Polyp resection – lessons learned

The ultimate benefit of colorectal cancer screening relies on the detection and resection of neoplastic polyps. Reports on cancers that are found following a colonoscopy illustrate that this test is far from perfect. There is little doubt that missed lesions contribute to the majority of these so-called interval cancers, with some authors estimating that 70 % – 80 % of interval cancers are a result of a missed lesion [1] [2]. Incomplete polyp resection has been blamed for 10 % – 27 % of interval cancers, while a minority may be related to fast growing de novo cancers [1] [3].

Finding a cancer or any lesion soon after a colonoscopy is disappointing and humbling, but it also reminds us of our responsibility as colonoscopists to do our utmost to detect all neoplastic lesions and to remove them completely. Whereas adenoma detection has been the main quality indicator for the performance of colonoscopy, little attention has been given to completeness of adenoma resection.

The study by Jung et al. in this issue of Endoscopy contributes to the knowledge on polyp resection [4]. The authors examined the rate of completely resected diminutive polyps ( ≤ 5 mm) using a cold biopsy forceps for 86 polyps in 65 patients. Overall, 91 % of polyps and 92 % of adenomatous polyps were completely resected. The complete resection rate appeared to be higher for polyps ≤ 3 mm than for polyps 4 – 5 mm. In total, 100 % (18 /18) of 1 – 3 mm and 89 % (42 /47) of 4 – 5 mm adenomas were completely removed (P = 0.311). The authors concluded that cold biopsy forceps removal of diminutive polyps appears to be adequate for most diminutive polyps, especially for those ≤ 3 mm.

This study has several strengths. First, the authors attempted to choose an optimal approach to ensure macroscopically complete polyp resection. Multiple bites were permitted and chromoendoscopy with indigo carmine was applied to visualize and remove any remaining polyp tissue. Second, to guarantee complete sampling, the resection margin was removed by endoscopic mucosal resection. Third, a thorough pathological examination with thin cuts through the resection specimen was performed to find any remaining polyp tissue in the margin.

At first glance, the results appear positive and suggest that diminutive polyps can be adequately removed with a cold biopsy forceps. However, it is important to provide a few notes of caution and point out some of the shortcomings of the study.

Our main interest should be complete resection of neoplastic polyps. Similar to previous studies, the authors focused their primary analysis on any polyp, and only 65 adenomas were available for analysis [6] [7] [8] [11] [12]. The computed complete adenoma resection rate therefore comes with a wide 95 % confidence interval, from 83 % to 97 % – a range that probably provides a more realistic idea of the true complete resection rate. Simply put, the wider the confidence interval, the less generalizable the finding.

Furthermore, the study was performed by one endoscopist only. Complete resection rates varied broadly across endoscopists in the recent complete adenoma resection (CARE) study, from 77 % to 94 % [5]. A single endoscopist study combined with a nonblinded design leads to potential bias. Even under study conditions and a meticulous examination of the resection site, a fairly large proportion of adenomatous polyps (8 %) were not completely removed.

The study suggested that smaller polyps were more often completely removed than larger polyps. Picturing a typical biopsy forceps of 2.4 mm (or 2.8 mm) in diameter, complete resection of 1 – 2 mm polyps with one bite seems plausible. Recognizing and engaging minor remaining polyp tissue after a first or second bite of larger polyps may be more of a challenge. Although the study suggests that forceps removal of 1 – 3 mm polyps is reasonable, it does not provide sufficient evidence to enable this practice to become the preferred approach.

Few studies have examined resection of adenomatous polyps ([Table 1]) [4] [5] [6] [7] [8] [9] [10]. The range of incomplete resection of diminutive adenomas using a cold forceps ranged from 8 % to 38 %. The CARE study examined 5 – 20-mm neoplastic polyps and found an incomplete resection rate of 10 % overall. Incomplete resection increased with size (7 % for 5 – 9 mm polyps and 17 % for ≥ 10 mm polyps), and was highest in sessile serrated adenomas/polyps (33 %). Although endoscopists were asked to confirm visibly complete resection before obtaining marginal biopsies, no special technique (e. g. with chromoendoscopy or narrow band imaging) was mandatory [5]. A smaller study, which published its data in abstract form, suggested that complete resection can be achieved in all adenomas, if the polyp is removed with a 1 – 2 mm rim of healthy mucosa [9]. In a recent study on 117 diminutive polyps, including non-neoplastic polyps, cold snare resection resulted in fewer incomplete resections than cold forceps resection (7 % vs. 24 %) [11]. Incomplete resection was also more common with larger 4-5 mm polyps as compared to 1 – 3 mm polyps (50 % vs. 8 %).

Table 1 Studies on polyp resection that provided data on resection of adenomatous polyps up to 20 mm.

Study [ref]	No. of polyp	Adenomas size, mm	Resection method	Assessment of resection	Incomplete resection
					%	(95 %CI*)
Woods 1989 [6]	2	≤ 5	Cold forceps	3-week follow-up	18	[5 – 40]
Humphris 2009 [10]	44	≤ 5	Cold snare	Marginal biopsies	11	[4 – 25]
Efthymiou 2011 [7]	21	≤ 5	Cold forceps	Marginal EMR	38	[18 – 62]
Liu 2012 [8]	37	Any	Any	Marginal biopsies	19	[8 – 35]
Urquhart 2012 [9] (abstract)	50	Any	Cold snare	Marginal biopsies	0	[0 – 7]
Pohl 2013 [5]	346	5 – 20	Hot snare	Marginal biopsies	10	[7 – 13]
Jung 2013 [4]	65	≤ 5	Cold forceps	Marginal EMR	8	[3 – 17]

CI, confidence interval; EMR, endoscopic mucosal resection.
95 %CI was calculated using the absolute numbers obtained from the original studies.

Several important questions should be considered in future studies.

• What is an accepted threshold for incomplete (or complete) resection?

• Does remaining polyp tissue matter? Minute polyp tissue may take longer to regrow, or might even regress. It is also plausible (yet unknown) that resecting tissue may accelerate growth.

• What is the best way to check for completeness of polyp resection? Some investigators have sampled the resection margin; others examined the site at follow-up ([Table 1]). Although pathology evaluation of the polyp margin for completeness of resection is mandatory in some countries, it is unclear whether this truly correlates with complete polyp removal.

• What size of polyps should we target in studies of complete resection? With improved abilities to diagnose polyps real time, some may argue that we should forgo resection in diminutive polyps altogether and focus on higher risk large polyps – those that really matter.

• What quality measures can we use to assess complete resection?

In summary, the study by Jung et al. contributes to the limited knowledge on complete resection of adenomatous polyps. The results, however, are too preliminary to formulate practice standards. Considering that incomplete resection likely contributes to interval cancers, future studies are needed to identify the best resection practice and introduce quality metrics to assure complete resection. To answer these questions, researchers should be clear about their primary outcome of interest (adenomas vs. any polyp), enroll a sufficient number of patients based on an adequate sample size calculation, and include several endoscopists to allow generalizability of the results.

• References

• 1 Robertson DJ, Lieberman DA, Winawer SJ et al. Colorectal cancers soon after colonoscopy: a pooled multicohort analysis. Gut 2013; DOI: 10.1136/gutjnl-2012-303796.
• 2 Pohl H, Robertson DJ. Colorectal cancers detected after colonoscopy frequently result from missed lesions. Clin Gastroenterol Hepatol 2010; 8: 858-864
• 3 Farrar WD, Sawhney MS, Nelson DB et al. Colorectal cancers found after a complete colonoscopy. Clin Gastroenterol Hepatol 2006; 4: 1259-1264
• 4 Jung YS, Park JH, Kim HJ et al. Complete biopsy resection of diminutive polyps. Endoscopy 2013; 45
• 5 Pohl H, Srivastava A, Bensen SP et al. Incomplete polyp resection during colonoscopy – results of the complete adenoma resection (CARE) study. Gastroenterology 2013; 144: 74-80
• 6 Woods A, Sanowski RA, Wadas DD et al. Eradication of diminutive polyps: a prospective evaluation of bipolar coagulation versus conventional biopsy removal. Gastrointest Endosc 1989; 35: 536-540
• 7 Efthymiou M, Taylor AC, Desmond PV et al. Biopsy forceps is inadequate for the resection of diminutive polyps. Endoscopy 2011; 43: 312-316
• 8 Liu S, Ho SB, Krinsky ML. Quality of polyp resection during colonoscopy: are we achieving polyp clearance?. Dig Dis Sci 2012; 57: 1786-1791
• 9 Urquhart P, Brown GJ. The effectiveness of cold snare polypectomy for the removal of small sessile solonic polyps. Gastrointest Endosc 2012; 75 : AB328
• 10 Humphris JL, Tippett J, Kwok A et al. Cold Snare Polypectomy for Diminutive Polyps: An Assessment of the Risk of Incomplete Removal of Small Adenomas. Gastrointest Endosc 2009; 69: AB207
• 10 Lee CK, Shim JJ, Jang JY. Cold snare polypectomy vs. Cold forceps polypectomy using double-biopsy technique for removal of diminutive colorectal polyps: a prospective randomized study. Gastroenterol 2013; 108: 1593-1600
• 10 Draganov PV, Chang MN, Alkhasawneh A et al. Randomized, controlled trial of standard, large-capacity versus jumbo biopsy forceps for polypectomy of small, sessile, colorectal polyps. Gastrointest Endosc 2012; 75: 118-126