Endoscopy 2021; 53(04): 392-393
DOI: 10.1055/a-1290-7696

Incomplete resection rate: incomplete training, imperfect measuring?

Referring to Pedersen IB et al. p. 383–391
John Anderson
Department of Gastroenterology, Cheltenham General Hospital, Cheltenham, United Kingdom
› Author Affiliations

Incomplete resection of a colonic polyp is referred to as a recurrence at subsequent endoscopic follow-up. Residual tissue may or may not be visible at the original resection. With time, the residual tissue grows, becoming more noticeable at subsequent colonoscopy. Incomplete polypectomy is a very important adverse outcome for the patient. Polyp recurrence is associated with fibrosis and tethering, increasing the difficulty and risk of subsequent removal. Several additional procedures may be necessary, with some patients referred to specialist units and occasionally for surgical intervention. This is an unnecessary inconvenience for the patient and is associated with increased health care and opportunity costs. Most importantly, during the interval between resection and surveillance, the lesion may become malignant: a missed opportunity for cancer prevention. It is possible that the original intervention stimulates growth factors as part of the mucosal healing process, accelerating malignant transformation.

“It may well be that improving the incomplete resection rate of the 10 – 20 mm sessile serrated lesions not amenable to en bloc resection is currently the greatest challenge in routine colonoscopy practice.”

Determining the incomplete resection rate (IRR) of polyps can be difficult and it is likely that reported rates are underestimates. In this issue of Endoscopy, Pederson et al. report on incomplete resection in 339 nonpedunculated polyps (> 5 mm) [1]. The group used the same approach as Pohl et al., taking biopsies from the resection edge to detect microscopic residual tissue [2]. Despite this, there will be sampling errors, as the whole of the resection margin is not sampled. Studies of recurrence at a subsequent procedure (polyp recurrence) are even more likely to underestimate the rate, not least because the resection site must be correctly identified first. The longer the interval between the index colonoscopy and polypectomy, the more likely residual polyp tissue will be visible.

The IRR could be a valuable patient-centered performance metric, but determining the appropriate denominator, and then defining and measuring the numerator (recurrence) is challenging. Any such metric would need to address the issue of those patients who have polypectomy but no surveillance. An alternative could be the World Endoscopy Organization Category D post-colonoscopy colorectal cancer (PCCRC) rate, defined as a cancer arising from incomplete resection of a polyp [3]. However, this would only include a small number of cases that develop subsequent cancer within a 3-year period after incomplete resection, and identifying PCCRCs reliably, categorizing them systematically, and adjusting for case mix is beyond current IT systems. Current guidelines attempt to balance the potential risk of developing colonic malignancy against the recommended surveillance interval.

The Pedersen paper reports an overall IRR of 16 % and a strong association with sessile serrated histology (IRR 36.9 %) and increasing size of polyp. It is recognized that en bloc resection is not always possible or safe to perform on larger polyps. Piecemeal endoscopic mucosal resection (EMR) is normally associated with larger polypectomies and was the only independent risk factor associated with IRR in a systemic review IRR [4]. It has already been recommended that larger lesions ( > 20 mm) are treated by expert endoscopists [5] [6]. It may well be that improving the IRR of the 10 – 20 mm sessile serrated lesions (SSLs) not amenable to en bloc resection is currently the greatest challenge in routine colonoscopy practice.

Performing complete polypectomy requires both cognition and technical competencies. Colonoscopy training has historically been focused on intubation and more recently on polyp detection, both of which have reliable, easily measured performance metrics. Polypectomy training typically occurs late in the training pathway (if at all). External drivers (e. g. performance metrics) are frequently the catalyst of quality improvement. Currently there is nothing stimulating individuals to achieve high quality polypectomy, during and after training. The IRR would be a reasonable immediate, proxy measure of polypectomy quality. It has the merit of being the sum of applying several polypectomy competencies and, very importantly, impacts directly on patient experience and outcome.

Published IRRs for trainees and board-certified colonoscopists were the same in both Norway and the USA [2]. Increasing experience would be expected to lead to improved manual dexterity and experiential learning from errors and to improved IRR. Why is this not the case? The IRR is dependent on a number of factors and individual competencies. These include tip control and scope handling ability, the choice of technique and accessories used for a particular type of lesion, decision making, leadership, etc. The technique of resection employed in the paper was left to the endoscopist, and trainers are likely to train techniques they themselves employ, which may account for the suboptimal IRR. Certain techniques may improve the IRR related to SSLs. Diathermy has the potential advantage that the thermal energy applied will destroy any potential residual tissue left at the edges; however, it can also make it difficult to determine whether there is any residual tissue and increases the potential of delayed bleeding and perforation. Submucosal injection will protect against thermal injury but can make the SSL border even less obvious. Adding a contrast agent to the lifting solution for SSLs improves delineation of the lesion and is superior to digital image-enhancing modalities. Current European guidelines advocate the use of hot snare for intermediate size lesions (10 – 20 mm), but there is increasing use of cold snares (with or without submucosal injection) for SSL resection [6] [7] [8]. While blood oozing may compromise post-resection evaluation, adding some adrenaline to the lifting mix provides a bloodless field and using irrigation to interrogate the resection defect helps clarify the margins. Soft tip coagulation has been shown to significantly reduce recurrence in large piecemeal EMR resections of adenomas, although its role in SSLs is not established [9]. If the endoscopists had performed resection of the SSLs using cold snares with a submucosal lift containing a contrast agent – arguably the optimal technique – would the IRR have been different?

Improvement is needed in the resection of SSLs, requiring a consistent technique that is based on best evidence or practice. Ideally, the technique should be supported by a competency framework to support training. A standardized performance metric is needed to drive this approach, which should also be monitored to determine whether implementation is effective. This process should apply to all polypectomy procedures in order to improve training and the associated outcomes of IRR and PCCRC.

Publication History

Publication Date:
29 March 2021 (online)

© 2021. Thieme. All rights reserved.

Georg Thieme Verlag KG
Rüdigerstraße 14, 70469 Stuttgart, Germany