Advertisement

ASGE review of adverse events in colonoscopy

Published:September 25, 2019DOI:https://doi.org/10.1016/j.gie.2019.07.033
      Colonoscopy is the most commonly performed endoscopic procedure and overall is considered a low-risk procedure. However, adverse events (AEs) related to this routinely performed procedure for screening, diagnostic, or therapeutic purposes are an important clinical consideration. The purpose of this document from the American Society for Gastrointestinal Endoscopy’s Standards of Practice Committee is to provide an update on estimates of AEs related to colonoscopy in an evidence-based fashion. A systematic review and meta-analysis of population-based studies was conducted for the 3 most common and important serious AEs (bleeding, perforation, and mortality). In addition, this document includes an updated systematic review and meta-analysis of serious AEs (bleeding and perforation) related to EMR and endoscopic submucosal dissection for large colon polyps. Finally, a narrative review of other colonoscopy-related serious AEs and those related to specific colonic interventions is included.

      Abbreviations:

      AE (adverse event), ASGE (American Society for Gastrointestinal Endoscopy), CI (confidence interval), ESD (endoscopic submucosal dissection), IBD (inflammatory bowel disease), PEDS-CORI (Pediatric Endoscopy Database System-Clinical Outcomes Research Initiative), PPES (postpolypectomy electrocoagulation syndrome), WE (water exchange)
      Colonoscopy is a commonly performed endoscopic procedure for various GI conditions and most routinely for the screening and surveillance of colorectal neoplasia. Overall, colonoscopy is considered a safe procedure, although a number of serious adverse events (AEs) have been reported. The definition of serious AEs is variable across studies but generally includes AEs that lead to an unplanned hospitalization, unplanned procedures or interventions, prolongation of an existing hospitalization, or death. Examples include bleeding, perforation, postpolypectomy syndrome, and cardiopulmonary AEs related to moderate or deep sedation.
      Few population-based colonoscopy registries provide the exact magnitude of AEs associated with colonoscopy. Estimates of AEs related to colonoscopy in various studies differ by indications, patient population, asymptomatic versus symptomatic individuals, length, and completeness of follow-up after the procedure. In a 2008 systematic review of 12 studies totaling 57,742 colonoscopies performed for average-risk screening, the pooled overall AE rate was 2.8 per 1000 procedures (95% confidence interval [CI], 1.5-5.2),
      • Whitlock E.P.
      • Lin J.S.
      • Liles E.
      • et al.
      Screening for colorectal cancer: a targeted, updated systematic review for the U.S. Preventive Services Task Force.
      whereas the reported incidence of AEs from mostly diagnostic colonoscopies performed in an integrated healthcare system in the United States was 5 per 1000 procedures (95% CI, 4.0-6.2).
      • Levin T.R.
      • Zhao W.
      • Conell C.
      • et al.
      Complications of colonoscopy in an integrated health care delivery system.
      In a 2016 evidence synthesis report by the Agency for Healthcare Research and Policy, the authors reported a pooled rate of major bleeding (22 studies; n = 3,347,101) of .8 per 1000 procedures (95% CI, .5-1.4) and rate of perforation of .4 per 1000 procedures (95% CI, .2-.5) for screening colonoscopy.
      • Lin J.S.
      • Piper M.A.
      • Perdue L.A.
      • et al.
      Screening for colorectal cancer: a systematic review for the U.S. Preventive Services Task Force.
      Although the risk of AEs in most studies are often not stratified by whether or not polypectomy was performed, according to 1 study, 85% of AEs are reported in patients undergoing colonoscopy with polypectomy.
      • Whitlock E.P.
      • Lin J.S.
      • Liles E.
      • et al.
      Screening for colorectal cancer: a targeted, updated systematic review for the U.S. Preventive Services Task Force.
      With the widespread application of advanced endoscopic techniques for removal of colorectal polyps, including EMR and endoscopic submucosal dissection (ESD), the AEs associated with these advanced techniques are highly relevant.
      The aims of this document are to provide evidence-based estimates of the 3 most common and important AEs of colonoscopy (bleeding, perforation, and mortality) from population-based studies, to provide evidence-based estimates of AEs related to EMR and ESD (bleeding and perforation) for large colon polyps, and to provide a narrative-based review of aspiration, splenic injury, and less common AEs. A narrative update of the previous Standards of Practice document
      • Fisher D.A.
      • Maple J.T.
      • Ben-Menachem T.
      • et al.
      ASGE Standards of Practice Committee
      Complications of colonoscopy.
      on other AEs, such as postpolypectomy syndrome, infection, and gas bloating, is addressed in this document. Risk of AEs as they relate to sedation and the pediatric population are also discussed. Details of various bowel preparations and their respective AEs are discussed in a separate American Society for Gastrointestinal Endoscopy (ASGE) document.
      • Saltzman J.R.
      • Cash B.E.
      • Pasha S.F.
      • et al.
      ASGE Standards of Practice Committee
      Bowel preparation before colonoscopy.

      Methods

      Panel composition and conflict of interest management

      The panel was composed of 2 primary authors (S.T.K., R.J.H.), a content expert (A.S.), committee chair (S.W.), and members of the Standards of Practice Committee. All panel members disclosed possible intellectual and financial conflicts of interest in concordance with ASGE policies (https://www.asge.org/docs/default-source/about-asge/mission-and-governance/asge-conflict-of-interest-and-disclosure-policy.pdf).

      Selection criteria

      A search for population-level studies that provided estimates for the major postprocedural endpoints of perforation, bleeding, and mortality was conducted by a professional librarian using Ovid MEDLINE: Epub Ahead of Print, In-Process & Other Non-Indexed Citations, Ovid MEDLINE Daily and Ovid MEDLINE 1946 to present, Embase Classic+Embase 1947 to January 2018, and Wiley Cochrane. In cases of multiple studies from the same group using the same data source (such as a conference proceeding followed by a manuscript), we included only the more recent and extensive of the studies.
      Only studies published in English were included for analysis. We included both retrospective and prospective cohort studies with data collected between January 2001 and March 2017 in the study. Prespecified medical subject headings, non–medical subject heading terms, and the search algorithm are shown in Appendix 1 (available online at www.giejournal.org). For estimates of perforation and bleeding after EMR and ESD, we chose case series and comparative trials published between January 2008 and January 2018; this decision was made given the rapid changes in advanced mucosal/submucosal resection techniques within the last decade. We adopted a search algorithm derived from Hassan et al,
      • Hassan C.
      • Repici A.
      • Sharma P.
      • et al.
      Efficacy and safety of endoscopic resection of large colorectal polyps: a systematic review and meta-analysis.
      which is available in Appendix 2 (available online at www.giejournal.org). For EMR, we restricted our analysis to polyps ≥20 mm in size.
      Two reviewers (S.T.K. and R.J.H.) independently screened all abstracts. Case reports, review articles, cost-effectiveness or modeling studies, and animal studies were excluded. The full text of the remaining articles was evaluated to determine if they met inclusion criteria in the study. For each study the first author, time period, and date of publication were extracted. For population-level studies mean age, percentage of females in the cohort, rates of perforation, bleeding, mortality, and percentage of colonoscopies with polypectomy were recorded. Because only a subset of population-level studies reported the indication for colonoscopy (eg, screening, surveillance, or diagnostic), this variable was not included in the meta-regression analysis. For EMR/ESD studies mean age, percentage of females in the cohort, location of the study (East Asian or Other), rate of perforation, rate of delayed bleeding, and mean polyp size (in mm) was recorded. Delayed bleeding was defined as any clinically significant bleeding that occurred after completion of the procedure up to 30 days postprocedure. Intraprocedural bleeding was not recorded as a separate outcome given the heterogeneity in definition and because almost all cases of reported intraprocedural bleeding were controlled endoscopically during the procedure.

      Statistical analysis

      A random-effects model was used to calculate the pooled perforation and bleeding rate for both population-level and EMR/ESD studies. Pooled estimates were reported with 95% CIs. Covariates analyzed in regression analysis included mean population age, percentage of females in the cohort, and percentage of polypectomies in the cohort for population-level studies. Covariates analyzed in regression analysis included mean population age, percentage of females in cohort, and size of polyp in EMR/ESD studies. Pooled rates of perforation and bleeding were calculated and grouped by EMR or ESD status. Heterogeneity between studies was measured using the I2 statistic. Analysis was performed using Comprehensive Meta-Analysis v 3.3.070 (Englewood, NJ).

      Population-level estimates of serious AEs

      Twenty-one population-level studies (11 from North America) reporting the rates of perforation, bleeding, or mortality after colonoscopy were identified (Supplementary Table 1, available online at www.giejournal.org). From these studies, data were extracted on 10,328,360 patients undergoing colonoscopy, of which 5,464,324 (54%) were women; the mean age of all patients was 62.3 years.

      Perforation

      Colonic perforation during colonoscopy may result from mechanical forces against the bowel wall, barotrauma, or a direct result of therapeutic procedures. Early symptoms include persistent abdominal pain and abdominal distention. Colonic perforation can be intraperitoneal or extraperitoneal. Intraperitoneal perforation leads to leak of air and colonic contents into the peritoneum. Plain radiographs of the chest and abdomen may demonstrate free air, although CT is superior to an upright chest film.
      • Stapakis J.C.
      • Thickman D.
      Diagnosis of pneumoperitoneum: abdominal CT vs. upright chest film.
      Therefore, an abdominal CT should be considered for patients with an unrevealing plain film in whom there is a high suspicion of perforation.
      • Fisher D.A.
      • Maple J.T.
      • Ben-Menachem T.
      • et al.
      ASGE Standards of Practice Committee
      Complications of colonoscopy.
      Rarely, colonic perforation can be extraperitoneal, leading to the passage of air into the retroperitoneal space, which can then diffuse along the fascial planes and large vessels, causing pneumo-retroperitoneum, pneumo-mediastinum, pneumopericardium, pneumothorax, and subcutaneous emphysema. Such patients can have an atypical presentation, including subcutaneous crepitus, neck swelling, chest pain, and shortness of breath after colonoscopy.
      • Tiwari A.
      • Sharma H.
      • Qamar K.
      • et al.
      Recognition of extraperitoneal colonic perforation following colonoscopy: a review of the literature.
      The pooled rate of perforations among 10,328,360 colonoscopies was 5.8 per 10,000 colonoscopies (95% CI, 5.7-6.0) (Fig. 1A). Reported population-level perforation rates ranged from a low of 1.6 per 10,000 to a high of 11.9 per 10,000 with significant heterogeneity between studies (I2 = 97.6%). This heterogeneity in studies may reflect differences in indication, population age, comorbidity, geographic location, and rates of polypectomy between studies. In a meta-regression analysis (Supplementary Fig. 1A, available online at www.giejournal.org), neither age nor gender was significantly associated with perforation rate. Moreover, after adjusting for differences in age and gender between different population-level studies, polypectomy was not significantly associated with risk for perforation (P = .9). A previous meta-analysis of population-level studies found a trend toward higher rate of perforation in colonoscopies with polypectomy (8 per 10,000) compared with those without polypectomy (4 per 10,000, P = .07).
      • Reumkens A.
      • Rondagh E.J.
      • Bakker C.M.
      • et al.
      Post-colonoscopy complications: a systematic review, time trends, and meta-analysis of population-based studies.
      These data suggest that a substantial proportion of the risks of perforation from colonoscopy are related to procedural characteristics independent of the performance of polypectomy, such as the amount of torque or pressure applied to the bowel wall during advancement of the colonoscope or barotrauma from insufflation of the colon. Notably, as discussed later in this article, performance of advanced mucosal resection techniques (EMR and ESD) increases the risk for perforation; however, on a population level, the numbers of these advanced procedures as a percentage of all polypectomies are small. These risk estimates therefore likely accurately reflect the risk that most average-risk patients face when undergoing an examination for screening or surveillance purposes.
      Figure thumbnail gr1a
      Figure 1Pooled rates (and 95% CIs) of postcolonoscopy perforation (A), bleeding (B), and mortality (C) from population-level studies. Additional covariates included are percentage of colonoscopies with polypectomy (% polypectomy), mean age, and percentage of cohort that is female (% Female). Data analyzed with a random-effects model. Data for mortality presented as logit of event rate given low event rates. CI, Confidence interval.
      Figure thumbnail gr1bc
      Figure 1Pooled rates (and 95% CIs) of postcolonoscopy perforation (A), bleeding (B), and mortality (C) from population-level studies. Additional covariates included are percentage of colonoscopies with polypectomy (% polypectomy), mean age, and percentage of cohort that is female (% Female). Data analyzed with a random-effects model. Data for mortality presented as logit of event rate given low event rates. CI, Confidence interval.
      Certain populations may face higher risks for perforation during colonoscopy, including patients with diverticulosis and inflammatory bowel disease (IBD).
      • Whitlock E.P.
      • Lin J.S.
      • Liles E.
      • et al.
      Screening for colorectal cancer: a targeted, updated systematic review for the U.S. Preventive Services Task Force.
      • Bielawska B.
      • Day A.G.
      • Lieberman D.A.
      • et al.
      Risk factors for early colonoscopic perforation include non-gastroenterologist endoscopists: a multivariable analysis.
      • Chukmaitov A.
      • Bradley C.J.
      • Dahman B.
      • et al.
      Association of polypectomy techniques, endoscopist volume, and facility type with colonoscopy complications.
      • Lohsiriwat V.
      Colonoscopic perforation: incidence, risk factors, management and outcome.
      • Rutter M.D.
      • Nickerson C.
      • Rees C.J.
      • et al.
      Risk factors for adverse events related to polypectomy in the English Bowel Cancer Screening Programme.
      Mukewar et al
      • Mukewar S.
      • Costedio M.
      • Wu X.
      • et al.
      Severe adverse outcomes of endoscopic perforations in patients with and without IBD.
      found that patients with IBD undergoing colonoscopy were at an 8-fold higher risk for endoscopy-associated perforations compared with patients without IBD (18.91 per 10,000 procedures vs 2.5 per 10,000 procedures). The use of corticosteroids is associated with a 13-fold greater risk for perforation associated with colonoscopy. Certain comorbid conditions also increase the risk for AEs. In a study of U.S. Medicare beneficiaries, Warren et al
      • Warren J.L.
      • Klabunde C.N.
      • Mariotto A.B.
      • et al.
      Adverse events after outpatient colonoscopy in the Medicare population.
      found that the presence of stroke, chronic obstructive pulmonary disease, atrial fibrillation, and congestive heart failure all significantly increased the risk of AEs due to colonoscopy. In addition to patient factors, provider factors may also influence the procedure risk. Ranasinghe et al
      • Ranasinghe I.
      • Parzynski C.S.
      • Searfoss R.
      • et al.
      Differences in colonoscopy quality among facilities: development of a post-colonoscopy risk-standardized rate of unplanned hospital visits.
      found significant variation (median, 12.3/1000; 5th to 95th percentile, 10.5 to 14.6/1000) in rates of AEs after outpatient colonoscopy between both hospital outpatient departments and free-standing ambulatory surgery centers, which could not be explained by case mix alone, raising the possibility that provider experience could be contributing to the variations in rates of AEs. Using administrative data from several large Canadian provinces, Rabeneck et al
      • Rabeneck L.
      • Paszat L.F.
      • Hilsden R.J.
      • et al.
      Bleeding and perforation after outpatient colonoscopy and their risk factors in usual clinical practice.
      found that endoscopists performing at volumes in the lowest quintile (<141 colonoscopies per year) had a 2.96 increase in odds of either perforation or bleeding compared with endoscopists performing at volumes in the highest quintile (>379 colonoscopies per year). In addition, Bielawska et al
      • Bielawska B.
      • Day A.G.
      • Lieberman D.A.
      • et al.
      Risk factors for early colonoscopic perforation include non-gastroenterologist endoscopists: a multivariable analysis.
      reported that colonoscopies performed by surgeons and endoscopists of unknown specialty had higher perforation rates when compared with gastroenterologists (odds ratio, 2.00; 95% CI, 1.30-3.08).

      Bleeding

      Unlike perforation, risk for bleeding during colonoscopy appears to be strongly associated with the performance of polypectomy. Postpolypectomy hemorrhage may occur immediately or can be delayed for up to 4 weeks after the procedure. In our systematic review, the rate of bleeding based on 15 population-level studies, including 5,544,454 patients, was 2.4 per 1000 colonoscopies (95% CI, 2.4-2.5) (Fig. 1B). In a meta-regression analysis (Supplementary Fig. 1B, available online at www.giejournal.org), the percentage of colonoscopies involving a polypectomy strongly predicted rates of bleeding, with a 2.7% increase in risk of bleeding for every 1% increase in rate of polypectomy (P < .001). This association remained significant after adjustment for age and gender (P = .016). The association between performance of polypectomy and risk for bleeding was also observed by Reumkens et al,
      • Reumkens A.
      • Rondagh E.J.
      • Bakker C.M.
      • et al.
      Post-colonoscopy complications: a systematic review, time trends, and meta-analysis of population-based studies.
      with findings of significantly more bleeding events after colonoscopies with polypectomy (9.8/1000) compared with colonoscopies without polypectomy (.6/1000, P < .001).
      Polyp size has been reported as a risk factor for postpolypectomy bleeding in several studies.
      • Consolo P.
      • Luigiano C.
      • Strangio G.
      • et al.
      Efficacy, risk factors and complications of endoscopic polypectomy: ten year experience at a single center.
      • Dafnis G.
      • Ekbom A.
      • Pahlman L.
      • et al.
      Complications of diagnostic and therapeutic colonoscopy within a defined population in Sweden.
      • Kim H.S.
      • Kim T.I.
      • Kim W.H.
      • et al.
      Risk factors for immediate postpolypectomy bleeding of the colon: a multicenter study.
      Additional risk factors may include the number of polyps removed,
      • Singh M.
      • Mehta N.
      • Murthy U.K.
      • et al.
      Postpolypectomy bleeding in patients undergoing colonoscopy on uninterrupted clopidogrel therapy.
      • Witt D.M.
      • Delate T.
      • McCool K.H.
      • et al.
      Incidence and predictors of bleeding or thrombosis after polypectomy in patients receiving and not receiving anticoagulation therapy.
      recent warfarin therapy,
      • Kim H.S.
      • Kim T.I.
      • Kim W.H.
      • et al.
      Risk factors for immediate postpolypectomy bleeding of the colon: a multicenter study.
      • Hui A.J.
      • Wong R.M.
      • Ching J.Y.
      • et al.
      Risk of colonoscopic polypectomy bleeding with anticoagulants and antiplatelet agents: analysis of 1657 cases.
      • Sawhney M.S.
      • Salfiti N.
      • Nelson D.B.
      • et al.
      Risk factors for severe delayed postpolypectomy bleeding.
      right-sided colon segment location,
      • Buddingh K.T.
      • Herngreen T.
      • Haringsma J.
      • et al.
      Location in the right hemi-colon is an independent risk factor for delayed post-polypectomy hemorrhage: a multi-center case-control study.
      • Kim J.H.
      • Lee H.J.
      • Ahn J.W.
      • et al.
      Risk factors for delayed post-polypectomy hemorrhage: a case-control study.
      and polyp histology.
      • Consolo P.
      • Luigiano C.
      • Strangio G.
      • et al.
      Efficacy, risk factors and complications of endoscopic polypectomy: ten year experience at a single center.
      Patient comorbidities, such as cardiovascular disease,
      • Consolo P.
      • Luigiano C.
      • Strangio G.
      • et al.
      Efficacy, risk factors and complications of endoscopic polypectomy: ten year experience at a single center.
      • Kim H.S.
      • Kim T.I.
      • Kim W.H.
      • et al.
      Risk factors for immediate postpolypectomy bleeding of the colon: a multicenter study.
      may increase the risk for bleeding but also may be a marker for antithrombotic use.
      • Sawhney M.S.
      • Salfiti N.
      • Nelson D.B.
      • et al.
      Risk factors for severe delayed postpolypectomy bleeding.
      Recommendations for the management of antithrombotic therapy in the periendoscopic period are discussed in detail in an ASGE guideline.
      • Acosta R.D.
      • Abraham N.S.
      • Chandrasekhara V.
      • et al.
      ASGE Standards of Practice Committee
      The management of antithrombotic agents for patients undergoing GI endoscopy.
      The prophylactic use of mechanical methods, such as clips or detachable snares, is commonly performed in practice; however, their efficacy in preventing delayed bleeding after non-EMR polypectomies has not been confirmed. Prospective, randomized studies and a meta-analysis have shown prophylactic clipping for polyps <2 cm does not prevent delayed bleeding,
      • Boumitri C.
      • Mir F.A.
      • Ashraf I.
      • et al.
      Prophylactic clipping and post-polypectomy bleeding: a meta-analysis and systematic review.
      • Matsumoto M.
      • Kato M.
      • Oba K.
      • et al.
      Multicenter randomized controlled study to assess the effect of prophylactic clipping on post-polypectomy delayed bleeding.
      • Shioji K.
      • Suzuki Y.
      • Kobayashi M.
      • et al.
      Prophylactic clip application does not decrease delayed bleeding after colonoscopic polypectomy.
      but in case of nonpedunculated polyps >2 cm, endoscopic clip closure of the mucosal defect has been demonstrated to reduce the incidence of delayed bleeding events in the proximal colon after resection (see Serious AEs Related to Advanced Resection Techniques, Postprocedural bleeding). Injection of epinephrine before polypectomy was reported to reduce the incidence of intraprocedural bleeding, although there was no demonstrated effect on delayed bleeding.
      • Di Giorgio P.
      • De Luca L.
      • Calcagno G.
      • et al.
      Detachable snare versus epinephrine injection in the prevention of postpolypectomy bleeding: a randomized and controlled study.
      • Hsieh Y.H.
      • Lin H.J.
      • Tseng G.Y.
      • et al.
      Is submucosal epinephrine injection necessary before polypectomy? A prospective, comparative study.

      Mortality

      Death after colonoscopy has been rarely reported. In a 2010 review of AEs based on prospective studies and retrospective analyses of large clinical or administrative databases, 128 deaths were reported among 371,099 colonoscopies, for an unweighted pooled death rate of .03%, or 3 in 10,000 colonoscopies
      • Ko C.W.
      • Dominitz J.A.
      Complications of colonoscopy: magnitude and management.
      ; all-cause mortality within 30 days occurred in .07% of patients, whereas colonoscopy-specific mortality occurred in .007% of patients. Our systematic review and meta-analysis included only colonoscopy-specific mortality, which was defined as death that could be directly attributable to a postprocedural AE (such as perforation) or the management of a postprocedural AE (such as surgery for a perforation). Nine studies reported colonoscopy-associated mortality rates. Thirty-six deaths occurred among 1,152,158 colonoscopies, for a pooled death rate of .003%, or 3 in 100,000 colonoscopies (Fig. 1C). Because of the small number of population-level studies reporting mortality data, meta-regression was not performed for the endpoint of mortality. Of the studies that reported both all-cause and colonoscopy-specific mortality, most deaths within 30 days of colonoscopy were not attributable to postcolonoscopy AEs but rather to underlying comorbidities such as cardiopulmonary disease, cirrhosis, and neurologic diseases. Most causes of death directly attributable to colonoscopy were either cardiopulmonary events that occurred during or immediately after the procedure or sequelae of bowel perforation.

      Serious AEs related to advanced resection techniques

      With enhancements in endoscopic technology, the role of the endoscopist has expanded to removal of large benign polyps and polyps harboring early cancers using advanced techniques such as EMR and ESD. As with standard polypectomy, bleeding and perforation are the most common AEs with EMR and ESD, but they occur more frequently with these advanced techniques. The reported AE rates vary. Lesion size, location, and histology and operator experience may all contribute to this variability.
      • Bae J.H.
      • Yang D.H.
      • Lee J.Y.
      • et al.
      Clinical outcomes of endoscopic submucosal dissection for large colorectal neoplasms: a comparison of protruding and laterally spreading tumors.
      • Burgess N.G.
      • Bassan M.S.
      • McLeod D.
      • et al.
      Deep mural injury and perforation after colonic endoscopic mucosal resection: a new classification and analysis of risk factors.
      • Swan M.P.
      • Bourke M.J.
      • Moss A.
      • et al.
      The target sign: an endoscopic marker for the resection of the muscularis propria and potential perforation during colonic endoscopic mucosal resection.
      We systematically analyzed the rates of the major endpoints of perforation and bleeding after both EMR of polyps ≥20 mm in size and ESD while also controlling for covariates of age, gender, location of the study, and polyp size. Rates of AEs were analyzed separately for EMR and ESD. Our search strategy yielded 29 studies, including 8237 unique procedures (Supplementary Table 2, available online at www.giejournal.org). Of the studies, 14 were reported from East Asia and 15 from either North America, Europe, or Oceania (Western).

      Perforation

      Twenty studies included in this analysis reported data on perforation rate after EMR. Of 6529 procedures, 59 were complicated by a perforation for a pooled rate of 1.1% (95% CI, .9%-1.4%) (Fig. 2A). There was substantial heterogeneity between studies (I2 = 83%), with reported rates ranging from .1% to 2.2%. Reasons for this heterogeneity include differences in definition of perforation, differences in polyp size and shape, center experience, and time period of study. Mortality appeared to be low, as no series reported fatalities as sequelae of perforation. Perforation is a term that may incorporate a spectrum of levels of deep mural injury; some have proposed a grading system for levels of mural injury ranging from muscularis propria exposure to full-thickness injury with a visualized hole and observed contamination.
      • Saito Y.
      • Fukuzawa M.
      • Matsuda T.
      • et al.
      Clinical outcome of endoscopic submucosal dissection versus endoscopic mucosal resection of large colorectal tumors as determined by curative resection.
      Although delayed perforation can occur, most cases of perforation captured in this review were immediately apparent to the endoscopist. The “target sign,” which represents concentric resection of progressively deeper layers of mural tissue and appears on the transected surface of the polypectomy specimen as a white to gray circular disk (the “target”) surrounded by a web of submucosal tissue that is then encircled by the white cauterized mucosa (Fig. 3), has been suggested as an early marker of perforation that can be visualized endoscopically.
      • Toyonaga T.
      • Man-i M.
      • Fujita T.
      • et al.
      Retrospective study of technical aspects and complications of endoscopic submucosal dissection for laterally spreading tumors of the colorectum.
      It represents a sign of muscle injury that can be either a full-thickness perforation or a partial muscle injury.
      Figure thumbnail gr2
      Figure 2Pooled rates (and 95% CIs) of EMR/ESD-related perforation (A) and delayed bleeding (B) stratified by performance of EMR or ESD. Additional covariates included are mean diameter of polyp and location of study (Asian or Western). Data analyzed with a random-effects model. CI, Confidence interval; ESD, endoscopic submucosal dissection.
      Figure thumbnail gr3
      Figure 3The “target sign” appears on the transected surface of the polypectomy specimen as a white to gray circular disk (the “target”) surrounded by a web of submucosal tissue that is then encircled by the white cauterized mucosa.
      Twelve studies reported data on perforation after ESD, and the rate of perforation after ESD was nearly 7-fold higher than after EMR, with a pooled rate of 7.2% (103/1708 procedures; 95% CI, 6.0%-8.7%) (Fig. 2A); importantly, however, no fatalities from perforation were reported from these studies. Because perforation is frequently encountered by endoscopists performing colonic ESD, prompt recognition of muscularis propria tissue along the dissection plane is essential. Most series captured in this meta-analysis reported that immediate perforations could usually be closed by deployment of endoscopic clips with very few patients requiring surgery.
      • Bae J.H.
      • Yang D.H.
      • Lee J.Y.
      • et al.
      Clinical outcomes of endoscopic submucosal dissection for large colorectal neoplasms: a comparison of protruding and laterally spreading tumors.
      • Burgess N.G.
      • Bassan M.S.
      • McLeod D.
      • et al.
      Deep mural injury and perforation after colonic endoscopic mucosal resection: a new classification and analysis of risk factors.
      • Swan M.P.
      • Bourke M.J.
      • Moss A.
      • et al.
      The target sign: an endoscopic marker for the resection of the muscularis propria and potential perforation during colonic endoscopic mucosal resection.
      • Xu M.D.
      • Wang X.Y.
      • Li Q.L.
      • et al.
      Colorectal lateral spreading tumor subtypes: clinicopathology and outcome of endoscopic submucosal dissection.
      Meta-regression was performed analyzing factors predicting perforation from advanced resection techniques (Supplementary Fig. 2A, available online at www.giejournal.org), pooling both EMR and ESD. The strongest and only predictor of perforation was the performance of ESD (as compared with EMR). In the subgroup of studies that reported perforation after both EMR and ESD, rates of perforation after ESD were between 5- and 8-fold higher compared with EMR.
      • Bae J.H.
      • Yang D.H.
      • Lee J.Y.
      • et al.
      Clinical outcomes of endoscopic submucosal dissection for large colorectal neoplasms: a comparison of protruding and laterally spreading tumors.
      • Lee E.J.
      • Lee J.B.
      • Lee S.H.
      • et al.
      Endoscopic treatment of large colorectal tumors: comparison of endoscopic mucosal resection, endoscopic mucosal resection-precutting, and endoscopic submucosal dissection.
      • Zhang Q.S.
      • Han B.
      • Xu J.H.
      • et al.
      Clip closure of defect after endoscopic resection in patients with larger colorectal tumors decreased the adverse events.
      The mean size of the polyps removed, location of study, and study date were not associated with risk of perforation.

      Postprocedural bleeding

      Because intraprocedural bleeding prompts immediate therapy and because all advanced tissue removal techniques are usually associated with some degree of intraprocedural bleeding, we only included rates of postprocedural bleeding in our systematic review. We found 27 studies that reported delayed bleeding rates after EMR (19 studies) and ESD (11 studies) (with a follow-up period of up to 30 days after colonoscopy), with a pooled rate of 3.7% (95% CI, 3.2%-4.2%) (Fig. 2B) and a range of .2% to 8.4%. As with perforation, substantial heterogeneity was noted (I2 = 66%). Unlike perforation, however, the rate of delayed bleeding was not statistically higher after ESD. Of the 11 studies reporting on rates of delayed bleeding after ESD (8 Asian and 3 Western), we found a pooled rate of 2.2% (95% CI, 1.5%-3.0%). By contrast, the EMR cohort experienced a pooled delayed bleeding rate of 4.0% (95% CI, 3.5%-4.5%). In a meta-regression analysis (Supplementary Fig. 2B, available online at www.giejournal.org), ESD was not associated with higher postprocedural bleeding rates compared with EMR. Polyp size, location, and year of study (temporal trend analysis) were also not associated with higher rate of delayed bleeding. The overwhelming majority of reported bleeding episodes were treated conservatively (blood transfusions, close observation with spontaneous resolution) or with endoscopic therapy. No fatalities from bleeding were reported. Other studies have reported that the following factors are associated with an increased risk for delayed bleeding: right-sided colon location, use of electrosurgical current not controlled by a microprocessor, intraprocedural bleeding at the time of polyp removal, exposed vessels of the post-EMR ulcer, signs of coagulation injury to the resection bed, and use of anticoagulants.
      • Witt D.M.
      • Delate T.
      • McCool K.H.
      • et al.
      Incidence and predictors of bleeding or thrombosis after polypectomy in patients receiving and not receiving anticoagulation therapy.
      • Burgess N.G.
      • Metz A.J.
      • Williams S.J.
      • et al.
      Risk factors for intraprocedural and clinically significant delayed bleeding after wide-field endoscopic mucosal resection of large colonic lesions.
      • Kim G.U.
      • Seo M.
      • Song E.M.
      • et al.
      Association between the ulcer status and the risk of delayed bleeding after the endoscopic mucosal resection of colon.
      • Metz A.J.
      • Bourke M.J.
      • Moss A.
      • et al.
      Factors that predict bleeding following endoscopic mucosal resection of large colonic lesions.
      The intraprocedural bleeding rate in the literature is over 10% in several large series, with delayed bleeding reported in 1.5% to 14% of cases.
      • Saito Y.
      • Uraoka T.
      • Yamaguchi Y.
      • et al.
      A prospective, multicenter study of 1111 colorectal endoscopic submucosal dissections (with video).
      • Kantsevoy S.V.
      • Adler D.G.
      • Conway J.D.
      • et al.
      ASGE Technology Committee
      Endoscopic mucosal resection and endoscopic submucosal dissection.
      In their large case series of 479 large sessile polyps that underwent EMR, Moss et al
      • Moss A.
      • Bourke M.J.
      • Williams S.J.
      • et al.
      Endoscopic mucosal resection outcomes and prediction of submucosal cancer from advanced colonic mucosal neoplasia.
      reported the following rates of AEs: hospitalization, 7.7%; postprocedural pain, 2.1%; serositis, 1.5%; bleeding (mostly immediate), 2.9%; and perforation, 1.3%. There were no deaths. Others have reported rates of immediate and delayed bleeding of 11% and 7%, respectively.
      • Thirumurthi S.
      • Raju G.S.
      How to deal with large colorectal polyps: snare, endoscopic mucosal resection, and endoscopic submucosal dissection; resect or refer?.
      Immediate bleeding can be treated with endoscopic therapy during the procedure but adds to procedure time. Most clinical bleeding AEs stop without intervention, although the need for transfusions has been reported.
      • Niimi K.
      • Fujishiro M.
      • Kodashima S.
      • et al.
      Long-term outcomes of endoscopic submucosal dissection for colorectal epithelial neoplasms.
      Previous studies on prophylactic hemoclip placement for minimizing risk of postpolypectomy bleeding in patients not on anticoagulation have not shown a reduction in bleeding rates.
      • Shioji K.
      • Suzuki Y.
      • Kobayashi M.
      • et al.
      Prophylactic clip application does not decrease delayed bleeding after colonoscopic polypectomy.
      • Feagins L.A.
      • Nguyen A.D.
      • Iqbal R.
      • et al.
      The prophylactic placement of hemoclips to prevent delayed post-polypectomy bleeding: an unnecessary practice? A case control study.
      However, recent data from a randomized controlled trial demonstrated that endoscopic clip closure of the mucosal defect after resection of large colon polyps (≥20 mm) reduces the risk of postpolypectomy bleeding (3.5% for the clip group vs 7.1% for the control group; absolute risk difference, 3.6%; 95% CI, .7-6.5), an effect that appeared to be restricted to large polyps in the proximal colon (3.3% in the clip group vs 9.6% in the control group; absolute risk difference, 6.3%; 95% CI, 2.5-10.1.
      • Pohl H.
      • Grimm I.S.
      • Moyer M.T.
      • et al.
      Clip closure prevents bleeding after endoscopic resection of large colon polyps in a randomized trial.
      Prophylactic clipping postresection of polyps <20 mm in size has not been shown to reduce the risk of postpolypectomy bleeding.
      • Matsumoto M.
      • Kato M.
      • Oba K.
      • et al.
      Multicenter randomized controlled study to assess the effect of prophylactic clipping on post-polypectomy delayed bleeding.

      Other AEs related to colonoscopy

      In this section we provide a narrative review of other AEs related to colonoscopy.

      Postpolypectomy electrocoagulation syndrome

      Postpolypectomy electrocoagulation syndrome (PPES) is the result of electrocoagulation-induced thermal injury during standard polypectomy, EMR, or ESD to the bowel wall that causes a transmural burn and localized peritonitis without evidence of frank perforation on radiographic studies. Typically, patients with PPES present 1 to 5 days after colonoscopy with fever, localized abdominal pain, localized peritoneal signs, and leukocytosis without any radiologic evidence of perforation. The reported incidence varies widely, from 1 in 100 (1%) to 3 per 100,000 (.003%).
      • Levin T.R.
      • Zhao W.
      • Conell C.
      • et al.
      Complications of colonoscopy in an integrated health care delivery system.
      • Ko C.W.
      • Dominitz J.A.
      Complications of colonoscopy: magnitude and management.
      • Singh H.
      • Penfold R.B.
      • DeCoster C.
      • et al.
      Colonoscopy and its complications across a Canadian regional health authority.
      • Cha J.M.
      • Lim K.S.
      • Lee S.H.
      • et al.
      Clinical outcomes and risk factors of post-polypectomy coagulation syndrome: a multicenter, retrospective, case-control study.
      The incidence is higher (7%-8%) in patients undergoing ESD.
      • Hirasawa K.
      • Sato C.
      • Makazu M.
      • et al.
      Coagulation syndrome: delayed perforation after colorectal endoscopic treatments.
      PPES usually is managed with intravenous hydration, broad-spectrum parenteral antibiotics, bowel rest, and nothing by mouth until symptoms subside.
      • Levin T.R.
      • Zhao W.
      • Conell C.
      • et al.
      Complications of colonoscopy in an integrated health care delivery system.
      • Cha J.M.
      • Lim K.S.
      • Lee S.H.
      • et al.
      Clinical outcomes and risk factors of post-polypectomy coagulation syndrome: a multicenter, retrospective, case-control study.
      In a large multicenter study, hypertension, large lesion size (>2 cm), and nonpolypoid morphology were found to be independent predictors of PPES. Thirty-four patients had PPES in this study, with 2.9% of patients requiring intensive care unit stay and no mortality.
      • Cha J.M.
      • Lim K.S.
      • Lee S.H.
      • et al.
      Clinical outcomes and risk factors of post-polypectomy coagulation syndrome: a multicenter, retrospective, case-control study.
      Full-thickness burns may result in bowel necrosis and delayed perforation, thereby requiring surgical management. Removal of right-sided lesions is likely to be associated with a higher risk of wall injury because of the thinner colonic wall.
      • Ferrara F.
      • Luigiano C.
      • Ghersi S.
      • et al.
      Efficacy, safety and outcomes of “inject and cut” endoscopic mucosal resection for large sessile and flat colorectal polyps.
      Submucosal injection of saline solution and other lifting agents are frequently performed to minimize the risk of PPES during endoscopic removal of large polyps.
      • Ferrara F.
      • Luigiano C.
      • Ghersi S.
      • et al.
      Efficacy, safety and outcomes of “inject and cut” endoscopic mucosal resection for large sessile and flat colorectal polyps.
      • Ferreira A.O.
      • Moleiro J.
      • Torres J.
      • et al.
      Solutions for submucosal injection in endoscopic resection: a systematic review and meta-analysis.
      Other techniques reported to decrease the risk of PPES include tenting the polyp away from the colonic wall before applying electrocautery, cutting the stalk of pedunculated polyps one-half or one-third of the distance from the base of the pedicle, and suctioning air out of the colon, especially in the right-sided colon segment, to minimize tension on the wall and increase colonic wall thickness.
      • Kedia P.
      • Waye J.D.
      Colon polypectomy: a review of routine and advanced techniques.

      Abdominal discomfort and/or bloating

      Postcolonoscopy abdominal pain can be because of a host of serious AEs as discussed above; however, abdominal discomfort and/or bloating is a less severe but very common AE postcolonoscopy and can affect patient compliance with future colonoscopies.
      • Ko C.W.
      • Dominitz J.A.
      Complications of colonoscopy: magnitude and management.
      • Park S.K.
      • Lee M.G.
      • Jeong S.H.
      • et al.
      Prospective analysis of minor adverse events after colon polypectomy.
      The discomfort may be caused by colonic spasm, gaseous distention of the colon, and mechanical or barotrauma. The most commonly reported minor AEs of colonoscopy are bloating (2.6%-25%) and abdominal pain and/or discomfort (2.5%-11%).
      • Park S.K.
      • Lee M.G.
      • Jeong S.H.
      • et al.
      Prospective analysis of minor adverse events after colon polypectomy.
      • Ko C.W.
      • Riffle S.
      • Shapiro J.A.
      • et al.
      Incidence of minor complications and time lost from normal activities after screening or surveillance colonoscopy.
      • Bini E.J.
      • Firoozi B.
      • Choung R.J.
      • et al.
      Systematic evaluation of complications related to endoscopy in a training setting: a prospective 30-day outcomes study.
      • Zubarik R.
      • Fleischer D.E.
      • Mastropietro C.
      • et al.
      Prospective analysis of complications 30 days after outpatient colonoscopy.
      Appropriate techniques, such as avoiding and reducing endoscope looping and minimizing air insufflation, should help reduce these symptoms.
      • Waye J.D.
      The most important maneuver during colonoscopy.
      In addition, a systematic review and meta-analysis of randomized controlled trials demonstrated less postprocedure pain with carbon dioxide compared with standard air insufflation.
      • Memon M.A.
      • Memon B.
      • Yunus R.M.
      • et al.
      Carbon dioxide versus air insufflation for elective colonoscopy: a meta-analysis and systematic review of randomized controlled trials.
      Water immersion and water exchange (WE) techniques that avoid air insufflation also may reduce pain, especially when minimal or no sedation is used. A meta-analysis and systematic review comparing air insufflation and water-aided methods (water immersion and WE) for procedure-related pain revealed both water immersion and WE to be superior to air insufflation.
      • Leung F.W.
      • Amato A.
      • Ell C.
      • et al.
      Water-aided colonoscopy: a systematic review.
      Studies comparing water-aided colonoscopy with CO2 insufflation have reported water immersion and WE significantly reduced colonoscopy insertion pain, and WE was the least painful technique but was the most time consuming.
      • Amato A.
      • Radaelli F.
      • Paggi S.
      • et al.
      Carbon dioxide insufflation or warm-water infusion versus standard air insufflation for unsedated colonoscopy: a randomized controlled trial.
      • Garborg K.
      • Kaminski M.F.
      • Lindenburger W.
      • et al.
      Water exchange versus carbon dioxide insufflation in unsedated colonoscopy: a multicenter randomized controlled trial.
      Postcolonoscopy abdominal discomfort because of gaseous distention is usually self-limited and rarely requires hospitalization.

      Gas explosion

      Explosive AEs from colonoscopy are rare but have serious consequences.
      • Fisher D.A.
      • Maple J.T.
      • Ben-Menachem T.
      • et al.
      ASGE Standards of Practice Committee
      Complications of colonoscopy.
      A 2007 review reported 9 cases, each resulting in colonic perforation and in 1 case, death.
      • Ladas S.D.
      • Karamanolis G.
      • Ben-Soussan E.
      Colonic gas explosion during therapeutic colonoscopy with electrocautery.
      Gas explosion can occur when combustible levels of hydrogen or methane gas are present in the colonic lumen, oxygen is present, and electrosurgical energy is used (eg, electrocautery or argon plasma coagulation). Suspected risk factors are use of nonabsorbable or incompletely absorbable carbohydrate preparations, such as mannitol, lactulose, or sorbitol,
      • Avgerinos A.
      • Kalantzis N.
      • Rekoumis G.
      • et al.
      Bowel preparation and the risk of explosion during colonoscopic polypectomy.
      • La Brooy S.J.
      • Avgerinos A.
      • Fendick C.L.
      • et al.
      Potentially explosive colonic concentrations of hydrogen after bowel preparation with mannitol.
      and incomplete colonic cleansing either because a sigmoidoscopy preparation was used (eg, enemas) or because of inadequate colonoscopic preparation.
      • Monahan D.W.
      • Peluso F.E.
      • Goldner F.
      Combustible colonic gas levels during flexible sigmoidoscopy and colonoscopy.
      Some authors have advocated use of CO2 during colonoscopy and avoiding enema-only bowel prep before applying argon plasma coagulation for treatment of radiation proctitis.
      • Ben-Soussan E.
      • Antonietti M.
      • Savoye G.
      • et al.
      Argon plasma coagulation in the treatment of hemorrhagic radiation proctitis is efficient but requires a perfect colonic cleansing to be safe.
      Bowel preps such as polyethylene glycol and sodium phosphate are reportedly safer before use of electrocautery and argon plasma coagulation because they do not result in combustible levels of hydrogen and methane.

      Infection

      Transient bacteremia after colonoscopy, with or with polypectomy, causing bacterial translocation of normal colonic flora to the bloodstream can occur in approximately 4% of procedures, with a range of 0% to 25%.
      • Nelson D.B.
      Infectious disease complications of GI endoscopy: part II, exogenous infections.
      However, signs or symptoms of infection are rare. Bacteremia is uncommon (6.3%) even after therapeutic colon procedures such as colonic stent insertion.
      • Chun Y.J.
      • Yoon N.R.
      • Park J.M.
      • et al.
      Prospective assessment of risk of bacteremia following colorectal stent placement.
      Although individual cases of infection after colonoscopy have been reported, there is no definite causal link with the endoscopic procedure and no proven benefit for antibiotic prophylaxis.
      • Khashab M.A.
      • Chithadi K.V.
      • Acosta R.D.
      • et al.
      ASGE Standards of Practice Committee
      Antibiotic prophylaxis for GI endoscopy.
      Therefore, current guidelines from the American Heart Association and ASGE recommend against antibiotic prophylaxis for patients undergoing colonoscopy.
      • Wilson W.
      • Taubert K.A.
      • Gewitz M.
      • et al.
      Prevention of infective endocarditis: guidelines from the American Heart Association: a guideline from the American Heart Association Rheumatic Fever, Endocarditis, and Kawasaki Disease Committee, Council on Cardiovascular Disease in the Young, and the Council on Clinical Cardiology, Council on Cardiovascular Surgery and Anesthesia, and the Quality of Care and Outcomes Research Interdisciplinary Working Group.
      The 2016 update of the multisociety guideline on reprocessing flexible GI endoscopes reported cases of transmission of infection resulting from defective equipment and/or failure to adhere to reprocessing guidelines.
      • Reprocessing Guideline Task Force
      • Petersen B.
      • Cohen J.
      • Hambrick R.
      • et al.
      Multisociety guideline on reprocessing flexible GI endoscopes: 2016 update.
      Pneumonia and perirectal abscess have been reported after colonoscopy.
      • Ko C.W.
      • Riffle S.
      • Michaels L.
      • et al.
      Serious complications within 30 days of screening and surveillance colonoscopy are uncommon.
      Pneumonia is mostly because of aspiration related to sedation and less commonly directly related to the procedure, as discussed in detail below in Sedation-related AEs.

      Splenic injury

      Splenic injury is a rare but serious AE of colonoscopy. It can develop immediately or up to several days after the procedure, making diagnosis difficult. The true incidence is unknown because of variability in reporting, but the reported rates from large series are 1 in 10,000 to 4.5 per 10,000 colonoscopies.
      • Bielawska B.
      • Hookey L.C.
      • Sutradhar R.
      • et al.
      Anesthesia assistance in outpatient colonoscopy and risk of aspiration pneumonia, bowel perforation, and splenic injury.
      • Piccolo G.
      • Di Vita M.
      • Cavallaro A.
      • et al.
      Presentation and management of splenic injury after colonoscopy: a systematic review.
      These patients are usually women (71.5%) in their sixth decade.
      • Piccolo G.
      • Di Vita M.
      • Cavallaro A.
      • et al.
      Presentation and management of splenic injury after colonoscopy: a systematic review.
      A high degree of suspicion is necessary because the clinical presentation can be nonspecific and variable in timing. Because of the high morbidity and mortality (up to 5%) associated with this entity, early identification and treatment are critical. Suggested mechanisms that may cause splenic injury during colonoscopy include direct trauma as the colonoscope traverses the splenic flexure of colon or rupture of the splenic capsule because of traction on the splenocolic ligament or adhesions. Suggested risk factors for splenic injury can be divided into patient-related and procedure-related factors. Suggested patient-related risk factors for splenic injury include prior abdominal surgery, presence of splenocolic adhesions, splenomegaly, endometriosis, inflammation (diverticular disease, IBD, pancreatitis), infection (malaria, typhoid, Epstein-Barr virus–induced infectious mononucleosis), and anticoagulant use.
      • Piccolo G.
      • Di Vita M.
      • Cavallaro A.
      • et al.
      Presentation and management of splenic injury after colonoscopy: a systematic review.
      • Ha J.F.
      • Minchin D.
      Splenic injury in colonoscopy: a review.
      Proposed procedure-related risk factors include difficult colonoscopy, deep sedation with propofol, operator inexperience, supine position, maneuvers such as hooking the splenic flexure to straighten the colon, “slide by” technique and alpha maneuver, and applying abdominal pressure in the left hypochondrium. The splenic injury can be intraparenchymal, subcapsular, or with intraperitoneal extension.
      There are 5 grades of splenic injury based on severity (Table 1).
      • Moore E.E.
      • Cogbill T.H.
      • Jurkovich G.J.
      • et al.
      Organ injury scaling: spleen and liver (1994 revision).
      Clinical presentation can vary, and the patient may present with nonspecific abdominal discomfort or abdominal pain that can be diffuse or localized to the left upper quadrant of the abdomen or left shoulder pain (Kehr’s sign) occurring within 24 hours of the procedure or, less commonly, delayed by days. Rarely, patients present with hemodynamic shock. Laboratory tests can reveal anemia. The criterion standard for diagnosis is contrast-enhanced CT of the abdomen that enables evaluation of the grade and extent of the splenic injury. US can be useful in assessing splenic injury in patients with hemodynamic instability and for those with contraindication to contrast-enhanced CT (allergy, renal insufficiency, etc). Management options include a conservative approach, splenic artery embolization, and surgery. The treatment option chosen is based on the presentation, underlying comorbidities, and imaging findings. Hemodynamically stable patients can be managed conservatively with close monitoring, intravenous fluids, blood transfusion, and antibiotics. However, some patients may fail a conservative approach and require surgery or splenic artery embolization. For hemodynamically stable patients with grades I to IV lacerations and no history of splenic disease, embolization of the splenic artery has been reported to be an effective therapeutic option.
      • Brennan I.M.
      • Faintuch S.
      • Sacks B.
      Superselective splenic artery embolization for the management of splenic laceration following colonoscopy.
      Splenectomy is usually reserved for cases with active bleeding and hemodynamic instability.
      Table 1Spleen injury scale
      GradeInjury description
      IHematoma subcapsular <10% surface area

      Laceration capsular tear <1 cm parenchymal depth
      IIHematoma subcapsular 10%-50% surface area; intraparenchymal <5 cm in diameter

      Laceration 1-3 cm parenchymal depth that does not involve a trabecular vessel
      IIIHematoma subcapsular >50% surface area or expanding; ruptured subcapsular or parenchymal hematomaIntraparenchymal hematoma >5 cm or expanding

      Laceration >3 cm parenchymal depth or involving trabecular vessels
      IVLaceration involving segmental or hilar vessels producing major devascularization (>25% of spleen)
      VCompletely shattered spleen

      Hilar vascular injury that devascularizes spleen

      Sedation-related AEs

      In hospital- and population-based studies, the incidence of aspiration events requiring hospitalization after colonoscopy with moderate or deep sedation is generally low (≤1 in 1000). In a population-based study, the incidence of aspiration requiring hospitalization during 165,527 outpatient diagnostic colonoscopies in 100,359 Medicare patients age 66 years and older (mean age, 76 years) was .14% for patients having colonoscopy under deep sedation with anesthesia assistance and .10% for patients under moderate sedation without anesthesia assistance (P = .02).
      • Cooper G.S.
      • Kou T.D.
      • Rex D.K.
      Complications following colonoscopy with anesthesia assistance: a population-based analysis.
      A study of 23,508 outpatient colonoscopies at 3 hospitals in Australia reported 1 case (.004%) of aspiration requiring hospitalization in a patient undergoing colonoscopy with general anesthesia.
      • Viiala C.H.
      • Zimmerman M.
      • Cullen D.J.
      • et al.
      Complication rates of colonoscopy in an Australian teaching hospital environment.
      A study of 3155 colonoscopies performed with patients under sedation managed by an anesthesiologist in adults at a single hospital in Italy reported that .16% of patients undergoing colonoscopy had an aspiration requiring “some intervention by an anesthesiologist.”
      • Agostoni M.
      • Fanti L.
      • Gemma M.
      • et al.
      Adverse events during monitored anesthesia care for GI endoscopy: an 8-year experience.
      Aspiration requiring hospitalizations was not reported. Others have investigated the risk of colonoscopy-associated aspiration using monitored anesthesia care/anesthesia assistance, with mixed results. Wernli et al
      • Wernli K.J.
      • Brenner A.T.
      • Rutter C.M.
      • et al.
      Risks associated with anesthesia services during colonoscopy.
      did not find an increased risk of aspiration pneumonia between anesthesia-assisted colonoscopy and standard sedation (odds ratio, 1.03; 95% CI, 1.00-1.06). However, their study population was limited to patients aged 40 to 64 years. A recent population-based cohort study of 3,059,045 outpatient colonoscopies, of which 862,817 were anesthesia assisted, reported that use of anesthesia assistance was associated with an increased risk of aspiration pneumonia (odds ratio, 1.63; 95% CI, 1.11-2.37) compared with colonoscopies performed without anesthesia assistance.
      • Bielawska B.
      • Hookey L.C.
      • Sutradhar R.
      • et al.
      Anesthesia assistance in outpatient colonoscopy and risk of aspiration pneumonia, bowel perforation, and splenic injury.
      It is commonly believed that aspiration risk is related to duration of nothing by mouth status before colonoscopy. A recent systematic review and meta-analysis
      • Shaukat A.
      • Malhotra A.
      • Greer N.
      • et al.
      Systematic review: outcomes by duration of NPO status prior to colonoscopy.
      found 6 studies (4 randomized controlled trials and 2 observational studies; sample size ranging from 115 to 1345) that reported risk of aspiration based on different durations of nothing by mouth status.
      • Gurudu S.R.
      • Ratuapli S.
      • Heigh R.
      • et al.
      Quality of bowel cleansing for afternoon colonoscopy is influenced by time of administration.
      • Huffman M.
      • Unger R.Z.
      • Thatikonda C.
      • et al.
      Split-dose bowel preparation for colonoscopy and residual gastric fluid volume: an observational study.
      • Manno M.
      • Pigò F.
      • Manta R.
      • et al.
      Bowel preparation with polyethylene glycol electrolyte solution: optimizing the splitting regimen.
      • Mathus-Vliegen E.M.
      • van der Vliet K.
      Safety, patient's tolerance, and efficacy of a 2-liter vitamin C-enriched macrogol bowel preparation: a randomized, endoscopist-blinded prospective comparison with a 4-liter macrogol solution.
      • Matro R.
      • Shnitser A.
      • Spodik M.
      • et al.
      Efficacy of morning-only compared with split-dose polyethylene glycol electrolyte solution for afternoon colonoscopy: a randomized controlled single-blind study.
      • Varughese S.
      • Kumar A.R.
      • George A.
      • et al.
      Morning-only one-gallon polyethylene glycol improves bowel cleansing for afternoon colonoscopies: a randomized endoscopist-blinded prospective study.
      In 5 studies no aspirations occurred during colonoscopy and none was reported within 30 days after colonoscopy.
      • Gurudu S.R.
      • Ratuapli S.
      • Heigh R.
      • et al.
      Quality of bowel cleansing for afternoon colonoscopy is influenced by time of administration.
      • Huffman M.
      • Unger R.Z.
      • Thatikonda C.
      • et al.
      Split-dose bowel preparation for colonoscopy and residual gastric fluid volume: an observational study.
      • Manno M.
      • Pigò F.
      • Manta R.
      • et al.
      Bowel preparation with polyethylene glycol electrolyte solution: optimizing the splitting regimen.
      • Mathus-Vliegen E.M.
      • van der Vliet K.
      Safety, patient's tolerance, and efficacy of a 2-liter vitamin C-enriched macrogol bowel preparation: a randomized, endoscopist-blinded prospective comparison with a 4-liter macrogol solution.
      • Varughese S.
      • Kumar A.R.
      • George A.
      • et al.
      Morning-only one-gallon polyethylene glycol improves bowel cleansing for afternoon colonoscopies: a randomized endoscopist-blinded prospective study.
      In 4 studies bowel preparation was completed at least 2 to 4 hours before colonoscopy. Overall, this systematic review found low-grade evidence that risk of aspiration is not related to duration of nothing by mouth status beyond 2 hours. Similarly, consumption of split-dose bowel prep within 3 to 4 hours of propofol sedation has not shown to increase risk of aspiration.
      • Alghamry A.
      • Ponnuswamy S.K.
      • Agarwal A.
      • et al.
      Split-dose bowel preparation with polyethylene glycol for colonoscopy performed under propofol sedation. Is there an optimal timing?.

      AEs in the pediatric population

      Overall rates of AEs from pediatric colonoscopy are uncommon, occurring in 1.1% to 2.4% of cases. Data collected prospectively from the Pediatric Endoscopy Database System-Clinical Outcomes Research Initiative (PEDS-CORI), a pediatric component of CORI, on 7792 colonoscopies reported 88 cases with at least 1 serious AE (1.1%).
      • Thakkar K.
      • El-Serag H.B.
      • Mattek N.
      • et al.
      Complications of pediatric colonoscopy: a five-year multicenter experience.
      From this database the most common AE was bleeding, occurring in .38% of cases. During colonoscopy, 71% of bleeding occurred in children under 10 years of age. Of 348 patients undergoing polypectomy, 5 (1.4%) had clinically significant bleeding. Hypoxemia was reported in 25% of those with an AE, with more cases using moderate sedation compared with general anesthesia (P < .001). Perforation during colonoscopy has been reported in several large series.
      • Thakkar K.
      • El-Serag H.B.
      • Mattek N.
      • et al.
      Complications of pediatric colonoscopy: a five-year multicenter experience.
      • Kramer R.E.
      • Narkewicz M.R.
      Adverse events following gastrointestinal endoscopy in children: classifications, characterizations, and implications.
      • Hsu E.K.
      • Chugh P.
      • Kronman M.P.
      • et al.
      Incidence of perforation in pediatric GI endoscopy and colonoscopy: an 11-year experience.
      • Iqbal C.W.
      • Askegard-Giesmann J.R.
      • Pham T.H.
      • et al.
      Pediatric endoscopic injuries: incidence, management, and outcomes.
      Kramer and Narkewicz
      • Kramer R.E.
      • Narkewicz M.R.
      Adverse events following gastrointestinal endoscopy in children: classifications, characterizations, and implications.
      reported a perforation rate of 3.1% during polypectomy but none during diagnostic colonoscopy. PEDS-CORI reported a single case (.1%) of perforation in a patient with active IBD. Similarly, in a retrospective single-center study based on surgical records, 3 colonoscopy-related perforations occurred out of 3269 procedures (.09%).
      • Kramer R.E.
      • Narkewicz M.R.
      Adverse events following gastrointestinal endoscopy in children: classifications, characterizations, and implications.
      All perforations were recognized within 24 hours, including 1 intraprocedure, and all were operatively repaired. In another large retrospective study based on physician recall of 7100 patients, 2 perforations occurred, both in patients with IBD: 1 was recognized periprocedure and the other 3 days postprocedure.
      • Hsu E.K.
      • Chugh P.
      • Kronman M.P.
      • et al.
      Incidence of perforation in pediatric GI endoscopy and colonoscopy: an 11-year experience.
      Several case reports also describe an increased risk of perforation in patients with Ehlers-Danlos type 4, because these patients may also have spontaneous perforation.
      • Soucy P.
      • Eidus L.
      • Keeley F.
      Perforation of the colon in a 15-year-old girl with Ehlers-Danlos syndrome type IV.
      • Fuchs J.R.
      • Fishman S.J.
      Management of spontaneous colonic perforation in Ehlers-Danlos syndrome type IV.
      One of the limitations of PEDS-CORI is that it predominantly reflects intraprocedure information and likely underestimates postprocedural events. A single-center prospective observational study of 9577 procedures (1819 diagnostic colonoscopies) with designated AE terminology tracked pediatric patients within 72 hours of an endoscopic procedure.
      • Kramer R.E.
      • Narkewicz M.R.
      Adverse events following gastrointestinal endoscopy in children: classifications, characterizations, and implications.
      The authors identified an AE rate of 2.4% (with a minimum of referral to an emergency department) and a polypectomy AE rate of 10.9%, taking into account events beyond the immediate postprocedure threshold. This disproportionate rate likely reflects a bias in children toward increased evaluation when there is a concern for a postprocedure AE. Bleeding risk and serious infection rates were .11% and .07%, respectively; however, data included both upper and lower endoscopic procedures. This study also provided more global AE rates rather than procedure-specific rates.
      At least in children under 3 years of age, there is a potential neurocognitive risk for prolonged and repeated procedures.
      • Andropoulos D.B.
      • Greene M.F.
      Anesthesia and developing brains—implications of the FDA warning.
      Thus, prolonged, incomplete, or nondiagnostic studies necessitating additional procedures may have longer term effects beyond standard 24-hour or 30-day AE rate statistics. Pediatric-specific colonoscopy quality indicators (eg, bowel preparation quality and ileal intubation rates) may indirectly assess these risks beyond costs, patient inconvenience, and other factors compared with adult patients.
      • Kramer R.E.
      • Walsh C.M.
      • Lerner D.G.
      • et al.
      Quality improvement in pediatric endoscopy: a clinical report from the NASPGHAN Endoscopy Committee.
      • Thakkar K.
      • Holub J.L.
      • Gilger M.A.
      • et al.
      Quality indicators for pediatric colonoscopy: results from a multicenter consortium.
      • Singh H.K.
      • Withers G.D.
      • Ee L.C.
      Quality indicators in pediatric colonoscopy: an Australian tertiary center experience.
      Taken together, overall rates of AEs during pediatric colonoscopy are uncommon (1.1%-2.4%) and most often associated with polypectomy. As in adults, significant AEs include bleeding and perforation (.02%-.1%). Pediatric patients with IBD and those with polyps may also be at higher risk of AEs. Research gaps include standardizing AE definitions in pediatric patients and assessing disease-specific risks, the relationship of AEs to pediatric quality indicators, and the role of postprocedural bleeding and closure techniques.

      Other rare AEs related to colonoscopy

      Other rare AEs associated with colonoscopy and interventions, such as acute appendicitis and acute diverticulitis, are summarized in the Supplementary Text (available online at www.giejournal.org).

      Conclusion

      Colonoscopy is the most commonly performed endoscopic procedure, and AEs are inherent to its performance. This document provides evidence-based estimates of AEs related to colonoscopy based on population-based studies, AEs related to advanced mucosal resection techniques (EMR and ESD), and a narrative update on other AEs associated with colonoscopy. Even though significant heterogeneity was seen in several analyses, the overall estimates of AEs have remained stable. Because endoscopy has assumed a more therapeutic role in the management of various colonic disorders, the potential for AEs increases. Improved knowledge of potential endoscopic AEs, their expected frequency, and the risk factors associated with their occurrence may help to minimize the incidence by careful selection of measures to help mitigate the risks associated with colonoscopy and other specific colonoscopic interventions[
      • Fisher D.A.
      • Maple J.T.
      • Ben-Menachem T.
      • et al.
      ASGE Standards of Practice Committee
      Complications of colonoscopy.
      Endoscopists are expected to carefully select patients for the appropriate intervention, be familiar with the planned procedure and available technology, and be prepared to manage any AE that may arise. Early recognition and prompt intervention are essential as soon as an AE occurs to help minimize associated morbidity and mortality. Tracking and monitoring AEs as part of a continuing quality improvement process may serve to educate endoscopists, help to reduce the risk of future AEs, and improve the overall quality of colonoscopy. Also, centralized reporting of AEs in the future may help in the development of a robust database, allowing for a better understanding of the true estimates of AEs and thereby serve for safety and quality targets in colonoscopy practice.

      Appendix 1. Search strategy for population-level studies

      Search date: March 19, 2017
      Databases searched: Ovid MEDLINE: Epub Ahead of Print, In-Process & Other Non-Indexed Citations, Ovid MEDLINE Daily and Ovid MEDLINE 1946 to present, Embase Classic+Embase 1947 to 2017 March 17, Wiley Cochrane
      Ovid MEDLINE, Embase
      Tabled 1
      #SearchesResults
      1*Colonoscopy/27,391
      2*endoscopic polypectomy/ use emczd or Colonic Polyps/su4512
      3colonoscop*.ti.22,610
      4(colon* adj2 polypectom*).ti,ab.2031
      5or/1-435,626
      6exp *Postoperative Complications/ use ppez243,109
      7exp *postoperative complication/ use emczd172,778
      8exp *Intraoperative Complications/ use ppez22,563
      9exp *peroperative complication/ use emczd6076
      10exp *Gastrointestinal Hemorrhage/68,570
      11((gi or gastrointestin* or intestin* or colon) adj2 (bleed* or hemorrag* or haemorrag*)).ti,ab.50,205
      12exp *Intestinal Perforation/ use ppez8581
      13exp *digestive system perforation/ use emczd18,243
      14((gi or gastrointestin* or intestin* or colon) adj2 perforation*).ti,ab.10,029
      15exp Mortality/1,339,415
      16mortality.ti,ab.1,467,173
      17exp Anesthesia/ae, mo [Adverse Effects, Mortality]24,321
      18exp *anesthesia complication/ use emczd4365
      19exp Colon/in [Injuries]1691
      20exp colon injury/ use emczd2802
      21(adverse or complication*).ti.351,154
      22or/6-212,814,327
      235 and 225594
      24limit 23 to english language4834
      25limit 24 to yr="1980 -Current"4690
      26animals/ not (humans/ and animals/)5,550,858
      2725 not 264667
      28limit 27 to (case reports or comment or editorial or letter or note) [Limit not valid in Ovid MEDLINE(R),Ovid MEDLINE(R) Daily Update,Ovid MEDLINE(R) In-Process,Ovid MEDLINE(R) Publisher,Embase; records were retained]1012
      29Case Report/4,116,817
      3027 not (28 or 29)3264
      31remove duplicates from 302452
      Wiley Cochrane
      Search Name: Colonoscopy-Adverse Events
      Date Run: 19/03/17 15:32:25.423
      Description:
      Tabled 1
      IDSearchHits
      #1medical subject headings (MeSH) descriptor: [Colonoscopy] explode all trees1914
      #2colonoscop*:ti1738
      #3#1 or #22718
      #4MeSH descriptor: [Postoperative Complications] explode all trees33,491
      #5MeSH descriptor: [Intraoperative Complications] explode all trees4022
      #6MeSH descriptor: [Gastrointestinal Hemorrhage] explode all trees1901
      #7((gi or gastrointestin* or intestin* or colon) near/2 (bleed* or hemorrag* or haemorrag*)):ti,ab1443
      #8MeSH descriptor: [Intestinal Perforation] explode all trees148
      #9((gi or gastrointestin* or intestin* or colon) near/2 perforation*):ti,ab188
      #10MeSH descriptor: [Mortality] explode all trees12981
      #11mortality:ti,ab37,315
      #12(adverse or complication*):ti7476
      #13#4 or #5 or #6 or #7 or #8 or #9 or #10 or #11 or #1287,576
      #14#2 and #1393

      Appendix 2. Search strategy for colonoscopies with EMR/endoscopic submucosal dissection

      Databases searched: Ovid MEDLINE: Epub Ahead of Print, In-Process & Other Non-Indexed Citations, Ovid MEDLINE Daily and Ovid MEDLINE 1946 to present, Embase Classic+Embase 1947 to 2017, Wiley Cochrane
      MEDLINE/Embase
      Tabled 1
      1exp Colon/ or exp Rectum/20,3534
      2(colon or rectum or colorectal).ti,ab.63,8072
      31 or 272,4789
      4exp Colonic Polyps/ use ppez7675
      5exp Colon Polyp/ use emczd19,639
      6exp Polyps/ use ppez29,762
      7exp Polyp/ use emczd67,959
      8exp Adenoma/198,527
      9exp Neoplasms/ use ppez3,034,373
      10exp Neoplasm/ use emczd4,099,596
      11(colon* and polyp*).ti,ab.51,289
      12(colonic polyps or colon polyp* or polyps or polyp* or lesion* or adenoma* or adenomatous or neoplasia*).ti,ab.2,574,943
      13or/4-128,798,252
      14(Polypectomy or removal or EMR or ESD).ti,ab.722,293
      15(endoscopic resection or mucosectomy or endoscopic submucosal resection).ti,ab.12,082
      16exp Colonoscopy/89,383
      17tu.fs.2,074,126
      18th.fs.3,105,206
      1916 and (17 or 18)8307
      2014 or 15 or 19738,992
      213 and 13 and 2017,176
      22limit 21 to yr="2014 - 2017"4424
      23remove duplicates from 223442
      24(30 mm or 3 cm or large or 20 mm or 2 cm).ti,ab.2,907,929
      2523 and 24938
      Wiley Cochrane
      Tabled 1
      IDSearchHits
      #1MeSH descriptor: [Colon] explode all trees1504
      #2MeSH descriptor: [Rectum] explode all trees1358
      #3(colon or rectum or colorectal):ti,ab15,101
      #4MeSH descriptor: [Colonic Polyps] explode all trees371
      #5MeSH descriptor: [Polyps] explode all trees832
      #6MeSH descriptor: [Adenoma] explode all trees1053
      #7MeSH descriptor: [Neoplasms] explode all trees61,680
      #8(colon* and polyp*):ti,ab1037
      #9(colonic polyps or colon polyp* or polyps or polyp* or lesion* or adenoma* or adenomatous or neoplasia*):ti,ab31,393
      #10#1 or #2 or #316,364
      #11#4 or #4 or #6 or #7 or #8 or #988,559
      #12(Polypectomy or removal or EMR or ESD):ti,ab11,648
      #13(endoscopic resection or mucosectomy or endoscopic submucosal resection):ti,ab750
      #14MeSH descriptor: [Colonoscopy] explode all trees1947
      #15#12 or #13 or #1413,971
      #16(30 mm or 3 cm or large or 20 mm or 2 cm):ti,ab70,321
      #17#10 and #11 and #15 and #16 Publication Year from 2014 to 2017, In trials105
      Supplementary Table 1Evidence table for the population studies
      Author(s)/Year of PublicationStudy DesignInclusion/ExclusionPatient CharacteristicsPerforation RateBleeding RateMortalityNotes
      Beilawska et al. (2014)Retrospective analysis using Clinical Outcomes Research Initiative National

      Endoscopic Database
      Inclusion Criteria:

      All complete and incomplete colonoscopies.

      Colonoscopies involving patients over 18 years of age from January 2000 through March 2011

      Exclusion criteria:

      Flexible sigmoidoscopies

      Procedures performed by Pediatric Gastroenterologists.
      N = 1,144,900

      Male 596220 (52.1)

      Female 548484 (47.9)

      Age =

      <60 years 566913 (49.5)

      60 - 74 yrs 426222 (37.2)

      >= 75 yrs 151140 (13.2)

      Unknown 433 (0.04)
      0.017% or 1 in 5882 procedures.Not reportedNot reported“ Early perforation” recorded referred to a perforation discovered before the procedure report is signed off at the end of the

      colonoscopy

      Patients aged 60-74 years had an OR of 2.69 (95% CI, 1.83 - 3.98) and those 75 years and older had an OR of 5.63 (95% CI, 3.73 -8.49) toward increased early perforation compared with those aged 60 years and younger.

      Increasing ASA class was also associated with increasing early perforation risk, with patients in ASA Class III and above having greatest risk.

      Female gender was also a significant independent risk factor (OR 2.00).

      Any therapy (OR 3.93) and large polyp size (OR 4.14) were highly significant risk factors for early perforation
      Blotiere et al. (2014)Retrospective study based on SNIIRAM and the PMSI databases.Inclusion Criteria:

      Total or partial colonoscopy in 2010

      Exclusion Criteria: patients with a history of chronic inflammatory bowel disease (ICD10 codes K50 and K51) or colorectal cancer (ICD10 code C16-C26)

      Patients who had already undergone colonoscopy during the 12 months preceding the first colonoscopy performed in 2010
      N = 947,061

      Men = 420,852 (44.4%)

      Age =

      0—39 92,188 (9.7%)

      40—49 143,604 (15.2%)

      50—59 249,746 (26.4%)

      60—69 252,689 (26.7%)

      70—79 155,861 (16.5%)

      80 and older 52,973 (5.6%)
      Between 4.5 and 9.7 per 10,000 proceduresBetween 9.9 and 11.0 per 10,000 proceduresNot reportedThe main risk factors associated with perforation and hemorrhage were patient’s age (over 80 years compared to under 40, OR = 7.51 and 3.23),

      resection of polyps larger than 1 cm or more than 4 polyps (compared to no polypectomy,

      OR = 2.72 and 5.12) and emergency colonoscopy (OR = 4.63 and 5.99). Colonoscopy performed by a gastroenterologist performing less than 244 colonoscopies per year was associated with an increased risk of perforation (OR = 2.29).
      Rutter et al. (2014)Population based studyInclusion criteria:

      All colonoscopies performed in the NHSBCSP between the start of the program in 2006 and January 2012

      Exclusion criteria:

      Not reported
      N = 130831 colonoscopies (167208 polypectomies) were analyzed, including 30 881 single polypectomies.

      Male = 60.7%

      Mean age = 65.7 years (59-92 years)
      Overall 0.06%

      Diagnostic colonoscopy

      0.03%

      Therapeutic colonoscopy

      0.09%
      Overall

      0.65%

      Diagnostic colonoscopy

      0.10%

      Therapeutic colonoscopy

      1.14%
      The overall rate of adverse events was 1.42% (1 in 71 procedures).

      Polypectomy increased bleeding risk 11.14-fold and perforation

      risk 2.97-fold. Cecal location (but not elsewhere

      in the proximal colon) and increasing polyp size

      were the two most important risk factors for bleeding and perforation
      Ranasinghe et al. (2016)Population based studyInclusion criteria:

      Medicare FFS patients aged ≥65 years undergoing outpatient colonoscopy at HOPDs, ASCs, and physician office settings.

      Common nonhigh-risk outpatient diagnostic and therapeutic colonoscopy procedures with or without biopsy, lesion ablation,

      and/or polypectomy.

      only patients with continuous enrollment in Medicare FFS Parts A and B in the 12 months before the procedure

      Exclusion criteria:

      High-risk patient groups undergoing colonoscopy

      Colonoscopies that occurred concurrently with high-risk upper gastrointestinal endoscopies, such as for control of bleeding; and colonoscopies for patients with a history

      of inflammatory bowel disease or diverticulitis in the year

      preceding the colonoscopy

      Colonoscopies for patients who lacked continuous enrollment in Medicare FFS Parts

      A and B in the 1 month after the procedure.
      N = 331,880

      Number of facilities = 8140

      Female =

      180,313 (54.3%)

      Age =

      65-69 = 98,248 (29.6%)

      70-74 = 101,555 (30.6%)

      75-79 = 72,984 (22 %)

      80-84 = 42,021 (12.7%)

      85 + = 17,072 (5.1%)

      Mean age = 74.2 years
      3 %6.4%6/331,880 . 0.002%Outpatient colonoscopies were followed by 5412 unplanned hospital visits within 7 days (16.3/1000 colonoscopies).

      A history of disorders of fluid and electrolyte balance (odds ratio [OR] . 1.43; 95% confidence interval

      [CI]: 1.29_1.58), psychiatric disorders (OR . 1.34; 95% CI:

      1.22_1.46), and, in the absence of a history of arrhythmia, increasing age >65 years (age >85 years vs 65 - 69 years;

      OR . 1.87; 95% CI: 1.54_2.28) were the strongest predictors of unplanned hospital visits.
      Saraste et al. (2016)Population based studyInclusion criteria:

      All patients with a positive FOBT followed by a colonoscopy performed between 1 January 2008 and 30 June 2012 . (Swedish study)

      Exclusion criteria:
      N = 2984Overall 1/1000

      2.5/1000 after colonoscopy with polypectomy
      14/10001/2984 (unrelated to colonoscopy)Mortality and complications within 30 days after colonoscopy or subsequent surgery were identified through national registers, and complications were assessed through review of medical charts

      Overall complications 1%
      Bielawska et al. (2017)Population basedInclusion criteria:

      Adults aged 18 and older who underwent outpatient colonoscopy in Ontario, Canada between January 1, 2005 and December 31, 2012

      Exclusion criteria:

      1) Colonoscopies performed during hospital admission

      2) Flexible sigmoidoscopy

      3) Other procedures like EGD performed with the colonoscopy
      N = 3,059,045 patients

      Male = Anesthesia assisted 421,352 (48.8%)

      Unassisted

      1,070,723 (48.8%)

      Total Anesthesia Assisted

      Colonoscopy N=862,817

      Total Unassisted

      Colonoscopy

      N=2,196,228

      Age:

      18 - 29 AA 18,699 (2.2%) UA 58,957 (2.7%)

      30 - 39 AA 39,434 (4.6%) UA 116,517 (5.3%)

      40 - 49 AA 125,883 (14.6%) UA 329,597 (15.0%)

      50 - 59 AA 293,419 (34.0%) UA 693,237 (31.6%)

      60 - 69 AA 227,585 (26.4%) UA 560,976 (25.5%)

      70 - 79 125,129 (14.5%) 339,402 (15.5%)

      80+ AA 32,668 (3.8%) UA 97,542 (4.4%)
      1396 colonoscopy related bowel perforations (0.046%)Not reportedNot reported138 splenic injuries (0.0045%)

      186 aspiration pneumonia (0.0061%)

      Risk of perforation was significantly associated with standard polypectomy (OR 1.78), large polypectomy (OR 7.60) and stenting or stricture

      dilation (OR 16.80), compared to no intervention

      AA not associated with perforation

      After propensity matching, use of AA was associated with a statistically significant (P = 0.012) 63% increase in the odds of aspiration pneumonia
      Forsberg et al. (2017)Retrospective population-based cohort studyInclusion criteria:

      Colonoscopies performed during the years 2001–2013 on adults ≥ 18 years identified in the Swedish health registers (National Hospital Discharge Register and the National

      Outpatient Register)

      Individuals with at least one colonoscopy during 2001–2013

      Exclusion criteria:

      Polypectomy performed in the colon or rectum (e.g., a rectoscopy) during a procedure other

      than a colonoscopy

      Individuals for whom there was only a code for polypectomy and no code for endoscopy
      N = 593,315 colonoscopies performed on the 426,560 individuals

      Female = 56%

      Median Age = 62.7 years (range 18-105 years)

      Inpatient colonoscopies = 14.9%
      0.11%

      For colonoscopy with polypectomy perforation rate was 0.25%

      The multivariate RR for perforation when general anaesthesia was employed was 2.65 (p<.001; 95%CI 1.71–4.12).
      0.17%

      For colonoscopy with polypectomy bleeding rate was 0.53%
      0.68%Adverse events were recorded for within 30 days of the procedure

      31 splenic injuries (1:20,000 colonoscopies) reported.

      1.ˆ% colonoscopies performed with general anesthesia

      In 47,492 of the colonoscopies performed (8%), polypectomies

      Were performed.

      There was a significant increased risk for perforation observed when general anesthesia was employed (RR 2.65;

      95% CI 1.71–4.12)

      Risk of bleeding and perforation were higher in age > 60 and when polypectomy performed.
      Wang et al. (2017)Population based studyInclusion criteria:

      Screening or surveillance colonoscopy and non-screening/non-surveillance colonoscopy performed in California hospital-owned and nonhospital owned ambulatory facilities, emergency departments, and hospitals from January 1, 2005 through December 31, 2011.

      For persons who underwent multiple colonoscopies, the first colonoscopy encounter was used.

      Exclusion criteria:
      N =

      S-Colo was performed on 1,580,318 patients and included biopsy/intervention in 59.2%,

      NS-Colo was performed on 1,222,070 patients and included biopsy/intervention in 39.7%.

      Average age =

      60.4 years for SC

      57.7 years for NSC

      Male =

      768,858 (48.7%) for SC

      490,826 (40.2%) for NSC

      Race =

      White =

      1,042,493 (66%) SC

      739,335 (60.5%) NSC
      After screening/surveillance

      2.9/10000 (95% CI, 2.5–3.3) without biopsy or intervention

      With biopsy or intervention

      6.3/10000 (95% CI, 5.8–6.8)

      Total SC

      4.9 (4.5–5.2)

      Nonscreening/nonsurveillance colonoscopy without intervention

      4.8 (4.3–5.3)

      Nonscreening/nonsurveillance colonoscopy with intervention

      10.8 (9.9–11.7)

      Total NSC

      7.2 (6.7–7.7)
      Lower bleeding after screening /surveillance 5.3/10000 (95% CI 4.8–5.9)

      Without biopsy or intervention

      With biopsy or intervention

      36.4/10000 (95%

      CI, 35.1–37.6)

      Total SC:

      23.7 (22.9–24.5)

      Nonscreening/nonsurveillance colonoscopy without intervention

      18.9 (17.9–19.8)

      Nonscreening/nonsurveillance colonoscopy with intervention

      50.2 (48.2–52.2)

      Total NSC

      31.3 (30.3–32.3)
      After screening /surveillance

      2.1 (1.7–2.4)

      Without biopsy or intervention

      With biopsy or intervention

      4.1 (3.6–4.5)

      Total SC:

      3.2 (3.0–3.5)

      Nonscreening/nonsurveillance colonoscopy without intervention

      5.0 (4.5–5.5)

      Nonscreening/nonsurveillance colonoscopy with intervention

      9.1 (8.3–10.0)

      Total NSC

      6.6 (6.2–7.1)
      30-day GI and non-GI complication rates in patients undergoing outpatient colonoscopy.

      Complications were documented within 30 days in 109 (95% CI 107–112) per 10,000 S-Colo-Diag patients, and in 218 (95% CI 215–221) per 10,000 S-Colo-Int patients

      Overall complications were seen in 239 (236–243) and in 400 (395–406) NS colonoscopy without and with interventions respectively

      Biopsy/intervention was an independent predictor of lower GI

      bleeding and perforation (adjusted odds ratios 3.07 [95% CI

      2.91–3.25] and 1.89 [95% CI 1.70–2.11], respectively).
      Rabeneck et al. (2008)Population basedInclusion criteria:

      Individuals 50 to 75 years old who underwent an outpatient colonoscopy during April 1, 2002, to March 31, 2003

      Exclusion criteria:

      In the 5 years preceding the index colonoscopy, had a colonoscopy, a diagnosis of colorectal

      cancer, a hospital admission with inflammatory bowel

      disease, or a colonic resection.

      In the 7 days prior to or on the day of the index colonoscopy, had an upper

      endoscopy

      Underwent colonoscopy for endoscopic hemostasis of a bleeding colonic lesion,

      insertion of a colonic stent, endoscopic colonic dilatation,

      or endoscopic reduction of a sigmoid volvulus.
      N = 97,091

      Mena age = 60.9 years

      Female = 52,641 (54.2%)
      Pooled rate = 1.64/1000Pooled rate = 0.85/10005/67,632 or 0.074/1000Bleed or a perforation within 30 days following the index colonoscopy recorded.

      23,623 (24.3%) had polypectomy performed during the procedure.

      Older age and having a polypectomy were associated with higher odds of bleeding

      or perforation.
      Radaelli et al. (2008)Cross-sectional, prospective multicenter study.Inclusion criteria:

      Data from 278 centers in Italy and 12,835 consecutive colonoscopies were evaluated

      The main features of each endoscopy center (structure indicators) were collected through the use of a standardized

      questionnaire, completed by the head of each participating

      centers (questionnaire No. 1).

      A second questionnaire was used to prospectively record details of all the consecutive colonoscopies performed

      in a 2-week study period in 2004 (process indicators).

      Exclusion criteria:

      Not reported
      N = 12835 colonscopies

      Gender =

      Male 6740 (52.5%)

      Female 5910 (

      Age (years) =

      >75 2005 (15.6%)

      65–75 3727

      36–64 6056

      ≤35 933
      2/12835 (0.02%)26/12835 (0.20%)Not reportedAbout 93% of colonoscopies were performed for diagnostic purpose; screening and surveillance accounted for 13.7% and 25.3% of the indications, respectively. Sedation and/or analgesia was administered in about half of the patients.

      Immediate complications related to the procedure included intestinal perforation and post-polypectomy bleedings

      either during the examination or before discharge, and cardiac and/or respiratory complications serious enough to oblige the endoscopist to stop the examination were recorded.

      32 (0.25%) cardiopulmonary complications were noted.
      Arora et al. (2009)Retrospective, population-based, cohort studyInclusion criteria:

      Colonoscopy performed in patients 18 years or older enrolled in the Medi-Cal program during the period from January 1, 1995, to June 30, 2005.

      Only one (first) colonoscopy per patient was studied.

      Exclusion criteria:

      Patients not enrolled in Medi-Cal continuously for the

      7 days after their date of first colonoscopy
      N = 277,434

      Control N = 1,072,723

      Mean (SD) age

      = 64.20 ± 14.80 years (range 18-107.8 years)

      Mean age control = 63.97 ± 14.99 years (range 18-107.9 years)

      Gender =

      Women: 175,816 (63.4%)

      Men: 101,

      618

      Race=

      White 108,946 (39.3%)

      AA: 26,824
      228/277434 (0.082%)

      82/100,00
      Not reportedNot reportedRisk of perforation 7 days post colonoscopy recorded

      The OR of getting a perforation from a colonoscopy compared with matched controls (n Z 1,072,723) who did not undergo a colonoscopy was 27.6 (95% CI, 19.04-39.92),

      P < .001.

      increasing age (> 65 yrs), significant comorbidity, obstruction as an indication for the colonoscopy, and performance of invasive interventions during colonoscopy were significant risk factors for perforation.
      Bokemeyer et al. (2009)Population studyInclusion criteria:

      Asymptomatic individuals over 55 years of age undergoing screening

      colonoscopy under outpatient conditions.

      Data were collected prospectively in 280 practices of gastroenterology

      and/or internal medicine as well as endoscopy units in hospitals between 1 October 2003 and 31 December 2006

      and included the entire number of complete investigations

      of the colon in the centers during the respective time

      period.

      Completeness of colonoscopy was a prerequisite for inclusion in the study

      Exclusion criteria:

      If participants reported symptoms suggestive of disease of the lower gastrointestinal tract, including rectal bleeding

      6 months earlier, marked changes in bowel habits or lower abdominal pain that would itself require colonoscopy.

      Other exclusion criteria were earlier bowel surgery, surveillance

      after previous polypectomy or surgery of colorectal

      adenomas or cancer or positive hemoccult tests.
      N = 269144 colonoscopies

      Male 119 264 (44.3%)

      Female 149 880 (55.7%)
      0.02% of the colonoscopies and 0.09% of polypectomies.0.16% of colonoscopies and 0.8% of polypectomiesNo mortality seen in study groupCardiopulmonary complication in 0.10% of the colonoscopies

      Most of the bleeding managed endoscopically 0.03% needed surgery

      90% procedures performed with conscious sedation
      Crispin et al. (2009)Population studyInclusion criteria:

      Age > 18 yrs

      Data from compulsory health insurance (CHI) members

      who underwent colonoscopies in 2006.

      Data were documented prospectively in the Electronic

      Colonoscopy Documentation of the Bavarian Association of CHI Physicians, a registry of outpatient

      colonoscopies performed in practices

      throughout Bavaria, Germany

      Exclusion criteria:

      Younger than 18 yrs

      Duplicate records
      N = 236087

      Median age = 61 years

      Male = 43.29%
      69/236087 (0.03%)520/236087 (0.22%)Not reportedLess than a quarter (23.72 %) were screening cases, the rest had specific indications (60.53% clinical signs and symptoms,

      10.98% adenoma surveillance, 4.78% cancer aftercare).

      A total of 735 patients (0.31 %) suffered complications

      152 (0.06%) cardiopulmonary complications seen

      50/69 (72%) perforations required surgery.

      Male sex, higher age, nonscreening indication, biopsies, polypectomies, and absence of sedation/analgesia were indicative of a higher bleeding

      risk. Perforations were also related to biopsies and polypectomies.

      Biopsy performed in 30.67% and polypectomy in 24.38%.

      Patients with large lesions > 3 cm exhibited an approximately 30-fold bleeding risk compared with patients

      with small ones < 0.5 cm (OR: 27.522, 17.198–44.049).

      The risk of perforation with the largest lesions (compared with the smallest

      ones) was augmented by a factor of approximately 30 (OR for lesions

      > 3 cm: 31.485, 95%CI 6.368–155.664)
      Singh et al. (2009)Retrospective studyInclusion criteria:

      Lower GI endoscopies performed on adult (aged > 16 years) outpatients

      at 6 Winnipeg hospitals between April 1, 2004, and March 31, 2006.

      Exclusion criteria:
      N = 24,509

      Mean age ±SD =

      59 ± 15 yrs

      Women = 56%
      Perforation post polypectomy

      1.8/1000

      Perforation after colonoscopy

      1/1000

      After stricture dilation

      58.8/1000

      After colonoscopy with biopsy

      0.5/1000

      Flex sig without additional procedure

      0.8/1000

      Flex sig with biopsy

      3.1/1000
      Bleeding post polypectomy

      6.4/1000
      1/24509

      Due to post op complications after surgery for perforation after snare polypectomy of small polyp in cecum
      General surgeons performed

      13,705 (56%), gastroenterologists 9618 (39%), and family physicians 1180 (5%) of the procedures.

      There were 303 admissions with potential complications

      The overall rate of complications was 2.9/1000

      procedures

      The complication rate was highest for endoscopists performing fewer than 200 procedures per year (5.4/1000 vs 2.7/1000 for the rest, P = .02, relative risk 2 [95% CI, 1.1-3.7]

      The mean (± SD) age of the individuals who had procedure-related complications was

      65 ±15 years; 65% were men.

      The rate of complications for individuals older than 50 years was 3.3 per 1000 procedures (60/17,918).
      Sewitch et al. (2012)Prospective cohort studyInclusion criteria:

      Between 2007 and 2009, individuals 50 to 75 years of age who underwent outpatient colonoscopy

      and were covered by the provincial health insurance plan (Régie de l’assurance maladie du Québec [RAMQ])

      Recruited from seven university-affiliated hospitals in Montreal (Quebec); they were approached by the research

      assistant before colonoscopy and explained the purpose of the study.

      Individuals who provided written informed consent were included

      Exclusion criteria:

      Not reported
      N = 2134

      Mean age = 60.9 yrs

      Male = 50.1%
      0% [95% CI 0.00% to 0.17%]).1/2134

      Post polypectomy
      0% [95% CI 0.00% to 0.17%]).33/2134 (1.55% [95% CI 1.06% to 2.16%]) were hospitalized within 30 days.

      Rate of serious complications was 0.05% (95% CI 0.00% to 0.26%) for all colonsocopies

      and 0.18% (95% CI 0.00% to 1.01%) for colonoscopies with at least one polypectomy
      Cooper et al. (2013)Population based studyInclusion criteria:

      Patients undergoing outpatient diagnostic colonoscopy was identified using the January 1, 2000, through November 30, 2009, Medicare Carrier and Outpatient files

      Patients 66 years or older (to allow measurement of comorbidities during the year prior to the colonoscopy

      procedure) and were receiving Medicare benefits

      through Part A and Part B for at least 1 year prior to and 30

      days after the colonoscopy

      Exclusion criteria:

      Patients enrolled in Medicare-sponsored managed care plans during the 1-year period

      prior to and 30-day period after the colonoscopy were excluded because of the high likelihood of incomplete claims
      N = 165 527 procedures in 100 359 patients, including 35 128 procedures with anesthesia services (21.2%) and 130399 colonoscopies (78.8%) without anesthesia.

      Mean (SD) age =

      75.5 (6.4) years

      Female = 55%

      White = 85.1%
      101/1655270Overall 30-day mortality 0.29%

      It was similar in the anesthesia (0.32%) and nonanesthesia (0.28%)

      groups (P = .29)

      overall 1-year mortality 2.68%

      It was similar in the anesthesia (2.82%) and nonanesthesia (2.64%) groups (P = .06).
      Patients were monitored from the index colonoscopy for up to 30 days after the procedure for the occurrence of specific

      complications and up to 1 year for death.

      Complications were documented after 284 procedures (0.17%), included aspiration (n=173), splenic injury (n=12).

      Overall complications were more common in cases with anesthesia assistance (0.22% [95% CI, 0.18%-0.27%]) than in others (0.16% [0.14%-0.18%]) (P<.001), as was aspiration (0.14% [0.11%-0.18%] vs 0.10% [0.08%-0.12%], respectively; P=.02).

      Predictors

      Of complications included age greater than 70 years, increasing comorbidity, and performance of the procedure

      in a hospital setting. In multivariate analysis, use of anesthesia services was associated with an increased complication

      risk (odds ratio, 1.46 [95% CI, 1.09-1.94]).
      Denis et al. (2013)Retrospective cohort studyInclusion criteria:

      Colonoscopies performed in patients aged 50–74 undergoing a colonoscopy

      for a positive guaiac-based fecal occult blood test between September 2003 and February 2010

      within the population-based CRC screening programme organized

      in Alsace, a region in eastern France

      Exclusion criteria:

      People with serious illness, recent CRC screening or high CRC risk
      N = 102777 colonoscopies10 (1.0‰, 95% CI 0.4–1.6)31 (3.0‰, 95% CI 2.0–4.1)0All events definitely, probably and possibly related to colonoscopy occurring within 30 days of the colonoscopy were taken into account.

      Overall 250 (24.3‰ colonoscopies)

      AEs were recorded in 249 patients. They were classed as mild (n = 202, 80.8%), moderate (n = 29, 11.6%) and severe (n = 19,

      7.6%)

      The rate of moderate and severe AEs was 4.7‰ (95%

      CI 3.4–6.0), 8.8‰ for therapeutic and 0.8‰ for diagnostic colonoscopies

      (p < 0.001)
      Hamdani et al. (2013)Retrospective cross-sectional studyInclusion criteria:

      18 years and older and had an inpatient or outpatient colonoscopy procedure code in any facility within the Geisinger Health System during the period from January 1, 2002 to August 25, 2010

      Exclusion criteria:

      Not reported
      N = 80118

      Age category (yr) =

      18-50 13698 (17.11) Perf 5 (10.00%)

      50-65 38695 (48.33) Perf 0 (20.00%)

      65-80 22954 (28.67) Perf 20 (40.00%)

      80+ 4271 (5.90%)

      Perf 15 (30.00%)

      Gender

      Male 38972 (46.68%) Perf 16 (32.00%)

      Female 41087 (51.32%)

      Perf 34 (68.00%)
      50/80118

      0.06% (95%CI: 0.05-0.08) or a rate of 6.2 per 10000 colonoscopies
      Not reportedNot reportedThe study outcome was the diagnosis of colonic

      Perforation using

      International Classification of Disease,

      9th revision (ICD-9) codes 569.83 and 998.2, defined as perforation of intestine and accidental puncture or laceration

      during a procedure, 7 d after the day of colonoscopy

      For every year increase in age, the risk of a perforation increased by 7% (95%CI: 5-9) with the

      incidence of perforation increasing from 2.6 cases per 10000 in the 50-64 year old age group to 31.7 cases per 10000 in the 80+ year old age group

      Significant risk factors for perforation: age, gender,

      BMI, albumin level, ICU patients, inpatient setting, and abdominal pain and Crohn’s disease as indications for colonoscopy
      Samalavicius et al. (2013)Retrospective multicenter case series studyInclusion criteria:

      Patients with iatrogenic

      Full thickness large bowel perforations resulting from colonoscopy within the period January 1, 2007, to December 31, 2011

      Representatives of 14 Lithuanian public and private hospitals participated in the survey.

      Exclusion criteria:

      Not reported
      N = 56882 colonoscopies

      Gender:

      Female 23 57.5%

      Male 17 42.5%

      Mean age 70 (39-85 years)
      Diagnostic :

      28 of 49,795 patients (0.056%)

      therapeutic colonoscopies : 12 of 7,087

      patients (0.169%)

      Total incidence:

      40/56882 (0.07%)
      Not reported6/40 (15%) deaths

      All in diagnostic colonoscopy group due to the perforation
      All patients underwent surgery either primary repair (70.0 %) or bowel resection (30.0 %).

      The most common site of perforation is sigmoid colon and rectosigmoid

      junction, at 70 %. Risk rises when colonoscopy is performed

      in low-volume practice centers.

      Urgent surgical management resulted in overall mortality rate of 15.0 %

      and morbidity of 37.5 %.
      Stock et al. (2013)Retrospective matched cohort.Inclusion criteria:

      Data collected between January 1, 2000, and December 31, 2008, from a random sample of individuals aged 20 years and older undergoing colonoscopy insured by the AOK in Hesse (N _ 326,652 [18.75% of AOK-insured persons in Hesse in 2000]).

      Exclusion criteria:

      The examined individual was not insured by the AOK over the 12 months before the colonoscopy

      date (this time frame was used to ascertain comorbidity)

      until 30 days after the colonoscopy date

      In case of multiple colonoscopies per individual, only the first procedure coded in the insurance data was considered
      N = 33,086 individuals who underwent colonoscopy as an outpatient (8658 screening, 24,428 nonscreening) and 33,086 matched controls who did not undergo colonoscopy.

      Mean age screening =

      66 years

      Mean age non screening =

      59 years

      Female screening =

      55%

      Female nonscreening =

      56%
      0.8 (95% CI 0.3-1.7) per 1000 screening colonoscopies

      0.7 (95% CI, 0.4-1.1) per 1000 nonscreening colonoscopies
      0.5 (95% CI, 0.1-1.2) per 1000 screening colonoscopies

      1.1 (95% CI, 0.8-1.7) per 1000 nonscreening colonoscopies
      0.6 (0.2-1.3) per 1000 screening colonoscopy

      1.6 (1.2- 2.2) per 1000 nonscreening colonoscopy

      Overall

      1.4 (1.0-1.8) per 1000
      Adverse events within 30 days of colonoscopy recorded
      Warren et al

      2009
      Population based matched cohort studyInclusion criteria:

      Random 5% sample of Medicare beneficiaries, age 66 to 95 years, who underwent outpatient colonoscopy between 1 July

      2001 and 31 October 2005, matched with beneficiaries who did not have colonoscopy.

      Inpatient procedures

      All patients to have continuous enrollment in Medicare Parts A and B

      and fee-for-service coverage during the year before their

      colonoscopy and for 30 days after the procedure

      Exclusion criteria:

      Outpatient procedures

      Procedures performed in patients at high risk of perforation ( diagnosis of diverticulitis,

      Crohn disease or ulcerative colitis, or colorectal cancer)

      Incomplete procedure coded by physician

      Persons younger than 66 years

      patients who had more

      than 2 colonoscopies during study period

      patients who had 2 colonoscopies less than 60 days apart

      persons who had their procedure

      outside of a SEER area

      Colonoscopies with no corresponding beneficiary in the match group
      N = 53220 colonoscopies

      10.1% screening, 33.6 % diagnostic and 56.3% procedures involving polypectomy.

      Male = 22174 (41.7%)

      Female 31046 (58.3%)

      Age

      66-69y 24.3%

      70-74y 31.2%

      75-79y 25%

      80-84y 14%

      ≥85 5.5%
      0.6 per 1000 procedures8.7 per 1000 procedures in polypectomy group

      2.1 per 1000

      Procedures in screening group
      Not calculatedall adverse events occurring within 30

      days after outpatient colonoscopy that were severe enough

      to require an emergency department visit or hospitalization were identified.

      Risk for paralytic ileus was also higher in the polypectomy group (4.8 per 1000 procedures) than in the diagnostic group (1.3 per 1000 procedures) or screening group (1.1 per 1000 procedures). Risk for cardiovascular events was higher in the polypectomy

      group (23.4 per 1000 procedures) or diagnostic group

      (15.4 per 1000 procedures) than in the screening group

      (9.9 per 1000 procedures). For all types of colonoscopies, the most common cardiovascular event was arrhythmia (10.2 per 1000 procedures).
      Supplementary Table 2Evidence table for the EMR/ESD studies
      Author(s)/Year of PublicationStudy DesignInclusion/ExclusionPatient CharacteristicsPerforation RateBleeding RateNotes
      Hiraishi et al. (2010)

      Location: Japan
      Prospective CohortInclusion Criteria: Patients that underwent ESD for colorectal neoplasia

      Exclusion Criteria:

      None explicitly stated
      N = 25 consecutive patients

      Age (mean) = 67.4 years

      Male = 64%

      Race = NR

      Mean tumor diameter = 32.2mm

      Mean removed specimen diameter = 38.2mm

      Mean procedure time = 42.6 minutes

      Follow-up period (mean) = 19.8 months
      0%0%2 patients required additional intervention (surgery).

      No complications observed.

      All patients in this cohort had ESD performed with SBK (stag beetle knife).
      Moss et al. (2010)

      Location: Australia
      RCT, double-blindedInclusion Criteria: Aged >18 years, able to give informed consent and were referred to the Endoscopy Unit for ER of treatment-naïve laterally spreading tumor or sessile lesion >20mm in size

      Exclusion Criteria: (1) Previous attempt at removal or hot biopsy of lesion, (2) personal hx of allergy to gelatin, SG or any other type of colloidal plasma expander, Haemaccel (3) use of clopidogrel or aspirin plus dipyridamole within 7 days, or use of warfarin within 5 days of procedure (4) therapeutic dose of unfractionated heparin within 6h or low-molecular weight heparin within 12h of procedure, (5) known clotting disorder, (6) pregnancy, or (7) lactation
      N = 80 patients

      Age (mean) = 69 years

      Male = 56.3%

      Race = NR

      Randomized to receive Normal Saline (NS) or Succinylated Gelatin (SG); patients = 39 NS vs. 41 SG

      Lesion Size (mean) = 40mm (SG) vs. 35mm (NS)

      Mean procedure time = 12.0 minutes (SG) vs. 24.5 minutes (NS); p = 0.006
      0%2% (SG) vs. 5% (NS)3 patients with significant post-procedure bleeding (1 SG / 2 NS) were hospitalized for treatment.

      Sydney Resection Quotient (SRQ) significantly higher in SG group (10.0 vs. 5.9, p = 0.004).

      No adverse events reported that were attributable to SG.
      Saito et al. (2010)

      Location: Japan
      Retrospective, Case-ControlInclusion Criteria:

      Noninvasive pattern, as determined by magnification chromoendoscopy:

      ESD = LST-NG lesion >20mm (definite), LST-G lesion >40mm (relative), or large villous tumor, intramucosal lesion, recurrent lesion or residual intramucosal lesion showing non-lifting sign after EMR

      EMR/EPMR = Any lesion <20mm (definite), or LST-G lesion >20mm and <40mm (definite)

      Exclusion Criteria = None explicitly stated
      N = 145 ESD, 228 EMR patients

      Age (mean) = 64 years

      Sex = NR

      Race = NR

      Tumor size (mean) = 28mm (EMR) vs. 37mm (ESD)

      Mean procedure time = 29 minutes (EMR) vs. 108 minutes (ESD)
      1.3% (EMR) vs. 6.2% (ESD)3.1% (EMR) vs. 1.4% (ESD)All complications treated endoscopically.

      ESD = higher en bloc resection rate (84% vs. 33%, p <0.0001).

      33 patients in the EMR group required additional EMR due to recurrence vs. 3 in the ESD group (p <0.0001).

      Despite a longer procedure time and higher perforation rate, ESD achieved higher en bloc and curative resection rates vs. EMR.
      Salama et al. (2010)

      Location: Australia
      Prospective

      Cohort
      Inclusion Criteria: Referred for endoscopic resection of a colorectal neoplasm 20mm or larger

      Exclusion Criteria: None explicitly stated
      N = 154 lesions in 140 consecutive patients

      Age (mean) = 68 years

      Male = 57.1%

      Race = NR

      Lesion type: Sessile (117) vs. Pedunculated (37)

      Lesion size (mean) = 29mm

      Mean procedure time = 55 minutes
      1.9% (all in Sessile lesions)1.3% (all in Sessile lesions)Complete endoscopic removal of lesion in 95% of cases.

      20 patients referred to surgery (14%).

      Endoscopic follow-up data was available in 90% of cases, revealing 5 patients to need endoscopic treatment of residual adenoma to achieve complete clearance of disease.
      Seo et al. (2010)

      Location: South Korea
      Retrospective CohortInclusion Criteria:

      Underwent EPMR using a submucosal saline injection technique

      Sessile polyp, >20mm

      Exclusion Criteria:

      Co-existence of synchronous advanced colorectal cancer

      Non-lifting tumor

      Encircling lesion > 70%

      Transfer to outside institution

      Suspicion of muscle invasion on EUS

      Recurrent tumor
      N = 50 lesions in 47 patients

      Age (mean) = 60 years

      Female = 51.1%

      Race = NR

      Mean polyp size = 30.1mm
      0%12% (5 during procedure, 1 post-procedure)All bleeding episodes were managed by endoscopic clip and/or APC.

      4 local recurrences were detected at 3-months post-EPMR; 1 occurred at 14-months post-EPMR.

      Recurrence was significantly higher in malignant polyps (33.3% vs. 3.1%, p = 0.05).
      Toyonaga et al. (2010)

      Location: Japan
      Retrospective CohortInclusion Criteria: Underwent ESD for LST of >20mm diameter

      Exclusion Criteria: None explicitly stated
      N = 99 LST-NG and 169 LST-G lesions

      Age (mean) = 68 years (LST-NG) vs. 69 years (LST-G)

      Male = 52.6%

      Race = NR

      Tumor diameter (median) = 28mm (LST-NG) vs. 36mm (LST-G); p <0.0001

      Resected specimen diameter (median) = 40mm (LST-NG) vs. 46mm (LST-G); p <0.0001

      Procedure time (median) = 69 min (LST-NG) vs. 60 min (LST-G)
      5.1% (LST-NG) vs. 0.59% (LST-G); p = 0.0270% (LST-NG) vs. 0.59% (LST-G); p = NSEn bloc resection rate, en bloc R0 resection rate and en bloc curative resection rate similar in both groups (LST-NG: 99%, 98%, and 88%; LST-G: 99%, 98%, and 91%).

      No recurrence seen in either group.

      5/6 perforations were seen intra-procedure and treated with endoscopic clip and antibiotics; one delayed perforation was treated surgically.

      ESD was effective for both LST-NG and LST-G; however, degree of technical difficulty is higher in LST-NG group.
      Khashab et al. (2009)

      Location: US
      Retrospective CohortInclusion Criteria:

      Underwent piecemeal resection of a large sessile polyp (>20mm)

      Attended subsequent follow-up examination at 3-6 months following the initial resection, and at least 1 year (up to 6-years post-initial resection) after the initial resection

      Exclusion Criteria:

      Pedunculated polyps

      Cancerous polyps that underwent subsequent surgical resection

      Polyps deemed endoscopically unresectable

      Patients that did not complete all follow-up at this institution, as outlined per inclusion
      N = 136 lesions in 132 patients

      Age (mean) = 67.4 years

      Male = 50.8%

      Race = NR

      Mean polyp size = 32.8mm
      0%4.5%All patients with bleeding (n = 6) were successfully managed by endoscopy.

      24 patients had macroscopically evident residual adenoma at follow-up – 18 at first follow-up, and 6 with a “late” recurrence.

      Normal macroscopic appearance of the polypectomy site and negative scar biopsy specimen at first follow-up is predictive of long-term eradication.
      Luigiano et al. (2009)

      Location: Italy
      Retrospective CohortInclusion Criteria: Patients that underwent EMR for sessile or flat polyps measuring 20mm or greater in size

      Exclusion Criteria:

      Lack of follow-up data and/or hospitalization at outside institutions (not enough information available)
      N = 148 patients

      Age (mean) = 69.2 years

      Male = 63.5%

      Race = NR

      Mean polyp size = 30.7mm

      Lesion Type = 74 Sessile vs. 74 Flat
      0.68%10.1% (13/15 bleeds occurred during procedure)One early bleed (10 hours post-procedure) and delayed bleed (36 hours post-procedure) were observed. All bleeds (including procedural bleeding) were treated endoscopically – clips (4), injection (7), and APC (4).

      Perforation required surgical management.

      Recurrence was observed in 4.2% of patients, and was more often found in patients with giant (>40mm) polyps (p = 0.014).
      Conio et al. (2010)

      Location: Italy
      Prospective CohortInclusion Criteria:

      Patients with Sessile polyps or Laterally Spreading Tumors (>20mm) that underwent EMR-C (colorectal)

      Only benign and resectable (as determined by the endoscopist) lesions were selected for inclusion

      Exclusion Criteria:

      Ulceration, Induration, depressed lesions and the absence of lifting after submucosal injection
      N = 282 polyps in 255 consecutive patients (146 sessile polyps vs. 136 laterally spreading tumors)

      Age (median) = 70 (SP), 64.5 (LST-NG), and 73 (LST-G) years

      Male = 56.0%

      Race = NR

      Median polyp size = 25mm (SG), 30mm (LST-NG), and 30mm (LST-G)

      Procedure time (median) = 40 minutes
      0%7% (all procedural bleeding)All bleeding occurred during the procedure and was treated endoscopically.

      Endoscopic follow-up was available in 200 patients with 216 adenomas and demonstrated a recurrence rate of 4%. Recurrences were treated with ablation and/or additional resection.
      Bahin et al. (2015)

      Location: Australia
      RCT, Multi-CenterInclusion Criteria:

      Patients referred for WF-EMR of a colorectal Sessile or Laterally Spreading lesion (>20mm)

      Exclusion Criteria:

      Lesion size less than 20mm

      Paris Classification of 0-1p appearance

      Suspected invasive disease

      WF-EMR of multiple lesions in 1 session

      Incompletely resected lesion

      Muscularis propria injury (suspected or confirmed)

      Occurrence of major intraprocedural bleeding requiring intervention for hemostasis (confounding factor)
      N = 347 patients, 172 PEC vs. 175 control

      Age (mean) = 67.1 years

      Male = 51.3%

      Race = NR

      Median lesion size = 39.5mm (WF-EMR + PEC group), 39.8mm (WF-EMR group), p = 0.13

      Mean procedure time = 35 minutes (WF-EMR + PEC group), 30 minutes (WF-EMR group)
      0%6.6% (5.2% PEC vs. 8.0% Controls)Patients were randomized 1:1 to receive prophylactic endoscopic coagulation (PEC) to determine if it reduced the incidence of clinically significant post-endoscopic mucosal resection bleeding (CSPEB).

      CSPEB occurred in 9 patients receiving PEC vs. 14 controls (p = 0.3). PEC does not significantly decrease the incidence of CSPEB after WF-EMR.
      Lee et al. (2015)

      Location: South Korea
      Retrospective CohortInclusion Criteria:

      Patients that underwent ESD with the indication of colorectal carcinoma or adenoma

      Exclusion Criteria:

      Pathologic result of hyperplastic polyps, inflammatory polyps, neuroendocrine tumors, or no tumor detected

      Resection with EMR or conventional polypectomy
      N = 173 lesions in 170 patients

      Age (mean) = 65.01 years

      Male = 61.3%

      Race = NR

      Mean tumor size = 25.95mm

      Resected specimen diameter (mean) = 31.76mm

      Procedure time (mean) = 77.46 minutes

      En bloc resection = 88.4%
      10.98%3.47%Perforation occurred in 19 cases, with 17 being classified as microperforations.

      Bleeding occurred in six cases, with half being minor and major bleeds.

      Complications occurred in 26.6% (n = 46) of cases (including post-coagulation syndrome). Surgical intervention was required in 22/46 cases.
      Wada et al. (2015)

      Location: Japan
      Prospective Cohort, Multi-CenterInclusion Criteria:

      Patients with colorectal lesions of 20mmn or larger that underwent endoscopic resection

      Exclusion Criteria:

      Patients that underwent ESD were excluded from the final analysis (all patients were consented to reduce selection bias; however, only interested in EMR or polypectomy outcomes)
      N = 1845 colorectal lesions; only 1029 included in the final analysis

      Age (mean) = 65.2 years

      Male = 61.9%

      Race = NR

      EMR vs. Polypectomy = 866 vs. 163 cases

      Tumor size (mean) = 26.4mm
      0.78%1.6%Post-procedure bleeding occurred in 18 cases, all treated endoscopically with endoclips or hemostatic forceps. One case required additional intervention (blood transfusion).

      Perforation occurred in 8 (EMR) cases, all diagnosed intra-procedure. 7/8 were managed with endoclips, with one case requiring surgical intervention.

      Risk factors for bleeding were in multivariate analysis were only patients under 60 years of age. Perforation risk factors were en bloc resection and Vienna Classification category 4-5 (high-grade dysplasia and submucosal carcinoma).
      Zhang et al. (2015)

      Location: China
      Prospective RCTInclusion Criteria:

      Patients referred for ESD or EMR for a colorectal tumor

      Tumor size between 10mm and 40mm

      Exclusion Criteria:

      Blood disorder

      Hx of colorectal surgical resection

      Taking anticoagulant medication

      Lesions displaying an invasive pattern and/or those lesions that were recurrent or residual tumors

      Coagulation dysfunction
      N = 348 patients, randomized to clip-closure (174) or no clip-closure (174) group

      Age (mean) = 67.9 (clip-closure group) vs. 64.2 (no-clip closure group)

      Male = 62.9%

      Race = NR

      Tumor size:

      10mm-20 mm (n = 111 clip-closure group vs. 107 non-clip closure group)

      20-40mm (n = 63 clip-closure group vs. 67 non-clip closure group)

      Procedure duration (mean) = 38.1 minutes (clip-closure group) vs. 30.9 minutes (non-clip closure group)
      0.57%4.0%1 patient in the non-clip closure group experienced a perforation during ESD, managed with endoscopic clips.

      1 patient in the clip-closure group was found to have a microperforation on CT post-procedure (free air) and was managed conservatively.

      14 patients (n = 12 non-clip closure v. 2 clip-closure) experienced delayed post-operative bleeding (p = 0.012). All bleeds were treated endoscopically, no patients in either group required blood transfusion or surgical intervention.
      Ahlawat et al. (2011)

      Location: US
      Retrospective CohortInclusion Criteria:

      Patients that underwent polypectomy of a >20mm colorectal polyp

      Exclusion Criteria:

      None explicitly stated
      N = 183 polyps in 174 patients

      Age (mean) = 64 years

      Male = 55%

      Race = NR

      Sessile polyps = 84%

      Polyp size = 73% were 20-30mm
      2.2%9.3%17 patients had post-polypectomy bleeding – 5 immediate, 12 delayed. All patients with immediate bleeding were treated with endoscopic clips. For those with delayed bleeds, all required hospitalization (3 needing blood transfusion).

      Of 4 patients with a post-polypectomy perforation, 3 required surgical intervention. 1 patient was managed with endoscopic clip closure of the perforation.

      Recurrence of adenoma was noted in 12% of patients on follow up (>1-year post-polypectomy), all managed endoscopically.
      Moss et al. (2011)

      Location: Australia
      Prospective Cohort, Multi-CenterInclusion Criteria:

      Patients referred for EMR of sessile colorectal polyps 20mm or larger

      Exclusion Criteria:

      None explicitly stated
      N = 514 lesions in 479 patients

      Age (mean) = 68.5 years

      Male = 53%

      Race = NR

      Lesion Size (mean) = 35.6mm

      Procedure Duration (mean) = 25.3 minutes

      Complete excision achieved in 89.2% of EMR cases
      1.3%2.9%14 patients were admitted for bleeding post-procedure. 7 were treated conservatively, 6 patients underwent repeat colonoscopy (4 received endoclips or coagulation application), and 1 required an extended right hemicolectomy.

      6 patients had a perforation post-procedure. Four patients were stabilized with endoscopic clip; however, ¼ required readmission and received conservative management (LOS = 9 days). Two patients required surgical intervention.

      Risk factors for submucosal invasion were as follows: Paris Classification 0-IIa+c morphology, non-granular surface, and Kudo pit pattern type V.

      Predictors of recurrence after effective EMR were lesion size greater than 40mm (p < .001), and use of APC (p = 0.017).
      Binmoeller et al. (2012)

      Location: US
      Prospective CohortInclusion Criteria:

      Referred for EMR of large sessile polyp (defined as a lesion with a thickness less than half of the maximum width)

      Benign adenoma on previous biopsy

      Benign appearance on high-definition colonoscopy without stigmata of malignancy (ulceration, bleeding, induration, Kudo pit pattern V)

      Size equal to or greater than 20mm

      Exclusion Criteria:

      None explicitly stated
      N = 60 consecutive patients with 62 sessile colorectal polyps

      Age (mean) = 65.4 years

      Female = 53.3%

      Race = NR

      Polyp size (mean) = 33.8mm

      Procedure Duration (mean) = 48.9 minutes

      Resection time (mean) = 21.4 minutes
      0%5%Evaluated novel technique of EMR performed with “water immersion” (UEMR).

      Delayed bleeding occurred in 3 patients and was managed conservatively.

      Mean follow-up time was 20.4 weeks, with one patient experiencing a recurrence of adenoma.
      Fasoulas et al. (2012)

      Location: Greece
      Prospective RCT, Multi-CenterInclusion Criteria:

      LST > 30mm detected during colonoscopy, benign appearance, and respectability of the lesion

      Exclusion Criteria:

      Firm consistency, ulceration, friability, appearance of expansion of normal tissue immediately surrounding the lesion indicating the presence of cancer spreading into the surrounding submucosal space, and converging fold (2 or more) toward the lesion predicting submucosal invasion by cancer cells, even with negative biopsies

      The presence of a “non-lifting” sign after submucosal injection of solution

      A lesion with biopsies suspicious for invasive cancer

      Patients with previous incomplete resections
      N = 49 patients (25 in Hydroxyethyl Starch vs. 24 in Normal Saline group)

      Age (mean) = 68 (HS group) vs. 67 (NS group)

      Male = 64% (HS) vs. 50% (NS)

      Race = NR

      Lesion size (median) = 45mm (HS) vs. 46mm (NS)

      Recurrence = 20% (HS) vs. 29.2% (NS)
      4% (HS) vs. 0% (NS)4% (HS) vs. 25% (NS)Evaluated hydroxyethyl starch (HS) vs. normal saline + epinephrine (NS) to determine if HS allowed for prolonged elevation of submucosal cushion for EMR of LSTs.

      Six patients had mild intraprocedural bleeding successfully controlled endoscopically with epinephrine injection and coagulation by coagrasper (4) or hemoclips (2). One patient had delayed bleeding (self-limited).

      One patient in the HS group presented with macroperforation following the procedure (42mm lesion) and was managed surgically.

      HS provided a more prolonged submucosal elevation (p = 0.001), and a shorter procedure time (p = 0.013) than NS (with similar safety profile).
      Lee et al. (2012)

      Location: South Korea
      Prospective CohortInclusion Criteria:

      Patients referred for resection of colorectal tumors 20mm or larger

      Exclusion Criteria:

      Colorectal tumors with endoscopic findings such as hardness, ulceration, friability, and spontaneous bleeding (indicative of massive submucosal invasion)
      N = 140 tumors in 135 patients (EMR group) vs. 69 tumors in 67 patients (EMR-P group) vs. 314 tumors in 309 patients (ESD group)

      Age (mean): 63 (EMR) vs. 62 (EMR-P) vs. 61 (ESD)

      Sex (ratio, male: female) = 1:0.57 (EMR) vs. 1:0.64 (EMR-P) vs. 1:0.80 (ESD)

      Tumor size (mean) = 21.7mm (EMR) vs. 23.5 (EMR-P) vs. 28.9 (ESD)

      En-bloc resection rates = 42.9% (EMR) vs. 65.2 (EMR-P) vs. 92.7% (ESD)

      Recurrence rates = 25.9% (EMR) vs. 3.2% (EMR-P) vs 0.8% (ESD)
      0% (EMR) vs. 2.9% (EMR-P) vs. 8.0% (ESD)0% (EMR) vs. 2.9% (EMR-P) vs. 0.64% (ESD)Perforation occurred in 2 cases of EMR-P and 25 in ESD. Endoscopic clipping was performed for 20/25 ESD cases, and both of the EMR-P cases. 2 ESD perforations required surgical intervention, with the remaining (3) cases managed conservatively.

      Delayed bleeding occurred in 2 patients in the EMR-P and 2 in the ESD groups. All bleeding was controlled with conservative management.
      Terasaki et al. (2012)

      Location: Japan
      Retrospective CohortInclusion Criteria:

      Patients that underwent endoscopic treatment of an LST greater than 20mm diameter

      Exclusion Criteria:

      None explicitly stated
      N = 267 consecutive patients (61 ESD vs. 28 Hybrid ESD vs. EMR vs. 108 EPMR)

      Age (mean) = 65.0 (ESD) vs. 70.3 (Hybrid ESD) vs. 65.5 (EMR) vs. 69.4 (EPMR) years

      Sex (ratio, male: female) = 38:23 (ESD) vs. 17:11 (Hybrid ESD) vs. 40:30 (EMR) vs. 59:49 (EPMR)

      Tumor size (mean) = 42.1mm (ESD) vs. 31.7mm (Hybrid ESD) vs. 24.2mm (EMR) 37.4mm (EPMR); p = 0

      0017

      Recurrence rates = 0% (ESD) vs. 0% (Hybrid ESD) vs. 1.4% (EMR) vs. 12.1% (EPMR)
      0% (ESD) vs. 7.1% (Hybrid ESD) vs. 1.4% (EMR) vs. 1.9% (EPMR)11.5% (ESD) vs. 0% (Hybrid ESD) vs. 7.1% (EMR) vs. 9.3% (EPMR)Hybrid ESD was defined as the use of a snare during the final stage of a procedure (following ESD for resection).

      Perforation occurred in 0 ESD, 2 Hybrid ESD, 1 EMR and 2 EPMR patients. All perforations were managed endoscopically.

      Delayed bleeding occurred in 7 ESD, 0 Hybrid ESD, 5 EMR and 10 EPMR patients. All delayed bleeding events were managed endoscopically and/or conservatively.
      Longcraft-Wheaton et al. (2013)

      Location: United Kingdom
      Prospective CohortInclusion Criteria:

      Patients referred for resection of large and difficult colonic polyps of at least 20mm or larger

      Exclusion Criteria:

      Polyps less than 20mm

      Polyps with features suggestive of malignancy
      N = 220 patients

      Age (mean) = 68 years

      Male = 61.4%

      Race = NR

      Polyp size (mean) = 43mm

      SMSA scoring system (size/morphology/site/access) = level 3 (9-12), level 4 (>12); levels 3 and 4 considered difficult lesions; SMSA level 3 (32%) vs. level 4 (63%)
      0.45%2.7%Six patients experienced delayed bleeding and required hospitalization, blood transfusion, or endoscopic intervention. No patients with post-EMR bleeding required surgical intervention.

      One patient experienced a microperforation, managed endoscopically.

      Complications were independent of lesion size and location, but were affected by the SMSA score (p = 0.018).

      Complete endoscopic clearance was achieved in 90% of this cohort with first endoscopic resection attempt.
      Repici et al. (2013)

      Location: Italy
      Prospective CohortInclusion Criteria:

      All LST’s 30mm or larger located within 15mm of the anal verge with no previous attempt at endoscopic resection in patients in patients 18-80 years of age

      Exclusion Criteria:

      Lesion infiltrating the submucosal layer as diagnosed by EUS, previous endoscopic resection attempts, ASA Class-III or higher, and disturbance of coagulative parameters
      N = 40 consecutive patients

      Age (mean) = 65.3 years

      Male = 67.5%

      Race = NR

      Polyp size (mean) = 46.8mm

      Procedure time (mean) = 86.1 minutes

      Recurrence = 2.5%
      2.5%5.0%Bleeding occurred in 2 patients, with intraprocedural hemostasis being successfully achieved with thermocoagulation and clipping.

      Perforation occurred in 1 patient and was endoscopically treated with clipping.

      En bloc resection rate was 90%, with curative resection in 80%.
      Xu et al. (2013)

      Location: China
      Retrospective CohortInclusion Criteria:

      Patients with LST’s that underwent ESD (lesions over 20mm, and those that were difficult to remove en bloc using EMR)

      Exclusion Criteria:

      Histology demonstrating cancer and invasion of the submucosa or permeating the vessels

      Presence of invasive type V pit pattern

      Lesion not amenable to ESD (no lifting elevation at submucosal injection)
      N = 137 lesions in 135 patients

      Age (mean) = 64.5 years

      Female = 50.4%

      Race = NR

      Tumor size (median) = 30mm

      Procedure duration (mean) = 52.3 minutes

      Recurrence = 0.84%

      En bloc resection = 94.9%

      Curative resection = 90.5%
      2.2%3.6%Three patients experienced perforation following ESD. 2/3 were treated successfully with endoscopic clips. The remaining case was treated surgically.

      Five patients experienced delayed postoperative bleeding. All cases were successfully managed endoscopically (clips).
      Bialek et al. (2014)

      Location: Poland
      Prospective CohortInclusion Criteria:

      Large, sessile polyp (type 1s) greater than 20mm which could not be removed in 1 piece

      LST-G regular type or mix type with a dominant nodule, greater than 20mm

      Non-granular LST, any size

      Exclusion Criteria:

      Lesions above 70mm in size
      N = 53 (37 ESD vs. 16 Hybrid ESD)

      Age (mean) = 64.0 years

      Sex = NR

      Race = NR

      Lesion size (mean) = 37mm (overall)

      Procedure time (median) = 70.0 (ESD) vs. 39.0 (Hybrid ESD) minutes

      En bloc resection rate = 86.5% (ESD) vs. 87.5% (Hybrid ESD)

      Curative resection rate = 81.1% (ESD) vs. 87.5% (Hybrid ESD)

      Recurrence Rate = 1.9%
      0%5.7%Three patients experienced bleeding post-procedure. 2/3 cases were controlled using electrosurgical hemostatic forceps or hemoclips. One case required admission (LOS = 7 days) and required 2 units of blood to stabilize.

      Recurrence occurred in one patient at 30-months post-procedure (Hybrid ESD).
      Knabe et al. (2014)Prospective Cohort, Multi-CenterInclusion Criteria:

      Patients with large, non-pedunculated lesions (larger than 20mm) that were referred for ER

      Exclusion Criteria:

      ASA class 3 or higher

      Patients receiving anticoagulation treatment
      N = 252 lesions in 243 consecutive patients

      Age (range, only) = 36 – 86 years

      Male = 58.0%

      Race = NR

      Resected lesion size (median) = 33mm

      En bloc resection rate = 11.5% vs. Piecemeal resection rate = 88.5%
      1.6%2.5% (4 delayed GIB in adenoma patients vs. 2 in polypectomy patients)Evaluated a standardized follow-up protocol following ER of large lesions.

      Four patients with resected adenomas experienced delayed GIB (>48h). All of these cases were successfully managed with clipping and epinephrine injection.

      Two patients (polypectomy) experienced major GIB (defined as a fall in hemoglobin >2g/dL). Both patients were managed blood transfusion and/or endoscopic therapy (epinephrine).

      Four perforations occurred that were all considered intraprocedural and were sealed with immediate clip placement.

      Evident residual neoplasia was noted after 3-6 months in 31.7% of lesions. After 12 months, 19 late recurrences were detected (16.4%). All residual adenomas were retreated with ER or treated with APC.
      Bae et al. (2016)

      Location: South Korea
      Retrospective CohortInclusion Criteria:

      Patients who underwent ESD for lesions greater than 30mm in diameter

      Exclusion Criteria:

      Submucosal or non-neoplastic tumors

      Neoplastic lesions smaller than 30mm
      N = 220 lesions in 218 patients

      Age (mean) = 63.0 years

      Male = 57.7%

      Race = NR

      Protruding tumor group = 30.5% vs. LST group (69.5%)

      Tumor size (mean) = 43.8mm (overall)

      En bloc resection rate = 87.7%

      Complete resection rate = 74.5%

      Procedure time (mean) = 75.5 minutes
      9.1%10.9%All intraprocedural bleeding events (n = 19) were managed successfully with endoscopic hemostasis during ESD procedure. 5 cases of post-procedural bleeding were treated conservatively after hemostasis with APC and/or hemoclip placement.

      Perforation occurred in 20 cases (no difference among groups). 18 cases were successfully treated using hemoclip, with the rest requiring antibiotic administration and admission for observation. No perforation required surgical management.

      En bloc resection and complete resection rates were lower in the protruding tumor group vs. LST (p = 0.001). Intra- and post-procedural bleeding were more frequent in protruding tumors vs. LST’s.

      Protruding tumors and size of greater than 60mm were independently associated with incomplete resection.
      Bahin et al. (2016)

      Location: Australia
      Prospective RCT, Multi-CenterInclusion Criteria:

      Referred for the management of LSL > 20mm

      Underwent wide-field endoscopic mucosal resection (WF-EMR)

      Exclusion Criteria:

      Suspicion of submucosal invasive carcinoma

      Surgery within 14 days of WF-EMR

      Technical failure of WF-EMR attempt
      N = 2012 patients

      Age (median) = 68 years

      Male = 52.4%

      Race = NR

      Lesion size (median) = 30mm

      Procedure duration (median) = 20 minutes

      En bloc resection rate = 15.5%
      NR6.7%A total of 135 cases of clinically significant bleeding (CSPEB) presented to the ED. 134/135 were hospitalized, and intervention was required in 54 patients (40.0%; colonoscopy in 52, angioembolization in 4 patients, surgery in 2 patients). Median time of CSPED = 24-48h.

      Patients were randomized to training cohort or test cohort (n = 1006 in each). Training cohort determined predictive factors to develop a scoring system to compare with the Test cohort.

      Predictors included lesion size >30mm (OR 2.5), and proximal colonic location (OR 2.3). The risk score scale comprised of lesion size >30mm (2 points), proximal colon (2 points), major comorbidity (1 point), and absence of epinephrine use (1 point). Probabilities of CSPEB of scores low (0-1), medium (2-4), and elevated (5-6) risk levels were 1.7, 7.1, and 17.5% in training cohort vs. 3.4, 6.2, and 15.7% in test cohort.
      Burgess et al. (2016)

      Location: Australia
      Retrospective CohortInclusion Criteria:

      Referred for the management (EMR) of sessile or laterally spreading colorectal lesions of greater than 20mm in size

      Exclusion Criteria:

      Patients that did not undergo EMR due to suspicion of malignancy or technical failure
      N = 912 lesions in 802 patients

      Age (mean) = 66.8 years

      Male = 51.4%

      Lesion size (mean) = 29.2

      Type III-V lesions = 28.9%

      En bloc resection rate = 13.0%

      Resected lesion size (en bloc, only) = 23.3mm
      0.1%NRLesions classified by deep mural injury (DM) class: I/II intact MP with/without fibrosis, III target sign, IV/V obvious transmural perforation with/without contamination). All type III-V were clipped, if possible.

      A delayed perforation (12-d post-procedure) presented in 1 patient (DM type II) following complete resection of 40mm Paris 0-IIa adenoma and admitted for 9 days.

      Type III-V lesions were associated with en bloc resection (OR 3.84, p = 0.005) and HGD/submucosal invasive cancer (OR 2.97, p = 0.014).
      Sauer et al. (2016)

      Location: Germany
      Prospective Cohort (Case-Series)Inclusion Criteria:

      Adults, aged 18+

      Sessile or laterally spreading adenomatous lesions larger than 20mm

      Exclusion Criteria:

      Coagulopathy (INR >1.5; thrombocytopenia <100g/L)

      Dual platelet inhibitor therapy or oral anticoagulation that could not be interrupted

      Pregnancy or lactation

      Signs of submucosal tumor invasion

      Life expectancy of less than 6 months
      N = 182 consecutive lesions in 178 patients

      Age (median) = 70 years

      Male = 57.9%

      Race = NR

      Lesion size (mean) = 41.0mm

      Procedure duration (mean) = 127.5 minutes

      En bloc resection rate = 88.4%

      Curative resection rate = 62.6%
      9.3%2.8%In 27 procedures (14.8%), technical difficulties (severe fibrosis, or lack of access to the lesion) resulted in conversion to EPMR (n = 24) or referral to surgery (n = 3).

      Microperforations occurred in 17 patients. All perforations were treated conservatively with hemoclips and antibiotics.

      Delayed bleeding occurred in 5 cases. All patients were treated endoscopically and successfully achieved hemostasis.
      Albeniz et al. (2016)

      Location: Spain
      Prospective Cohort, Multi-CenterInclusion Criteria:

      Patients referred for EMR of non-pedunculated colorectal lesions 20mm or larger

      Exclusion Criteria:

      Lesions smaller than 20mm
      N = 1255 lesions in 1214 patients

      Age (mean) = 67.9 years

      Male = 63.4%

      Race = NR

      ASA of III or IV = 30.5%

      Lesion size (mean) = 30.5mm

      En bloc resection rate = 24.6%
      1.4%3.7%17 cases of perforation were reported, 16/17 were successfully treated endoscopically. One patient was referred for surgical management.

      Delayed bleeding presented in 46 cases, with most (29; 63%) occurring 3-7+ days post-procedure. Patients were treated endoscopically and/or with blood transfusion. No patient required surgery.

      Predictors significantly associated with delayed bleeding presentation were age 75+ (OR 2.36), aspirin use (OR 3.16), ASA classification III or IV (OR 1.90), and lesion size of 40mm or larger (OR 1.91).

      Based on these factors, a predictive score was created indicating low risk (0-3) vs. average risk (4-7) vs. high risk (8-10). High-risk patients experienced a probability of 40% chance of presenting with delayed bleed.
      Figure thumbnail fx1a
      Supplementary Figure 1Meta-regression of postcolonoscopy perforation (A) and bleeding (B) from population-level studies with percentage of cohort undergoing polypectomy (% polypectomy) as a mediator covariate, following adjustment for age and gender. Data analyzed with a random-effects model, and 95% confidence intervals shown.
      Figure thumbnail fx1b
      Supplementary Figure 1Meta-regression of postcolonoscopy perforation (A) and bleeding (B) from population-level studies with percentage of cohort undergoing polypectomy (% polypectomy) as a mediator covariate, following adjustment for age and gender. Data analyzed with a random-effects model, and 95% confidence intervals shown.
      Figure thumbnail fx2a
      Supplementary Figure 2Meta-regression of EMR/ESD-related perforation (A) and bleeding (B) with type of procedure (ESD vs EMR), location (Asian vs Western), and mean diameter of polyp as mediator covariates. Data analyzed with a random-effects model. ESD, Endoscopic submucosal dissection.
      Figure thumbnail fx2b
      Supplementary Figure 2Meta-regression of EMR/ESD-related perforation (A) and bleeding (B) with type of procedure (ESD vs EMR), location (Asian vs Western), and mean diameter of polyp as mediator covariates. Data analyzed with a random-effects model. ESD, Endoscopic submucosal dissection.

      References

        • Whitlock E.P.
        • Lin J.S.
        • Liles E.
        • et al.
        Screening for colorectal cancer: a targeted, updated systematic review for the U.S. Preventive Services Task Force.
        Ann Intern Med. 2008; 149: 638-658
        • Levin T.R.
        • Zhao W.
        • Conell C.
        • et al.
        Complications of colonoscopy in an integrated health care delivery system.
        Ann Intern Med. 2006; 145: 880-886
        • Lin J.S.
        • Piper M.A.
        • Perdue L.A.
        • et al.
        Screening for colorectal cancer: a systematic review for the U.S. Preventive Services Task Force.
        JAMA. 2016; 315: 2576-2594
        • Fisher D.A.
        • Maple J.T.
        • Ben-Menachem T.
        • et al.
        • ASGE Standards of Practice Committee
        Complications of colonoscopy.
        Gastrointest Endosc. 2011; 74: 745-752
        • Saltzman J.R.
        • Cash B.E.
        • Pasha S.F.
        • et al.
        • ASGE Standards of Practice Committee
        Bowel preparation before colonoscopy.
        Gastrointest Endosc. 2015; 81: 781-794
        • Hassan C.
        • Repici A.
        • Sharma P.
        • et al.
        Efficacy and safety of endoscopic resection of large colorectal polyps: a systematic review and meta-analysis.
        Gut. 2016; 65: 806-820
        • Stapakis J.C.
        • Thickman D.
        Diagnosis of pneumoperitoneum: abdominal CT vs. upright chest film.
        J Comput Assist Tomogr. 1992; 16: 713-716
        • Tiwari A.
        • Sharma H.
        • Qamar K.
        • et al.
        Recognition of extraperitoneal colonic perforation following colonoscopy: a review of the literature.
        Case Rep Gastroenterol. 2017; 11: 256-264
        • Reumkens A.
        • Rondagh E.J.
        • Bakker C.M.
        • et al.
        Post-colonoscopy complications: a systematic review, time trends, and meta-analysis of population-based studies.
        Am J Gastroenterol. 2016; 111: 1092-1101
        • Bielawska B.
        • Day A.G.
        • Lieberman D.A.
        • et al.
        Risk factors for early colonoscopic perforation include non-gastroenterologist endoscopists: a multivariable analysis.
        Clin Gastroenterol Hepatol. 2014; 12: 85-92
        • Chukmaitov A.
        • Bradley C.J.
        • Dahman B.
        • et al.
        Association of polypectomy techniques, endoscopist volume, and facility type with colonoscopy complications.
        Gastrointest Endosc. 2013; 77: 436-446
        • Lohsiriwat V.
        Colonoscopic perforation: incidence, risk factors, management and outcome.
        World J Gastroenterol. 2010; 16: 425-430
        • Rutter M.D.
        • Nickerson C.
        • Rees C.J.
        • et al.
        Risk factors for adverse events related to polypectomy in the English Bowel Cancer Screening Programme.
        Endoscopy. 2014; 46: 90-97
        • Mukewar S.
        • Costedio M.
        • Wu X.
        • et al.
        Severe adverse outcomes of endoscopic perforations in patients with and without IBD.
        Inflamm Bowel Dis. 2014; 20: 2056-2066
        • Warren J.L.
        • Klabunde C.N.
        • Mariotto A.B.
        • et al.
        Adverse events after outpatient colonoscopy in the Medicare population.
        Ann Intern Med. 2009; 150: 849-857
        • Ranasinghe I.
        • Parzynski C.S.
        • Searfoss R.
        • et al.
        Differences in colonoscopy quality among facilities: development of a post-colonoscopy risk-standardized rate of unplanned hospital visits.
        Gastroenterology. 2016; 150: 103-113
        • Rabeneck L.
        • Paszat L.F.
        • Hilsden R.J.
        • et al.
        Bleeding and perforation after outpatient colonoscopy and their risk factors in usual clinical practice.
        Gastroenterology. 2008; 135: 1899-1906
        • Consolo P.
        • Luigiano C.
        • Strangio G.
        • et al.
        Efficacy, risk factors and complications of endoscopic polypectomy: ten year experience at a single center.
        World J Gastroenterol. 2008; 14: 2364-2369
        • Dafnis G.
        • Ekbom A.
        • Pahlman L.
        • et al.
        Complications of diagnostic and therapeutic colonoscopy within a defined population in Sweden.
        Gastrointest Endosc. 2001; 54: 302-309
        • Kim H.S.
        • Kim T.I.
        • Kim W.H.
        • et al.
        Risk factors for immediate postpolypectomy bleeding of the colon: a multicenter study.
        Am J Gastroenterol. 2006; 101: 1333-1341
        • Singh M.
        • Mehta N.
        • Murthy U.K.
        • et al.
        Postpolypectomy bleeding in patients undergoing colonoscopy on uninterrupted clopidogrel therapy.
        Gastrointest Endosc. 2010; 71: 998-1005
        • Witt D.M.
        • Delate T.
        • McCool K.H.
        • et al.
        Incidence and predictors of bleeding or thrombosis after polypectomy in patients receiving and not receiving anticoagulation therapy.
        J Thromb Haemost. 2009; 7: 1982-1989
        • Hui A.J.
        • Wong R.M.
        • Ching J.Y.
        • et al.
        Risk of colonoscopic polypectomy bleeding with anticoagulants and antiplatelet agents: analysis of 1657 cases.
        Gastrointest Endosc. 2004; 59: 44-48
        • Sawhney M.S.
        • Salfiti N.
        • Nelson D.B.
        • et al.
        Risk factors for severe delayed postpolypectomy bleeding.
        Endoscopy. 2008; 40: 115-119
        • Buddingh K.T.
        • Herngreen T.
        • Haringsma J.
        • et al.
        Location in the right hemi-colon is an independent risk factor for delayed post-polypectomy hemorrhage: a multi-center case-control study.
        Am J Gastroenterol. 2011; 106: 1119-1124
        • Kim J.H.
        • Lee H.J.
        • Ahn J.W.
        • et al.
        Risk factors for delayed post-polypectomy hemorrhage: a case-control study.
        J Gastroenterol Hepatol. 2013; 28: 645-649
        • Acosta R.D.
        • Abraham N.S.
        • Chandrasekhara V.
        • et al.
        • ASGE Standards of Practice Committee
        The management of antithrombotic agents for patients undergoing GI endoscopy.
        Gastrointest Endosc. 2016; 83: 3-16
        • Boumitri C.
        • Mir F.A.
        • Ashraf I.
        • et al.
        Prophylactic clipping and post-polypectomy bleeding: a meta-analysis and systematic review.
        Ann Gastroenterol. 2016; 29: 502-508
        • Matsumoto M.
        • Kato M.
        • Oba K.
        • et al.
        Multicenter randomized controlled study to assess the effect of prophylactic clipping on post-polypectomy delayed bleeding.
        Dig Endosc. 2016; 28: 570-576
        • Shioji K.
        • Suzuki Y.
        • Kobayashi M.
        • et al.
        Prophylactic clip application does not decrease delayed bleeding after colonoscopic polypectomy.
        Gastrointest Endosc. 2003; 57: 691-694
        • Di Giorgio P.
        • De Luca L.
        • Calcagno G.
        • et al.
        Detachable snare versus epinephrine injection in the prevention of postpolypectomy bleeding: a randomized and controlled study.
        Endoscopy. 2004; 36: 860-863
        • Hsieh Y.H.
        • Lin H.J.
        • Tseng G.Y.
        • et al.
        Is submucosal epinephrine injection necessary before polypectomy? A prospective, comparative study.
        Hepatogastroenterology. 2001; 48: 1379-1382
        • Ko C.W.
        • Dominitz J.A.
        Complications of colonoscopy: magnitude and management.
        Gastrointest Endosc Clin North Am. 2010; 20: 659-671
        • Saito Y.
        • Fukuzawa M.
        • Matsuda T.
        • et al.
        Clinical outcome of endoscopic submucosal dissection versus endoscopic mucosal resection of large colorectal tumors as determined by curative resection.
        Surg Endosc. 2010; 24: 343-352
        • Toyonaga T.
        • Man-i M.
        • Fujita T.
        • et al.
        Retrospective study of technical aspects and complications of endoscopic submucosal dissection for laterally spreading tumors of the colorectum.
        Endoscopy. 2010; 42: 714-722
        • Bae J.H.
        • Yang D.H.
        • Lee J.Y.
        • et al.
        Clinical outcomes of endoscopic submucosal dissection for large colorectal neoplasms: a comparison of protruding and laterally spreading tumors.
        Surg Endosc. 2016; 30: 1619-1628
        • Burgess N.G.
        • Bassan M.S.
        • McLeod D.
        • et al.
        Deep mural injury and perforation after colonic endoscopic mucosal resection: a new classification and analysis of risk factors.
        Gut. 2017; 66: 1779-1789
        • Swan M.P.
        • Bourke M.J.
        • Moss A.
        • et al.
        The target sign: an endoscopic marker for the resection of the muscularis propria and potential perforation during colonic endoscopic mucosal resection.
        Gastrointest Endosc. 2011; 73: 79-85
        • Xu M.D.
        • Wang X.Y.
        • Li Q.L.
        • et al.
        Colorectal lateral spreading tumor subtypes: clinicopathology and outcome of endoscopic submucosal dissection.
        Int J Colorectal Dis. 2013; 28: 63-72
        • Lee E.J.
        • Lee J.B.
        • Lee S.H.
        • et al.
        Endoscopic treatment of large colorectal tumors: comparison of endoscopic mucosal resection, endoscopic mucosal resection-precutting, and endoscopic submucosal dissection.
        Surg Endosc. 2012; 26: 2220-2230
        • Zhang Q.S.
        • Han B.
        • Xu J.H.
        • et al.
        Clip closure of defect after endoscopic resection in patients with larger colorectal tumors decreased the adverse events.
        Gastrointest Endosc. 2015; 82: 904-909
        • Burgess N.G.
        • Metz A.J.
        • Williams S.J.
        • et al.
        Risk factors for intraprocedural and clinically significant delayed bleeding after wide-field endoscopic mucosal resection of large colonic lesions.
        Clin Gastroenterol Hepatol. 2014; 12: 651-661
        • Kim G.U.
        • Seo M.
        • Song E.M.
        • et al.
        Association between the ulcer status and the risk of delayed bleeding after the endoscopic mucosal resection of colon.
        J Gastroenterol Hepatol. 2017; 32: 1846-1851
        • Metz A.J.
        • Bourke M.J.
        • Moss A.
        • et al.
        Factors that predict bleeding following endoscopic mucosal resection of large colonic lesions.
        Endoscopy. 2011; 43: 506-511
        • Saito Y.
        • Uraoka T.
        • Yamaguchi Y.
        • et al.
        A prospective, multicenter study of 1111 colorectal endoscopic submucosal dissections (with video).
        Gastrointest Endosc. 2010; 72: 1217-1225
        • Kantsevoy S.V.
        • Adler D.G.
        • Conway J.D.
        • et al.
        • ASGE Technology Committee
        Endoscopic mucosal resection and endoscopic submucosal dissection.
        Gastrointest Endosc. 2008; 68: 11-18
        • Moss A.
        • Bourke M.J.
        • Williams S.J.
        • et al.
        Endoscopic mucosal resection outcomes and prediction of submucosal cancer from advanced colonic mucosal neoplasia.
        Gastroenterology. 2011; 140: 1909-1918
        • Thirumurthi S.
        • Raju G.S.
        How to deal with large colorectal polyps: snare, endoscopic mucosal resection, and endoscopic submucosal dissection; resect or refer?.
        Curr Opin Gastroenterol. 2016; 32: 26-31
        • Niimi K.
        • Fujishiro M.
        • Kodashima S.
        • et al.
        Long-term outcomes of endoscopic submucosal dissection for colorectal epithelial neoplasms.
        Endoscopy. 2010; 42: 723-729
        • Feagins L.A.
        • Nguyen A.D.
        • Iqbal R.
        • et al.
        The prophylactic placement of hemoclips to prevent delayed post-polypectomy bleeding: an unnecessary practice? A case control study.
        Dig Dis Sci. 2014; 59: 823-828
        • Pohl H.
        • Grimm I.S.
        • Moyer M.T.
        • et al.
        Clip closure prevents bleeding after endoscopic resection of large colon polyps in a randomized trial.
        Gastroenterology. 2019; 157: 977-984
        • Singh H.
        • Penfold R.B.
        • DeCoster C.
        • et al.
        Colonoscopy and its complications across a Canadian regional health authority.
        Gastrointest Endosc. 2009; 69: 665-671
        • Cha J.M.
        • Lim K.S.
        • Lee S.H.
        • et al.
        Clinical outcomes and risk factors of post-polypectomy coagulation syndrome: a multicenter, retrospective, case-control study.
        Endoscopy. 2013; 45: 202-207
        • Hirasawa K.
        • Sato C.
        • Makazu M.
        • et al.
        Coagulation syndrome: delayed perforation after colorectal endoscopic treatments.
        World J Gastrointest Endosc. 2015; 7: 1055-1061
        • Ferrara F.
        • Luigiano C.
        • Ghersi S.
        • et al.
        Efficacy, safety and outcomes of “inject and cut” endoscopic mucosal resection for large sessile and flat colorectal polyps.
        Digestion. 2010; 82: 213-220
        • Ferreira A.O.
        • Moleiro J.
        • Torres J.
        • et al.
        Solutions for submucosal injection in endoscopic resection: a systematic review and meta-analysis.
        Endosc Int Open. 2016; 4: E1-E16
        • Kedia P.
        • Waye J.D.
        Colon polypectomy: a review of routine and advanced techniques.
        J Clin Gastroenterol. 2013; 47: 657-665
        • Park S.K.
        • Lee M.G.
        • Jeong S.H.
        • et al.
        Prospective analysis of minor adverse events after colon polypectomy.
        Dig Dis Sci. 2017; 62: 2113-2119
        • Ko C.W.
        • Riffle S.
        • Shapiro J.A.
        • et al.
        Incidence of minor complications and time lost from normal activities after screening or surveillance colonoscopy.
        Gastrointest Endosc. 2007; 65: 648-656
        • Bini E.J.
        • Firoozi B.
        • Choung R.J.
        • et al.
        Systematic evaluation of complications related to endoscopy in a training setting: a prospective 30-day outcomes study.
        Gastrointest Endosc. 2003; 57: 8-16
        • Zubarik R.
        • Fleischer D.E.
        • Mastropietro C.
        • et al.
        Prospective analysis of complications 30 days after outpatient colonoscopy.
        Gastrointest Endosc. 1999; 50: 322-328
        • Waye J.D.
        The most important maneuver during colonoscopy.
        Am J Gastroenterol. 2004; 99: 2086-2087
        • Memon M.A.
        • Memon B.
        • Yunus R.M.
        • et al.
        Carbon dioxide versus air insufflation for elective colonoscopy: a meta-analysis and systematic review of randomized controlled trials.
        Surg Laparosc Endosc Percutan Tech. 2016; 26: 102-116
        • Leung F.W.
        • Amato A.
        • Ell C.
        • et al.
        Water-aided colonoscopy: a systematic review.
        Gastrointest Endosc. 2012; 76: 657-666
        • Amato A.
        • Radaelli F.
        • Paggi S.
        • et al.
        Carbon dioxide insufflation or warm-water infusion versus standard air insufflation for unsedated colonoscopy: a randomized controlled trial.
        Dis Colon Rectum. 2013; 56: 511-518
        • Garborg K.
        • Kaminski M.F.
        • Lindenburger W.
        • et al.
        Water exchange versus carbon dioxide insufflation in unsedated colonoscopy: a multicenter randomized controlled trial.
        Endoscopy. 2015; 47: 192-199
        • Ladas S.D.
        • Karamanolis G.
        • Ben-Soussan E.
        Colonic gas explosion during therapeutic colonoscopy with electrocautery.
        World J Gastroenterol. 2007; 13: 5295-5298
        • Avgerinos A.
        • Kalantzis N.
        • Rekoumis G.
        • et al.
        Bowel preparation and the risk of explosion during colonoscopic polypectomy.
        Gut. 1984; 25: 361-364
        • La Brooy S.J.
        • Avgerinos A.
        • Fendick C.L.
        • et al.
        Potentially explosive colonic concentrations of hydrogen after bowel preparation with mannitol.
        Lancet. 1981; 1: 634-636
        • Monahan D.W.
        • Peluso F.E.
        • Goldner F.
        Combustible colonic gas levels during flexible sigmoidoscopy and colonoscopy.
        Gastrointest Endosc. 1992; 38: 40-43
        • Ben-Soussan E.
        • Antonietti M.
        • Savoye G.
        • et al.
        Argon plasma coagulation in the treatment of hemorrhagic radiation proctitis is efficient but requires a perfect colonic cleansing to be safe.
        Eur J Gastroenterol Hepatol. 2004; 16: 1315-1318
        • Nelson D.B.
        Infectious disease complications of GI endoscopy: part II, exogenous infections.
        Gastrointest Endosc. 2003; 57: 695-711
        • Chun Y.J.
        • Yoon N.R.
        • Park J.M.
        • et al.
        Prospective assessment of risk of bacteremia following colorectal stent placement.
        Dig Dis Sci. 2012; 57: 1045-1049
        • Khashab M.A.
        • Chithadi K.V.
        • Acosta R.D.
        • et al.
        • ASGE Standards of Practice Committee
        Antibiotic prophylaxis for GI endoscopy.
        Gastrointest Endosc. 2015; 81: 81-89
        • Wilson W.
        • Taubert K.A.
        • Gewitz M.
        • et al.
        Prevention of infective endocarditis: guidelines from the American Heart Association: a guideline from the American Heart Association Rheumatic Fever, Endocarditis, and Kawasaki Disease Committee, Council on Cardiovascular Disease in the Young, and the Council on Clinical Cardiology, Council on Cardiovascular Surgery and Anesthesia, and the Quality of Care and Outcomes Research Interdisciplinary Working Group.
        Circulation. 2007; 116: 1736-1754
        • Reprocessing Guideline Task Force
        • Petersen B.
        • Cohen J.
        • Hambrick R.
        • et al.
        Multisociety guideline on reprocessing flexible GI endoscopes: 2016 update.
        Gastrointest Endosc. 2017; 85: 282-294
        • Ko C.W.
        • Riffle S.
        • Michaels L.
        • et al.
        Serious complications within 30 days of screening and surveillance colonoscopy are uncommon.
        Clin Gastroenterol Hepatol. 2010; 8: 166-173
        • Bielawska B.
        • Hookey L.C.
        • Sutradhar R.
        • et al.
        Anesthesia assistance in outpatient colonoscopy and risk of aspiration pneumonia, bowel perforation, and splenic injury.
        Gastroenterology. 2018; 154: 77-85
        • Piccolo G.
        • Di Vita M.
        • Cavallaro A.
        • et al.
        Presentation and management of splenic injury after colonoscopy: a systematic review.
        Surg Laparosc Endosc Percutan Tech. 2014; 24: 95-102
        • Ha J.F.
        • Minchin D.
        Splenic injury in colonoscopy: a review.
        Int J Surg. 2009; 7: 424-427
        • Moore E.E.
        • Cogbill T.H.
        • Jurkovich G.J.
        • et al.
        Organ injury scaling: spleen and liver (1994 revision).
        J Trauma. 1995; 38: 323-324
        • Brennan I.M.
        • Faintuch S.
        • Sacks B.
        Superselective splenic artery embolization for the management of splenic laceration following colonoscopy.
        Acta Radiol Short Rep. 2014; 3 (2047981614524199)
        • Cooper G.S.
        • Kou T.D.
        • Rex D.K.
        Complications following colonoscopy with anesthesia assistance: a population-based analysis.
        JAMA Intern Med. 2013; 173: 551-556
        • Viiala C.H.
        • Zimmerman M.
        • Cullen D.J.
        • et al.
        Complication rates of colonoscopy in an Australian teaching hospital environment.
        Intern Med J. 2003; 33: 355-359
        • Agostoni M.
        • Fanti L.
        • Gemma M.
        • et al.
        Adverse events during monitored anesthesia care for GI endoscopy: an 8-year experience.
        Gastrointest Endosc. 2011; 74: 266-275
        • Wernli K.J.
        • Brenner A.T.
        • Rutter C.M.
        • et al.
        Risks associated with anesthesia services during colonoscopy.
        Gastroenterology. 2016; 150 (quiz e18): 888-894
        • Shaukat A.
        • Malhotra A.
        • Greer N.
        • et al.
        Systematic review: outcomes by duration of NPO status prior to colonoscopy.
        Gastroenterol Res Pract. 2017; 2017: 3914942
        • Gurudu S.R.
        • Ratuapli S.
        • Heigh R.
        • et al.
        Quality of bowel cleansing for afternoon colonoscopy is influenced by time of administration.
        Am J Gastroenterol. 2010; 105: 2318-2322
        • Huffman M.
        • Unger R.Z.
        • Thatikonda C.
        • et al.
        Split-dose bowel preparation for colonoscopy and residual gastric fluid volume: an observational study.
        Gastrointest Endosc. 2010; 72: 516-522
        • Manno M.
        • Pigò F.
        • Manta R.
        • et al.
        Bowel preparation with polyethylene glycol electrolyte solution: optimizing the splitting regimen.
        Dig Liver Dis. 2012; 44: 576-579
        • Mathus-Vliegen E.M.
        • van der Vliet K.
        Safety, patient's tolerance, and efficacy of a 2-liter vitamin C-enriched macrogol bowel preparation: a randomized, endoscopist-blinded prospective comparison with a 4-liter macrogol solution.
        Dis Colon Rectum. 2013; 56: 1002-1012
        • Matro R.
        • Shnitser A.
        • Spodik M.
        • et al.
        Efficacy of morning-only compared with split-dose polyethylene glycol electrolyte solution for afternoon colonoscopy: a randomized controlled single-blind study.
        Am J Gastroenterol. 2010; 105: 1954-1961
        • Varughese S.
        • Kumar A.R.
        • George A.
        • et al.
        Morning-only one-gallon polyethylene glycol improves bowel cleansing for afternoon colonoscopies: a randomized endoscopist-blinded prospective study.
        Am J Gastroenterol. 2010; 105: 2368-2374
        • Alghamry A.
        • Ponnuswamy S.K.
        • Agarwal A.
        • et al.
        Split-dose bowel preparation with polyethylene glycol for colonoscopy performed under propofol sedation. Is there an optimal timing?.
        J Dig Dis. 2017; 18: 160-168
        • Thakkar K.
        • El-Serag H.B.
        • Mattek N.
        • et al.
        Complications of pediatric colonoscopy: a five-year multicenter experience.
        Clin Gastroenterol Hepatol. 2008; 6: 515-520
        • Kramer R.E.
        • Narkewicz M.R.
        Adverse events following gastrointestinal endoscopy in children: classifications, characterizations, and implications.
        J Pediatr Gastroenterol Nutr. 2016; 62: 828-833
        • Hsu E.K.
        • Chugh P.
        • Kronman M.P.
        • et al.
        Incidence of perforation in pediatric GI endoscopy and colonoscopy: an 11-year experience.
        Gastrointest Endosc. 2013; 7: 960-966
        • Iqbal C.W.
        • Askegard-Giesmann J.R.
        • Pham T.H.
        • et al.
        Pediatric endoscopic injuries: incidence, management, and outcomes.
        J Pediatr Surg. 2008; 43: 911-915
        • Soucy P.
        • Eidus L.
        • Keeley F.
        Perforation of the colon in a 15-year-old girl with Ehlers-Danlos syndrome type IV.
        J Pediatr Surg. 1990; 25: 1180-1182
        • Fuchs J.R.
        • Fishman S.J.
        Management of spontaneous colonic perforation in Ehlers-Danlos syndrome type IV.
        J Pediatr Surg. 2004; 39: e1-e3
        • Andropoulos D.B.
        • Greene M.F.
        Anesthesia and developing brains—implications of the FDA warning.
        N Engl J Med. 2017; 376: 905-907
        • Kramer R.E.
        • Walsh C.M.
        • Lerner D.G.
        • et al.
        Quality improvement in pediatric endoscopy: a clinical report from the NASPGHAN Endoscopy Committee.
        J Pediatr Gastroenterol Nutr. 2017; 65: 125-131
        • Thakkar K.
        • Holub J.L.
        • Gilger M.A.
        • et al.
        Quality indicators for pediatric colonoscopy: results from a multicenter consortium.
        Gastrointest Endosc. 2016; 83: 533-541
        • Singh H.K.
        • Withers G.D.
        • Ee L.C.
        Quality indicators in pediatric colonoscopy: an Australian tertiary center experience.
        Scand J Gastroenterol. 2017; 52: 1453-1456