Robot-assisted lateral pelvic lymph node dissection in patients with advanced rectal cancer: a single-center experience of 65 cases

The treatment of lateral pelvic lymph node (LPLN) metastasis of rectal cancer has evolved because of technical difficulties from open surgery to laparoscopy and, recently, robot-assisted surgery. This study aimed to evaluate the technical feasibility and short- and long-term outcomes of robot-assisted LPLN dissection (LPND) following total mesorectal excision (TME) in advanced rectal cancer. Clinical data of 65 patients who underwent robotic-assisted TME with LPND from April 2014 to July 2022 were reviewed. Data regarding operative details, postoperative morbidity (within 90 postoperative days) for short-term outcomes and lateral recurrence as long-term outcomes were analyzed. Among the 65 patients with LPND, preoperative chemoradiotherapy was performed in 49 (75.4%). The mean operative time was 306.8 (range 191–477) min, and the mean time of unilateral LPND was 38.6 (range 16–66) min. LPND was bilaterally performed in 19 (29.2%) patients. The mean number of each side of harvested LPLNs was 6.8. Lymph node metastasis was observed in 15 (23.0%) patients, and 10 (15.4%) patients had postoperative complications. Lymphocele (n = 3) and pelvic abscess (n = 3) were the most common, followed by voiding difficulty, erectile dysfunction, obturator neuropathy, and sciatic neuropathy (all n = 1). During the 25 months of median follow-up, no lateral recurrence of the LPND site was noted. Robot-assisted LPND following TME is safe and feasible and showed acceptable short- and long-term outcomes. Despite some study limitations, we may be able to apply this strategy more widely through subsequent prospective controlled studies.


Introduction
Lateral pelvic lymph node (LPLN) metastasis of rectal cancer is reported to be associated with poorer survival outcomes [1][2][3][4]. LPLN dissection (LPND), a challenging technique with several reports of increased morbidity, has only lately been accepted as a standard treatment in Western medicine. As surgical devices evolve, new approaches are introduced to overcome the difficulties of the procedure, for example, laparoscopy and, more recently, robotics. Especially, robot-assisted surgery is becoming more popular because of its numerous advantages, particularly working in a confined area such as the male pelvic cavity [5]. Other advantages of robot-assisted surgery include a magnified three-dimensional view, fewer tremors, and better dexterity [6,7].
Several studies have analyzed robot-assisted LPND; however, these studies did not aim to evaluate the procedure itself. Studies have compared robot-assisted LPND with laparoscopic LPND but did not precisely assess the complication and treatment, which is directly related to LPND [8][9][10][11][12]. Thus, this study aimed to evaluate the short-and long-term outcomes of robot-assisted LPND following total mesorectal excision (TME) in advanced rectal cancer.

Patients
Medical records of patients who underwent robotic-assisted TME with LPND at Asan Medical Center (Seoul, Korea) from April 2014 to July 2022 were retrospectively reviewed. Patients suspected of having LPLN metastasis were determined by preoperative magnetic resonance imaging (MRI). LPND was decided if the short-axis diameter of the LPLN was > 7 mm with a heterogenic pattern on pretreatment rectal MRI [13] regardless of the response to neoadjuvant treatment. Patients receiving bilateral LPND and those with synchronous colorectal cancers were included in the study. For the integrity of the study, patients who had the LPLN sampling instead of a full dissection and cases of palliative resection were excluded. Patients with coexisting malignancy, hereditary colorectal cancer, emergency operation, history of other malignancy or pelvic radiotherapy before rectal cancer diagnosis, and those who did not undergo preassessment MRI were excluded. The study protocol was approved by institutional review board of Asan Medical Center (IRB no. 2022-1264).
Preoperative evaluation of rectal cancer included digital rectal examination, endoscopy with histologic confirmation, chest computed tomography (CT), abdominal CT, and rectal MRI. Before any treatment, patients were assessed for clinical T and N stages. Neoadjuvant chemotherapy was recommended for patients with clinical stage II or stage III low rectal cancer according to the TNM stage based on the American Joint Committee on Cancer (AJCC) classification, 8th edition. Patients on neoadjuvant radiotherapy received a total dose of 50.4 Gy, with 25 fractions five times a week for 5 weeks in the pelvis and a boost dose of 5.4 Gy of primary tumor irradiation over 3 days. Concurrent chemotherapy was composed of either two cycles of intravenous injection for 3 days of 5-fluorouracil (5-FU 375 mg/m 2 /day) and leucovorin (20 mg/ m 2 /day) (FL) in the first and fifth weeks of radiation or oral intake of capecitabine (825 mg/ m 2 ) twice a day during radiation therapy. Intravenous injection of oxaliplatin was combined for some patients by the discretion of the oncologist on the first day of radiotherapy (85 mg/m 2 for patients receiving FL; 130 mg/m 2 for patients taking capecitabine orally). Surgical resection of the primary tumor and LPND were conducted 6-8 weeks after neoadjuvant chemoradiotherapy. All medically fit patients had adjuvant chemotherapy within 4 weeks after surgery. Adjuvant chemotherapy consisted of four cycles of FL or six cycles of capecitabine.

Surgical technique
All procedures were performed by three surgeons (YS Yoon, IJ Park, and JC Kim) with sufficient experience for open and laparoscopic LPND. Bilateral LPND was performed only in patients with suspicious bilateral LPLN metastasis.
The patients were placed in the Lloyd-Davies position with right-lateral tilt (right side lowered). A standard four-port incision was made for the insertion of the 8-mm trocars with an addition of a 12-mm trocar for assistant access (Fig. 1). LPND was performed after pelvic TME, and the distal end of the rectum was transected. For leftsided LPNDs, the docking arms remained the same as with the rectal dissection phase. For right-sided LPNDs, the arms and the camera port shifted toward the left side, with the third arm utilizing the 12-mm assistant port for better exposure and manipulation (Fig. 1b).
LPND was performed in two sections: the obturator lymph nodes and the internal iliac lymph nodes (Fig. 2). Key anatomical landmarks were identified, dissecting along the planes of the lateral pelvic compartment, and the autonomic nerves were preserved. The bifurcation of the internal and external iliac arteries was first dissected, later forming the proximal boundary of the internal iliac lymph node dissection plane. Then, the obturator lymph nodes were dissected from the lateral side of the internal iliac vessels and further along the obturator foramen and Alcock's canal distally, external iliac vessels and obturator internus laterally, perivesical fat anteriorly, and sacral nerve and piriformis muscle posteriorly (Fig. 3). The obturator nerve was preserved, and the obturator artery and vein were mainly preserved unless there is suspicion of metastatic lymph nodes in proximity. Finally, internal iliac lymph node dissection was performed between the medial side of the internal iliac artery and ureterohypogastric fascia; the latter of which forms one plane including the hypogastric nerve, pelvic splanchnic nerves, pelvic plexus, and ureter.

Data collection and statistical analysis
Data included patient demographics including sex, age, body mass index (BMI); tumor characteristics including tumor height (distance from the anal verge), clinical T and N stages; and treatment (neoadjuvant therapy and adjuvant therapy). Operative details were collected including the operation type, unilateral/bilateral LPND, ileostomy/ colostomy formation, length of the procedure, and estimated blood loss. Pathologic results included TNM stages based on the AJCC classification, 8th edition, number of harvested lymph nodes, and positive lateral pelvic and mesorectal lymph nodes. Short-term outcomes included the length of hospital stay, postoperative morbidity with regard to LPND, treatment and recovery of morbidity, and lateral recurrence. Postoperative morbidities were defined as morbidities occurring within the first 90 postoperative days, such as lymphocele, pelvic abscess, voiding difficulty, erectile dysfunction, obturator neuropathy, and sciatic neuropathy. LPND-related morbidities were collected, and ileus or other complications that were more related to anastomosis or the main rectal procedure were not analyzed. Lateral recurrence, as a long-term outcome, was defined as recurrence in the LPND surgical site only.
Owing to the low number of cases and the absence of direct comparison between groups, we provided descriptive results in the form of number and proportions (%) and/or standard deviation for numerical variables and median and range for non-numerical variables. Data analysis was performed using Microsoft Excel (Microsoft Inc., Redmond, WA, USA).

Results
We analyzed 65 consecutive patients within the study period. The patients were predominantly male, with a mean age of 58 years, a mean BMI of 24.2 kg/m 2 , and a mean tumor distance from the anal verge of 4 cm. Moreover, 48 (73.8%) patients received neoadjuvant chemoradiation therapy before surgery (Table 1).  The operative details of patients who underwent roboticassisted TME with LPND are shown in Table 2. Most of the patients had ultra-low anterior resection (LAR) (76.9%), and 19 (29.2%) patients had bilateral LPND. The mean operative procedure time was 306.8 min, and the mean unilateral LPND time was 38.6 min. The mean number of harvested lymph nodes per unilateral LPLN was 6.8, and 15 (23.0%) patients had positive metastasis in LPLNs. No patients had positive circumferential resection margin.
Overall, morbidities were observed in 10 (15.4%) patients ( Table 3). The most common complication was lymphocele (n = 3) and pelvic abscess (n = 3). All three patients who had lymphocele were treated with antibiotics orally. All three patients who had pelvic abscesses were treated with percutaneous drainage and antibiotics given intravenously. One patient had voiding difficulty, who recovered within 6 months with oral medication. One patient suffered from erectile dysfunction and recovered without medication. One patient who had obturator neuropathy caused by obturator nerve injury event during the operation was treated with medication and recovered within 6 months of follow-up. A patient with sciatic neuropathy did not recover until his most recent 6-month follow-up despite treatment with medication and rehabilitation. No mortality occurred within 90 postoperative days. During the mean follow-up of 25 months, no recurrence on the LPND site was noted.

Discussion
In this study, we conducted 65 cases of full dissection of LPLNs with TME of rectal cancer using the Da-Vinci Si and Xi platforms and found favorable safe and feasible outcomes.
The robotic platform has potential benefits, allowing for a fine dissection through magnification and three-dimensional views and articulated arm movement in a narrow operation space [6,7,[14][15][16]. This has consequently increased the number of robot-assisted rectal cancer surgery worldwide [17]. We believe that with the increasing adaptation of robotic surgery and the accumulation of surgical techniques, the number of patients undergoing surgery for advanced low rectal cancer with LPLN metastasis will be increasing.
Our data were comparable to the literature in three main aspects. The first aspect is the procedure time. Procedure   Fig. 3 a Left-sided and b right-sided lateral pelvic lymph node dissection using robot-assisted surgery time may have a potential effect on short-come complications [18]. Our mean operation procedure time of 306.8 min is comparable to the mean of 476 min (320-683) min reported by Kagawa et al. [9], a median of 420 (313-530) min reported by Peacock et al. [11], and a mean of 260 min reported by Choi et al. [5]. The unilateral LPND time was 38.6 min, which is comparable with 41.0 min in the report by Choi et al. This study also reported laparoscopic LPND time of 35 min. The second aspect is the number of retrieved LPLNs. Better oncologic results strongly correlated with the number of retrieved LPLNs [12]. Our study reported an average of 6.8 (1-24) lymph nodes, and 15 (23.0%) patients were positive for metastasis, which is comparable to results from studies of patients undergoing open and laparoscopic surgery [5,11,19,20]. In other reports, the median number of retrieved LPLNs ranged from 4 to 10, and the positivity of the lymph nodes ranged from 10.5% to 37%, which was comparable to our results.
The final aspect refers to the postoperative morbidity within 90 days. Our value of 15.4% was comparable to 10%-22.9% in other reports [5,9,11] although the definition of complication in these reports included those arousing from mesorectal excision. In this study, we focused on the complications directly related to LPND, such as urogenital dysfunction, lymphocele, pelvic abscess, and neuropathies, found within 90 days after the operation. Lymphocele was the most common postoperative morbidity, followed by pelvic abscesses. However, the occurrence rate of lymphocele is quite difficult to compare with those in other studies because only a few studies have focused on lymphocele in low rectal cancer [21]. Considering that the occurrence rate of infected lymphocele after lymphadenectomy for gynecologic malignancy was 4.4% [22], the occurrence rate of lymphocele and pelvic abscess in this study is thought to be comparable. Generally, asymptomatic lymphoceles can be followed without special treatment, but if they are accompanied by symptoms or infections, they may need to be percutaneously drained or surgically unroofed so they can be discharged into the abdominal cavity. To prevent lymphatic fluid leakage, it is recommended to avoid blunt dissection of lymphatic vessels as much as possible, peel lymphatic vessels off gently and ligate them with a sealing device.
One technical tip that we utilized to shorten the operation time and assure clear and safe dissection of LPLNs was the shift of docking ports when performing rightsided LPND. Several docking methods of robotic LPND are reported in the literature, that is, some methods were similar to laparoscopic approaches [14], and other methods were based on placements of the robot arm [9]. In the case of the docking method of our institute for right-sided LPND, the camera arm is shifted to the third port to have two left arms. This may provide a mirror-like image to the  This study has some limitations. First, selection bias may exist as this study was held retrospectively. To mitigate this, we included all consecutive patients undergoing robotic LPND during the study period. Second, our data was not validated from laparoscopic or open LPND. Although our center also performs open and laparoscopic LPND, a systemic registry is lacking, and there is heterogeneity in the treatment and results. Thus, the only way to validate our study was to compare it with the literature. Finally, complications such as urogenital dysfunction may be underreported because the outcomes were not actively surveyed and were collected from electronic medical record data.

Conclusion
Robot-assisted LPND following TME is safe and feasible and showed acceptable short-and long-term outcomes. Despite some study limitations, we may be able to apply this strategy more widely through subsequent prospective controlled studies.
Author contributions All authors contributed to the study design and analysis. Material preparation was performed by Eon Bin Kim, Yong Sik Yoon, In Ja Park, Jin Cheon Kim. Data collection, analysis and the first draft of the manuscript was written by Eon Bin Kim and all authors revised critically and commented on the previous version of the manuscript. All authors participated and approved the final manuscript.
Funding No funding was received to assist in the preparation of this manuscript.

Data availability statement
The dataset are available from corresponding author, Yong Sik Yoon, upon reasonable request.

Conflict of interest
The authors declare no competing interests.

Ethical approval
The study protocol was approved by the Institutional Review Board in accordance with the 1964 Helsinki Declaration. The study protocol was approved by institutional review board of Asan Medical Center (Date 18 Sep 2022/IRB no. 2022-1264).

Informed consent
The review board waived the requirements for informed consent, as this study was a retrospective analysis.