Safety of first surveillance colonoscopy at 12 months after piecemeal endoscopic mucosal resection of large non-pedunculated colorectal lesions.

BACKGROUND AND AIMS: After piecemeal endoscopic mucosal resection (pEMR) of nonpedunculated colorectal lesions ≥ 20 mm, guidelines recommend first endoscopic surveillance at 6 months. However, initial surveillance at 12 months may be adequate for selected low-risk lesions, and could save the cost, risk and inconvenience of one surveillance examination. METHODS: We retrospectively examined a prospectively collected database of all colorectal lesions referred to our center for endoscopic resection between August 2019 and April 2023. We report recurrence rates of colorectal lesions ≥ 20 mm removed by pEMR who were assigned to 6-month first surveillance or assigned to 12-month first surveillance (or assigned to 6-month but did not return until after 10 months). RESULTS: There were 561 nonpedunculated lesions ≥ 20 mm that underwent first follow-up, including 490 lesions in 443 patients assigned to 6-month, and 71 lesions in 65 patients assigned to 12-month surveillance. Lesions assigned to 12-month surveillance were smaller (mean size 25.9 ± 6.1mm vs. 37.0 ± 17.4mm), more likely serrated (63.4% vs. 9.6%), and more often removed by cold pEMR (74.6% vs 20.4%). Twenty-nine lesions in 24 patients assigned 6-month surveillance presented after 10 months and their recurrence data were included in the group assigned 12-month surveillance. Overall recurrence rates at 6 months and 12 months were 10.0% (46/461) and 9.0% (9/100), respectively. Mean recurrence sizes at 6 and 12 months were 10.9 ± 6.2mm and 4.2 ± 1.9mm, respectively. One patient in the 6-month surveillance group had cancer at the pEMR site, but no other recurrences at 6 or 12 months had either cancer or high-grade dysplasia. CONCLUSION: Twelve-month surveillance appears acceptable for selected colorectal lesions ≥ 20 mm removed by pEMR. A randomized trial comparing initial 6-month to 12-month surveillance is warranted for selected lesions.

Audit of hemostatic clip use after colorectal polyp resection in an academic endoscopy unit.

Background and study aims Prophylactic closure of endoscopic resection defects reduces delayed hemorrhage after resection of non-pedunculated colorectal lesions ≥ 20 mm that are located proximal to the splenic flexure and removed by electrocautery. The risk of delayed hemorrhage after cold (without electrocautery) resection is much lower, and prophylactic clip closure after cold resection is generally unnecessary. The aim of this study was to audit clip use after colorectal polyp resection in routine outpatient colonoscopies at two outpatient centers within an academic medical center. Patients referred for resection of known lesions were excluded. Patients and methods Retrospective chart analysis was performed as part of a quality review of physician adherence to screening and post-polypectomy surveillance intervals. Results Among 3784 total lesions resected cold by 29 physicians, clips were placed after cold resection on 41.7% of 12 lesions ≥ 20 mm, 19.3% of 207 lesions 10 to 19 mm in size, and 2.8% of 3565 lesions 1 to 9 mm in size. Three physicians placed clips after cold resection of lesions 1 to 9 mm in 18.8%, 25.5%, and 45.0% of cases. These physicians accounted for 8.1% of 1- to 9-mm resections, but 69.7% of clips placed in this size range. Electrocautery was used for 3.1% of all resections. Clip placement overall after cold resection (3.9%) was much lower than after resection with electrocautery (71.1%), but 62.4% of all clips placed were after cold resection. Conclusions Audits of clip use in an endoscopy practice can reveal surprising findings, including high and variable rates of unnecessary use after cold resection. Audit can potentially reduce unnecessary costs, carbon emissions, and plastic waste.

One-device colonoscopy: feasibility, cost savings, and plastic waste reduction by procedure indication, when performed by a high detecting colonoscopist.

BACKGROUND: Cold forceps and snares are each effective for removing polyps of 1-3 mm, while snares are more effective for polyps of 4-10 mm in size. If, in the same patient, polyps of 1-3 mm are removed with forceps and those of 4-10 mm with snares, two devices are used. If cold snares are used to resect all lesions of 1-10 mm (one-device colonoscopy), there is a potential for lower costs and less plastic waste. METHODS: A single high detecting colonoscopist prospectively measured the feasibility of cold snaring all colorectal lesions of ≤10 mm in size, along with the associated costs and plastic waste reduction. RESULTS: 677 consecutive lower gastrointestinal endoscopies (not for inflammatory bowel disease) were assessed. Of 1430 lesions of 1-3 mm and 1685 lesions of 4-10 mm in size, 1428 (99.9%, 95%CI 99.5%-100%) and 1674 (99.3%, 95%CI 98.8%-99.7%), respectively, were successfully resected using cold snaring. Among 379 screening and surveillance patients, universal cold snaring of lesions ≤10 mm saved 35 and 47 cold forceps per 100 screening and surveillance patients, respectively. CONCLUSION: Cold snare resection of all lesions ≤10 mm (one-device colonoscopy) was feasible, and reduced costs and plastic waste.

Snare Tip Soft Coagulation vs Argon Plasma Coagulation vs No Margin Treatment After Large Nonpedunculated Colorectal Polyp Resection: a Randomized Trial.

BACKGROUND & AIMS: Thermal treatment of the defect margin after endoscopic mucosal resection (EMR) of large nonpedunculated colorectal lesions reduces the recurrence rate. Both snare tip soft coagulation (STSC) and argon plasma coagulation (APC) have been used for thermal margin treatment, but there are few data directly comparing STSC with APC for this indication. METHODS: We performed a randomized 3-arm trial in 9 US centers comparing STSC with APC with no margin treatment (control) of defects after EMR of colorectal nonpedunculated lesions ≥15 mm. The primary end point was the presence of residual lesion at first follow-up. RESULTS: There were 384 patients and 414 lesions randomized, and 308 patients (80.2%) with 328 lesions completed ≥1 follow-up. The proportion of lesions with residual polyp at first follow-up was 4.6% with STSC, 9.3% with APC, and 21.4% with control subjects (no margin treatment). The odds of residual polyp at first follow-up were lower for STSC and APC when compared with control subjects (P = .001 and P = .01, respectively). The difference in odds was not significant between STSC and APC. STSC took less time to apply than APC (median, 3.35 vs 4.08 minutes; P = .019). Adverse event rates were low, with no difference between arms. CONCLUSIONS: In a randomized trial STSC and APC were each superior to no thermal margin treatment after EMR. STSC was faster to apply than APC. Because STSC also results in lower cost and plastic waste than APC (APC requires an additional device), our study supports STSC as the preferred thermal margin treatment after colorectal EMR. (Clinicaltrials.gov, Number NCT03654209.).

Frequency of serrated polyposis syndrome recognition by community endoscopists.

Background and study aims Some data indicate serrated polyposis syndrome (SPS) is underdiagnosed. We determined the frequency of SPS diagnosis by community endoscopists prior to referral to a tertiary center. Patients and methods We performed a retrospective analysis of a prospectively collected database of SPS patients at a tertiary academic hospital. There were 212 patients who were referred to our center for resection of one or more lesions detected at a prior colonoscopy and who had records available that allowed determination of whether SPS was diagnosed before referral. Results Only 25 of 212 patients (11.8%) had a diagnosis or suspicion of a polyposis syndrome prior to referral, and only 12 patients (5.7%) had a specific SPS diagnosis made prior to referral. Among 187 patients diagnosed at our center, 39 had sufficient serrated lesions removed and documented in outside records to meet SPS criteria prior to referral, but the diagnosis was not made by the referring physician despite adequate numbers of lesions resected. The remaining cases required lesions removed at our center to meet SPS diagnostic criteria. Limitations were a single center, single expert endoscopist. Conclusions SPS is the most common colorectal polyposis syndrome, but it remains underdiagnosed by community endoscopists. Underdiagnosis may contribute to post-colonoscopy colorectal cancer in patients with SPS.

Characterization of endoscopic features and histology of a distinct mucosal transition zone on the ileocecal valve (with video).

BACKGROUND AND AIMS: We have endoscopically encountered a zone of transitional mucosa between the colonic and ileal mucosa located in a 3- to 10-mm-wide ring around the ileocecal valve (ICV) orifice. We aimed to describe the features of the ICV transitional zone mucosa. METHODS: We used videos and photographs from normal ICVs and biopsy samples from normal colonic mucosa, transitional zone mucosa, and normal ileal mucosa to characterize the endoscopic and histologic features of the ICV transitional zone mucosa. RESULTS: The ICV transitional zone is identifiable on every ICV without a circumferential adenoma or inflammation that obliterates the zone. The zone is characterized endoscopically by an absence of villi, which distinguishes it from the ileal mucosa, but the pits are more tubular and with more prominent blood vessels compared with normal colonic mucosa. Histologically, the villi of the transitional zone are blunted, and the amount of lymphoid tissue is intermediate between the colonic mucosa and ileal mucosa. CONCLUSIONS: This is the first description of the normal transitional zone of mucosa on the ICV. This zone has unique endoscopic features that should be recognized by colonoscopists and that can potentially create difficulty in identifying the margins of adenomas located on the ICV.

Impact of changing diagnostic criteria on the diagnosis of serrated polyposis syndrome.

Background and study aims The World Health Organization criteria for serrated polyposis syndrome (SPS) were established in 2010 and modified in 2019. Neither set of criteria have been validated against genetic markers or proven to be the optimal criteria for defining colorectal cancer risk in patients with serrated colorectal lesions. In this study, we sought to gain insight into how frequently the change in SPS criteria in 2019 impacted the diagnosis of SPS. Patients and methods We reviewed 279 patients with SPS diagnosed between 2010 and 2019 using the 2010 criteria (n = 163) or since 2019 using the 2019 criteria (n = 116). We reviewed whether patients in each group met the diagnosis of SPS by the alternative criteria. Results Of those diagnosed using 2010 criteria, 5.5 % did not meet 2019 criteria. Of those diagnosed by 2019 criteria, 10.3 % did not meet 2010 criteria. Conclusions Most patients with SPS in our database met the diagnosis of SPS by both 2010 and 2019 criteria, with only 5 % to 10 % of patients in each cohort not meeting the alternative diagnostic criteria.

Impact of margin thermal treatment after EMR of giant (≥40 mm) colorectal lateral spreading lesions.

BACKGROUND AND AIMS: Increasing lesion size is a risk factor for recurrence after piecemeal EMR (pEMR). Snare-tip soft coagulation (STSC) treatment of the normal-appearing margin after pEMR of lesions ≥ 20 mm has been shown to reduce recurrence rates by 75% to 80%. We sought to evaluate the impact of STSC on giant (≥ 40 mm) lateral spreading lesions treated by pEMR. We describe the relative risk and absolute risks of recurrence with and without STSC margin treatment after EMR of ≥ 40-mm lesions. METHODS: We performed a retrospective evaluation of a prospectively collected database on large lesions describing lesion size, location, and methods of resection. We excluded lesions < 40 mm in maximum dimension, those that did not undergo follow-up care at our center, and those in which argon plasma coagulation was used for either ablative treatment of residual polyp or margin treatment. Propensity score analysis was used to account for potential differences between patients treated with and without STSC. RESULTS: There were 68 lesions ≥ 40 mm removed by pEMR without STSC treatment and 133 removed and treated with STSC. There were no differences between groups in demographics, polyp size, location, histologic features, and mean follow-up time. The recurrence rate in the no-treatment group was 35% versus 9% with STSC (P < .00001 by direct comparison and P = .008 by using the propensity score analysis). CONCLUSION: STSC treatment after pEMR of large lateral spreading lesions in the colorectum reduced recurrences by 75%. However, the absolute recurrence rate of 9% remained clinically significant in the STSC-treated group. Short-term follow-up care after STSC of lesions ≥ 40 mm is still warranted, and additional study of technical factors that eliminate recurrence after pEMR of giant lateral spreading lesions is warranted.

Prevalence of synchronous neoplasia in patients with large pedunculated colorectal polyps.

BACKGROUND: Large (≥ 20 mm) nonpedunculated colorectal lesions have high rates of synchronous neoplasia and advanced neoplasia. Synchronous neoplasia prevalence in patients with large pedunculated lesions is uncertain. We describe synchronous neoplasia in patients with large pedunculated colorectal polyps, using a cohort of patients with large nonpedunculated lesions as controls. METHODS: This study was a retrospective assessment of a prospectively recorded database listing synchronous findings in patients with ≥ 20 mm colorectal lesions referred to a tertiary center for endoscopic resection. RESULTS: At least one synchronous precancerous lesion was identified in 66/78 patients with large pedunculated index lesions (84.6 %, 95 %CI 74.9-91.1) and 726/814 patients with large nonpedunculated index lesions (89.2 %, 95 %CI 87.1-91.3). Patients with a large pedunculated index lesion had mean of 4.8 synchronous conventional adenomas, 56.4 % had ≥ 1 synchronous high risk lesion (advanced adenoma or advanced serrated lesion), 48.7 % had ≥ 1 synchronous advanced conventional adenoma, and 19.2 % had a synchronous neoplastic lesion ≥ 20 mm. Compared with patients with nonpedunculated index lesions, patients with large pedunculated index lesions had comparable rates of synchronous polyps, adenomas, and sessile serrated lesions, and higher rates of synchronous adenomas with villous elements (15.6 % [95 %CI 13.3-18.3] vs. 26.9 % [95 %CI 18.3-37.7]; P = 0.01) and synchronous pedunculated polyps (9.5 % [95 %CI 7.6-11.7] vs. 33.3 % [95 %CI 23.8-44.4]; P < 0.001). CONCLUSION: In patients with large (≥ 20 mm) pedunculated colorectal lesions, rates of synchronous neoplasia and advanced synchronous neoplasia were high and comparable to or higher than rates of synchronous neoplasia in patients with large nonpedunculated colorectal lesions.

2021 Patient Perceptions Regarding Colonoscopy Experience.

GOAL: We sought to document patient perceptions in 2021 regarding colonoscopy experience and potential deterrents to repeat colonoscopy. BACKGROUND AND AIM: Bowel preparation has been previously considered by patients to be the worst part of a colonoscopy. MATERIALS AND METHODS: We conducted a prospective survey of consecutive patients age 18 years and older who had just completed colonoscopy at 2 outpatient endoscopy centers at a tertiary academic hospital. The short survey was conducted in the recovery area. The main outcome measure was patient perceptions of the worst part of their colonoscopy experience and which factor would most deter them from a future colonoscopy. RESULTS: Four hundred patients completed the survey of 405 approached. Average patient age was 64 years, and 48% were women. Seventy-five percent of patients used low-volume preparations. Bowel preparation was considered the worst part of colonoscopy by 71% of patients. Women were more likely to choose laxatives as the worst part of a colonoscopy. Bowel preparation was chosen most often (55%) as the most likely deterrent to a future colonoscopy. There were minimal differences in responses between those receiving low-volume versus high-volume (4 L) preparations. CONCLUSION: Bowel preparation remains the worst part of the colonoscopy experience for patients, and the most likely deterrent to future colonoscopy.

Frequency and nature of endoscopic and pathologic errors leading to referral for endoscopic resection to a tertiary center.

Background and study aims We anecdotally encounter cases where referring endoscopists made errors in endoscopic interpretation of a colorectal lesion, sometimes combined with pathology errors at the referring centers, resulting in referral to our center for endoscopic resection. In this paper, we describe the frequency and nature of endoscopic and pathology errors leading to consultation for endoscopic resection. Patients and methods Review of 760 consecutive referrals to our center over a 26-month interval. Results In total, 28 (3.7 %) of all referred patients had ≥ 1 lesion that did not require any resection after investigation. There were 12 cases (1.6 % of all referrals) involving errors by both the referring endoscopist and the pathologist at the referring center. Errors commonly involved the ileocecal valve, lipomas, and mucosal prolapse changes. There were 15 additional referrals (2.0 % of all referrals) where no neoplastic lesion was identified at our center and either no biopsy was taken at the referring center (n = 9 patients, 10 lesions), the patient was referred although biopsy showed no neoplasia (n = 6), or the referring doctor correctly interpreted the lesion (lipoma), but the outside pathologist incorrectly reported adenoma (n = 1). Conclusions Endoscopists at tertiary centers should expect referrals to clarify the nature of colorectal lesions as neoplastic or non-neoplastic. Community endoscopists with equivocal endoscopic findings and unexpected or equivocal pathology results can consider pathology review at their center or at an expert center before referral for endoscopic or surgical resection.

Most large colorectal polyps missed by gastroenterology fellows at colonoscopy are sessile serrated lesions.

Background and study aims Data on adenoma and sessile serrated lesion (SSL) miss rates for gastroenterology fellows during colonoscopy are limited. We aimed to describe the miss rate of fellows based on a second examination by a colonoscopist with a high rate of detection. Patients and methods Second- and third-year gastroenterology fellows at a single, tertiary center performed initial examinations. A single experienced attending doctor then performed a complete examination of the colon. We recorded the size and pathology of all lesions found at both examinations and calculated the adenoma and SSL miss rates for fellows. Results Ten trainees performed 100 examinations. Miss rates for conventional adenomas and SSLs were 30.5 % and 85.7 %, respectively. Among pre-cancerous polyps ≥ 10 mm, 10 of 14 lesions missed were SSLs. Conclusions While conventional adenoma detection skills of gastroenterology fellows are acceptable, SSL detection is poor.

Assessment of submucosal distortion and mass effect seen at follow-up after colorectal EMR with ORISE (with video).

BACKGROUND AND AIMS: ORISE (Boston Scientific, Marlborough, Mass, USA) is a viscous gel used for submucosal injection. We noted anecdotally that ORISE is associated with submucosal distortion of EMR scars at follow-up. METHODS: We blindly reviewed photographs of 30 consecutive EMR scars at follow-up after resections using ORISE and 30 resections using other agents. Distortion was scored on a scale of 0 (no submucosal distortion) to 5 (overt submucosal distortion). RESULTS: The median submucosal distortion score at follow-up in ORISE cases was 3 (range, 0-5) versus 0 (range, 0-2) with other fluids (P < .001, Mann-Whitney U test) by 1 reviewer and 3 (range, 0-5) versus 2 (range, 0-5), respectively, by a second reviewer (P = .018). The kappa value for agreement in the submucosal distortion scores between the 2 experts was .148 for all photographs and .214 for the ORISE cases (overall minimal agreement). CONCLUSIONS: ORISE can cause submucosal distortion in the region of EMR scars when they are viewed at follow-up. Such distortions must not be mistaken for submucosal tumor growth.

Cold versus hot snare resection with or without submucosal injection of 6- to 15-mm colorectal polyps: a randomized controlled trial.

BACKGROUND AND AIMS: Cold snare resection of colorectal lesions has been found to be safe and effective for an expanding set of colorectal lesions. In this study, we sought to understand the efficacy of simple cold snare resection and cold EMR versus hot snare resection and hot EMR for colorectal lesions 6 to 15 mm in size. METHODS: At 3 U.S. centers, 235 patients with 286 colorectal lesions 6 to 15 mm in size were randomized to cold snaring, cold EMR, hot snaring, or hot EMR for nonpedunculated colorectal lesions 6 to 15 mm in size. The primary outcome was complete resection determined by 4 biopsy samples from the defect margin and 1 biopsy sample from the center of the resection defect. RESULTS: The overall incomplete resection rate was 2.4% (95% confidence interval [CI], .8%-7.5%). All 7 incompletely removed polyps were 10 to 15 mm in size and removed by hot EMR (n = 4, 6.2%), hot snare (n = 2, 2.2%), or cold EMR (n = 1, 1.8%). Cold snaring had no incomplete resections, required less procedural time than the other methods, and was not associated with serious adverse events. CONCLUSIONS: Cold snaring is a dominant resection technique for nonpedunculated colorectal lesions 6 to 15 mm in size. (Clinical trial registration number: NCT03462706.).

Implications of stable or increasing adenoma detection rate on the need for continuous measurement.

BACKGROUND AND AIMS: Measurement of the adenoma detection rate (ADR) is resource intensive, and the benefit of continuous measurement for colonoscopists with high ADR is unclear. We examined the ADR trends at our center to determine whether continuous measurement for consistently high ADR is warranted. METHODS: Among colonoscopies performed between January 1999 and November 2019 at a tertiary center, we analyzed data from colonoscopists performing at least 50 screening colonoscopies annually for 5 consecutive years. ADR trends for individual colonoscopists were examined using Joinpoint regression models. RESULTS: Eleven colonoscopists performed screening colonoscopies on 14,047 patients, and 5912 among them had at least 1 conventional adenoma removed (42.0%). Of 25,829 polyps, 13,585 (52.6%) were conventional adenomas or adenocarcinomas and contributed to ADR calculation. All but 1 colonoscopist included met the recommended minimum threshold ADR of 25% continuously over the study period. Of the 11 colonoscopists, 5 had an increase in their ADR and the remaining 6 had stable ADRs over the study period. CONCLUSIONS: For colonoscopists consistently performing above the minimum threshold, diversion of resources toward improvement of quality measures other than ADR is justified.

Impact of Expanding Screening Colonoscopy Interval on Providers Choice of Prescribing Colonoscopy for Colorectal Cancer Screening.

GOALS: We investigated if increasing the colonoscopy screening interval from 10 to 15 years would increase provider preferences for colonoscopy as a screening test. We further examined whether having colonoscopy performed at a 15-year interval by an endoscopist with a high adenoma detection rate would influence preferences. BACKGROUND: Colonoscopy is recommended every 10 years in average risk individuals without polyps for colorectal cancer (CRC) screening. The use of a 15-year interval offers substantial protection, increases cost-effectiveness, and might make colonoscopy more attractive to patients and health care providers who order CRC screening tests. STUDY: An anonymous online survey of health care providers across a health care system that serves a single US state and encompasses both academic and community physicians was conducted. Physicians and nurse practitioners in family medicine, obstetrics-gynecology, and internal medicine were included. Providers were asked to indicate their preference for CRC screening tests as a proportion of tests they prescribe among 5 common screening tools. Responses were compared for current colonoscopy screening intervals and if the screening intervals are increased to 15 years. RESULTS: One hundred and twelve (34%) responded of 326 providers. Colonoscopy was the most frequently ordered test for CRC screening. Increasing screening interval from 10 to 15 years increased the choice of colonoscopy from 75.2% to 78.6% ( P =0.003). CONCLUSIONS: Expanding colonoscopy screening interval to 15 years could produce an increase in physicians and nurse practitioners choice of using colonoscopy for CRC screening, but the clinical impact appears minor. Additional surveys of patients and providers are needed.

Yield of a second right colon examination during colonoscopy after a first examination using a mucosal exposure device.

BACKGROUND AND AIMS: Double right colon examination during colonoscopy has been advocated to reduce the risk of interval cancer in the right colon. Whether 2 examinations are necessary when the first examination is performed with a mucosal exposure device is uncertain. We documented the rates of missed adenomas, sessile serrated lesions, and hyperplastic polyps after an initial right colon examination by a high-level detector using a mucosal exposure device. METHODS: At a single tertiary hospital outpatient practice, we prospectively collected data on the yield of a second examination of the right colon after an initial examination by a single high-detecting colonoscopist using a mucosal exposure device. RESULTS: During the study period, 1331 eligible consecutive patients underwent colonoscopy. Right colon adenoma, sessile serrated lesion, and hyperplastic polyp miss rates were 15.8%, 14.1%, and 16.7%, respectively. Four percent of patients had adenomas detected in the right colon only with a second examination. CONCLUSIONS: A second examination of the right colon is warranted, even when using a distal mucosal exposure device to perform colonoscopy.