Endoscopic submucosal dissection (ESD) is a minimally invasive treatment option for local resection of early gastric cancer with a low risk of nodal metastasis. The clinical indications have been expanded based on the favorable long-term outcomes, although the level of recommendations varies among guidelines.¹ ESD allows for high en bloc and R0 resection rates regardless of lesion size and location. A recent multicenter prospective cohort study, the J-WEB EGC study, which included more than 10,000 gastric ESDs, showed that intraoperative perforation and delayed bleeding rates were 4.4% and 2.3%, respectively.² It is well known that most adverse events could be managed endoscopically without surgical intervention. However, delayed perforation is a rare but serious adverse event that often requires emergency surgery. The J-WEB EGC study reported that the delayed perforation occurred in 0.4% (40/9,616) and emergency surgery for delayed perforation was performed in 0.15% (14/9,616).² Because delayed perforation can be caused by ischemia due to cautery effect rather than sharp tissue laceration by electrosurgical knife. Although the detailed pathogenesis is unclear, a retrospective study suggested that gastric tube reconstruction after esophagectomy, which can alter the blood supply around the stomach, is the risk factor for delayed perforation.³ It suggests that the size of the necrotic area around the perforation site may be extensive.

In this issue of Gut and Liver, Kim et al.⁴ conducted a single-center retrospective observational study and reported that delayed perforation occurred in 15 of 11,531 patients (0.13%) undergoing gastric ESD. Of these, nine patients (60%) required emergency surgery. A perforation less than 1 cm in size was observed more frequently in the six patients who underwent successful nonsurgical treatments than in the nine patients who underwent surgery (p<0.001).

To the best of my knowledge, this is the largest cohort study and the data are consistent with previous studies.^2,3 The results are appreciated because the authors presented the clinical course after delayed perforation and provided an interesting discussion despite the limitations of a retrospective study. First, all five patients with small perforation sites <1 cm could be managed with conservative treatment or simple clip closure, while the other nine patients with large perforations ≥1 cm required emergency surgery (the size of perforation was not shown in one patient).⁴ In the present study, the indications for emergency surgery would be determined at the discretion of the referring endoscopists and surgeons. Emergency surgery is generally indicated for delayed perforation. However, it is clinically acceptable to attempt endoscopic intervention in the first instance when inflammation is localized and the consensus is reached after multidisciplinary discussion with surgical colleagues. The results of the present study suggest that simple clip closure is helpful in avoiding emergency surgery for small perforations <1 cm. In addition, some case reports showed that over-the-scope clip closure and tissue shielding with a bioabsorbable polyglycolic acid sheet were effective in avoiding emergency surgery even for a large perforation site.^5,6 Further studies are warranted to investigate the efficacy of endoscopic closure of a large perforation site, although the indications should be limited to stable general conditions as well as localized peritonitis.

Moreover, the time to event and the location of the delayed perforation site could also be related to the risk of emergency surgery. Early onset and prompt endoscopic intervention before diet initiation may be beneficial to avoid emergency surgery. Some case reports showed successful conservative management of delayed perforation with endoscopic or spontaneous drainage of perigastric abscess without perforation closure.^7,8 It could be explained that the perforation site could be anatomically covered with omentum, which could keep the peritonitis localized. Prophylactic closure of the mucosal defect after ESD is a future innovation to prevent delayed perforation. Currently, some suturing techniques have been developed to achieve complete and sustained closure of a large mucosal defect after gastric ESD, using overstitch device and endoscopic hand suturing. These novel approaches have potential to prevent delayed bleeding as well as perforation, allowing ESD to be in an outpatient setting.^9,10 However, they are costly and time consuming, and it is difficult to prove whether the arduous work is worth to prevent the rare adverse event.

Further investigation is warranted to analyze the relationship between the size of the perforation site and the risk of emergency surgery. Clinicians should keep in mind the rare but serious adverse events after gastric ESD and try not to miss the chance of endoscopic approach to achieve nonsurgical treatment.

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