Sa1582 Learning Curve for Endoscopic Submucosal Dissection (ESD) of Colorectal Lesions in a Western Endoscopic Center
Federico Iacopini,Patrizia Rigato,Takuji Gotoda,Walter Elisei,Cristina Grossi,Agostino Scozzarro
DOI: https://doi.org/10.1016/j.gie.2011.03.317
IF: 10.396
2011-01-01
Gastrointestinal Endoscopy
Abstract:IntroductionEndoscopic submucosal dissection (ESD) has developed as a minimal invasive procedure for en bloc removal of large superficial neoplasms to overcome problems of EMR. However, ESD is infrequent in Western countries due to technical difficulty and safety concerns.AimsTo calculate the learning curve of colorectal ESD in a western endoscopic center.MethodsProspective single center study. A single endoscopist, novice in ESD, not supervised by experts, and competent in other endotherapies was involved. Training before practicing on humans comprised of 5 ESDs on isolated stomach, an observer period at the NCC in Tokyo during which 1 ESD on isolated stomach was performed under expert supervision. Insulated and non-insulated knives and standard gastroscopes were used. After 5 consecutive rectal ESDs, one 20 mm colonic ESD was attempted. Morphology, size, location, pathology of all lesions were recorded. Primary outcomes: en bloc and histologically complete (R0) resection rates; secondary outcomes: circumferential incision and submucosal dissection times, complications, local recurrences during a scheduled follow-up at 3, 6, and 12 months.ResultsFrom February 2009 to September 2010, 24 lesions in 24 patients (median 68 yrs; range 64-86) were removed: 20 in the rectum, 4 in the colon. Procedures were performed under conscious and deep sedation in 19 and 5 cases, respectively. Rectal lesions had a median size of 30 mm (range 7-60), and were LST-G in 17 (71%). ESD was interrupted in 5 (25%) rectal cases and in 3 (75%) colonic cases. ESD interruption occurred for patient intolerance during conscious sedation in 3 of the first 15 rectal procedures, and for difficult lesion position and orientation in the first 3 colonic ESDs.En bloc and R0 resection rates of rectal ESDs were 75% (15/20) and 73% (11/15), respectively, with positive vertical margins in 2 cases. Median rectal ESD operative times were: 20 min (range 5-26) for circumferential incision, and 170 min (range 30-270) for submucosal dissection. Circumferential incision time showed a trend toward decrease in the last 5 ESDs, whereas submucosal dissection time correlated to lesion size and not changed during the learning curve. Complication: a perforation occurring during the third colonic ESD was managed by multiple clips. Median in-hospital stay was 4 days (range 0-7). No recurrence/residual tissue was observed after en-bloc ESDs during the follow-up (median 9 months; range 3-12).ConclusionTraining on animal models and in Japanese expert centers could be the starting point to perform rectal ESD in humans by western endoscopists competent in operative endoscopy. Although technically demanding, en-bloc R0 rectal ESD is mostly possible even in the early phase of the learning curve. Conversely, a high competence in rectal ESD seems to be required before approaching colonic lesions. IntroductionEndoscopic submucosal dissection (ESD) has developed as a minimal invasive procedure for en bloc removal of large superficial neoplasms to overcome problems of EMR. However, ESD is infrequent in Western countries due to technical difficulty and safety concerns.