The transhiatal tunnel valvuloplasty technique following laparoscopic proximal gastrectomy: the single-center experience in a retrospective cohort

Abstract

Background

The debate over the optimal reconstruction technique following proximal gastrectomy continues. Transhiatal tunnel valvuloplasty (ThTV) is a novel esophagogastrostomy method. This study aimed to evaluate the feasibility and safety of ThTV.

Methods

A cohort with upper early gastric cancer or Siewert type II tumors who underwent laparoscopic proximal gastrectomy was retrospectively reviewed in a single center. The ThTV esophagogastrostomy procedure involved placing a lengthy gastric tube into the lower mediastinum and firmly binding it to the esophagus. Demographic and surgical morbidity data were extracted from the medical records.

Results

Between March 2023 and November 2023, 15 patients underwent laparoscopic proximal gastrectomy using ThTV. The cohort consisted of 13 males and 2 females, with a median age of 68 years (range 49-77). The median operative time was 213 minutes (range 171-370). The median times for tunnel construction and anastomosis were 7 minutes (range 4-30) and 17 minutes (range 10-29), respectively. The median tumor size was 2.0 cm (range 1.0-5.0), and the median number of lymph nodes dissected was 29 (range 13-49). TNM staging revealed 9 cases of stage I, 5 cases of stage II, and 1 case of stage III. As of January 1, 2025, the median follow-up duration was 16.8 months (range 13.8-22.2). No tumor recurrence was observed. No patients reported severe reflux symptoms (Visick score ≥III). Gastroscopy confirmed reflux esophagitis (Los Angeles classification Grade A) in one patient, and an anastomotic stricture requiring endoscopic balloon dilation was observed in another patient.

Conclusions

Transhiatal tunnel valvuloplasty is a simple and reliable anti-reflux method following laparoscopic proximal gastric surgery. The further verification of the esophageal function is warranted.

Introduction

Despite a global decline in gastric cancer, the incidence rates of cardia gastric cancer have been gradually rising recently [1]. Traditionally, total gastrectomy (TG) has been the standard procedure for proximal gastric cancer. However, accumulating evidence has demonstrated that proximal gastrectomy (PG) offers advantages over TG, such as preserving gastric capacity and preventing hormonal and nutritional deficiencies [2, 3]. A prospective study also confirmed the low risk of lymph node metastases in the distal residual stomach for esophagogastric junction tumors [4]. Consequently, the Japanese gastric cancer treatment guidelines recommend PG for tumors of the proximal stomach and esophagogastric junction [5].

The standard approach for reconstruction following PG remains debated. Typically, either esophagogastrostomy (EG) or esophagojejunostomy (EI) is performed [6,7,8,9,10]. Given that it preserves normal gut integrity and is technically simpler, esophagogastrostomy is often preferred. However, the incidence of reflux esophagitis can increase significantly with EG, ranging from 21.8% to 71.6% [11]. Consequently, various modified EG techniques with additional anti-reflux measures have been developed. The gastric tube's low compliance allows for quick food transit, further enhancing anti-reflux effects when correctly positioned in the lower mediastinum [12]. The double-flap technique, also known as the Kamikawa approach, reconstructs the gastric fundus to form a one-way valve, simulating the natural anti-reflux mechanism of the gastric cardia [7]. However, this technique is considered difficult and time-consuming, especially when performed laparoscopically.

We hypothesize that a one-way valve and a sufficiently sized pseudo-fornix are essential for controlling reflux. Based on this strategy, we developed a new, simple esophagogastrostomy by placing the gastric tube into the lower mediastinum along the esophagus. Our findings demonstrate that transhiatal tunnel valvuloplasty (ThTV) effectively reduces gastroesophageal reflux. In this retrospective series, we report the feasibility and safety of ThTV and describe the technique in detail.

Methods

Study design and patient selection

Between March 2023 and November 2023, a cohort of patients who underwent laparoscopic PG for Siewert type II tumors or upper stomach cancer in a teaching hospital were retrospectively reviewed. The advanced operators in laparoscopic surgery from the same surgical team performed all patients. The inclusion criteria were as follows: 1) pathology biopsy confirmation of esophagogastric junction or upper stomach adenocarcinoma; 2) tumor size <4 cm without distant metastases; 3) clinical stage T1-3N0-1M0; and 4) retention of 50% of gastric volume after PG. The exclusion criteria were: 1) neoadjuvant chemotherapy before surgery; 2) severe cardiorespiratory disorders or other conditions that preclude laparoscopic surgery; 3) multiple stomach lesions; and 4) diffuse gastritis. This retrospective study received approval from the Ethics Committee of our center (Approval No. 24/077-4357). Written informed consent was obtained from all patients. The work has been reported in line with the STROCSS criteria [13].

Surgical procedure

Laparoscopic proximal gastrectomy

No special examination or medication were preoperatively administered. The surgical protocol adhered to the Japanese Gastric Cancer Association guidelines [14], including proximal gastrectomy and perigastric lymph node dissection. Briefly, the patient was positioned with legs apart in the reverse Trendelenburg position. The surgeon stood on the patient's left, and the first assistant stood on the right. Trocar placement followed the five-port method.

A 10-mm trocar was first inserted beneath the umbilicus, and pneumoperitoneum was established using carbon dioxide at a pressure of 10 to 12 mmHg. After dissecting the perigastric and suprapancreatic lymph nodes, the crura of the diaphragm were pulled bilaterally to improve visualization of the lower mediastinal space. A 6 to 8 cm section of the esophagus was released upon dissection to the inferior border of the pericardium. The esophagus was then cut with a linear stapler (Powered Plus Articulating Endoscopic Linear Cutter, Ethicon Endo-Surgery, USA) to ensure a sufficient margin. Finally, a 6 cm midline incision was made to remove the specimen.

The establishment of the flap tunnel

The gastric tube was designed as follows. The right gastroepiploic artery was retained on the greater curvature, and the first branch of the right gastric artery was preserved on the lesser curvature. A gastric tube measuring 4 cm in width and 20 cm in length was created. When necessary, the specimen's proximal and distal margins were examined pathologically. Continuous sutures using 3-0 barbed thread (Stratafix™, knotless tissue control device, Ethicon, or Polysorb™, braided absorbable suture, Covidien) were employed to reinforce the stapled line of the gastric tube. Two centimeters below the top of the gastric tube, a muscle flap tunnel measuring 2.5 x 2.0 cm was carefully constructed (Figure 1A). The specimen was reinserted into the abdominal cavity once it was confirmed that no tumor remained.

Fig 1

Reconstruction of the Flap Tunnel and Gastroesophageal Anastomosis. A A 4×20 cm gastric tube was prepared extracorporeally. The flap tunnel was 2.5×2 cm in size, with careful dissection between the muscular and submucosal layers. During laparoscopy, a small hole was made 1 cm below the tunnel's lower edge. B The esophageal stump was pulled down through the tunnel, and the gastric tube was inserted into the lower mediastinum. C Three to four stitches were placed between the esophagus and the tunnel to prevent retraction. Continuous suturing of the gastroesophageal anastomosis was then performed. D The muscular flap was pulled down and sutured to the stomach to comprehensively cover the gastroesophageal anastomosis. E. The gastroesophageal anastomosis was completed.

Gastroesophageal anastomosis

The pneumoperitoneum was re-established to perform intracorporeal anastomosis. A stay suture was placed at the stump of the esophagus. The esophagus was pulled down through the tunnel as the gastric tube was inserted into the lower mediastinum (Figure 1B). To prevent the esophagus from retracting, the tunnel and the esophagus were sutured with three to four stitches (Figure 1C). Next, a small incision was made laparoscopically, 1 cm away from the tunnel's lower edge. The opening was about the width of the esophagus stump (Figure 1A). The stapled esophageal stump was excised. Continuous suturing using 3-0 absorbable barbed thread was performed on the posterior esophageal wall and the upper edge of the incision. Similarly, the anterior esophageal wall and the lower edge of the incision were sutured with another barbed thread (Figure 1C). Beginning with the tenth case, the anterior wall anastomosis was sutured using interrupted stitches instead of continuous sutures. Finally, the anastomosis was covered by pulling down the tunnel's seromuscular flap, which was then sutured to the stomach wall (Figure 1D, E). Details are shown in the video.

Data collection

Perioperative variables, including age, sex, pathological findings, and operative outcomes, were extracted from the case files. TNM staging was graded according to the eighth edition of the AJCC guidelines [15]. The Clavien-Dindo classification was used to grade complications [16]. Complications were classified as early or late, depending on whether they occurred before or after one month following surgery. Postoperative anastomotic stricture was defined as the need for endoscopic balloon dilatation. The surgical video was reviewed to document several vital parameters. The extracorporeal tunnel construction time and anastomotic time were recorded separately. Tunnel construction time was measured from the incision of the muscular flap to the tunnel's completion. Anastomotic time was quantified from the initial suture of the remnant stomach and esophagus to the completion of the flap suturing.

Follow-up protocol

All patients underwent follow-ups at 3, 6, and 12 months after the operation. CT imaging and laboratory blood tests were part of the routine checkups. At the first follow-up and after a year, double-contrast upper gastrointestinal barium X-ray radiography (UGI-XR) and endoscopy were performed. Endoscopic evaluations were conducted by Dr. Yong Liu, an advanced endoscopist. The Los Angeles classification was used to determine the reflux esophagitis (RE) grade [17]. The Visick score was utilized to evaluate the quality of life following surgery [18]. The data were retrieved via the medical record, mail, telephone, etc.

Statistical analysis

All continuous variables were expressed as medians and ranges. All statistical analyses were performed using SPSS version 13.0 (SPSS Inc., Chicago, Illinois, USA).

Results

Patient characteristics

The demographic data of the cohort are summarized in Table 1. Between March 2023 and November 2023, 18 patients underwent ThTV esophagogastrostomy following laparoscopic proximal gastrectomy. Two cases received neoadjuvant chemotherapy, and one case with a gastric smooth muscle tumor was excluded. The final cohort comprised 13 males and two females, with a median age of 68 years (range 49-77). The median Body Mass Index (BMI) was 24.6 kg/m² (range 21.0-33.2). All patients were scored 0 or 1 according to Eastern Cooperative Oncology Group Performance Status.

Table 1 Patient characteristics and operative outcomes

The median operative time was 213 minutes (range 171-370), including one case involving concomitant repair of an inguinal hernia. The entire cohort experienced no postoperative bleeding, anastomotic leakage, or perioperative deaths. The median times required for tunnel construction and anastomosis were 7 minutes (range 4-30) and 17 minutes (range 10-29), respectively. In 12 out of 15 cases, the anastomotic site was positioned above the diaphragm. The median blood loss was 30 ml (range 10-500). Postoperative removal of the gastric tube and initiation of oral intake occurred at a median of 4 days (range 1-5) and 4 days (range 2-10), respectively. The median postoperative hospital stay was 8 days (range 6-13).

Table 2 summarizes the pathological outcomes. One patient was diagnosed with mucinous adenocarcinoma, one with neuroendocrine carcinoma, and the remaining 13 patients with tubular adenocarcinoma. The differentiation types were as follows: 8 cases of moderately differentiated adenocarcinoma and 7 cases of poorly differentiated adenocarcinoma. According to the Lauren classification, one case was categorized as mixed type, while the remaining 14 cases were identified as intestinal type. The median tumor size was 2.0 cm, ranging from 1.0 to 5.0 cm. The median upper margin was 2.0 cm, ranging from 0.5 to 5.0 cm. TNM staging revealed 9 cases of T1, 3 of T2, 2 of T3, and 1 of T4. Lymph node involvement was noted in 3 cases of N1 and 1 of N2. There were 9 cases of Stage I, 5 of Stage II, and 1 of Stage III. The median number of lymph nodes dissected was 29, with a range of 13 to 49. Four out of 15 patients received adjuvant chemotherapy.

Table 2 Pathological Information

All patients underwent a gastroscopic examination during the follow-up period. Only one patient exhibited reflux esophagitis (Grade A) at six months postoperatively. The gastroscopic examination showed no signs of reflux esophagitis in the lower esophagus and anastomotic site (Figures 2A, B). The distinct dome-shaped pseudo-gastric fundus was also clearly visible (Figure 2C). No food residue was found in the residual stomach (Figure 2D).

Fig 2

The Gastroscopic View of the Anastomosis. A, B The lower esophagus and anastomotic site show no signs of reflux esophagitis. C The pseudo-gastric fundus, with a dome shape, is clearly visible upon retroflection. D. The stomach is completely free of food.

Double-contrast upper gastrointestinal barium X-ray radiography (UGI-XR) demonstrated smooth passage of the contrast agent through the esophagus (Figure 3A). After oral administration of effervescent agents, gastric distension occurred, increasing the contact area between the pseudo-gastric fundus and the esophagus, resulting in a bird's beak appearance and closure of the lower esophagus (Figures 3B, C). In the supine position, the pseudo-gastric fundus occupied the mediastinal space, displacing the esophagus to the anterior left side (Figure 3D). Notably, the remainder of the gastric tube was still unfilled (Figure 3E). In the coronal view, the pseudo-gastric fundus was apparent in the lower mediastinum, and the esophageal stump was closed (Figure 3F).

Fig 3

Pseudo-Gastric Fundus in Double-Contrast Upper Gastrointestinal Barium X-Ray Radiography (UGI-XR) and CT. A The barium passes easily, indicating the esophagus's natural shape. B, C After the oral administration of gas-producing medications, there was a notable increase in the contact area between the esophagus and the pseudo-gastric fundus, resulting in a bird's beak appearance. The lower esophagus is indicated by the white and black arrows, respectively. D. When the patient was supine, the esophagus was pushed to the anterior left side by the inflated gastric fundus in the lower mediastinum. The white arrow points to the lower esophagus. E. Simultaneously, the residual stomach in the abdominal cavity is not filled. F. In the coronal view, the esophagus is closed. The pseudo-gastric fundus envelops the lower esophagus, as indicated by the white arrow. G. The natural course of the esophagus and the pseudo-gastric fundus is evident. H. The expanded pseudo-gastric fundus causes a 90° bend in the esophagus following the administration of gas-producing drugs (indicated by the white arrow).

As of January 1, 2025, the median follow-up duration was 16.8 months (range 13.8-22.2), with no signs of tumor recurrence. Eight patients reported no change in their quality of life (Visick grade I), while seven experienced mild symptoms (Visick grade II). One patient reported nighttime heartburn six months postoperatively, and reflux esophagitis (Los Angeles classification Grade A) was confirmed by gastroscopy. Symptomatic relief was achieved with oral proton pump inhibitors (PPIs). No other patients reported reflux symptoms. Four patients experienced swallowing discomfort. Two of these patients reported discomfort upon discharge, which spontaneously resolved within one month. One patient developed difficulty swallowing two months postoperatively and was diagnosed with anastomotic stenosis, requiring endoscopic balloon dilation. The fourth patient reported swallowing discomfort without any signs of anastomotic stenosis on gastroscopy. Upper gastrointestinal contrast imaging revealed a well-filled shape of the pseudo-gastric fundus and a 90° curvature of the esophagus (Figures 3G, H). The common reconstructive procedures following proximal gastrectomy were listed in Table 3.

Table 3 The reconstructive procedures following proximal gastrectomy

Discussion

The lower esophageal sphincter (LES), the diaphragmatic crus, and the gastric fundus collaborate to form the gastroesophageal anti-reflux system [21]. The His angle, created by the stomach fundus and the abdominal esophagus, also plays a crucial role in preventing regurgitation. Redo et al. [22] replicated this anti-reflux mechanism following proximal gastrectomy in a canine model by constructing a tunnel through the muscular layer of the stomach wall. Notably, a 4-cm flap tunnel effectively prevented reflux, with none of the 40 experimental dogs developing esophagitis. The double-flap technique (DFT) was first introduced by Kamikawa et al. [23] to create a one-way valve. During postoperative follow-up, a 4.2% incidence of grade B or higher reflux esophagitis was observed. However, the DFT involves intricate suturing, which is challenging during laparoscopy. Operating in the lower mediastinum is particularly difficult, especially when treating a Siewert type II tumor, sometimes necessitating a thoracoabdominal approach [24, 25]. Several modifications of the DFT have been developed to simplify the hand-sewn process, such as mechanical anastomosis and the single flap method [26,27,28]. Despite these advancements, completing the flap closure remains time-consuming. Additionally, due to the unidirectional blood supply to the flap, there is a risk of anastomotic leakage or stenosis with both single and double flaps.

We devised a novel esophagogastrostomy termed transhiatal tunnel valvuloplasty (ThTV), which offers several advantages. First, as previously reported [22, 23], the flap tunnel recreates the His angle and functions as a one-way valve (Figure 4A). Second, the gastroesophageal complex, in conjunction with the esophagus and pseudo-gastric fornix, ensures consistent pressure transmission from the gastric conduit to the esophagus. The close contact between the esophagus and pseudo-gastric fornix is further solidified within the narrow lower mediastinal space. Consistent with earlier publications [12, 29,30,31], we believe that a sufficiently long overlap length (4-5 cm) is essential to quickly raise the pressure in the lower esophagus when reflux occurs (Figure 4B). Third, by introducing a short tunnel and abandoning the double-flap closure, ThTV simplifies the gastroesophageal anastomosis, even for Siewert type II tumors, in the limited mediastinal space. Finally, the tunnel preserves a bidirectional blood supply to the flap, enhancing the safety of the anastomosis.

Fig 4

The Scheme of the Anti-Reflux Gastroesophageal Complex. A The gastroesophageal complex consists of the esophageal stump, pseudo-gastric fornix, and lower mediastinum. During periods of resting, a one-way valve maintains pressure in the lower esophagus. The pseudo-gastric fornix acts as a food reservoir, relieving pressure. B During episodes of increased intragastric pressure, such as vomiting, the enlarged pseudo-gastric fornix creates a high-pressure zone by pushing against the esophagus (indicated by black arrows). The narrow space of the lower mediastinum reinforces this effect. Thus, the gastroesophageal complex functions as a self-adjusting anti-reflux system.

The pseudo-fornix is a crucial structure for preventing reflux. However, its function can be compromised by the insufficient volume of the gastric fundus or a loose connection between the fornix and the esophagus [12, 22, 32]. In this series, the stomach remnant was transformed into a gastric tube and inserted into the lower mediastinum. The pseudo-fornix and lower esophagus were encircled by the pericardium, pleura, and aorta, further strengthening the integrity of the gastroesophageal complex (Figure 1C). Additionally, the narrow gastric tube facilitates the rapid passage of food, reducing retention [6]. Conversely, a large residual stomach is more susceptible to intrathoracic negative pressure. Animal studies have shown that the initial volume of the gastric residual and the compliance of the stomach wall are positively correlated [33]. To minimize the reflux induced by breathing, the narrow gastric conduit is a good alternative due to its low compliance. Our results showed that in the resting state, the gastric tube was pressed against the esophagus and occupied the lower mediastinal space (Figures 3D, 3F). The lower esophagus was compressed even more as intragastric pressure rose (Figure 3B). Thus, the narrow gastric tube with low compliance can quickly increase lower mediastinal pressure. In the supine position, the posterior pseudo-fornix serves as a food reservoir, preventing stomach contents from passing into the esophagus. Meanwhile, the gastric tube mitigates the impact of negative thoracic pressure and maintains proper resting esophageal pressure (Figure 4A). The lower mediastinum works with the gastric tube to quickly raise esophageal pressure during episodes of elevated abdominal pressure, such as belching, thereby enhancing the anti-reflux effect (Figure 4B). Together, the lower mediastinum, esophagus, and gastric conduit form the gastroesophageal complex, which functions as a self-adjusting anti-reflux mechanism.

Here are some surgical technique tips. To protect the muscle flap from thermal injury caused by the electric scalpel, the power should be reduced to 30 W or less when constructing the tunnel. The separation of the flap tunnel can be effectively enhanced using both blunt and sharp dissection techniques. To minimize accidental harm, a small amount of diluted methylene blue solution can be injected beneath the muscle layer to guide the visual separation plane. Ensuring reliable liver retraction is crucial for maintaining adequate space in the lower mediastinum. If necessary, the left diaphragm can be dissected to improve the surgical field. Three to four sutures are recommended to secure the esophagus to the gastric tube and maintain the esophageal stump. Performing the anastomosis with continuous sutures using 3-0 barbed threads is highly efficient. However, care must be taken to avoid anastomotic stricture by not pulling the threads too tightly. Alternatively, interrupted suturing of the anterior wall can be considered.

This study reported no mortality and an overall complication rate of 6.7%. The most frequent complaint was difficulty swallowing, with endoscopy revealing anastomotic stenosis in one patient. After DFT, multicenter retrospective research showed an overall anastomotic stricture rate of 5.5%, with a higher rate of 16.7% in the laparoscopic subgroup [19]. Notably, the stricture rate dramatically decreased as the learning curve improved. According to Shibasaki et al. [34], there is a positive correlation between the number of sutures and stenosis. We hypothesize that the anastomotic position and the surgeon's experience impact the stricture rate. About 80% of the anastomoses in this series were located over the diaphragm. Assistants frequently used more force to draw the thread due to the limited mediastinal space, potentially decreasing the anastomotic opening. No anastomotic strictures have occurred since we modified the suturing technique after the tenth case. Additionally, a pseudo-fornix can create the illusion of swallowing difficulty. The fourth patient reported difficulty swallowing, but endoscopy revealed no anastomotic stricture. An inflated pseudo-gastric fundus in UGI-XR caused the esophagus to twist nearly 90 degrees (Figure 3H), which could explain the patient's discomfort.

There are several limitations to the present study. First, the follow-up period needs to be extended to validate the long-term outcomes of the procedure. For instance, during the initial 3-month follow-up, the second patient in our group, an elderly male, did not report reflux symptoms. However, by the 6-month follow-up, he complained of nighttime heartburn, and endoscopy revealed grade A reflux esophagitis. The patient later acknowledged that he had resumed smoking, which, according to epidemiological research, increases the incidence of gastroesophageal reflux disease by 2.09 times [35]. Healthcare education should be emphasized during follow-up for these patients. Second, the study is limited by its small sample size and single-institution setting. The results should be validated through multicenter prospective clinical trials. Finally, there is a lack of objective functional evaluation in this study. Key functional outcomes, such as esophageal manometry and 24-hour impedance-pH monitoring, have not been thoroughly investigated. Esophageal reflux is a complex condition involving anatomical and neural factors. While ThTV achieved satisfactory short-term results, its physiological effects require confirmation through objective data.

Conclusion

In conclusion, our findings suggest that transhiatal tunnel valvuloplasty is an effective technique for esophagogastric anastomosis following laparoscopic proximal gastrectomy. The anti-reflux advantage of ThTV in this cohort deserve further investigation.