Endoscopic submucosal dissection (ESD) is a well-established advanced mucosal resection technique used in Japan, where it originated, and some other Asian countries. The technical difficulty and potentially higher rates of significant complications have been obstacles for its dissemination across the United States, on top of the lack of available equipment, candidate lesions in the gastrointestinal tract, and adequate training programs. Yet American physicians are becoming increasingly aware of the benefits of ESD. Simplification of technique, modification of tools and materials, and improved availability of training opportunities are essential in order to accelerate the adoption of ESD in the United States.
Key points
- •
Main hurdles for adoption of endoscopic submucosal dissection (ESD) in the United States have been the lack of necessary equipment, target lesions suitable for ESD, and training programs.
- •
ESD demands high technical proficiency and therefore requires a dedicated training program prior to performing ESD in clinical cases.
- •
Further advancement of techniques and equipment may further facilitate the adoption of ESD by US physicians.
Endoscopic submucosal dissection (ESD) was first reported in Japan in the late 1990s. It quickly gained popularity due to improvement of techniques and available equipment, and before long it spread to other Asian countries. The main indication for ESD in the gastrointestinal tract was early gastric cancer, for which excellent outcomes were reported with endoscopic mucosal resection (EMR) if invasion depth and size were limited.
Many well-trained physicians from the United States have visited high-volume ESD centers in Japan to observe and learn the technique, clearly demonstrating the enthusiasm of American physicians willing to adopt this technique. However, there was originally limited availability of the knives that were developed specifically for the procedure (US Food and Drug Administration [FDA] approval was only achieved later in 2009). Moreover, in North America, there is a lack of target pathology in the gastrointestinal tract (ie, early gastric cancer). In some Asian countries, national mass screening programs have been developed (which started in Japan in 1983 and subsequently in Korea in 1999). Gastric cancer screening programs shifted the discovery of gastric cancers to those in early stage, and further endoscopic treatment and surveillance were performed for those patients. In the United States, gastric cancer incidence is lower, and universal screening is not recommended.
Colon cancer is the second leading cause of cancer death in the United States, and mass screening colonoscopy has been shown to improve colon cancer mortality. Large colorectal lesions or intramucosal cancers are usually treated by EMR with proven benefit and efficacy, and in 2001, the Center for Medicare and Medicaid Services approved colonoscopy as the primary screening modality for colorectal cancer. Also, the recent rise in incidence of adenocarcinoma of the esophagus alarmed physicians, and Barrett’s related dysplasia surveillance was initiated and is now widely practiced. Nodular dysplasia containing high-grade dysplasia or intramucosal cancer is not an uncommon finding during screening and surveillance endoscopies for Barrett esophagus, and EMR has become the standard diagnostic and therapeutic modality. Naturally, those lesions in the colon and the esophagus would be ideal targets for advanced endoscopic resection (ie, ESD), especially for larger lesions. However, both the colon and esophagus are considered high-risk locations for complications (especially perforation), and therefore in Asia it is recommended to start learning ESD in the stomach.
The American Society of Gastrointestinal Endoscopy (ASGE) published a technology status evaluation report on EMR and ESD in 2008, but at the time, coverage of ESD was limited because of unavailability of FDA-approved ESD knives in the United States. Furthermore, there were no training centers in the United States for ESD-specific education, and to date, there are no guidelines on how to be trained in ESD in the United States. Therefore, pilgrimage to Japan (and more recently to Korea or China) continues to be the major pathway for American physicians to learn ESD.
Issues for ESD training in the United States
The lack of training pathways is well recognized as a hurdle for adopting ESD in the United States. Traditional training in Japan consists of learning the basic technical knowledge of ESD (indication, equipment, and techniques), observing expert endoscopists performing ESD in live cases or live demonstration courses, participating in ESD procedures as an assistant (typically 5, but up to 40 cases, providing opportunity for additional learning prior to actually performing ESD), and then finally performing supervised ESD procedure with hands-on assistance from experts. All of these training tiers suggest initiation with gastric neoplasms in the distal third of the stomach, because ESD is technically easier with fewer complications at this location compared with the colon or the esophagus.
Several Japanese groups have reported on the gastric ESD learning curve, but opinions vary on the matter, with reports of ESD sufficiency to proficiency requiring experience with 30 to 80 ESDs. Thirty cases have been shown to be required for beginners to successfully acquire the ESD techniques, with en bloc resection rates increasing to 85% from 45% after performance of 40 cases. Excellent clinical outcomes during vigorous, progressive upper ESD training periods have been reported (30 cases, standard resection criteria) ; however, further study has shown that trainees must complete 40 to 80 cases to show proficiency at removing difficult lesions, and over 80 cases in order to reach expert level with fewer complications.
The main difficulties for trainees are related to submucosal dissection and control of bleeding. Some training programs aim to minimize these difficulties by incorporating hands-on training using animal models. Before practicing any ESD techniques on humans, either explant tissues or live animal models can be used for practice ( Figs. 1 and 2 ). The opportunity to practice submucosal dissection is beneficial, but live animal models are needed to simulate the actual human condition of constant luminal wall motions and to practice effective hemostasis during ESD.
Currently, only a handful of physicians perform ESD routinely in the United States, while expert endoscopists are more readily available in Japan. Because practice opportunities under direct supervision are not as abundant here, animal model training would be the preferred method to train physicians in the United States. In fact, in low-volume centers in both Europe and South America, use of a porcine model for practicing ESD without expert supervision has been shown to strengthen technical skills. Regardless of whether it is to include animal model training or not, the step-wise method of learning ESD that has been established in Japan should be adopted by physicians in the United States.
Another setback to widespread use of ESD in the United States is the infrequent finding of early gastric cancer. The standard Japanese model for learning ESD, which is to gain proficiency in gastric ESD before attempting either the colon or the esophagus, is not feasible here. Although not yet established, a direct approach to organ-specific ESD training would be more beneficial in the United States. For instance, Iacopini and colleagues have shown a potential method for fast-track training of colonic ESD. An experienced endoscopist (ESD trainee) first performed five ESDs on explant porcine stomach, without the supervision of an expert, before traveling to an expert center to observe 40 gastric and colonic ESDs. After the observation period, the trainee performed one ESD similar to his original practice, but this time under the supervision of his expert trainer. This same trainer then supervised the performance of one rectal ESD procedure (retraining at 11th procedure). The en bloc resection rate reached 80% after five procedures in the rectum and after 20 procedures in the colon. This study suggested that colorectal ESD can be successfully learned after training in an explant stomach model. It should be noted that colonic ESD was only performed after achieving 80% en bloc resection rate in the rectum.
Adequate preclinical training is extremely important, as complications during ESD can be severe, possibly resulting in need for surgery, and an inability to achieve en bloc resection would nullify the clinical benefit of ESD. ESD is technically demanding, and even experienced endoscopists face high complication rates as they begin learning ESD. In fact, when a large group of ESD-novice Asian endoscopists was trained in ESD using a porcine model during an effort to examine the main difficulties experienced during training, bleeding and perforation occurred in over half of the cases (56% and 65%, respectively), proving that learning ESD is difficult, and not only for those in the West.
Issues for ESD training in the United States
The lack of training pathways is well recognized as a hurdle for adopting ESD in the United States. Traditional training in Japan consists of learning the basic technical knowledge of ESD (indication, equipment, and techniques), observing expert endoscopists performing ESD in live cases or live demonstration courses, participating in ESD procedures as an assistant (typically 5, but up to 40 cases, providing opportunity for additional learning prior to actually performing ESD), and then finally performing supervised ESD procedure with hands-on assistance from experts. All of these training tiers suggest initiation with gastric neoplasms in the distal third of the stomach, because ESD is technically easier with fewer complications at this location compared with the colon or the esophagus.
Several Japanese groups have reported on the gastric ESD learning curve, but opinions vary on the matter, with reports of ESD sufficiency to proficiency requiring experience with 30 to 80 ESDs. Thirty cases have been shown to be required for beginners to successfully acquire the ESD techniques, with en bloc resection rates increasing to 85% from 45% after performance of 40 cases. Excellent clinical outcomes during vigorous, progressive upper ESD training periods have been reported (30 cases, standard resection criteria) ; however, further study has shown that trainees must complete 40 to 80 cases to show proficiency at removing difficult lesions, and over 80 cases in order to reach expert level with fewer complications.
The main difficulties for trainees are related to submucosal dissection and control of bleeding. Some training programs aim to minimize these difficulties by incorporating hands-on training using animal models. Before practicing any ESD techniques on humans, either explant tissues or live animal models can be used for practice ( Figs. 1 and 2 ). The opportunity to practice submucosal dissection is beneficial, but live animal models are needed to simulate the actual human condition of constant luminal wall motions and to practice effective hemostasis during ESD.
