A Pilot Randomized Trial of Polypectomy Techniques for 4-6 mm Colonic Polyps

Colonoscopic polypectomy is one of the most commonly performed procedures in the United States. Despite the frequency with which polypectomy is performed, there is a remarkable absence of data regarding which polypectomy techniques most effectively remove polyps and minimize complications. As a result, polypectomy practice is based on the observational experience of experts and is how it is taught by attending physicians in fellowship training programs. It is not surprising therefore, that polypectomy practice is not consistent across the U.S., particularly for polyps in the 4-6 mm range, with 19% of endoscopists using cold forceps, 21% using hot forceps, 31% using a hot snare, 15% using a cold snare, and the remainder using a combination of these techniques.

The effectiveness of colonoscopy in the prevention of colorectal cancer (CRC) depends on polypectomy; most polyps removed during colonoscopy are < 1 cm in size. Because of their greater prevalence relative to larger polyps, polyps < 1 cm are responsible for most cases of post-polypectomy bleeding (even though the absolute risk is higher with larger polyps). Perhaps more concerning, several studies of surveillance colonoscopy have reported finding CRC within a few years of a complete colonoscopy. Nearly 30% of these cancers have occurred in colonic segments where prior polypectomy was performed, suggesting that incomplete removal may be responsible for the recurrence of neoplasia; in more than half of these cases, the initial polyp removed was 1 cm or less in size. Unfortunately, these studies are limited because the precise location of the polypectomy site relative to the subsequent cancer is uncertain.

Given the prevalence of the 4-6 mm polyp, the variation in polypectomy technique among endoscopists, and the concern about interval cancers at polypectomy sites, there is a clear and significant need to determine which technique(s) are most appropriate for clinical practice in terms of having the lowest risk for recurrent neoplasia and for major adverse effects (major bleeding, perforation).

The goal of this trial is to determine the superior method for polypectomy [cold biopsy forceps (CF), cold snare (CS), or hot snare (HS)] in the removal of adenomatous colonic polyps that are 4-6 mm in maximal diameter for the subsequent risk of recurrent neoplasia at the polypectomy site. A randomized trial comparing these three polypectomy techniques for the outcome of recurrent neoplasia at the polypectomy site requires at least 700 persons per group (at least 2100 persons total). (A trial powered to compare complication rates would require a sample size that is an order of magnitude greater than this and is beyond the feasibility and scope of both pilot and definitive trials.) Further, patients would need to undergo 3- or 5-year surveillance colonoscopy, which requires a research infrastructure to ensure complete and accurate data collection. Before embarking on this large-scale trial to examine effectiveness of polypectomy techniques, we must first demonstrate feasibility of conducting such a trial and determine the time frame, sampling frame, and resources required.

The specific aim of this proposal is to conduct a pilot randomized trial to establish:

the feasibility of conducting a definitive, single-institution, multi-site clinical trial comparing the three modalities for removing adenomatous colonic polyps 4-6 mm. The processes that will be assessed are patient enrollment, recruitment, exclusion, and determination of final eligibility for the trial.
the numbers of patients at each site (IU Hospital, Wishard Memorial Hospital/Eskenazi Health Hospital) who would be candidates for the larger, definitive study (This information will help estimate the time line and resources required for the definitive study).
the proportions of recruited patients that fall into the 3- and 5-year surveillance groups (or other surveillance interval, along with the reason(s) for the interval).

Hypotheses:

A single-institution, three-site clinical trial is feasible: patients can be recruited and enrolled as potential subjects; final eligibility based on endoscopic and histological findings can be established within a week of enrollment.
Investigators can identify the number of patients at each site who would be eligible for participation in a larger, more definitive study, from all patients undergoing colonoscopy. Investigators can establish refusal and exclusion rates. (Knowing both rates is necessary for planning the larger study.)
The proportion of recruited patients in the 5-year surveillance group is approximately 80% of all persons with neoplasia; the remaining 20% comprise the 3-year surveillance group.

MATERIALS AND METHODS:

The investigators conducted an randomized controlled trial (RCT) comparing polypectomy using CF, CS and HS. This trial was approved by the Institutional Review Board at Indiana University on March 6, 2008. It was conducted at three endoscopy units staffed by Indiana University Medical Center Gastroenterology faculty: Indiana University Hospital, Springmill endoscopy center and Eskenazi Health Hospital. Patients were recruited between September 2009 and May 2013 and were followed with surveillance colonoscopy until October 2019. Sample size for the study was estimated based on the primary outcome of recurrent neoplasia at the polypectomy site at surveillance.

Outpatients between ages 18- and 70 years undergoing outpatient screening, surveillance, or diagnostic colonoscopy and able to provide informed consent were eligible. Eligible patients presenting for colonoscopy, at 1 of 3 endoscopy units, from July 2009 to May 2013 were invited to participate in the study by research coordinators (RC).

Colonoscopy was performed in standard fashion with Olympus 180 series colonoscopes. If the subject was found to have ≥1 adenomatous colon polyps (as predicted by endoscopist) 4-6 mm in size with Paris morphology of types I or Paris morphology IIa in the colon, randomization to one of three polypectomy methods was done in a 1:1:1 ratio.

The following parameters for each study polyp (SP) were recorded: a) location, b) size, c) morphology (flat, sessile, or pedunculated), d) method of removal and e) pathology. For CF polypectomy, standard-sized radial jaw 4 biopsy forceps with a 2.8 mm needle were used and the number of bites for complete polypectomy was recorded. Polypectomy time was measured from time of appearance of forceps through the endoscope channel to polyp retrieval. For CS and HS polypectomy, 11 mm snares were used and the number of times the snare was used to complete the polypectomy was recorded. The size of the polyps was assessed using the span of the open radial jaw 4 forceps (7 mm) and the diameter of the snare catheter (2.5 mm) or the open snare diameter (11 mm). For any single patient, up to 5 SPs were removed under research protocol. HS polypectomy was performed using monopolar forced coagulation current, wattage range 18-20. There was no limit to the non-study polyps (NSP) resected.

The time to perform CS polypectomy was recorded from the time of appearance of snare through endoscope channel to polyp retrieval or the determination that the polyp was not retrievable. For HS, time was recorded from initiation of cautery set up to polyp retrieval or the determination that the polyp was not retrievable. For any single patient all SPs were removed using the same technique. Patients with non-neoplastic polyps were excluded post hoc from the analytical sample. Patients were also excluded post randomization if the assigned polypectomy method could not be used or if additional methods were required.

Following polyp resection and retrieval, 2-5 ml of tattoo ink (SPOTⓇ Ex stain) was injected 1-2 cm to the left and proximal to the polypectomy site. The surveillance interval was determined according to guidelines as follows: a) For 1-2 tubular adenomas (TA), 5-10 years; b) For 3 or more TAs, 3 years; c) For a TA >= 1 cm, a polyp with villous histology, or one with high-grade dysplasia, 3 years; d) For a large sessile polyp removed piecemeal, 3-6 months.

During surveillance colonoscopy, the previous polypectomy site was examined. Adequacy of polypectomy was determined by absence of recurrent adenoma tissue. Biopsy of the previous polypectomy site was not required unless neoplasia recurrence was suspected based on visual inspection.

Data were collected using standardized data collection forms, and the information was transferred to an Excel spreadsheet (Version 16) for analysis.