Prospective Study Comparing Clinical vs Indocyanine Green Fluorescence–Based Assessment of Line of Transection in Robotic Rectal Cancer Surgery—Indian Study

Abstract

Anastomotic leakage continues to be the most feared postoperative complications in rectal surgery with negative impact on both short- and long-term outcomes. Fortunately, new surgical strategies have helped to offset this complication and improve surgical outcomes. Traditionally, perfusion is assessed by intraoperative visual judgment by the surgeon. These subjective methods lack predictive accuracy resulting in either excess or insufficient colonic resection. Indocyanine green (ICG) fluorescence has shown promise in identifying the adequacy of perfusion. After injection of ICG, the system projected high-resolution near-infrared real-time images of blood flow in mesentery and bowel wall. This novel imaging method is used intraoperatively for taking real-time informed decisions. We conducted a single institutional prospective study to identify the feasibility of ICG identification of vascularity of anastomotic site and its impact on the change of plan of surgical management in robotic rectal cancer surgery. Between September 2017 and April 2019, fifty patients undergoing robotic rectal cancer surgery were included in the study. The aim was to analyze the feasibility and clinical benefit of intraoperative near-infrared fluorescence imaging in determining the line of transection in comparison with the traditional method. Line of proximal transection of the bowel subjectively assessed by the surgical team was marked point B and that after ICG injection was marked point A if moved proximally and point C if moved distally. The vascular anatomy was clearly identified with no intraoperative or injection-related adverse effects. Of the 50 patients, the line of transaction remained the same in 6 patients (12%). Based on the fluorescence imaging, the surgical team opted for further proximal change of the transection line up to an “adequate” fluorescent portion in 3 patients (6%) and distally in 41 patients (82%). ICG-based infrared image–guided localization gives a real-time image of colon vascularity possibly affecting anastomotic leak. The ICG fluorescence imaging system is a simple, safe, and useful technique, performed within a short time, and it enables visual evaluation of the blood flow in the intestinal tract prior to anastomosis. Larger studies are needed before this can become the standard of care.

INTRODUCTION

In modern oncological practice, radical surgery means resection of cancer without compromising on the principles while achieving superior outcomes with reduced complication rates. Surgical resection is established as the gold standard of treatment for rectal cancer surgery [1]. Advances in medical equipment and surgical techniques have enabled surgeons to offer patients better oncological and clinical outcomes.

Anastomotic leakage continues to be the most feared postoperative complication in rectal surgery. The reported leak rates range from 1 to 30% and increase as the anastomosis is more distal [2, 3]. The rate of patients with an anastomotic leak that requires surgical revision is from 10 to 35% with a mortality rate of 6 to 22% [4].

So it is imperative to adopt a strategy to prevent this dreaded complication. Common methods of assessing the anastomotic integrity include mechanical patency (air-leak and dye tests) or endoscopic assessment, but none of them is fool proof [5].

Although several factors have been identified as possible causes of anastomotic leakage (i.e., surgical techniques, patient risk factors, suture material or devices), the pathogenesis is still unclear [8].

Studies have found that poor local tissue oxygenation secondary to inadequate anastomotic vascular perfusion is the key in the determination of anastomotic viability [6]. The adequacy of bowel perfusion is related to the location of the tumor, pre-operative bowel obstruction, radiation therapy, and perioperative transfusion [7]. Anatomically, absence of Riolan arterial arcade and sacrifice of left colic artery during resection lead to compromised perfusion [8, 9].

One of the biggest issues is that none of the existing traditional methods is able to detect microvascular impairment at the anastomotic site. The most widely used method to assess perfusion is clinical judgment by the surgeon based on arterial pulsations, temperature, and bleeding edges. This is not accurate, thereby resulting in either excess or insufficient colonic resection. In fact, two studies have suggested that the clinical judgment of the operating surgeon underestimated the risk of anastomotic leakage in colorectal surgery [10, 11]. These subjective clinical signs are not only prone to misinterpretations even by experienced surgeons. They are not feasible while performing minimally invasive surgery.

Objective methods of assessing vascularity like pulse oximetry, Doppler ultrasound of marginal arteries, laser Doppler flowmetry, ultraviolet fluorescence studies, bowel wall contractility, and rapid sampling microdialysis have been used but with limited usage due to lack of reproducibility and practicality [12–14].

ICG fluorescence with high-definition 3D imaging systems is emerging as the latest strategy to reduce trauma and improve surgical outcomes during oncosurgery. Indocyanine green (ICG) is a vital dye with the distinctive feature of being fluorescent and is used with near-infrared light (NIR). It is a sterile, anionic, water-soluble but relatively hydrophobic molecule and when injected into the tissues binds to plasma proteins and emits an infrared signal when excited by laser light. The technique of fluorescence image–guided system has been applied in several fields of medicine with promising results and is continuously evolving [15, 16].

The da Vinci robotic surgical system is integrated with fluorescence imaging (firefly technology). The fluorescence imaging systems aim to improve the surgical outcomes by providing detailed anatomical and vascular information. Intraoperatively, real-time image of vascularity and avascular areas can be visualized and line of transection identified [17, 18].

Studies have shown that ICG-based detection is technically possible and has changed the plan of extensive bowel resection in 16.7% of patients and significantly reduced the anastomotic leak rates [19–21].

We present our experience using ICG fluorescence imaging during robotic rectal cancer surgery. We used intraoperative ICG fluorescence angiography to objectively assess bowel perfusion before anastomosis.

MATERIAL AND METHODS

We evaluated the feasibility and utility of ICG fluorescence–guided intraoperative assessment of bowel perfusion in determining the exact colonic transection line during robotic rectal surgery. After obtaining a written informed consent and approval from institutional ethics committee, patients with biopsy-proven colorectal cancer were included in the study. Exclusion criteria were pregnant or lactating women and patients with known allergy to ICG. This study was performed from September 2017 to April 2019 and includes 50 patients.

This prospective clinical trial was approved by the institutional review board and ethics board.

The robotic surgery was performed using the four arm da Vinci X surgical system (Intuitive Surgical Inc., Sunnyvale, CA, USA). Single docking total robotic colorectal mobilization and resection was performed in all using modified port placement [22, 23].

Total mesorectal excision (TME) with autonomic nerve preservation is a technically demanding procedure consisting of both abdominal and pelvic dissections. The abdominal part entails a “medial-to-lateral” approach where high vascular ligation of inferior mesenteric artery and inferior mesentric vein is performed with care taken to preserve the hypogastric nerves, ureter and gonadal vein, followed by splenic flexure mobilization. The pelvic part entails TME and anastomosis. The anterior resection/low anterior resection/ultra-low anterior resection was performed based on the location of cancer. ICG angiography was performed with the da Vinci firefly technology. The perfusion images were recorded and assessed in real time. Proximal colon transection followed by intra-corporeal anastomosis using circular staplers was performed. The integrity of doughnuts was checked after completion of the anastomosis followed by an air-leak test.

All the procedures were performed using ICG (Aurogreen, Aurolab) diluted with saline solution. The usual dose for standard clinical use (0.1–0.5 mg/ml/kg) [6] is below the toxicity level. Once the solution was prepared in the operating room, 3 ml of ICG in a concentration of 2.5 mg/ml was injected intravenously by the anesthetist by peripheral line followed by 10 ml of distilled water flush prior to the bowel resection (Fig. 1).

Fig. 1.

Indocyanine green dye (Aurogreen, Aurolab)

Visualization of structures due to fluorescence released by ICG was detected by specifically designated scopes and camera having near-infrared technology. The da Vinci X robot with firefly fluorescence imaging (LED illuminator and high-definition 3Dcamera system) has a near-infrared ray with a wavelength of 800 nm emitted from the robotic camera. This causes ICG in the blood vessel to emit a wavelength of 803 nm, and the fluorescence image was visualized on the monitor. Colonic blood perfusion was observed for 2 min after ICG injection, and the colon was observed for up to 5 min for adequacy of perfusion (Fig. 2).

Fig. 2.

Prior to the intestinal resection: comparison between line of transection subjectively determined by the surgical team through gross visual inspection technique and objective line of transection identified through ICG-based fluorescence in the firefly mode

Prior to the intestinal resection, the line of transection was first subjectively determined by the surgical team through gross visual inspection technique and marked.

The objective line of demarcation was identified through ICG-based fluorescence in the firefly mode. Comparison of the visually inspected line of transection and NIR-guided line of transection was performed. The complete lack of fluorescence or the macroscopically less fluorescence in comparison with the proximal portion of bowel was used as criteria to detect the poor perfusion.

Comparison of the visual transection line and fluorescence-guided line was performed. The perfusion transition area was named Line A if it was proximal to surgeon’s clinical transection line. It was named Line B if it coincided with surgeon marked line and named as Line C if it was distal to surgeon’s point (Fig. 3).

Fig. 3.

Comparison of the visual transection line and fluorescence-identified transection line. The perfusion transition area was named line A if it was proximal to surgeon’s clinical transection line. It was named Line B if it coincided with surgeon marked line and named as Line C if it was distal to surgeon’s point

The surgical team opted for change of transection line either proximally or distally in case of any discrepancy. ICG-enhanced fluorescence was detected in all the cases. None of these patients had clinical anastomotic leaks. After completion of the anastomosis, another bolus of 0.2 mg/kg of ICG was injected and a second evaluation of perfusion was made.

RESULTS

This study illustrates an initial experience with the NIR fluorescence imaging in rectal surgery.

During the study period, ICG-enhanced fluorescence was used to assess bowel perfusion in 50 patients with rectal cancer (mean age 54.52 ± 12.6 years, 18 females and 32 males). All patients with locally advanced disease received pre-operative chemo-radiation. Total mesorectal excision was done in a “medial-to-lateral” approach with high ligation of the main feeding vessels. Intra-corporeal stapled anastomosis was performed in all cases. There were no intraoperative adverse events or conversion to open surgery. Protective ileostomy was carried out in all low and ultra-low anterior resection patients. There were no side effects related to the injection of ICG. ICG-enhanced fluorescence was detected in 100% of the cases (Table 1).

Table 1.

Patient demographics, basal metabolic index (BMI), location of the tumor, and operative procedure

Patient characteristics	No. of patients (n = 50)
AGE (years), median ± SD	54.52 ± 12.6 years
Male:Female	32:18
BMI (≥ 25 kg/m2)	30
Co-morbidities	28
Pre-operative chemo-radiation	41
Cancer obstruction	3
TNM staging
I	0
II	3
III	47
Rectal cancer location
Upper 3rd (10–15 cm)	13
Mid 3rd (5–10 cm)	28
Lower 3rd (< 5 cm)	9
Operative procedure
Anterior resection	14
Low anterior resection	28
Ultra-low anterior resection	8
Ligation of vessel
High	44
Low	6
Splenic mobilization	48
Postoperative leak	0

On comparison, only in 6 patients there was no discrepancy between the visually inspected line of transection and NIR-guided line of transection (i.e., transection coincided with surgeon-marked line).These patients had no change in the surgical plan. In 88% of the patients, the fluorescence-guided transection line was either more proximal or distal to the visually determined transection by the surgeon. The vascularity of distal stump was adequate in all the patients.

ICG fluorescence therefore resulted in revision of transection line in 44 (88%) patients. In 3 patients (6%), the bowel was judged to be insufficiently perfused and the surgical team opted for further proximal “re-resection” up to an adequate fluorescent segment. Interestingly, the ICG-determined line of transection moved distally in 41 patients (82%) indicating unnecessary extra bowel resection by visual technique. Following transection, intra-corporeal stapling technique anastomosis was performed in all cases. There were no intraoperative adverse events or conversion to open surgery. None of these patients had anastomotic leaks. After completion of the anastomosis, another bolus of 0.2 mg/kg of ICG was injected and a second evaluation of perfusion was made to confirm the vascularity of the anastomotic leak. Reversal of ostomy was performed in the patients, the anastomotic evaluation prior to reversal was always made either by means of rigid endoscopy, and no sign of leaks or stenosis was reported (Table 2).

Table 2.

Results

Particulars	n	%
Total no. of cases	50	100
Identification rate	50	100
Line of transection
Line of transection moved proximally (Line A)	3	6
Line of transection moved distally (Line C)	41	82
Line of transection remained the same (Line B)	6	12
Change of plan	44	88
Conversion to open	0	0
Anastomotic leakage	1	2

DISCUSSION

The location of an anastomosis is the most consistent determinant of the anastomotic leak. The more distal the anastomosis is, the higher the likelihood of failure, with resection of a distal rectal cancer having almost a fivefold increased risk of anastomotic leak compared with resection for colon cancer [3]. Multiple studies have found that men undergoing rectal resection have a higher anastomotic leak rate, presumably due to the narrow confines of a male pelvis and the resultant increased technical difficulty of the operation [24, 25]. In our patients, low anterior resection and ultra-low anterior resection were performed in 28 and 8 patients respectively.

The surgeons’ ability to predict anastomotic leakage is low in gastrointestinal surgery, with a sensitivity of 61.3% and a specificity of 88.5%. Thus, objective and reliable intraoperative methods to assess bowel viability are required [10]. In colorectal surgery, studies have shown that intraoperative fluorescence-based perfusion angiography is a safe and feasible method to assess anastomotic perfusion, and its use might affect the incidence of anastomotic leaks. Controlled trials are ongoing to validate these conclusions.

Up to 19% of all colorectal resections develop clinically apparent insufficiencies. Due to lack of predictive accuracy for anastomotic leakage by the conventional visual inspection method, it is unreliable. Tissue perfusion can be objectified intraoperatively using laser fluorescence angiography and provides a real-time assessment of intestinal perfusion and reliably defines the point of resection by demonstrating the presence of an ischemic or “non-optimal” perfusion prior to anastomosis. Kudszus et al. reported that ICG imaging led to a change of the initially planned proximal transection line in 13.9% (28/201) of cases. It significantly reduced anastomotic leakage rate in colorectal surgery by 4% compared with a historical control group (3.5% vs. 7.5%) [15]. These data have been confirmed by Jafari et al. during a robotic-assisted laparoscopic rectal surgery [13, 26].

Intraoperative ICG angiography and the quantification of ICG kinetics can be used to intraoperatively reveal the tissue perfusion status during rectal surgery. This may be useful for intraoperatively changing a previously planned resection/anastomotic level and conceivably decreasing the degree of anastomotic leakage A recent systematic review showed that ICG assessment of colorectal anastomosis was associated with a significantly lower risk of anastomotic leakage compared with a control group (3.8% vs. 7.6%; P = 0.0055) [27]. A recent retrospective case-matched study by Kin et al. [28] on colorectal resection demonstrated that, once fluorescence angiography was performed, surgeons decided to change the proximal resection margin in 8/173 patients (4.6%).

Adequate vascular supply of the left colon depends on the patency of the inferior mesenteric artery, the left colic artery, but also relies on patency of the middle colic artery and the marginal Drummond and eventually Riolan arcades [29, 30]. Anatomic variations are frequent, and aberrations such as absence of the middle colic artery or inadequate vascularization of the splenic flexure are frequent (up to 25%) [31–33]. High ligation of the inferior mesenteric artery has been documented as a risk factor for anastomotic leakage [27]. For the colorectal cancer patients with IMA type III and Riolan artery arcade absence, selective low IMA ligation with root lymph node dissection should be recommended [34]. In our study, 88% of patients had high ligation of the inferior mesenteric artery (44 of 50 patients). In the study by Hellan et al. [18], all patients undergoing resection for malignant disease had high ligation. At present, there is no study comparing the same. Since the ICG fluorescence technique in laparoscopy is similar to the firefly, we expect the results to be similar.

Fluorescence evaluation, in contrast to the surgeons’ clinical assessment, led to a more proximal level of division in 3/50 (6%) of patients and distal level in 41 (82%) of patients. Forty-eight patients underwent splenic flexure mobilization. Although splenic flexure mobilization is widely considered to be an essential component of anterior resection for rectal cancer, it has been found to be unnecessary. Avoiding splenic flexure mobilization results in shorter operative times, reduced complications, and does not increase postoperative morbidity, anastomotic leakage, or local recurrence [35]. Eighty-two percent our patients where the line of transection was incorrectly deemed to be proximal underwent unnecessary splenic mobilization.

As per retrospective analysis of our records, the anastomotic leak was 5.6% before usage of ICG. We found that the anastomotic leaks reduced to 2% after the use of ICG-based firefly technique.

The effect of repeated injections of ICG is not known and has not been investigated [29]. Another drawback is that there is still no strict analytic measure to objectively quantify the signal intensity, and the evaluation of images still depends on the surgeon’s judgment.

Fluorescence imaging has been increasingly used as an intraoperative tool in routine practice to ensure adequate perfusion at the time of anastomosis. Recent literature shows the potential benefit of fluorescence imaging in lowering leak rates by changing the bowel transection line [36–38].

The predictive accuracy of the conventionally followed visual method to judge the optimal anastomotic site is low and unreliable. Quantitative analysis of ICG perfusion patterns can be applied to detect poor perfusion, thereby reducing anastomotic complications during colorectal surgery. The fluorescence functional imaging for assessing anastomotic perfusion in colorectal surgery can result in more precise operative decisions tailored for an individual patient. Large multicentric studies comparing anastomotic leaks in ICG vs no ICG group of patients are needed. This is the first study in Indian patients and we found this method of evaluating vascularity to be simple and safe and can be easily reproducible.

CONCLUSIONS

Our experience with the intraoperative application of fluorescence shows promising outcomes. The ICG fluorescence imaging system is a simple, safe, useful, and readily achievable technique with minimal added intraoperative time for accurate determination of the resection margin of the viable bowel and helps to reduce anastomotic leakage with subsequent improvement of outcome. ICG is easy to use logistically and provides important vascular information at low surgical expense.

This easily applied novel technique for intraoperative and interventional visualization has the potential to revolutionize rectal cancer surgery. Larger further randomized prospective trials like PILLAR-III are awaited to validate this new technique.

Abbreviations

TME

total mesorectal excision

ICG

indocyanine green

NIR

near-infrared

LAR

low anterior resection

CTRT

chemo-radiation

Authors’ Contributions

All authors read and approved the final manuscript.

Compliance with Ethical Standards

Competing Interests

The authors declare that they have no competing interests.

Approval and Consent

Institutional Review Board approval for the randomized study was obtained.

Consent for Publication

No individual person’s data contained in the publication.