In operations for gastrointestinal cancer, laparoscopic surgery and robot-assisted surgery are widely accepted in Japan [7]. Generally, we performed thoracoscopic esophagectomy in the prone position and retromediastinal gastric reconstruction for esophageal cancer before robotic surgery was covered by national insurance in Japan. Despite this minimally invasive surgery, half of the patients who underwent esophagectomy experienced postoperative functional disorders, such as delayed gastric emptying (DGE), dumping syndrome, and reflux with dysphagia [8], especially those who developed postoperative complications [9]. Jezerskyte et al. reported that a decrease in short- or long-term quality of life after esophagectomy is unrelated to postoperative complications but to the nature of esophagectomy and reconstructive procedures [10]. Various interventions have been performed to resolve these issues; however, the ideal solution is still unclear.

The most important factors in achieving improvement in patients’ postoperative conditions are the route of the reconstructed gastric tube and the size of the preserved stomach. We used sub-whole stomach with retromediastinal reconstruction, and we retract the stomach to the abdominal side and fix the stomach to the diaphragm around the hiatus after anastomosis to avoid having the stomach drawn into the thoracic cavity by negative pressure. Regarding the size of the reconstructed gastric tube, thoracic stomach syndrome can occur and cause chest discomfort after eating following esophagectomy with whole-stomach reconstruction [11,12,13]. Additionally, the causes of DGE vary and include bilateral vagotomy and pylorus paralysis due to automatic nerve imbalance. Furthermore, it is difficult to treat a soft stricture at the upper side of the pylorus, as in our case. We have rarely experienced deflection and torsion of a sub-whole stomach in the thoracic cavity leading to obstruction of the passage of food. Zhan et al. reported that a narrow gastric tube has a slight advantage regarding the avoidance of DGE compared with whole stomach in reconstruction [11]. We chose sub-whole stomach with retromediastinal route in reconstruction previously, but we have changed the reconstruction route from retromediastinal to retrosternal, and we use a narrow gastric tube.

Erythromycin (motilin receptor agonist) as a treatment for DGE after esophagectomy is effective [14]; however, this treatment is unsuitable for long-term use. If the cause of DGE is pylorospasm, endoscopic dilatation is effective. Additionally, gastrojejunal bypass may resolve this issue, but when the site of the stricture is in the thoracic cavity, as in our case, the bypass approach is not feasible. Re-operation such as removal of gastric tube with re-reconstruction was too invasive for benign stricture, and we considered to perform partial gastrectomy to make gastric tube straight in thoracic cavity. However, since the dilated area was the greater curvature side, we decided not to perform thoracic surgery to avoid injury of feeding artery of gastric tube. Our patient had deflection of the gastric tube into the thoracic cavity. However, she was able to eat for 7 years without symptoms. Therefore, we considered that the cause of the benign stricture was a change in the shape of the gastric tube over time with decreased peristaltic activity due to aging. Initially, we attempted decompression using a nasogastric tube and medication for 2 weeks to increase peristalsis; however, these measures were ineffective. We then attempted to correct the shape of the gastric tube using an ileus tube, but this procedure also failed. In this case, we tried medication of acotiamide hydrochloride hydrate, metoclopramide, and some other Japanese herbal Kampo medicines including ghrelin potentiator Rikkunshito using nasogastric tube or ileus tube through the stricture for over two weeks, but they were not effective. Regarding the loss of peristalsis in the gastrointestinal tract, peristalsis is also present in the gastric tube; therefore, age-related decrease in peristaltic activity appears likely, in our case [14, 15]. In such cases, retrievable metallic stent [16] is not suitable as they do not address the cause of the weakened peristalsis, and no improvement is expected after the stent removal. Generally, higher numbers of older patients are undergoing surgery compared with previous years; therefore, a similar situation to that in our patient could become more frequent in the near future.

Stent placement is considered a promising approach for strictures in the gastrointestinal tract. It is the most common modality performed worldwide to obtain lifelong relief of dysphagia [17]. However, there is no long-term evidence regarding stent efficacy for benign strictures; therefore, temporary stenting using a covered stent with removal after improvement is ideal [5, 6]. According to the European Society of Gastrointestinal Endoscopy guidelines, temporary placement of self-expandable stents for refractory benign esophageal strictures is weakly recommended [18]. However, Dan et al. reported that removable esophageal stents have poor efficacy in the treatment of benign esophageal strictures owing to the high rate of migration (53%) and subsequent chest pain [19]. In our case, the stricture was not severe; therefore, stent fixation after placement was very difficult. Moreover, stent migration occurs in approximately 28.6% of patients, even in those with malignancy [16]. In fact, in our patient, the stent migrated to the upper side of the gastric tube before we considered a surgical approach.

As a solution, we performed LSF as a less invasive approach compared with thoracoscopic bypass surgery. Over-the-scope clip (OTSC) [20], or endoscopic suturing [21] could be feasible for stent fixation endoscopically. However, when the stricture develops in the sub-whole gastric tube, there are few places where the stent can securely hit to the wall. Regarding endoscopic suturing, the gastric tube in retromediastinal route is surrounded by lung, aorta, pericardial sac, and liver, then it is not absolutely safe. In such gastric tube cases, we believe this LSF is better approach to fix the stent than OTSC or endoscopic suturing because we can choose the point to fix freely and safely. With retromediastinal gastric tube reconstruction, fistulae may develop between the gastric tube and aorta or the membranous part of the trachea, as adverse events [5]. Full informed consent from the patient and scheduled imaging follow-up is necessary with this type of adverse event.