Conventional Laparoscopy versus Robotic-Assisted Aortic Lymph-Nodal Staging for Locally Advanced Cervical Cancer: A Systematic Review and Meta-Analysis

Aortic lymph node metastases are a relative common finding in locally advanced cervical cancer. Minimally invasive surgery is the preferred approach to perform para-aortic lymph nodal staging to reduce complications, hospital stay, and the time to primary treatment. This meta-analysis (CRD42022335095) aimed to compare the surgical outcomes of the two most advanced approaches for the aortic staging procedure: conventional laparoscopy (CL) versus robotic-assisted laparoscopy (RAL). The meta-analysis was conducted according to the PRISMA guideline. The search string included the following keywords: “Laparoscopy” (MeSH Unique ID: D010535), “Robotic Surgical Procedures” (MeSH Unique ID: D065287), “Lymph Node Excision” (MeSH Unique ID: D008197) and “Aorta” (MeSH Unique ID: D001011), and “Uterine Cervical Neoplasms” (MeSH Unique ID: D002583). A total of 1324 patients were included in the analysis. Overall, 1200 patients were included in the CL group and 124 patients in the RAL group. Estimated blood loss was significantly higher in CL compared with RAL (p = 0.02), whereas hospital stay was longer in RAL compared with CL (p = 0.02). We did not find significant difference for all the other parameters, including operative time, intra- and postoperative complication rate, and number of lymph nodes excised. Based on our data analysis, both CL and RAL are valid options for para-aortic staging lymphadenectomy in locally advanced cervical cancer.


Introduction
Cervical cancer is one of the most common malignancies and is the most frequent cause of death from gynecological cancers worldwide [1]. Approximately more than onethird of patients present with locally advanced cervical cancer (LACC) at diagnosis, FIGO (International Federation of Gynecology and Obstetrics) stage IIB to IVA. This presentation is associated with 18-50% of lymph node metastases [2]. Para-aortic lymph-node status represents the most important prognostic factor in patients with LACC, with a severe negative impact on survival [3,4]. The recent FIGO 2018 classification of cervical cancer has included lymph node disease in the staging system to improve treatment allocation and inform about prognosis.
The current standard treatment in this population is concomitant chemoradiation, while the lymph node status assessment has an important staging role. The detection of para-aortic lymph node involvement is crucial to define the extent of the irradiation field and to personalize specific treatment protocols [5]. Para-aortic lymph nodal staging can be assessed using imaging, but the detection of lymph node metastases remains unsatisfactory. Even advanced imaging techniques, such as positron emission tomography-computed tomography (PET-CT), include a risk of 15% of false negative and 5-10% of false positive [6]. Computerized axial tomography (CT) and nuclear magnetic resonance (MRI) have shown lower sensitivity and specificity [7]. In this context, laparoscopic staging has been proposed as a valid alternative to the radiological nodal assessment, achieving more accurate results [8]. On the one hand, the survival benefit of surgical staging is still controversial due to conflicting results in available studies. On the other hand, the minimally invasive approach is known to improve peri-and postoperative outcomes, avoiding delay in the primary treatment. In a recent meta-analysis conducted by our group [9], we compared the two laparoscopic techniques for para-aortic surgical staging in cervical cancer: the transperitoneal laparoscopic lymphadenectomy (TLL) versus the extraperitoneal laparoscopic lymphadenectomy (ELL). Our results showed that TLL approach was associated with a higher rate of intraoperative complications, while no significant difference was found between the two techniques regarding postoperative complications. Since the introduction of the robotic surgery for gynecological oncological procedures, a significant increase of this approach has been reported following the hypothesis that the robotic route may represent an improvement compared with laparoscopic technique. Recently, the robotic approach has also been introduced for para-aortic staging in LACC patients, showing encouraging results [10].
In this scenario, the aim of our systematic review and meta-analysis was to compare the peri-and postoperative outcomes of conventional laparoscopy (CL; including either TLL and ELL) versus robotic-assisted laparoscopy (RAL) for surgical nodal assessment in LACC patients.

Materials and Methods
This systematic review was submitted in the PROSPERO international database (CRD42022335095) and conducted according to the PRISMA guideline (Preferred Reporting Items for Systematic Reviews and Meta-Analyses) [11]. We screened PubMed, Scopus, and Web of Science search engines from inception to March 2022. The search string included the following Medical Subject Headings (MeSH): "Laparoscopy" (MeSH Unique ID: D010535), "Robotic Surgical Procedures" (MeSH Unique ID: D065287), "Lymph Node Excision" (MeSH Unique ID: D008197) and "Aorta" (MeSH Unique ID: D001011), and "Uterine Cervical Neoplasms" (MeSH Unique ID: D002583).
Titles and/or abstracts of studies retrieved using the search strategy, and those from additional sources were screened independently by three review authors (M.C.D.D., V.G., G.L.B.) to identify studies that potentially met the aims of this systematic review. We excluded articles in languages other than English and reviews. The full text of these potentially eligible articles was retrieved and independently assessed for eligibility by other two review team members (G.S., A.A.). Any disagreement between them over the eligibility of particular articles was resolved through discussion with a third (external) collaborator. The references of the included studies were carefully evaluated to identify any potential additional source. A final review of the included articles was performed by the review supervisor (V.C.). In case of redundant articles or data used in previous studies, only the most recent articles with more comprehensive data were included in the analysis.
Prospective, retrospective, and pilot studies reporting surgical outcomes in women undergoing CL or RAL aortic lymph node dissection for surgical staging of LACC were con-sidered and included. In case of studies reporting direct comparison between CL and RAL, each arm was separately included in the pooled analysis. Bibliographical and technical data extracted from the articles using a pre-piloted standardized form to collect the following elements: authors, publication year, the type of surgery, FIGO stage, median age and Body Mass Index (BMI), operative time (OT), estimated blood loss (EBL), hospitalization time (HT), intra-and postoperative complications, conversion to another technique, and the number of lymph nodes retrieved. The common terminology criteria for adverse events (CTCAE) grade >3 were considered for complications [12].

Statistical Analysis
Data were expressed as standard deviation (SD) or as number (percentage). Categorical variables were compared using the chi-square or Fisher exact test. Between-group comparison of continuous variables was undertaken using the t-test and the Mann-Whitney nonparametric equivalent test. Two-sided p-values were calculated, and p-values < 0.05 were considered as statistically significant. Meta-analyses of proportions were used to combine data. Between-study heterogeneity was explored using the I 2 statistic, which indicates the percentage of between-study variation that is due to heterogeneity rather than chance. A value of I 2 of 0% indicates no observed heterogeneity, whereas values > 50% indicate a substantial level of heterogeneity. Given the small sample size of the included studies, a random effect model was preferred regardless of I 2 . StatsDirect 3.0.17 (StatsDirect Ltd., Altrincham, UK) statistical software was used for all data analyses.
Three studies described the two different laparoscopic approaches, TLL and ELL [14,19,26]. One study included robotic trans-and extra-peritoneal approach [29]. Furthermore, all studies were retrospective by design except one prospective randomized trial [18] and one prospective observational preliminary study [30]. The characteristics of the studies are showed in Appendix A; inclusion and exclusion criteria for each study are reported in Appendix B. The PRISMA flow chart is shown in Figure 1.
A total of 1324 patients were included in the analysis. Of these, 1200 patients were included in the CL group and 124 patients in the RAL group. The median age was 49.8 for CL and 51 for RAL. The median BMI was 25.5 and 25 for CL and RAL, respectively. The median of the OT was 129 min for patients who underwent CL aortic lymph node staging and 121.7 min for RAL aortic lymph node staging procedure. The median EBL was 81.1 mL and 26.9 mL in CL and RAL, respectively. The median length of HT was 1.9 and 3.3 days for CL and RAL, respectively. The median number of lymph nodes excised was 12.7 in the CL group and 15.7 in the RAL group. No significant differences were found between groups for BMI (p = 0.33), number of lymph node excised (p = 0.38), age (p = 0.62), and intraoperative time (p = 0.8). Conversely, EBL was significantly higher (p = 0.02) and HT significantly lower (p = 0.02) in the CL group compared with RAL group (Table 1) A total of 1324 patients were included in the analysis. Of these, 1200 patients were included in the CL group and 124 patients in the RAL group. The median age was 49.8 for CL and 51 for RAL. The median BMI was 25.5 and 25 for CL and RAL, respectively. The median of the OT was 129 min for patients who underwent CL aortic lymph node staging and 121.7 min for RAL aortic lymph node staging procedure. The median EBL was 81.1 mL and 26.9 mL in CL and RAL, respectively. The median length of HT was 1.9 and 3.3 days for CL and RAL, respectively. The median number of lymph nodes excised was 12.7 in the CL group and 15.7 in the RAL group. No significant differences were found between groups for BMI (p = 0.33), number of lymph node excised (p = 0.38), age (p = 0.62), and intraoperative time (p = 0.8). Conversely, EBL was significantly higher (p = 0.02) and HT significantly lower (p = 0.02) in the CL group compared with RAL group (Table 1).   Intraoperative complications were reported in 23 patients (2%) in the CL group and in 4 patients (3.2%) in the RAL group. The most frequent intraoperative complications were vascular and urinary ones. The type of intraoperative complications occurred in the two groups are shown in Table 2. As shown in Figure 2, the intraoperative complications pooled proportion is 4.1% (I 2 = 0%) for the RAL and 1.5% (I 2 = 47.7%) for CL. Intraoperative complications were reported in 23 patients (2%) in the CL group and in 4 patients (3.2%) in the RAL group. The most frequent intraoperative complications were vascular and urinary ones. The type of intraoperative complications occurred in the two groups are shown in Table 2. As shown in Figure 2, the intraoperative complications pooled proportion is 4.1% (I 2 = 0%) for the RAL and 1.5% (I 2 = 47.7%) for CL. Table 2. Analysis of intraoperative complications in conventional laparoscopy and robotic groups.

Type of Intraoperative Complication Laparoscopic (n = 1200) Robotic (n = 124) p
Vascular injuries In total, 104 patients (8.6%) of the CL group developed postoperative complications, while 13 patients (9.7%) of the RAL group reported a complication after surgery. Postoperative complications are reported in Table 3. In the two groups, sixty-three lymphatic complications occurred, ten to urinary compartment, three trocar site hernia, and one bowel complication. The postoperative complications pooled proportion is 11.1% (I 2 = 0%) for the RAL group and 7.7% (I 2 = 43.9%) for CL. Pooled proportion of post-operative complications among the two groups are shown in Figure 3. In 14 (1.2%) cases of the CL group and in 2 (1.6%) of the RAL group, a conversion to laparotomy was required ( Figure 4). In total, 104 patients (8.6%) of the CL group developed postoperative complications, while 13 patients (9.7%) of the RAL group reported a complication after surgery. Postoperative complications are reported in Table 3. In the two groups, sixty-three lymphatic complications occurred, ten to urinary compartment, three trocar site hernia, and one bowel complication. The postoperative complications pooled proportion is 11.1% (I 2 = 0%) for the RAL group and 7.7% (I 2 = 43.9%) for CL. Pooled proportion of post-operative complications among the two groups are shown in Figure 3. In 14 (1.2%) cases of the CL group and in 2 (1.6%) of the RAL group, a conversion to laparotomy was required ( Figure 4).

Discussion
Aortic lymph-node metastases are common findings in LACC patients, reaching a rate of about 40-70% for stages III and IV, and the diagnostic value of surgical para-aortic lymph node dissection has been widely investigated [32]. Although the increased morbid-

Discussion
Aortic lymph-node metastases are common findings in LACC patients, reaching a rate of about 40-70% for stages III and IV, and the diagnostic value of surgical para-aortic lymph node dissection has been widely investigated [32]. Although the increased morbidity is associated with surgical procedures, minimally invasive surgery (MIS) for surgical staging in cervical cancer provides several advantages without compromising the course of the disease [3]. Reduction in peri-and postoperative complications and reduction of HT represent the main benefits of this surgical approach. Moreover, MIS allows to perform para-aortic lymphadenectomy even in the difficult case of abdominal vascular and urinary anomalies [33].
Considering the only staging intent, in order to avoid delay of the primary chemoradiation, it is essential to propose the most effective and technologically advanced surgical approach. Laparoscopic aortic staging is one of the most challenging procedures in gynecologic oncology: in this scenario, RAL can have advantages over CL with faster learning curve, technical improvements such as a 3D imaging, elimination of physiological tremor, and increased precision due to the seven-degree instrument's articulation [34]. Accumulating evidence suggests the overall feasibility of robotic para-aortic lymph node staging for cervical cancer [10,27]; in addition, some authors have compared the RAL surgical staging to the CL and have found better perioperative outcomes and similar survival outcomes [10,27]. Our meta-analysis showed that RAL surgical staging in LACC patients is significantly associated with less EBL than conventional laparoscopy. Furthermore, OT and the number of lymph nodes excised are in favor of robotic approach although the difference is not statistically significant. Similar to previous series, our data analysis supports RAL as an appropriate alternative to CL for para-aortic lymph node dissection in LACC patients.
Although not significant, the higher number of lymph nodes excised using RAL compared with CL could be due, at least in part, to the greater precision of the robotic procedure and the possibility of being more radical with fine dissection in difficult anatomical spaces. Interestingly, we found shorter HT for the CL group compared with RAL: although we cannot explain with absolute certainty this data, this is probably due to the type of women who are addressed to RAL; indeed, patients undergoing robotic approach usually have more comorbidities (e.g., obese, more prior abdominal surgery, higher ASA score) [10], and this may be associated with longer HT.
Lymphatic complications represent the most frequent postoperative complication, especially for the robotic group. One possible and reliable explanation of this difference is related to using different instruments in the two surgical routes. During a laparoscopic lymphadenectomy, multifunction instruments seal and cut the lymphatic vessels, unlike the robotic approach, in which bipolar energy and scissors are commonly used. Moreover, some authors suggest systematically clipping any large lymphatic vessel to avoid lymphatic complications [3,16]. Vascular injuries and ureteral lesions were the most frequent intraoperative complications, especially in the robotic group. On the one hand, some authors suggested that during RAL, pneumoperitoneal pressure is lower than laparoscopy, and very often, the patients selected for RAL are already subjected to previous surgery, obese, and with comorbidities [35,36]: this challenging surgical scenario may limit the visualization of the ureter and the great vessels, increasing intraoperative complications [34,35]. On the other hand, the RAL approach allows a better exposition of the pre-cava and inter-aortocaval field than the CL. Indeed, robotic surgery is associated with a higher Trendelenburg inclination and a better range of motion of the instruments [10,27]. In addition, the learning curve for laparoscopic aortic lymphadenectomy [37] is longer than robotic procedure [38] due to limited rigid instruments and the 2-dimensional view of the laparoscope's video camera, which requires greater surgeon skills to perform this procedure. Furthermore, aortic lymphadenectomy is a single-quadrant surgery and perfectly matches with the robotic approach, using its increasing precision that would otherwise suffer in case of re-docking [39]. However, the selection of the patients for lymph node staging surgery should be careful, considering that the benefits of diagnostic surgery should justify its possible morbidities.
Limits of the present meta-analysis are represented by the small sample size of robotic group, the heterogenicity of the studies included, the retrospective nature of most of the articles analyzed. In addition, most of the included studies reported insufficient information on the pre-operative characteristics of the patients, which did not allow us to perform a robust sub-analysis (meta-regression) of surgical outcomes based on these parameters. Finally, only one study reported either the transperitoneal or the extra peritoneal technique for the robotic aortic lymph node dissection. However, good heterogenicity of the studies is showed by the pooled analysis.

Conclusions
Based on our data analysis, both CL and RAL can be considered valid options for para-aortic staging lymphadenectomy in women with LACC, with comparable safety and surgical outcomes. In particular, the two techniques allowed similar operative time, intraand postoperative complication rate, and number of lymph nodes excised.

Conflicts of Interest:
The authors declare no conflict of interest.    -Patients with cervical carcinoma with clinical FIGO stage IB2-IIIB who underwent a laparoscopic retroperitoneal para-aortic lymphadenectomy as staging procedure -Coagulation disorders -Presence of metastatic para-aortic lymph nodes on PET/PET-CT and/or CT (preoperative presence of metastatic nodes was defined as para-aortic lymph nodes larger than 1 cm with uptake of contrast on CT and/or PET scan)