Article Text
Abstract
Introduction Endoscopic resection (ER) of caecal lesions remains technically challenging due to instability of the scope, bowel preparation and thinner bowel wall. Such limitations can affect completeness of ER. Currently there are very few data on endoscopic outcomes following ER of caecal lesions ≥10 mm. Our aim was to assess completeness and complication of ER.
Methods Retrospective data was collected between 2011–2015 from documentation system ENDOBASE for all patients who had caecal ER. All lesions ≥10 mm were included. Morphology was categorised as per Paris classification. The following variables were recorded: lesion size and shape, type of ER (en bloc vs piecemeal), completeness of ER, histology, early complications, endoscopic follow up and remnant neoplasia on follow up.
Results Mean (SD). A total of 111 caecal resections were performed. See table 1 for demographic details. Average lesion size was 17.2 (8.3) mm; Ip 15.0 (7.0), Is 14.7 (4.6), non-polypoid lesion 17.9 (8.8) mm. En bloc resection was achieved in 63%. In lesions <20 mm, en bloc resection was 82% vs 26% in ≥20 mm lesion (p < 0.001). Complete endoscopic resection was in 93% (97% in lesions <20 mm vs 31% in ≥20 mm; p < 0.02). However definite histological confirmation of complete resection was in 23%. There was no early major complication. 68% of these lesions showed low-grade dysplasia, 7% high grade dysplasia, 5% serrated lesions and 21% were hyperplastic. Only 58 of 111 were followed up with colonoscopy in 3–36 months. In this group, 17% had endoscopic recurrent neoplasia on follow up. Average follow up was 7 months. Recurrence was treated with either or a combination of repeat EMR (42%), APC (17%) or biopsy forceps (42%). On average, it took 3.5 colonoscopies to achieve complete clearance. In patients with recurrent caecal lesions, there was no difference if their first ER was either by piecemeal or en bloc resection (p = 0.97). Recurrence was detected in 9% for all completely resected caecal lesions and 11% recurrence in ≥20 mm lesions. Recurrence was detected in all incomplete resected lesions (n = 5). 1 additional patient had remnant tissue detected only on biopsy on normal looking scar. The only variable associated with remnant neoplasia was an incomplete resection.
Conclusion Remnant tissue post ER was similar to the UK national guidelines for colorectal polyps.1 Endoscopic resection of caecal lesions remains a challenge. Incomplete resection lesion invariably has remnant neoplasia on follow up. Strategies such as ESD/hybrid EMR should be considered.
Reference 1 Rutter et al. doi:10.1136/gutjl-2015-309576
Disclosure of Interest None Declared