Patient 1

A 43-year-old man (78.2 kg; body mass index [BMI], 25.5 kg/m²) presented with a 3-month history of dysphagia and rapid weight loss of 5 kg within 1 month. He had no significant past medical history, and neither physical examination nor laboratory data revealed abnormalities.

Esophagogastroduodenoscopy (EGD) revealed an ulcero-infiltrative mass with unclear margins extending from the cardia to the antrum and infiltrative changes in the fundus. Endoscopic biopsy confirmed the presence of a poorly cohesive carcinoma. A computed tomography (CT) scan showed wall-thickening involving the stomach, cardia, gastroesophageal (GE) junction, and distal esophagus, with suspected lymph node metastasis near the lesser curvature. Advanced gastric cancer (cT4N1M0), specifically a Siewert type II GE-junction cancer, was diagnosed. Following multidisciplinary evaluation, a treatment plan was established, which included an upfront total gastrectomy, esophagectomy, and intrathoracic esophago–colon interposition surgery.

The patient underwent laparoscopic total gastrectomy, followed by left colonic conduit preparation using a robotic approach. Subsequently, robotic esophagectomy and intrathoracic esophago–colon anastomosis were performed. During the colon preparation phase, the colon was initially assessed using a laparoscopic approach. The left colon vessels were ligated, and the end of the descending colon was divided (Fig. 1A). The colon was then extracted through the primary incision, allowing a direct visual evaluation of its vascularity (Fig. 1B). The colon was then pulled up to the sternal angle to determine an adequate conduit length and subsequently inserted into the peritoneum. The proximal end of the colon was divided and laparoscope-assisted colo–jejunal and colo–colostomy anastomoses were performed. Intrathoracic esophago–colostomy was performed using a circular stapler. Following the anastomoses, the perfusion of the conduit was evaluated using indocyanine green (ICG) solution with a robotic camera (da Vinci Xi Surgical System; Intuitive Surgical Company, Sunnyvale, CA, USA) (Supplementary Video 1).

Figure 1. (A) The left colic vessels were ligated, and the end of the descending colon was divided. (B) The colon was extracted through the primary incision; we assessed its vascularity by direct visual evaluation.

The patient was transferred to the general ward on the 1st postoperative day. EGD performed on the 8th postoperative day confirmed the absence of extraluminal leakage at the anastomosis sites, allowing the patient to begin sips of water. On the 9th postoperative day, the patient started a liquid diet. By the 12th postoperative day, the patient was discharged. Postoperative pain, assessed using the Numeric Rating Scale (NRS), was rated at 7 on the first 2 postoperative days, decreased to NRS 3 on the 3rd postoperative day, and remained at that level until discharge. We have provided an explanation about the case report submission and obtained informed consent from the patients.

Patient 2

A 78-year-old man (67.8 kg; BMI, 27.5 kg/m²) visited our clinic after esophageal cancer was discovered during EGD. He reported no accompanying symptoms or abnormal laboratory findings.

The EGD revealed a mass located just above the GE junction, 35 cm from the upper incisor. A subsequent endoscopic biopsy confirmed the presence of a poorly differentiated adenocarcinoma. A CT scan demonstrated irregular wall-thickening in the distal esophagus, while positron emission tomography–CT revealed a diffuse hypermetabolic lesion near the mid-to-lower thoracic esophagus. Clinical staging indicated cT1bN0M0, prompting the recommendation of total gastrectomy and colon interposition surgery after a multidisciplinary consultation.

The entire surgical procedure was performed through a minimally invasive robotic approach. First, robot-assisted total gastrectomy was performed. The trocar designs are described in Fig. 2. Second, a colon conduit was prepared using the transverse colon and partial descending colon, and robot-assisted colon–jejunal anastomosis and colo–colostomy were performed. Next, robot-assisted transthoracic esophagectomy was conducted, followed by intrathoracic esophago-colon anastomosis and colo–jejunostomy. During the colon preparation phase, we temporarily clamped the left colic arteries, which we planned to ligate (Fig. 3A). We then injected ICG solution to assess perfusion to the colon (Fig. 3B). After evaluating colon vascularity, we estimated the length from the sternal notch to the abdomen using a U-tape. We then extracted a sufficient length of colon conduit (Fig. 3C, Supplementary Video 2).

Figure 2. We made 5 trocar ports for the abdominal phase in this operation. ^a)An umbilical port for camera port. ^b)A right upper quadrant port near the costal cartilage. ^c)A right upper quadrant port along the right midclavicular line. ^d)A left upper quadrant port along the left midclavicular line. ^e)A left upper quadrant port near the costal cartilage. Written informed consent for the publication of this image was obtained from the patient.

Figure 3. (A) The left colic vessels, which we planned to ligate, were temporarily clamped. (B) Perfusion to the colon was assessed using indocyanine green solution. (C) A sufficient length of colon conduit was measured using U-tape.

The patient was also transferred to the general ward on the 1st postoperative day. On the 3rd postoperative day, esophagography showed no extraluminal leakage or passage disturbance. From the 4th postoperative day onward, the patient initiated water intake. On the 11th postoperative day, the patient was discharged. Postoperative pain was initially rated at NRS 3 on the 1st postoperative day but improved to NRS 2 or less afterward. We have provided an explanation about the case report submission and obtained informed consent from the patients.

In this report, we present 2 surgical cases for GE-junction cancer, involving totally minimally invasive esophagectomy and colon interposition surgery. In the standard approach to esophageal reconstruction following an esophagectomy, the gastric conduit is typically the first choice [1]. However, for cases necessitating total gastrectomy, as in these patients, colon interposition becomes a viable alternative.

Adequate perfusion to the conduit is the most important factor for a successful anastomosis [2]. Ensuring an adequate length of the colon conduit and evaluating its vascularity are significant challenges in this procedure. Conventional colon interposition requires laparotomy, allowing direct visualization of colic vessels and evaluation of vascularity. However, when adopting a totally minimally invasive approach, determining the sufficient conduit length and assessing its blood supply become critical issues.

In patient 1, we assessed the colon length directly by pulling it up at the sternal angle to estimate the necessary conduit length. In contrast, for patient 2, we estimated the necessary conduit length indirectly using a U-tape. We chose this technique due to the difficulty of determining the optimal feeding vessel using a scope approach; specifically, we needed to choose either the middle or left colic artery. To guide this decision, we conducted superior mesenteric artery/inferior mesenteric artery angiography before surgery to evaluate the vascularity of the colic arteries; normal vascularity was confirmed. Ultimately, we opted for the middle colic artery because we empirically believed that it would allow for a longer conduit. A previous study reported that preoperative angiography can assist in the selection of feeding vessels for colon interposition, resulting in favorable outcomes [3]. Considering the limited visibility within the surgical field caused by the small incisions, angiography before surgery aids in selecting patients who are suitable candidates for minimally invasive procedures.

In both cases, we selected the ICG perfusion technique to assess vascularity because it is the only way to evaluate vascularity and ensure availability of the vessel in a minimally invasive approach. Previous studies have demonstrated the applicability of this technique in esophageal surgery. Meta-analyses have reported significant reductions in anastomotic leakage and shorter hospital stays after colorectal surgery or esophagectomy when ICG–fluorescence angiography was employed [4,5]. Considering its accessibility through a small surgical wound, we anticipate that this technique will suggest a new way of evaluating vessels for totally robotic colon interposition. However, there are situations where applying this technique may be challenging; for example, severe peritoneal adhesions can render the procedure impossible. Additionally, in extremely obese patients, identifying colon vessels and assessing vascularity can be difficult because of a fatty mesocolon and excessive pericolic fat tissue.

Prior studies have consistently shown that robot-assisted surgery is associated with improved postoperative outcomes. In a previous report, robot-assisted minimally invasive esophagectomy resulted in a lower incidence of pneumonia, reduced wound complications, and improved survival rates [6]. Robot-assisted colon surgery showed better outcomes in anastomotic leakage and time to regular diet than laparoscopic colon surgery [7]. Additionally, in our experiences, the robotic approach offers advantages, as it allows more precise colon manipulation and facilitates the evaluation of colonic conduit vascularity. Furthermore, it is useful in performing multiple anastomoses during the intra-abdominal phase.

In our routine practice, we typically initiate sips of water for patients without additional tests, such as esophagography or EGD. However, in cases where the perfusion of the colon conduit appears suboptimal or the conduit length is insufficient, we postpone starting sips of water until after performing EGD or esophagography. In these particular pilot cases, we adopted this approach as a precautionary measure to verify adequate perfusion, given their distinctive nature. Consequently, no issues were identified with the conduit in these cases.

We suggest that totally robotic colon interposition, combined with preoperative angiography and the ICG perfusion technique, can offer significant benefits to patients in specific circumstances. It can be a favorable choice when gastric conduits cannot be used, especially in cases of GE junction cancer or lower-third esophageal cancer. However, it is important to note that totally robotic colon interposition is not yet widely practiced, and as a result, established indications and contraindications are lacking. Therefore, the decision to employ this technique should be based on individual patient conditions, considering factors such as adhesions and obesity. Additionally, a preoperative evaluation of colic vessel vascularity is essential. Further research is necessary to refine the criteria for selecting patients and optimizing outcomes.

Author contributions

Conceptualization: YHK. Data curation: IHK, YHJ, KS. Formal analysis: IHK. Funding acquisition: none. Investigation: YHK. Methodology: YHK, CSG, CWK. Project administration: YHK. Resources: YHK, CSG, CWK. Software: none. Supervision: YHK, CSG, CWK. Validation: YHK, IHK, YHJ, KS. Visualization: IHK. Writing–original draft: KS. Writing–review & editing: YHK, CSG, CWK, IHK, YHJ, KS.

Conflict of interest

No potential conflict of interest relevant to this article was reported.

Funding

This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.

Supplementary materials

Supplementary materials can be found via https://doi.org/10.5090/jcs.23.131. Supplementary Video 1. Patient 1 underwent a laparoscopic total gastrectomy, followed by left colonic conduit preparation using a robotic approach. Supplementary Video 2. Patient 2 underwent completely robotic procedures. Perfusion to the colon was assessed using an indocyanine green solution, and the adequate conduit length was estimated using a U-tape.