Randomized Controlled Trial Investigating Cold Snare and Forceps Polypectomy Among Small POLYPs in Rates of Complete Resection: The TINYPOLYP Trial

INTRODUCTION

Colonoscopy is the most commonly used method of colorectal cancer screening (1). Colonoscopy with polypectomy has been shown to reduce risk of death from cancer (2). Optimizing complete resection is important because residual neoplastic tissue may play a role in interval cancers (3). The US Multi-Society Task Force on Colorectal Cancer and the European Society of Gastrointestinal Endoscopy in 2020 and 2017, respectively, have provided updated guidelines on polypectomy (4,5). In particular, the USMSTF and European Society of Gastrointestinal Endoscopy recommend that both diminutive (≤5 mm) and small (6–9 mm) polyps be removed by cold snare polypectomy (CSP), citing higher complete resection rates (4,5).

However, the evidence is less clear whether CSP retains a significant advantage over cold forceps polypectomy (CFP) for polyps ≤3 mm (6). Most studies have not specifically evaluated polyps ≤3 mm (7–9). Kim et al. performed a randomized controlled trial evaluating CSP vs CFP for polyps ≤7 mm and found that although complete resection was higher for CSP for polyps 5–7 mm compared with CFP, complete resection was not statistically different for polyps ≤4 mm (8).

Furthermore, there has been conflicting evidence regarding efficacy of CFP for polyps ≤3 mm (9–11). In 1 study by Lee et al. evaluating CSP compared with CFP for polyps ≤5 mm, the complete resection rate was significantly higher for CSP (93.2%) compared with CFP (75.9%) (P = 0.009). In subanalysis provided by the group, the complete resection rate was 91.7% for polyps 1–3 mm in size (10). Another study by Park et al. evaluating CSP and CFP (with narrow-band imaging) for polyps ≤5 mm found a similar complete resection rate between CSP (93.0%) and CFP (90.5%). When evaluating specifically polyps ≤3 mm, the complete resection rate was similar between CSP (92.6%) and CFP (95.8%), meeting their noninferiority definition (12). However, in work by Efthymiou et al. (9), CFP had a complete resection rate of just 50% for polyps <3 mm and 36% for polyps 3–5 mm. Given the small studies with mixed results, to better understand optimal management of polyps ≤3 mm, we sought to pursue a randomized clinical trial, the TINYPOLYP trial.

METHODS Study design

TINYPOLYP is a single-center, prospective randomized noninferiority clinical trial, evaluating the use of cold forceps and cold snare in performing polypectomy for nonpedunculated polyps ≤3 mm. The study protocol was performed in accordance with the Declaration of Helsinki and approved by the institutional review board of Stanford University. The study is registered at ClinicalTrials.gov (NCT04551001).

Patient selection

Adults (aged 18–80 years) who presented to Veterans Affairs Palo Alto Health Care System for outpatient screening, surveillance, or diagnostic colonoscopy were invited to participate between September 16, 2020, and October 15, 2021. Inclusion criteria were patients who had 1 or more polyps of the correct size that were removed. We included patients on anticoagulation and antiplatelet therapy if they were held before the procedure in accordance with American Society for Gastrointestinal Endoscopy guidelines (13). Exclusion criteria were patients aged <18 or >80 years, patients with inflammatory bowel disease or polyposis syndromes, and patients who were pregnant. In addition, patients who did not complete the consent form and who had colonoscopies without polypectomy were not included. Patients with poor bowel preparation (defined as Boston Bowel Preparation Scale [BBPS] <6) were included provided that the patient had polyps ≤3 mm that could be removed without severe interference from the poor bowel preparation.

In our analysis, polyps that were not retrieved or polypectomy site biopsies that were not retrieved were excluded. Patients with only polyps that were not retrieved or whose polypectomy site biopsies were not retrieved were also excluded (Figure 1). A total of 4 polyps were not retrieved, affecting 4 different patients. Two of those patients had only 1 polyp removed, so they were excluded from the study.

F1Figure 1.:

Patient selection for study.

Randomization and concealment

On encountering each polyp of the correct size (≤3 mm), our research coordinator (Y.C.) opened an opaque envelope that revealed whether the polypectomy would be performed through cold snare or cold forceps (14). Polypectomy technique was randomized by block randomization, with random block sizes of 4, 6, and 8, using a computer-generated randomization sequence (15). The assigned technique was concealed in opaque, sealed, sequentially numbered envelopes.

Endoscopists

All procedures were performed or supervised by full-time endoscopists at the VA Palo Alto Health Care System. Each of the staff (S.F., J.Y.P., S.Y.Q, and R.W.) have at least 5 years of experience after completion of fellowship. Fellow involvement was recorded. The attending was present throughout the entirety of the procedure.

Polypectomy and randomization

The endoscopist measured the size of the polyp using the open jaws or shaft of the biopsy forceps (Radial Jaw 4 large capacity with needle, Boston Scientific) or snare (Exacto cold snare, STERIS, US Endoscopy) as reference. The location of the polyp was also recorded during the procedure. Polyps were described using Paris classification (16). En bloc or piecemeal polypectomy was recorded. All polyps were placed in individual containers. After the endoscopist felt they had completed their polypectomy, 2 biopsies were performed at the polypectomy margins and placed in a separate jar (17). Each colonoscopy was video recorded. One author (M.T.W.) reviewed each colonoscopy video and measured the time between the forceps or snare being introduced into the screen until the polyp was completely removed. The number of forceps or snare passes to complete polypectomy was recorded.

Histopathology examination

All polyps were reviewed and classified by an expert gastrointestinal pathologist and her team (C.Y.L., R.B., M.C., K.J., and H.L.). All margin biopsies were reviewed for residual polyp tissue.

The pathology requisition forms had details regarding the polypectomy technique redacted to limit bias. Non-neoplastic polyps included hyperplastic polyps. Neoplastic polyps were categorized as conventional adenomas (tubular, tubulovillous, or villous), traditional serrated adenomas, sessile serrated lesions, or cancer (3).

Study outcomes

The primary outcome measured was completeness of resection. Complete resection was defined as absence of polyp tissue in both polypectomy site margin biopsies. This calculation was applied only to polyps identified on pathology review. Polyps that under pathology appeared to be normal colonic mucosa were excluded. The primary outcome was tested for noninferiority. Secondary outcomes included time required for polypectomy, number of cold forceps or snare attempts to remove polyp completely, use of hemostatic clips, and rates of complications. Complications evaluated included perforation, postpolypectomy bleeding, and postpolypectomy syndrome. Postpolypectomy bleeding was assessed based on visit to the emergency department, need for admission, transfusion, need for repeat colonoscopy (with or without intervention), surgery, and mortality.

Sample size calculations

In pooling the results of studies looking at CFP ≤3 mm, the complete resection rate was 94.1% (10–12,17,18). Similarly, the pooled results for CSP were also 94.1%. Using the pooled results, if we assumed CFP and CSP had the complete resection rate of 94.1%, using alpha of 5%, power of 90%, and noninferiority limit of −10%, we would need a sample size of 96 per group (total 192). If the lower bound of the 95% confidence interval was greater than −10%, noninferiority between CSP and CFP could be concluded (19). Patients were enrolled continuously, and results were tabulated every 2 weeks until minimum enrollment was achieved. The trial ended once the required number of polyps was reached.

RESULTS Patient and procedure characteristics

A total of 179 patients were included in the trial, 106 of whom had polyps removed by CSP, 119 patients who had polyps removed by CFP, and 46 patients who had polyps removed by both CSP and CFP (Table 1). Patients who had polyps removed with CSP had similar age, proportion of men, and race/ethnicity compared with patients who had polyps removed with CFP. Patients in the 2 groups had similar height, weight, and body mass index (BMI), as well as use of anticoagulation or antiplatelet therapy. In addition, there was a similar proportion of patients with cirrhosis (1.9 vs 2.5%, P = 0.747).

T1Table 1.:

Patient characteristics by intervention (with cold snare vs with cold forceps)

Patients who had polyps removed by CSP had similar proportions of procedure indications to patients who had polyps removed by CFP (68.9 vs 73.9% for surveillance colonoscopy; 2.8 vs 2.5% for screening, P = 0.826). A similar proportion of patients underwent colonoscopy with moderate sedation (85.8 vs 90.8%, P = 0.170). Patients in both groups had similar mean BBPS (7.7 vs 7.9, P = 0.209) as well as mean withdrawal time (27.2 vs 25.4 minutes, P = 0.383).

Polyp characteristics

A total of 279 nonpedunculated polyps ≤3 mm in size were identified, 138 removed by CSP and 141 by CFP (Table 2). Mean polyp size was similar between the CSP and CFP groups (2.5 vs 2.6 mm, P = 0.161). Distribution of polyps was similar between the 2 groups (33.3 vs 26.2% in the ascending colon; 26.8 vs 24.8% in the transverse colon, P = 0.119). There was higher proportion of tubular adenomas removed by CSP compared with CFP (79.7 vs 66.0%, P = 0.009). Overall, there was a high proportion of non-neoplastic tissue (20.3% [CSP] vs 34.0% [CFP]). A similar proportion of polyps were classified as Is (93.5 vs 93.6%, P = 0.962). One of the 2-mm polyps (Is) randomized to CFP was found to be Schwann cell on pathology review.

T2Table 2.:

Polyp characteristics by intervention (with cold snare vs with cold forceps)

Outcomes

Polypectomy by cold snare required significantly longer time to perform compared with cold forceps (42.3 vs 23.2 seconds, P < 0.001). However, CFP was significantly more likely to require removal of polyp in more than 1 piece (15.6 vs 3.6%, P < 0.001). Hemostatic clip was deployed for 1 polyp in the CFP group (0.7%) compared with none in the CSP group, although this was not statistically significant (P = 0.322). Excluding polyps in which the pathology returned as normal mucosa, there was no significant difference in positive margin biopsy with CSP (1.7%) or CFP (1.7%) (P = 0.973). There was no significant difference in complete resection (difference in complete resection rates was 0.057%, 95% confidence interval: −4.30% to 4.53%), demonstrating noninferiority of cold forceps compared with cold snare.

On multivariate logistic regression, none of the variables (use of cold snare or forceps, polyp size, polyp location, time of polypectomy, piecemeal resection, polyp pathology, and fellow involvement in polyp resection) were found to be statistically significant factors for complete resection.

Complications

There were no 30-day complications experienced in patients in either group, including perforation, postpolypectomy bleeding, and postpolypectomy syndrome. No patient required management of postpolypectomy bleeding including visit to emergency department, admission, transfusion, need for colonoscopy, or surgery. There were no deaths within 30 days of colonoscopy.

DISCUSSION

CFP is commonly used during polypectomy because of its efficiency (less set-up required, less time required to get the forceps into position, and fewer issues with tissue retrieval). Despite these advantages, The US Multi-Society Task Force recommends use of cold snare over cold forceps in removal of polyps <10 mm and reserves jumbo or large-capacity forceps for polyps ≤2 mm if CSP is technically challenging (20,21). This is due to concern for incomplete resection of polyps, with studies finding low complete resection rates of polyps ≤6 mm (36%–82.4%) and <3 mm (50%) (Table 3) (9,22). However, these studies suffer from low sample size, with Efthymiou et al. including just a sample size of 10 polyps <3 mm. By contrast, several studies including larger sample sizes have identified a complete CFP resection rate >90% for polyps ≤3 mm (8,10–12,17,18). Although further work is needed to better analyze the data yielded in these studies (6), trials comparing CFP with CSP have find that CFP of polyps ≤3 mm achieves a high complete resection rate (>90%), comparable with CSP (8,10,12,17). However, when CFP is applied to polyps 4–5 mm in size, the complete resection rate drops compared with CSP. Overall, although these studies are of great quality, the biggest criticism is the limited sample size. Park et al. features the largest CFP sample size ≤3 mm but is limited to 48 polyps analyzed (12). Given the current guidelines and the existing variable evidence, we embarked on the largest trial to date to compare CFP and CSP for polyps ≤3 mm (6).

T3Table 3.:

Studies evaluating cold forceps polypectomy of polyps ≤3 mm

In this randomized clinical trial, in which we evaluate 279 polyps ≤3 mm, use of CFP (N = 141) has shorter mean polypectomy time compared with CSP (N = 138) while maintaining a high complete resection rate (98.3% in both groups). Specifically, there was no significant difference in complete resection between CFP and CSP, demonstrating noninferiority in CFP compared with CSP.

Based on these results, when an endoscopist encounters a diminutive polyp ≤3 mm, either a cold forceps or cold snare can be used during the procedure. This can help reduce cost and time, especially when the cold forceps is used for other purposes during the procedure (such as ruling out microscopic colitis and performing inflammatory bowel disease surveillance). Our study measured polypectomy time based on the time when the snare or forceps were introduced into view till when the polyp was removed and does not account for time required to set up the polyp trap. Although our study does not address this question directly, given that use of cold forceps allowed for faster polypectomy compared with cold snare, one can extrapolate that overall procedure time is likely to be shorter, leading to reduced sedation risk of the patient as well as improved cost savings for the endoscopist. Another benefit of using cold forceps is its effective retrieval of tissue. Every endoscopist has encountered the frustrating experience of performing a snare polypectomy but not successfully retrieving the tissue. This scenario can be successfully eliminated by use of the forceps if within appropriate size limits.

There were several possible limitations to our study. Our study was performed at a single-center veterans affairs (VA) population, which loses the patient diversity (≥95% male) of a multicenter trial. However, our study uses 4 experienced board-certified endoscopists with at least 4 years of experience after completion of fellowship. Further by pursuing a randomized controlled trial, we were able to control for patient differences, as evidenced by similar patient characteristics including age, race/ethnicity, height, weight, BMI, and procedural indication. As such, we feel the results identified should be generalizable to the general patient population. An additional limitation was that the endoscopists were not completely blinded to the technique performed because they were told by our research coordinator (Y.C.) which technique (CFP or CSP) to be used for polypectomy. Nevertheless, blinding was provided whenever feasible in this study. The endoscopists only knew which polypectomy technique to perform after they identified the polyp ≤3 mm, at which point an opaque envelope was opened describing the assigned polypectomy technique. Because the allocation to CFP vs CSP was performed after identification of the polyp, we believe that the difference in histology between the 2 arms—specifically, the higher proportion of hyperplastic polyps in the CFP group—are likely due to chance. Furthermore, the pathology requisition form was redacted of information describing the polypectomy technique, helping to limit bias during histologic review. An additional limitation was the concern that polyp sizing may be inaccurate (23,24). Similar to other polypectomy studies, we attempt to mitigate this issue by using the open forceps or open snare as a ruler to compare the size of the polyp.

The prespecified noninferiority margin of 10% was another possible limitation. If a noninferiority margin of 5% were considered, then a total of 762 polyps would be needed. This would have significantly extended the study and increased resources, while likely not changing the outcome of the study, given the high complete resection rate of the polyps using CFP and CSP. One concern was the technique used to confirm completeness of resection. In our study, we used the double biopsy technique, which has been used in several other studies (10,12,17,22). In other studies, endoscopic mucosal resection of the polypectomy site was performed and sent for pathologic analysis (8,9,11,18). With the exception of Efthymiou et al. (9) and Draganov et al. (22), studies 2013 and onward find a complete resection rate >90%, regardless of histologic criteria used. As such, we felt either technique could be reasonable for evaluation of completeness of polyp resection.

Theoretical mechanisms for false-positive errors in the assessment of incomplete resection include the chance presence of an unnoticed microscopic second polyp very near the resected polyp or contamination of the margin biopsy tissue sample by leftover tissue from the original resection in the biopsy channel or forceps. We believe that these types of events would be quite rare and unlikely to influence the results of the study significantly; in particular, care was taken to clear the biopsy channel by suctioning water after each resection and to thoroughly clean the forceps after polypectomy. Although large-capacity forceps were used in this study, other studies have shown that standard-capacity forceps may also be reasonable in the resection of ≤3-mm polyps. However, evidence for ≥4 mm is less clearly favoring standard-capacity forceps, with Lee et al. (10) identifying just 50.0% complete resection rate, although the sample size was 22 polyps. Although work by Huh et al. found that jumbo forceps achieve a comparable resection rate compared with CSP for polyps 4–5 mm at around 90%, we sought to investigate the utility of large-capacity forceps in polypectomy of ≤3-mm polyps (17). Based on our results, the larger jumbo forceps are likely to also achieve high complete resection rates. Further studies are needed to evaluate the use of different types of forceps and different sizes of polyps. Overall, the procedures performed featured good BBPS (only 5 patients with inadequate BBPS), mean withdrawal time >25 minutes, and involved endoscopists with an adenoma detection rate >40%. Although these are considered high-quality examinations, we feel the results for diminutive polypectomy should be generalizable to colonoscopies performed by all endoscopists, given how fundamental cold forceps usage is.

In the largest randomized clinical trial to date evaluating cold forceps and cold snare removal of nonpedunculated polyps ≤3 mm, we found that CFP met the noninferiority criteria for complete resection compared with CSP. Furthermore, CFP required significantly less time for resection compared with cold snare. The results of TINYPOLYP suggest that polypectomy of ≤3-mm polyps can be safely performed with cold forceps or cold snare. Further studies are needed to better evaluate utility of jumbo cold forceps in the removal of polyps 4–5 mm.

CONFLICTS OF INTEREST

Guarantor of the article: Mike T. Wei, MD.

Specific author contributions: All authors participated meaningfully to the production of this manuscript.

Financial support: None to report.

Potential competing interests: S.F. is a consultant for Capsovision and Intuitive Surgical, and M.T.W. is a consultant for Neptune Medical and AgilTx.

Study Highlights

WHAT IS KNOWN ✓ The USMSTF and ESGE recommend that both diminutive (<5 mm) and small (6–9 mm) polyps be removed by cold snare polypectomy, citing higher complete resection rates. ✓ The evidence remains unclear whether cold snare polypectomy retains a significant advantage over cold forceps polypectomy for polyps <3 mm.

WHAT IS NEW HERE ✓ In the largest randomized clinical trial to date evaluating polyps ≤3 mm, we find that high complete resection rate in both CFP and CSP (98.3% in both groups). ✓ We find no significant difference in complete resection rates between CFP and CSP, and demonstrate non-inferiority of cold forceps compared to cold snare in resection of polyps ≤3 mm polyps. ✓ We find that mean polypectomy time was longer for CSP compared to CFP. ✓ We find a higher proportion of polyps removed by cold forceps were removed in more than one piece compared to cold snare. REFERENCES 1. Franco DL, Leighton JA, Gurudu SR. Approach to incomplete colonoscopy: New techniques and technologies. Gastroenterol Hepatol (N Y) 2017;13(8):476–83. 2. Zauber AG, Winawer SJ, O'Brien MJ, et al. Colonoscopic polypectomy and long-term prevention of colorectal-cancer deaths. N Engl J Med 2012;366(8):687–96. 3. Pohl H, Srivastava A, Bensen SP, et al. Incomplete polyp resection during colonoscopy-results of the complete adenoma resection (CARE) study. Gastroenterology 2013;144(1):74–80.e1. 4. Kaltenbach T, Anderson JC, Burke CA, et al. Endoscopic removal of colorectal lesions: Recommendations by the US Multi-Society Task Force on Colorectal Cancer. Am J Gastroenterol 2020;115(3):435–64. 5. Ferlitsch M, Moss A, Hassan C, et al. Colorectal polypectomy and endoscopic mucosal resection (EMR): European Society of Gastrointestinal Endoscopy (ESGE) clinical guideline. Endoscopy 2017;49(3):270–97. 6. Djinbachian R, Iratni R, Durand M, et al. Rates of incomplete resection of 1- to 20-mm colorectal polyps: A systematic review and meta-analysis. Gastroenterology 2020;159(3):904–14.e12. 7. Matsuura N, Takeuchi Y, Yamashina T, et al. Incomplete resection rate of cold snare polypectomy: A prospective single-arm observational study. Endoscopy 2017;49(3):251–7. 8. Kim JS, Lee BI, Choi H, et al. Cold snare polypectomy versus cold forceps polypectomy for diminutive and small colorectal polyps: A randomized controlled trial. Gastrointest Endosc 2015;81(3):741–7. 9. Efthymiou M, Taylor AC, Desmond PV, et al. Biopsy forceps is inadequate for the resection of diminutive polyps. Endoscopy 2011;43(4):312–6. 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. Am J Gastroenterol 2013;108(10):1593–600. 11. Jung YS, Park JH, Kim HJ, et al. Complete biopsy resection of diminutive polyps. Endoscopy 2013;45(12):1024–9. 12. Park SK, Ko BM, Han JP, et al. A prospective randomized comparative study of cold forceps polypectomy by using narrow-band imaging endoscopy versus cold snare polypectomy in patients with diminutive colorectal polyps. Gastrointest Endosc 2016;83(3):527–32.e1. 13. Acosta RD, Abraham NS, Chandrasekhara V, et al. The management of antithrombotic agents for patients undergoing GI endoscopy. Gastrointest Endosc 2016;83(1):3–16. 14. Kim J, Shin W. How to do random allocation (randomization). Clin Orthop Surg 2014;6(1):103–9. 16. The Paris endoscopic classification of superficial neoplastic lesions: Esophagus, stomach, and colon: November 30 to December 1, 2002. Gastrointest Endosc 2003;58(6 Suppl):S3–43. 17. Huh CW, Kim JS, Choi HH, et al. Jumbo biopsy forceps versus cold snares for removing diminutive colorectal polyps: A prospective randomized controlled trial. Gastrointest Endosc 2019;90(1):105–11. 18. O'Connor SA, Brooklyn TN, Dunckley PD, et al. High complete resection rate for pre-lift and cold biopsy of diminutive colorectal polyps. Endosc Int Open 2018;6(2):E173–8. 19. Papastergiou V, Paraskeva KD, Fragaki M, et al. Cold versus hot endoscopic mucosal resection for nonpedunculated colorectal polyps sized 6-10 mm: A randomized trial. Endoscopy 2018;50(4):403–11. 20. Kaltenbach T, Anderson JC, Burke CA, et al. Endoscopic removal of colorectal lesions: Recommendations by the US Multi-Society Task Force on Colorectal Cancer. Gastroenterology 2020;158(4):1095–129. 21. Kaltenbach T, Anderson JC, Burke CA, et al. Endoscopic removal of colorectal lesions: Recommendations by the US Multi-Society Task Force on Colorectal Cancer. Gastrointest Endosc 2020;91(3):486–519. 22. 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(1):118–26. 23. Anderson BW, Smyrk TC, Anderson KS, et al. Endoscopic overestimation of colorectal polyp size. Gastrointest Endosc 2016;83(1):201–8. 24. Utsumi T, Horimatsu T, Nishikawa Y, et al. Factors associated with inaccurate size estimation of colorectal polyps: A multicenter cross-sectional study. J Gastroenterol Hepatol 2021;36(8):2224–9.

留言 (0)

沒有登入
gif