Editorial: Optimizing the success of cold snare polypectomy in colonoscopy practice

Abstract

Most colonic polyps detected at screening colonoscopy are smaller than 10 mm. Effective, efficient, and safe removal of those small polyps plays a pivotal role in ensuring colonoscopy quality. Cold snare polypectomy (CSP) outperforms hot snare polypectomy in several aspects, including shorter procedure time, higher histological eradication rate, and a lower risk of delayed post-polypectomy bleeding, and this has been demonstrated in several large-scale randomized controlled trials.1-3 In fact, clinical practice guidelines have affirmed the use of CSP for sessile polyps 6–9 mm in size because of its superior safety profile.4, 5 Although CSP has become the mainstay for resecting small polyps at colonoscopy, there remain several issues that still require further study. First, its long-term efficacy in clearing neoplastic lesions remains unclear. This relates to the variable complete resection rates in published literature and the fact that the vertical plane of resection during CSP usually does not include the entire muscularis mucosa, which therefore cannot ensure completeness of resection for lesions with advanced histology. Although Kawamura et al. reported a complete resection rate of as high as 98.2% by CSP in the CRESCENT study,3 and Sidhu et al. reported a complete resection rate of 98.5%,2 in a meta-analysis by Qiu et al., the complete resection rate ranged from 77.3% to 98.2%.6 Admittedly, these authors have used variable definitions for complete resection, including endoscopic observation, biopsies of resection margins, and clearance of histological margins. Long-term outcomes such as post-colonoscopy colorectal cancer rate after CSP has never been reported. Second, the horizontal margin was seldom reported in previous studies. This is important, as there have been an increasing number of studies reporting the application of CSP in lesions larger than 10 mm or even larger than 20 mm.7 Because the risk of high-grade dysplasia (HGD) or invasive cancer increases along with size, an unclear horizontal or vertical margin may increase the risk of PCCRC. Though CSP does not apply electrocautery and reduces the likelihood of cautery effect affecting margin assessment, reconstruction of multiple tumor specimens after piecemeal CSP is still challenging, and failure of specimen retrieval may cause underdiagnosis of HGD or invasive cancer. Although a recent multicenter study from Japan revealed that the local recurrence rate after CSP for HGD was 13.7%, the CSP scar could be identified in only 46.8% of the cases; this may impact on the ability to precisely localize the exact site for close surveillance or repeat resection during follow-up endoscopy.8 That said, it is crucial to achieve a complete resection rate as high as possible at the initial procedure. Moreover, endoscopic morphology, either macroscopic or with magnification, of the lesion with the aid of image-enhanced endoscopy (IEE) provides useful information on the likelihood of the existence of advanced histology. In such a case, hot polypectomy after submucosal injection is a more appropriate practice as it can obtain a larger amount of submucosal tissue. Third, when resection of the lesion could not be easily achieved by CSP, it is sometimes resected by forcefully pulling on the snare, leading to cold snare defect protrusions (CSDP). Histopathological studies have shown that CSDPs are in the muscularis mucosa, and it may be an indicator of incomplete resection and is often associated with specimen fragmentation.7 In this issue of Journal of Gastroenterology and Hepatology, Horii et al. performed a randomized controlled study and reported that CSP using a thinner diameter of snare wire could achieve a significantly higher histological complete resection rate compared with that using a normal-diameter snare (70% vs 81%, P = 0.04).9 In a subanalysis of results based on level of experience, it is noteworthy that more experienced endoscopists appeared to have a higher complete resection rate than trainees. This subanalysis did not show a difference between snares when the analysis was stratified by experience. In addition, the normal-diameter wire snare group had significantly more cases of unclear horizontal margin evaluation than the thin-wire snare group (28.1% vs 15.9%, P = 0.02). Both groups were similar in terms of depth of vertical resection achieved, although there was a trend for the specimens in the thin-wire group to contain submucosa (23% vs 16.4%, P = 0.24). It is also noteworthy that despite the use of IEE and size to select lesions for CSP, eight out of the 254 lesions had either HGD or were intramucosal carcinoma, which meant that curative resection could not be established by the histopathological assessment from CSP. The key strengths of this study are that it is a well-designed randomized controlled trial, with further blinding of the pathologist, and that it sheds further light on the impact of CSP on the assessment of horizontal margins for clearance. The assessment of margins is particularly important for assurance of completeness of resection, especially in the presence of advanced histology. Otherwise repeat endoscopic assessment would be required, adding to the burden of resource utilization and patient inconvenience. Although one could argue that by suctioning out the resected specimen, the histological assessment is affected, this reflects the real-world practice and adds to the convenience of the technique of CSP. As mentioned earlier, different authors have used different definitions of incomplete resection. Unlike the studies by Sidhu et al.2 and Kawamura et al.,3 which defined complete resection based on negative biopsies of the resection margin and thus had very high complete resection rates, this study used histological margin as the definition. It may well be that if a similar approach had been used, the results would have been similar, but for the clinician and pathologist, what matters most would most likely be the certainty of complete resection as evidenced by clear margins. This is a similar approach as an earlier study by Horiuchi et al., who had a higher complete resection rate of 91%. However, an important difference is that retrieval by suction was used only for lesions smaller than 6 mm, whereas larger lesions were retrieved by forceps.10 Although this study by itself would not be sufficient to mandate the use of only thin-wire snares for CSP, it does provide additional information for decision making in clinical practice, in terms of the choice of devices and the corresponding potential benefits. This would need to be weighed against other considerations such as cost, and the role of dedicated CSP devices compared with those that can be used for both hot and cold polypectomy. CSP is already being widely embraced. To optimize success, careful attention to the correct technique is crucial, in particular, ensuring a normal rim of mucosa is consistently captured prior to resection.11 The choice of snare type may be important. CSP must be avoided when advanced histology is suspected, but it may not always be possible to predict this reliably prior to resection, and thus, close follow-up, or even further resection, may be required. There remain uncharted frontiers, such as the suitability of CSP for resection of larger lesions and the use of cold snare for piecemeal endoscopic mucosal resection with submucosal injection.