Is It Necessary to Endoscopically Evaluate the Anastomosis in Robotic or Laparoscopic Surgical Procedures for Colorectal Cancer?

Introduction

According to Globocan 2022, colorectal cancers are the third most common cancer in the world and the third most common cause of cancer-related mortality. ¹ Treatment of colon cancer consists of surgical and oncologic treatment steps. ² The most important complication in the early postoperative period is anastomotic leakage. In colorectal surgery, anastomotic leakage is observed with a rate of 2.6%–12.3%, and mortality and morbidity rates after leakage increase considerably.^3,4 As a result, oncologic treatment of patients with anastomotic leakage is delayed, and overall survival (OS) and disease-free survival (DFS) are negatively affected due to increased morbidity. ⁵

The causes of anastomotic leakage are diverse and include patient-related factors, tumor characteristics, surgical technique, type of anastomosis, number and type of staples used in the anastomosis, but tension-free anastomosis and adequate blood supply are the most important factors in preventing leakage. ⁶ Therefore, to minimize the risk of leakage, intravenously injected indocyanine green (ICG), which shows that the blood supply of the anastomosis line is sufficient, is frequently preferred during surgery. In addition, the intraoperative air leak test (IALT) or methylene blue injection rectally are the most commonly used methods to assess anastomotic integrity. Intraoperative colonoscopy (IOC), in contrast, is increasingly preferred by surgeons because it provides the opportunity to detect bleeding that may be seen in the anastomosis and to detect possible inadequate anastomotic integrity.^7–10

In this study, we aimed to compare the air test, which is frequently preferred in the anastomotic evaluation of minimally invasive (robotic and laparoscopic) anterior resection patients with sigmoid and rectosigmoid junction cancer, with the increasingly used colonoscopic evaluation.

Materials and Methods

This study was conducted with Antalya Training and Research Hospital Ethics Committee number 2024/077. Between January 2018 and February 2024, preoperative and postoperative data of patients who underwent minimally invasive (robotic and laparoscopic) anterior resection for sigmoid and rectosigmoid junction cancer in the Gastroenterology Surgery Clinic of Antalya Training and Research Hospital were retrospectively evaluated.

Patients who were metastatic at the time of diagnosis and started systemic treatment, patients who were operated on for urgent reasons such as obstruction or bleeding, and patients who had an ileostomy and colostomy in the first surgery were excluded from the study. Demographic data (age, gender, body mass index [BMI], American Society of Anaesthesiologists [ASA] score, comorbid diseases, surgical data (robotic/laparoscopic, number of linear staples used and size of circular staples, intraoperative ICG use, intraoperative leak test type, and leak management) were recorded retrospectively.

All operations were performed by 4 surgeons experienced in colorectal surgery. In all surgical procedures, the inferior mesenteric artery was ligated at a distance of 1 cm from the aorta, and the inferior mesenteric vein was ligated at the level of the pancreatic corpus. The splenic flexure was fully mobilized in all cases. The ICG test was performed in all patients except for the period of dye supply problems, and the transection line was made at least 10 cm proximal to the colon in an area with adequate blood supply with ICG. In patients in whom ICG could not be used, the transection line was made at a minimum distance of 10 cm from the tumor and proximal to the IMV. The inferior resection line was performed at the level of the pelvic peritoneum in sigmoid colon tumors and 5 cm distal to the tumor in upper rectum tumors using the ECHELON FLEX™ Articulating Endoscopic Linear Cutter ECR60G—Reload Green 60 mm cartridge. The incision was removed from the Pfannenstiel incision in all patients after preventing wound contamination with a protective dressing, and the same brand of circular stapler but with different diameters preferred according to the colon diameter size was used.

No diversion ostomy was opened in any of the patients. From January 2018 to June 2021, IALT was used for anastomosis control, while anastomotic leakage test was performed with IOC since July 2021, when an endovision system was provided in the operating room. After the colorectal anastomosis was completed, the pfannenstiel incision was completely closed, and the laparoscopic procedure was continued. Approximately 10 cm proximal to the anastomosis, the colon was closed with a wide-tipped laparoscopic handpiece. The pelvic anastomosis was filled with isotonic to cover the anastomosis, and if IALT was to be applied, air was given until sufficient tension was reached in the colon with the help of 50 mL catheter tip syringes placed from the anal canal to the anastomosis neighborhood. In the colonoscopic evaluation, the anastomosis was evaluated with a colonoscope after the colon was closed proximal to the anastomosis, and isotonic was given into the pelvis. In both groups, the presence of air bubbles from the anastomosis was accepted as intraoperative leakage, and primary repair was performed with laparoscopic 3.0 absorbable suture. Reanastomosis was performed if the anastomotic separation was too large to be closed with primary suturing (PS). If the leak area could not be repaired with PS, a Hartman colostomy procedure was performed. In all patients, an 18 F urinary catheter was inserted rectally proximal to the anastomosis to reduce intracolonic pressure and withdrawn on postoperative day 3.

In postoperative follow-up, rectal and IV optic imaging were performed in patients with suspected leakage (patients with suspected intestinal leakage from the drain, elevated blood infection parameters), and leakage was detected. Colonoscopy was performed in patients in whom leakage could not be detected on this examination. In patients in whom leakage was detected, conservative follow-up was performed if the leakage could be controlled with a drain and did not cause an abscess in the abdomen and did not cause deterioration in the clinical follow-up of the patient. Patients who could not be followed conservatively were operated, and treatment was planned.

Patients were divided into two groups according to the intraoperative leak test method, and patients who underwent IALT were Group 1, and patients who underwent IOC test were Group 2. Both groups were compared in terms of demographic and operation-related data.

Results

Between 2018 and 2024, 273 patients with sigmoid colon or rectosigmoid junction tumors with a median age of 64.2 years underwent minimally invasive (laparoscopic or robotic) surgery. Anastomotic leakage testing was performed with IALT in 148 patients and IOC in 125 patients. These two groups were similar in terms of age, gender, BMI, and comorbidities. Demographic data of the patients are given in Table 1.

Table 2 compares the data related to the surgical procedure. The DaVinci robotic surgical system was used in 55.4% of cases in the IALT group and 58.4% in the IOC group, while laparoscopic surgical procedures were performed in the remaining patients. The number of linear staplers used for rectal transection was similar in both groups, and two staplers were used in 68% of cases. In 19 patients (6.9%), three or more staplers were used, and there was no statistical difference between the groups. In 225 patients (82.4%), ICG was used to evaluate the blood supply of the rectal stump, and the colon was planned for anastomosis after rectal transection. Circular staplers with diameters of 29 mm, 31 mm, or 33 mm were preferred for colorectal anastomosis, and the utilization rates were 14.2%, 66.9%, and 18.9% in the IALT group and 6.4%, 76.8%, and 16.8% in the IOC group, respectively.

Leakage was detected in 7 (4.7%) patients in the IALT group and 14 (11.2%) in the IOC group (P = .06). In the first group, 5 of 7 patients were repaired with PS, and in 2 patients the anastomosis was reconstructed. In the second group, 10 of 14 patients were repaired with PS, 2 patients underwent reanastomosis, and 2 patients needed colostomy. The postoperative anastomotic leakage rate was 5.9% in the whole patient group and developed in 10 patients in the first group and in 5 patients in the second group (after removal of 2 patients with colostomy), respectively. Intraoperative leakage occurred in 4 of these 15 patients with postoperative leakage (2 patients from the IALT group and 2 patients from the IOC group), and all of them were repaired with PS. Of the remaining 11 patients, 3 were in the IOC group, and 8 were in the IALT group.

Intensive care unit (ICU) hospitalization and subsequent ward length of stay were longer in the IALT group than in the IOC group (ICU length of stay 2.07 versus 0.8 days; P = .049, ward length of stay 7.44 versus 6.1 days; P = .008, respectively).

After excluding the 4 patients with both intraoperative and postoperative leakage, the demographic data, including age, BMI, gender, and comorbidities, as well as the method of surgery, number of linear staplers, and number of circular staplers used in the operation, were similar when these two groups of patients were compared (Table 3). The ICG utilization rate was 88.2% in the group with intraoperative leakage and 26.7% in the group with postoperative leakage, which was statistically significantly lower (P = .001). In addition, it was statistically determined that colonoscopy was used as a leak test at a higher rate (70.6%) in patients with intraoperative leakage, while this rate was very low (33.3%) in the postoperative leakage group (P = .035).

Discussion

Minimizing complications in colorectal cancers can increase OS and DFS.^11,12 In surgical treatment of colon cancer, anastomotic leaks are the most important complication affecting mortality and morbidity. In addition, many factors related to the patient and surgical technique are responsible for anastomotic leakage. In order to minimize anastomotic leaks that may occur due to surgical technique, surgeons use a number of test methods to evaluate colonic perfusion during surgery and to detect anastomotic leaks. ¹³ IALT and IOC are the most commonly used anastomotic evaluation tests. These tests provide early detection of anastomotic leaks and provide the advantage of preventing serious postoperative complications. ¹⁴ In our study, we found that in patients with sigmoid colon and rectosigmoid junction tumors operated with minimally invasive methods (robotic or laparoscopic), the leak test performed with IOC detected more leaks than IALT (P = .06). We concluded that with a limited number of patients, the leakage rate with PS could be reduced to 20%, but if there is a serious doubt in the anastomosis, this risk can be reduced to 0% with reanastomosis and/or diversion ostomy (RA±DO).

There is no consensus among surgeons about the repair of the leak detected during surgery. The main techniques used are repair of the anastomosis with PS and addition of RA±DO. ¹⁵ Especially detection of minimal leaks may provide surgeons with the opportunity to repair with PS during surgery. However, the question of whether the risk of leakage decreases with repair should be answered. In the review by Wu et al. it was argued that intraoperative leak tests did not affect the postoperative leak results. ¹⁶ In a retrospective cohort study by JB Mitchem et al. in a sample of 2360 patients, leakage was detected with IALT in 119 patients, and 68 patients underwent repair with PS only. In addition, 51 patients underwent RA±DO. The clinically significant leak rate was 9% (6/68; 95% CI: 2%–15%) in the PS-only repair group and 0% (0/51) in the RA±DO group, respectively. In this study, repair with PS alone was less effective than RA±DO. ¹⁷ Valdes-Hernandez et al. showed that intraoperative intervention of the leak detected after intraoperative leak test with transanal repair of the detected leak completely prevented the leak and decreased the risk of leak. ¹⁸ In our study, 11.2% (n: 14) in the IOC group and 7% (n: 5) in the IALT group were detected intraoperatively. In our study, 71.4% (n: 5) of the patients with IOC leakage and 71.4% (n: 10) of the patients with IALT leakage were controlled with PS repair. Postoperative leakage was detected in 20% (n: 2) of IOC patients and 40% (n: 2) of IALT group patients who underwent PS repair. However, no leak was detected in any of the 4 patients who underwent reanastomosis (0%). This shows that leak management with intraoperative PS repair is inadequate, and these leaks should be intervened with more satisfactory methods, as in the study results of Li et al. As in the series evaluated by JB Mitchem et al, it shows that the addition of RA±DO will decrease the leak rate.^12,17

Wheeler JM et al. argued that IOC may threaten anastomotic integrity. ¹⁹ Furthermore, the probability of detecting smaller leaks increases due to the controlled pressure provided by the IOC. As found in the review by R. Castano et al., the detection rate of small leaks increases even more with IOC. In addition, although 72–166 mmHg of force can be applied to the anastomosis at most with colonoscopy, it has been found that the colorectal anastomosis can withstand up to 200 mmHg in evaluations, ²⁰ which is evidence that the possibility of damaging the integrity of the anastomosis with colonoscopy is very low. However, since IALT is much lower than these rates, small leaks due to insufficient pressure formation may be missed. ²¹ Although IOC has the advantage of detecting and intervening in existing leaks during surgery, ²² Shibuya et al. and Shamiyeh et al. ²³ found that bleeding occurred in the anastomosis in patients who underwent IOC and had normal anastomotic findings. This resulted in contradictions with the opposite of the efficacy target in the detection of early bleeding with IOC. In our study, both intraoperative bleeding and postoperative anastomotic bleeding were not observed, and larger series are needed to examine the results of these studies.

In colorectal surgery, a tension-free and well-fed anastomosis is the golden rule to prevent anastomotic leakage. ¹³ With the further development of the current stapler technology, leakage rates due to inadequate technical reasons are decreasing. ¹¹ As surgeons, we think that these leaks we detect during surgery will result in anastomotic leaks after surgery, but we do not know how much the body can limit these minimal leaks with mucosal repairs. It is obvious that the nutritional status of the patient, the catabolic effects of the tumor, or the primary comorbidities of the patient have an effect on this healing process. In our study, we did not open a diversion ostomy in the patients who were found to have no leakage by repeat leakage test after repair with PS. This led us to conclude that the leakage rates with PS repair were higher than normal leakage rates. We came to the conclusion that in anastomoses where leakage was detected in the tests, it is not correct to repair a single focus of leakage, but in cases of leakage, it should be considered that there is a general problem in the integrity of the anastomosis. Therefore, we think that RA±DO should be added in this high-risk patient group. The fact that diverter ostomy is frequently used after rectal cancer, especially in patients receiving neoadjuvant chemoradiotherapy, and the minimal leaks formed are closed by the body,^24,25 and the complications of the patient are minimized can be considered as evidence of this.

The fact that we could not use ICG in all patients in our study and the lack of standardization in the use of staplers is one of our important shortcomings. Although there was no statistically significant difference between the number of staplers used in transection or the differences in the diameter of the circular stapler used (P = .777 / P = .355), the limited number of patients may have affected the difference. In the study by Balciscueta et al., the result that increasing the number of staplers used in rectal transection increased the leakage rate ²⁶ supports this concern. There is no clear consensus on the diameter of the circular stapler, and no difference was found between diameter and anastomotic leakage in the study by Shibutani et al., ²⁷ whereas a difference was found in the study by Jiang et al. ²⁸ Although there was no statistically significant difference in the demographic characteristics of our patients, preoperative nutritional status and tumor stage were not included in our study. The difference that these factors may cause in tissue elasticity and thickness in patients could not be standardized, and it could not be determined whether they affect tissue integrity in leaks detected by intraoperative tests.

Conclusion

Intraoperative detection and intervention of intraoperative leaks due to technical reasons reduces the likelihood of postoperative leakage. In our study comparing IOC with IALT, the likelihood of intraoperative leak detection with IOC increased even more. This supports the conclusion that IOC will decrease the rate of postoperative leakage as supported by prospective studies in the literature.^29,30 In addition, intraoperative IOC can also detect anastomotic integrity and bleeding foci, if any. However, we think that the complications can be further reduced with the addition of RA±DO methods considering the general loss of integrity in this anastomosis instead of PS repair of leaks detected during surgery. With the use of IOC in patients with colorectal cancer, anastomotic complications will decrease, oncologic treatment of patients will not be delayed, and the possibility of achieving the goal of prolonging DFS/OS will increase.

Original research has not been previously published, in any language, in whole or in part, and the article is not currently under consideration for publication elsewhere.