Modified abdominal wall retractor for total laparoscopic esophagojejunostomy in gastric cancer

Abstract

BACKGROUND

Complete intracorporeal esophagojejunostomy remains technically demanding following total gastrectomy. Conventional specialized stapling accessories are not easily available clinically, which hinders wider use of total laparoscopic anastomosis. A modified abdominal retractor was used in our center to simplify relevant surgical steps.

AIM

To compare perioperative results and learning curve between open and modified laparoscopic anastomosis.

METHODS

Eighty-four patients receiving radical total gastrectomy were enrolled retrospectively, consisting of 51 patients treated with traditional open reconstruction and 33 with the modified laparoscopic anastomosis. Baseline features and surgical data were compared, and multivariate regression was used to adjust potential confounding factors. Moving average and cumulative sum methods were utilized to analyze the learning curve of the novel surgical procedure.

RESULTS

Patients receiving laparoscopic surgery had shorter surgical incision, although the overall operation and anastomosis took longer. No significant differences were found in intraoperative bleeding, lymph node retrieval, postoperative rehabilitation indicators or total complication rates. Operating efficiency improved obviously after surgeons finished the initial learning stage of 11 consecutive operations.

CONCLUSION

The modified retractor helps implement safe and feasible total laparoscopic esophagojejunostomy, with a short learning cycle. Further multicenter prospective research is needed to confirm its clinical value.

Key Words: Gastroesophageal junction cancer; Laparoscopic esophagojejunostomy; Circular stapling anastomosis; Minimally invasive surgery; Surgical learning

Core Tip: We designed a customized abdominal wall retractor to facilitate totally intracorporeal esophagojejunostomy after total gastrectomy without costly dedicated stapling accessories. Compared with open anastomosis, this laparoscopic procedure shortens surgical incision while maintaining comparable perioperative safety, and novice surgeons can become proficient with this operative skill after about 11 clinical cases.

INTRODUCTION

Laparoscopic total gastrectomy combined with D2 Lymph node dissection is the most common surgical procedure for advanced esophagogastric union tumors[1,2]. Postoperative reconstruction of the digestive tract is often accomplished with the aid of a small trans-epigastric incision; however, because of the deep location of the esophagus, anastomosis is difficult and often requires thoracotomy[3,4]. Newer anastomosis methods, such as Overlap, require a longer esophagus, which partially limits their application[5]. Therefore, new types of anastomosis need to be explored.

Du et al[6] proposed complete laparoscopic circular anastomosis in 2014[6]. It is currently used for esophagojejunal anastomosis after total gastrectomy. This technique is characterized by a laparoscopic circular anastomosis of the esophagus and jejunum using special instruments. Compared to traditional open anastomosis, this method allows full visualization of the entire anastomosis, less trauma, and precise anastomosis, and is favored by many surgeons. However, this technique requires special anterior abdominal wall openers and laparoscopic purse-stringers, limiting its popularization.

In the present study, we adopted a novel modified abdominal wall retractor to facilitate total laparoscopic intracorporeal circular-stapled esophagojejunostomy. We systematically evaluated the clinical efficacy, short-term outcomes, and learning curve of this modified technique, aiming to provide a simplified, low-cost, and reproducible alternative for digestive tract reconstruction after total gastrectomy.

MATERIALS AND METHODS

Study design, and inclusion and exclusion criteria

This retrospective single-center study enrolled 84 patients who underwent radical total gastrectomy with D2 Lymph node dissection for gastroesophageal junction cancer between November 2023 and April 2025. All procedures were performed by a single experienced surgical team.

Inclusion criteria: (1) Pathologically confirmed gastric or gastroesophageal junction adenocarcinoma; (2) Siewert type II/III tumors requiring total gastrectomy; (3) pTNM stage I or II patients directly enrolled; pTNM stage III patients included only after multidisciplinary team discussion and confirmation of surgical resectability; (4) Complete radical resection with negative surgical margins; and (5) Complete perioperative and follow-up data.

Exclusion criteria: Previous upper gastrointestinal surgery, palliative resection, combined thoracotomy, alternative anastomotic reconstruction methods, preoperative adjuvant radiotherapy or chemotherapy, emergency surgery, incomplete clinical data, or loss to follow-up.

This study complied with the Declaration of Helsinki. All patients provided written informed consent, and the protocol was approved by the institutional Ethics Committee of Shaanxi Provincial People’s Hospital.

Outcome measures

Primary outcomes: Total operating time, esophagojejunostomy duration, and overall incidence of postoperative complications (including anastomotic leakage, bleeding, and abdominal infection).

Secondary outcomes: Total abdominal incision length, intraoperative blood loss, number of harvested lymph nodes, postoperative exhaust time, defecation time, drainage tube removal time, postoperative hospital stay, and learning curve proficiency threshold.

Follow-up and complications assessment

All patients received standardized postoperative management. Upper gastrointestinal contrast radiography was routinely performed on postoperative day 6 to screen for anastomotic leakage. Postoperative complications were diagnosed based on clinical manifestations, laboratory examinations, and imaging findings, and classified according to the Clavien-Dindo grading system. Regular outpatient and telephone follow-up was conducted after discharge to monitor long-term recovery and tumor recurrence.

Surgical methods

All patients underwent total gastrectomy with D2 Lymph node dissection, after which the esophageal margins were frozen and sent for examination to determine whether the margins were negative before esophagojejunal anastomosis.

Traditional anastomosis group: After excision of the stomach and dissection of the stomach and duodenum, a 10-15 cm incision was made in the middle of the epigastrium, and the esophagus was dissected using a purse-stringer 2 cm above the tumor after the abdomen was opened. The anvil of the circular stapler was inserted into the esophageal stump and a purse-string suture was secured to prepare for anastomosis. The jejunum was dissected 15 cm below the ligament of Treitz, and a Roux-en-Y anastomosis was performed. The circular stapler was inserted via the distal jejunal stump, and an end-to-side jejunoesophageal anastomosis was created 5 cm from the stump. The stump was closed using a linear stapler. The small bowel was further anastomosed laterally, 40 cm from the anastomosis to the small bowel at the ligament of Treitz, and the anastomosis was closed with a linear stapler. All anastomoses were closed using reinforced 3.0 interrupted sutures. The mesenteric fissure was closed after suturing.

Modified laparoscopic circular-stapled anastomosis group: The surgical procedure was standardized and divided into five stages for improved reproducibility.

Step 1: Modified retractor placement and trimming: After complete gastric mobilization and dissection, a 3-cm mini-incision was created in the upper abdomen, and a modified abdominal wall retractor was placed. One instrument port of the retractor was trimmed to a suitable diameter to allow smooth insertion of the laparoscopic purse-string clamp.

Step 2: Anvil preparation and esophageal purse-string fixation: The anvil was disassembled from the circular stapler, and a 10-cm 7-silk suture was tied to the metallic end to facilitate laparoscopic manipulation. After retractor assembly and pneumoperitoneum re-establishment, a purse-string suture was placed 2 cm above the esophageal tumor margin under laparoscopic visualization, followed by esophageal transection. The anvil was inserted into the esophageal lumen, and the purse-string suture was tightened and securely fixed.

Step 3: Jejunal mobilization and specimen extraction: The jejunum was transected 15 cm distal to the ligament of Treitz, and adequate mesenteric dissection was performed to ensure tension-free anastomosis. The retractor cover was opened, and the gastric specimen was extracted.

Step 4: Stapler assembly and protective T-tube placement: The mobilized jejunum was exteriorized, and the transected end was opened. A circular stapler was inserted and advanced 5 cm into the jejunal lumen. An 18F T-tube was cut into an 8-cm segment and sleeved over the stapler head as a protective barrier to prevent intraoperative mucosal injury during abdominal insertion.

Step 5: Intracorporeal anastomosis completion: The trimmed retractor opening was enlarged to accommodate the stapler. After stapler insertion, the retractor was resealed to restore stable pneumoperitoneum. The protective T-tube was removed laparoscopically, and intracorporeal circular-stapled esophagojejunostomy was completed. The residual jejunal stump was closed with a linear stapler, and side-to-side jejunojejunostomy was performed 40 cm distal to the esophagojejunal stoma. All anastomoses were reinforced with 3-0 sutures.

Learning curve

The learning curve of the modified laparoscopic circular-stapled anastomosis (LCA) technique was evaluated sequentially based on anastomosis time, which was defined as the primary evaluation indicator. Two validated methods were adopted: Moving average analysis and cumulative sum (CUSUM) analysis. The CUSUM value was calculated using the formula: (Xi) = Σ (Xi - μ), where Xi represents the anastomosis time of each case, and μ indicates the mean anastomosis time of all enrolled LCA patients. The inflection point of the curve was defined as the end of the learning phase.

Statistical analysis

Statistical analysis was performed using SPSS 26.0. The Shapiro-Wilk test was applied to assess data normality. Continuous variables with normal distribution were expressed as mean ± SD and compared using the independent t test; non-normally distributed data were analyzed using the Mann–Whitney U test. Categorical variables were compared using the Fisher’s exact test instead of the χ² test to improve statistical accuracy for small sample sizes. To reduce retrospective selection bias, multivariate regression analysis was performed to adjust for baseline confounding factors, including age, body mass index (BMI), tumor stage, Siewert type, perineural invasion, and vascular tumor thrombus. Statistical significance was defined as P < 0.05.

RESULTS

Comparison between LCA and traditional anastomosis groups

This study included 84 patients, 51 of whom underwent traditional anastomosis (TA) and 33 underwent laparoscopic circular stapling. We compared the baseline characteristics of the two groups, including age, sex, BMI, American Society of Anesthesiologists (ASA) score, preoperative white blood cell count, preoperative hemoglobin, preoperative albumin, Siewert type classification, tumor size, tumor stage, tumor differentiation degree, and the presence of vascular tumor thrombus invasion. No significant intergroup differences were observed (Table 1).

Table 1 Baseline characteristics of the two groups, n (%)/mean ± SD.

Factor	TA group (n = 51)	LCA group (n = 33)	t/χ2	P value
Age (year)	60.78 ± 8.57	62.03 ± 10.08	-0.607	0.546
Sex			0.359	0.549
Male	40	24
Female	11	9
BMI (kg/m2)	22.98 ± 1.85	22.29 ± 2.32	1.524	0.131
ASA score			0.082	0.775
I/II	40	25
III	11	8
WBC before surgery (× 109/L)	5.36 ± 1.90	5.41 ± 1.58	-0.105	0.912
Hb before surgery (g/L)	112.22 ± 13.04	115.22 ± 14.24	-0.986	0.327
ALB before surgery (g/L)	35.73 ± 3.83	35.00 ± 4.28	0.809	0.421
Siewert type			2.218	0.136
II	38	29
III	13	4
Tumor diameter (cm)	4.54 ± 1.13	4.62 ± 1.25	-0.323	0.784
Degree of differentiation			0.000	1.000
Well/moderate	7	5
Worse	44	28
TNM stage			2.254	0.133
I/II	25	27
III	26	6
Presence of vascular thrombus invasion			0.703	0.402
Yes	31	17
No	20	16
Perineural invasion			1.348	0.246
Yes	15	6
No	36	27

LCA: Laparoscopic circular-stapled anastomosis; TA: Traditional anastomosis; ALB: Albumin; ASA: American Society of Anesthesiologists; BMI: Body mass index; Hb: Hemoglobin; TNM: Tumor, node, metastasis; WBC: White blood cell.

We compared surgical and postoperative recovery-related indicators between the two groups. The results revealed that the LCA group had longer operating and anastomosis times. However, the total abdominal incision length was shorter than that of the TA group. There were no significant differences between the two groups in terms of blood loss, lymph node dissection rate, postoperative exhaust time, postoperative defecation time, postoperative drainage tube removal time, postoperative hospital stay, incidence of anastomotic leakage, or postoperative surgery-related complications (Table 2).

Table 2 Comparison of surgical and postoperative recovery-related indicators between the two groups, mean ± SD.

Factor	TA group (n = 51)	LCA group (n = 33)	t/χ2	P value
Operation time (minute)	260.47 ± 62.05	333.58 ± 90.77	-2.767	0.007
Anastomosis time (minute)	60.39 ± 18.38	72.73 ± 24.01	-2.659	0.009
Blood loss (mL)	101.76 ± 95.03	113.03 ± 133.57	-0.452	0.653
Lymph node number	25.39 ± 7.16	27.88 ± 9.68	-1.352	0.180
First exhaust time (day)	2.96 ± 0.75	3.00 ± 0.71	-0.240	0.811
First defecation time (day)	4.61 ± 1.46	4.70 ± 1.31	-0.285	0.777
Drainage tube removal time (day)	7.39 ± 1.91	7.73 ± 2.13	-0.751	0.455
Postoperative hospitalization (day)	8.55 ± 1.59	8.42 ± 1.35	0.372	0.711
Total length of incision (cm)	16.27 ± 1.95	7.79 ± 1.43	21.513	< 0.01
Anastomotic leakage (yes/no)	1/50	1/32		1.000
Clavien-Dindo grade			0.483	0.923
I	11	6
II	12	8
III	8	4

LCA: Laparoscopic circular-stapled anastomosis; TA: Traditional anastomosis.

Learning curve for laparoscopic circular stapling anastomosis

We sequentially arranged all patients who underwent laparoscopic circular stapling anastomosis in chronological order. Initially, the moving average method was applied, revealing that the curve began to plateau around the 11^th procedure (Figure 1). Further analysis using the CUSUM method indicated that the peak of the curve occurred at approximately the 11^th procedure. Therefore, we defined completion of the learning curve at 11 procedures and proceeded with further analysis based on this threshold (Figure 2).

Figure 1 The learning curve using the moving average method. The solid line represents the anastomosis duration for each patient in chronological order, and the dashed line indicates the learning curve derived from the moving average method. The curve began to stabilize after performing approximately 11 procedures, suggesting the proficiency of this technique.

Figure 2 The learning curve using the cumulative sum method. The curve peaks at the 11^th procedure, indicating the inflection point of the learning phase. CUSUM: Cumulative sum.

Impact of laparoscopic circular stapling anastomosis on surgical outcomes

We compared the differences in surgical outcomes between patients treated before and after overcoming the learning curve for laparoscopic circular stapling anastomosis. Based on the learning curve, we divided the patients into two groups in chronological order: A pre-learning curve group (11 patients) and a post learning curve group (22 patients). There were no significant differences between the two groups in terms of baseline characteristics, including age, sex, BMI, ASA classification, preoperative hemoglobin level, preoperative albumin level, preoperative white blood cell count, tumor size, degree of differentiation, and Siewert type classification.

Further comparison of the surgical outcomes revealed that the pre-learning curve group had longer operating and anastomosis times. However, no significant differences were observed between the two groups in terms of intraoperative blood loss, number of lymph nodes dissected, postoperative exhaust time, postoperative defecation time, postoperative drainage tube removal time, postoperative hospital stay, total abdominal incision length, incidence of postoperative complications, and anastomotic leakage (Tables 3 and 4).

Table 3 Baseline characteristics of patients treated before and after overcoming the learning curve, mean ± SD.

Factor	Pre-learning curve (n = 11)	Post-learning curve (n = 22)	t/χ2	P value
Age (year)	66.64 ± 8.51	60.23 ± 10.50	1.48	0.149
Sex			0.000	1.000
Male	8	16
Female	3	6
BMI (kg/m2)	21.78 ± 1.88	22.60 ± 2.45	-0.966	0.342
ASA score				0.586
I/II	9	20
III	2	2
WBC before surgery (× 109/L)	5.71 ± 1.63	5.23 ± 1.57	0.776	0.444
Hb before surgery (g/L)	117.82 ± 13.17	114.73 ± 15.10	0.577	0.568
ALB before surgery (g/L)	36.00 ± 5.31	34.50 ± 3.70	0.948	0.351
Siewert type				0.375
II	8	19
III	3	3
Tumor diameter (cm)	4.14 ± 1.82	4.87 ± 0.80	-1.622	0.115
Degree of differentiation				1.000
Well/moderate	2	3
Worse	9	19
TNM stage				1.000
I/II	9	18
III	2	4

ALB: Albumin; ASA: American Society of Anesthesiologists; BMI: Body mass index; Hb: Hemoglobin; TNM: Tumor, node, metastasis; WBC: White blood cell.

Table 4 Surgical and postoperative recovery-related indicators in patients treated before and after overcoming the learning curve, mean ± SD.

Factor	Pre-learning curve (n = 11)	Post-learning curve (n = 22)	t/χ2	P value
Operation time (minute)	386.63 ± 91.02	307.05 ± 81.00	2.458	0.020
Anastomosis time (minute)	95.45 ± 25.05	61.36 ± 13.20	5.158	< 0.01
Blood loss (mL)	163.64 ± 218.65	87.73 ± 50.33	1.574	0.126
No. of lymph nodes	24.27 ± 9.71	29.68 ± 9.36	-1.546	0.132
First exhaust time (day)	2.91 ± 0.54	3.05 ± 0.79	-0.516	0.609
First defecation time (day)	4.91 ± 1.30	4.59 ± 1.33	0.651	0.520
Drainage tube removal time (day)	7.36 ± 2.16	7.91 ± 2.14	-0.689	0.496
Postoperative hospitalization (day)	8.36 ± 1.63	8.45 ± 1.22	-0.180	0.858
Total length of incision (cm,)	8.27 ± 1.56	7.55 ± 1.34	1.397	0.172
Anastomotic leakage (yes/no)	1/10	0/22		0.333
Clavien–Dindo grade
I	2	4		1.000
II	3	5	0.000	1.000
III	2	2		0.586

DISCUSSION

The ideal anastomotic method after total gastrectomy remains a key research focus, particularly for esophagojejunal anastomosis. This procedure poses significant technical challenges because of the deep anatomical location of the esophagus and its proximity to vital organs. Consequently, exploring novel, convenient, and safe anastomotic techniques has been a persistent pursuit among surgeons. Recently, laparoscopic circular stapling anastomosis was proposed by Professor Du et al[6] from Xijing Hospital, China[6]. This technique has gained some clinical adoption as it enables complete intracorporeal circular stapling under laparoscopy. However, its widespread application has been limited by the requirement for specialized instruments, particularly customized retractors. In our study, we modified the existing abdominal wall retractors to perform this procedure and achieve favorable outcomes. Additionally, we evaluated the associated learning curve and provided valuable insights to facilitate further clinical implementation.

Currently, most circular stapling techniques for esophagojejunal anastomosis require open assistance. Owing to the deep operating field, a large incision is necessary, and the narrow surrounding space makes the procedure technically demanding. However, laparoscopic circular stapling anastomosis addresses these challenges by enabling fully visualized intracorporeal anastomosis under laparoscopy, enhancing the safety and ease of subsequent suture reinforcement of the anastomosis. Compared to the commonly used Overlap method for esophagojejunostomy, this approach minimizes the extent of esophageal resection[2-8]. Particularly in cases where intraoperative frozen sections reveal a positive margin, the circular stapling technique can still be applied for anastomosis, avoiding the need for thoracotomy. The Overlap method requires about 4 cm of the esophagus for anastomosis making thoracotomy unavoidable if the margin is positive.

Laparoscopic circular stapling is technically difficult and involves multiple steps that require meticulous attention to detail. When using a modified traditional abdominal wall retractor, two trimming steps are necessary because of the diameter mismatch between the laparoscopic purse-string forceps and the circular stapler. The first trimming creates an appropriately sized opening for the purse-string forceps shaft, whereas the second adjustment is made to accommodate the diameter of the stapler. The outer edge of the retractor can be secured with materials, such as surgical gloves, to ensure an airtight seal for pneumoperitoneum maintenance. Given the proximity of the gastroesophageal junction to the liver and spleen, there is a significant risk of iatrogenic injury. Therefore, care must be taken when introducing the stapler and its components and during esophageal purse-string suturing to avoid damage to the adjacent organs. Team coordination is crucial during anastomosis, particularly to prevent excessive traction, which can lead to anastomotic or mesenteric tears. All anastomosis and reinforcement procedures are performed under laparoscopic visualization to ensure a secure connection. Although the process is complex, mastering this technique offers greater precision and operational ease than for open anastomosis.

Currently, the main methods for calculating learning curves include group analysis, CUSUM analysis, and moving average analysis; each with its own advantages[9-12]. In this study, the CUSUM and moving average methods were used to determine that the learning curve for laparoscopic circular stapling anastomosis was about 11 procedures. Further comparative group analysis revealed a significant difference in the operating time before and after overcoming the learning curve, whereas other related indicators showed no significant differences. A possible explanation is that the surgeons had already performed > 100 esophagogastric junction tumor operations before adopting this technique. Additionally, apart from the anastomosis method, no significant differences in the baseline characteristics between the two patient groups were observed, ensuring that the surgical quality remained unaffected.

This study had some limitations. This was a single-center, retrospective, nonrandomized investigation with a small sample size. Potential temporal selection bias cannot be completely excluded, as the LCA procedures were performed within a concentrated time period. Although multivariate adjustment was applied, there may have been residual confounding factors. In addition, only short-term postoperative outcomes were analyzed. Therefore, large-sample, prospective, multicenter trials with long-term oncological follow-up are required to validate the clinical value of this modified technique[13-16].

CONCLUSION

Total laparoscopic circular-stapled esophagojejunostomy assisted by a modified abdominal wall retractor is safe, feasible, and minimally invasive, with a manageable learning curve. It provides a simplified and low-cost alternative for digestive tract reconstruction after total gastrectomy, Further largesample, prospective, multicenter studies are warranted to validate its general clinical applicability.