robotic-assisted versus conventional laparoscopic colorectal resection for colorectal cancer

Figure 5: GRADE assessment for clinical outcomes of robotic-assisted versus conventional laparoscopic colorectal resection for colorectal cancer

Figure 6: sensitivity analysis: A) female sexual function, B) lymph nodes harvested, and 6.3 length of hospital stay, secondary outcomes of robotic-assisted versus conventional laparoscopic colorectal resection for colorectal cancer with heterogeneity>50%

References