Authors
A. LEMMERS (1), V. HUBERTY (1), C. MUSALA (1), A. BUCALAU (1), P. DEMETTER (2), L. VERSET (2), J. VAN LAETHEM (1), A. BUGGENHOUT (3), I. EL NAKADI (3), J. DEVIÈRE (1) / [1] Erasme Hospital, Brussels, Belgium, Gastroenterology, Hepatopancreatology and Digestive Oncology, [2] Erasme Hospital, Brussels, Belgium, Pathology, [3] Erasme Hospital, Brussels, Belgium, Abdominal Surgery
Introduction
Launching a professional program of endoscopic submucosal dissection (ESD) for the treatment of early gastrointestinal neoplasia in Western countries is fastidious and debated.
Aim
After practicing few ESDs in selected indications, we sough to develop a structured ESD program by sending two advanced endoscopists (AL, VH) for 7 weeks to Japan (2015, Keio Cancer Center, Pr Yahagi) for lesion recognition and ESD learning, watching 95 ESD procedures, and performing 18 ESD on isolated pig stomach models. Further hands-on was performed on living pigs before treating patients. All ESDs were from then concentrated on these 2 operators, starting treating patients following European guidelines in June 2015. Here, we analyse the safety and efficacy of ESD from the beginning of our program.
Methods
Clinical and technical data were prospectively collected from June 18th 2015 to October 10th 2017, excluding ESD performed by foreign experts during live demonstrations. R0 resection rate was defined as clear margins (no dysplasia / no adenoma for lateral margins and clear vertical margin). Curative resection was defined following European ESD guidelines. All ESD were performed under general anaesthesia using a 20% glycerol submucosal injection solution and for 98% of them a Dual-knife (Olympus).
Results
85 ESD were performed in 83 patients (41% female; aged 69(27-98) years old) by two operators (AL,VH). Lesions were located for 28% in the oesophagus (10/24 squamous cell carcinoma), 21% in the stomach, 42% in the rectum and 9% in the colon. En-bloc resection rate was 99%, complete endoscopic resection rate was 98%. R0 resection rate was 68% globally (6% with positive vertical margin; 2% with carcinoma in the lateral margin). In details, R0 resection rate was 83%, 88% and 57% for oesophageal, gastric and colorectal lesions respectively. Median specimen size was 40 (15-110)mm. Median procedure duration was 120 (IQR 90-180)min. In 91% of cases, there were none or conservatively managed complications. Two patients needed endoscopic hemostasis for delayed bleeding, 4 presented secondary stenosis needing dilations, one urgent surgery for sepsis after colonic perforation. Pathological analysis revealed a carcinoma in 79% of oesophageal lesions (14 pTis/pT1a; 6 pT1b), in 16% of gastric lesions (2 pT1a; 1 pT1b) and 30% of colorectal lesions (7 pTis/6 pT1). A neuroendocrine tumor was present in 1 oesophageal, 5 gastric and 2 rectal cases. Curative ESD was obtained in 75% of cases and 14 patients needed complementary oncological surgery with 50% of them having no residual tumor in the organ and negative lymph nodes. When endoscopic follow-up was recommended, data were obtained in 69% of cases with a median of 6 (range 1-24) months length and no recurrence of the lesion observed in 96% of cases. Pathological specimen processing was revised after the first 6 months implementing Japanese standards and increasing the rate of free lateral margins from 47 to 75%. For the rectum, despite a 31% positive (adenoma) lateral margin rate, 100% of patients were free of residual adenoma at the end of endoscopic follow-up, suggesting coagulation artefact effect on the specimen.
Conclusions
Nowadays, launching an ESD program in an academic European Center by experienced therapeutic endoscopists, after an observation period in an expert Japanese center, is possible, safe and quick with good results in terms of en-bloc resections and outcomes. Technical and pathological analysis efforts must be done to decrease positive lateral margins.
