Pa Patient Saf Advis 2008 Jun;5(2):57-63.
Colon Perforations Complicating Colonoscopies: What is the Best Known Evidence for Prevention?
Gastroenterology; Internal Medicine and Subspecialties; Surgery
Expand All Collapse All

Abstract

A systematic qualitative review was done of all modifiable risk factors reported in the medical literature to have a possible association with perforations associated with colonoscopies and of the recommendations in the medical literature and clinical practice guidelines for reducing the risks of these perforations. Inclusion criteria were defined a priori. Ten databases were searched, plus the National Guideline Clearinghouse and the National Quality Measures Clearinghouse. Only three clinical studies specifically addressed modifiable risk factors for perforations associated with colonoscopies. Another four publications addressed recommendations for avoiding perforation. The modifiable risk factors identified were the use of pediatric colonoscope in an adult and the duration and strength of coagulation. Seventeen recommendations were reported for reducing perforation rates associated with colonoscopy. The evidence used to support these recommendations was weak. The recommendations have not been definitively shown to reduce the risk of colon perforation. Critical prospective reporting and comparisons may be beneficial in providing sufficient evidence to support current conclusions.

Introduction

Colonoscopy is considered the best way to screen for colorectal cancer, the second leading cause of cancer death in the United States.1 Colonoscopy is used to remove precancerous polyps. It is also used to diagnose and treat other diseases of the colon. One of the most serious complications of colonoscopy is perforation. Perforation of the colon almost always requires operative repair of the resulting hole. As noted in a previous Patient Safety Advisory article,  Pennsylvania healthcare facilities reported 125 to 152 perforations following colonoscopies through PA-PSRS over a one-year period.2 The PA-PSRS clinical analysts estimated that the rate of this complication was between 39 and 47 per 100,000 colonoscopies, assuming all perforations for the 322,867 colonoscopies done that year were reported as “unanticipated” injuries.

The risk of a perforation during a colonoscopy may vary according to the patient’s medical conditions and the techniques used by the endoscopist. In the previous Advisory article, some of the risk factors reported in the literature were mentioned.2 Some risk factors are invariable. Others are modifiable. Identifying modifiable risk factors could lead to fewer perforations.

This article is a systematic qualitative review of all modifiable risk factors reported in the medical literature to have a possible association with perforations associated with colonoscopies and of the recommendations in the medical literature and clinical practice guidelines for reducing the risks of these perforations. 

Methods

The analysts defined two questions to be answered by a systematic qualitative review of the medical literature and clinical practice guidelines:

  1. What are the modifiable risk factors for perforations associated with colonoscopies?
  2. What are the published recommendations for controlling modifiable risk factors?

For both questions, the analysts investigated the importance of modifiable factors related to the patient, the endoscopist, and the procedures or techniques.

The methods for answering those questions were defined a priori to preclude bias in selection or review of the literature and practice guidelines.

Inclusion Criteria

The clinical studies included in this systematic review were based on inclusion criteria that were determined a priori to reduce the risk of bias because the decision to include or exclude each study is independent of the results of the study. All publications had to meet the following criteria to be included in the review:

  • The patients had either conventional colonoscopies or colonoscopies through enterostomy stomas. Publications reporting more than 15% of patients having endoscopies only involving the rectum, rigid or flexible sigmoidoscopies, or intraoperative colonoscopies were not included.
  • The publications were in English.
  • Only data collected since 1990 were included, in order to relate the findings to modern practice.
  • In addition to clinical studies, literature reviews, reports, letters, editorials, clinical practice guidelines, and other publications were included.
  • Only publications that provided an answer to either of the questions were included.

Literature Searches and Article Selection

The clinical studies included in this systematic review were identified using a multistaged study selection process. The analysts performed a comprehensive literature search using broad criteria and retrieved all articles that appeared to meet inclusion criteria, based on their titles and published abstracts. The analysts reviewed the full text of each retrieved article to verify that it met the inclusion criteria.

Ten databases were searched through November 1, 2007, including the following: the Cumulative Index to Nursing and Allied Health Literature, the Cochrane Central Register of Controlled Trials, the Cochrane Database of Methodology Reviews, the Cochrane Database of Systematic Reviews, the Database of Abstracts of Reviews of Effects, EMBASE (Excerpta Medica), Healthcare Standards, Health Technology Assessment Database, MEDLINE/PubMed, and the NHS Economic Evaluation Database. The National Guideline Clearinghouse and the National Quality Measures Clearinghouse were also searched through the same time period. To supplement the electronic searches, the analysts reviewed the bibliographies of the retrieved publications, the contents of new issues of selected journals, and relevant gray literature (reports and studies produced by local government agencies, private organizations, educational facilities, and corporations that do not appear in the peer-reviewed literature). The advisory panel (see “Advisory Panel to Minimize Colonoscopy Perforations in Pennsylvania”) provided guidance for the search process.

Results of the Literature Search and Article Selection

The literature searches identified 172 publications, of which 3 specifically addressed modifiable risk factors for perforations associated with colonoscopies. Another four publications addressed recommendations for avoiding perforation. These seven articles that were found to have met the inclusion criteria were used for the systematic review (see Table 1).

 

Table 1. Included Studies and Key Questions they Address

Study

Year
Published

Year(s) Data Collected

What Are The Risk Factors For Colonic Perforation?

What Are The Best Practices For
Controlling Modifiable Risk Factors?

Clinical Studies,
Retrospective Chart/
Record Reviews

Vokura

2004

1998 to 2002

x

 

Wexner et al.

2001

April 1998 to
September 1999

x

 

Cobb et al.

2004

January 1997 through December 2003

x

 

Literature Reviews

Fatima and Rex

2007

Not reported

 

x

Reickert and Beck

2001

Not reported

 

x

Clinical Practice Guidelines

Rex et al. "Quality indicators . . .  "

2006

Not reported

 

x

Rex et al. "Quality in the technical . . . "

2002

Not reported

 

x

Sources: Cobb WS, Heniford BT, Sigmon LB, Hasan R, et al. Colonoscopic perforations: incidence, management, and outcomes. Am Surg 2004 Sep;70(9):750-7; discussion 757-8;  Fatima H, Rex DK. Minimizing endoscopic complications: colonoscopic polypectomy. Gastrointest Endosc Clin N Am 2007 Jan;17(1):145-56;  Reickert CA, Beck DE. Complications of colonoscopy. Clin Colon Rectal Surg 2001;14(4):379-85;  Rex DK, Bond JH, Winawer S, et al. Quality in the technical performance of colonoscopy and the continuous quality improvement process for colonoscopy: Recommendations of the U.S. Multi-Society Task Force on Colorectal Cancer. Am J Gastroenterol 2002 Jun;97(6):1296-1308;  Rex DK, Petrini JL, Baron TH, et al. Quality indicators for colonoscopy. Am J Gastroenterol 2006 Apr;101(4):873-85;  Vokurka J. Iatrogenic perforation during an endoscopic examination of the gastrointestinal tract. Bratisl Lek Listy 2004;105(10-11):387-9;  Wexner SD, Garbus JE, Singh JJ; SAGES Colonoscopy Study Outcomes Group. A prospective analysis of 13,580 colonoscopies. Reevaluation of credentialing guidelines. Surg Endosc 2001 Mar;15(3):251-61.

 

Results of the Systematic Review

1.  What Are the Modifiable Risk factors for Perforations Associated with Colonoscopies?

A total of three publications were identified that reported modifiable risk factors for colonoscopy-associated perforation. These clinical studies are summarized in Table 2.

 

Table 2. Characteristics of Clinical Studies Used in the Systemic Review

 

Wexner et al.

Vokura

Cobb et al.

Methodological Characteristics

Year of publication

2001

2004

2004

Years data collected

April 1998 to
September 1999

1998 to 2002

January 1997 through December 2003

Median year of data collection

1999

2000

2000

Method of data collection

Volunteer center reporting to centralized database

Hospital database
review

Hospital database

Study type

Prospective database investigation

Retrospective record review

 

Retrospective quality assurance database review of single center

Method of patient selection

Consecutive

Consecutive

Consecutive

Site and Procedural Characteristics

Location(s)

 

United States

 

Brno, Czech Republic

Not reported (probably Charlotte, North Carolina, United States)

Setting

 

Various centers, types not reported

University hospital gastroenterology or surgery department

Large, urban teaching hospital

 

Patient Characteristics

Reason for colonoscopy

 

62.4% diagnostic,
37.6% therapeutic

Not reported

 

Not reported

 

Provider Characteristics

Specialty

 

Not reported; some training fellows and surgeons

Gastroenterology and surgery departments

Not reported; cites surgeons and gastroenterologists

Perforation Rates

Number of perforations

       10

       6

       14

Number of colonoscopies

13,580

3,897

43,609

Comments

 

No association with practitioner experience and complications (including perforations).

 

The authors concluded that polypectomy of thick, rigid, polyps with wider bases was more likely to result in perforation, owing to longer time duration and strength of coagulation.

Rate of perforation with general surgeons 0.080% (1 of 1,243), with gastroenterologists 0.031% (13 of 42,366). Mechanical stress was considered the most common mechanism of perforation, accounting for 6 perforations. Remaining perforations were associated with cone biopsy (3), electrocautery (3), and pneumatic causes (2); 57% of perforations were in males; mean age of patients with perforations was 65.9 years (range 31 to 83). Six of the 43,609 colonoscopies were attempted using a pediatric scope in an adult patient and 4 of them resulted in perforation. Both pneumatic injuries were associated with pediatric colonoscope use.

Sources: Cobb WS, Heniford BT, Sigmon LB, et al. Colonoscopic perforations: incidence, management, and outcomes. Am Surg 2004 Sep;70(9):750-7; discussion 757-8;  Vokurka J. Iatrogenic perforation during an endoscopic examination of the gastrointestinal tract. Bratisl Lek Listy 2004;105(10-11):387-9;  Wexner SD, Garbus JE, Singh JJ; SAGES Colonoscopy Study Outcomes Group. A prospective analysis of 13,580 colonoscopies. Reevaluation of credentialing guidelines. Surg Endosc 2001 Mar;15(3):251-61.​ ​ ​

  

All factors identified as modifiable risk factors for perforation, or modifiable factors identified as not being risk factors, are reported in Table 3, along with the type of evidence supporting each. These factors are grouped as follows:

Procedural factors. Only two of the publications reported modifiable procedural factors associated with colonoscopy. Cobb et al. described six colonoscopies in which the colonoscopy was attempted with a pediatric colonoscope in an adult.3 Four of the six resulted in perforation. Both cases of pneumatic perforation in the entire sample were associated with pediatric colonoscope use. Vokura et al. concluded that longer duration and strength of coagulation was associated with increased risk of perforation.4

Provider characteristics. Only two of the publications reported association between perforation and provider characteristics. Wexner et al.  found that practitioner experience was not associated with perforation.5 Cobb et al.  reported that the rate of perforation was higher for general surgeons (1 of 1,243 or 0.080%) than for gastroenterologists (13 of 43,609 or 0.031%); however, the difference was not statistically significant.3

Table 3. Identification of Modifiable Risk Factors for Perforation

Factor

Study

 

Year

 

Found To Be A Risk Factor?

Evidence Type

 

Comments

 

Procedural Factors

Type of colonoscope

 

Cobb et al.

 

2004

 

Yes

 

Retrospective quality assurance database review of single center

Four of six colonoscopies attempted using a pediatric colonoscope resulted in perforation (age range of patients with perforation was 31 to 83). Both cases of pneumatic perforation in the entire sample were associated with the use of a pediatric colonoscope.

 

Polypectomy/biopsy

 

Vokura

 

2004

 

Yes

 

Retrospective record review

Longer time duration and strength of coagulation were associated with increased perforations.

Provider Characteristics

Practitioner experience

 

Wexner et al.

 

2001

 

No

 

Prospective database investigation

How experience was determined was unclear.

 

Provider specialty

 

Cobb et al.

 

2004

 

No

 

Retrospective quality assurance database review of single center

Higher rate reported for general surgeons (1 of 1,243) than gastroenterologists (13 of 42,366) was not statistically significant.

 

Sources: Cobb WS, Heniford BT, Sigmon LB, et al. Colonoscopic perforations: incidence, management, and outcomes. Am Surg 2004 Sep;70(9):750-7; discussion 757-8;  Vokurka J. Iatrogenic perforation during an endoscopic examination of the gastrointestinal tract. Bratisl Lek Listy 2004;105(10-11):387-9;  Wexner SD, Garbus JE, Singh JJ; SAGES Colonoscopy Study Outcomes Group. A prospective analysis of 13,580 colonoscopies. Reevaluation of credentialing guidelines. Surg Endosc 2001 Mar;15(3):251-61.

 

Summary of Modifiable Risk Factors Associated with Perforation following Colonoscopy

The small number of studies reporting on modifiable risk factors precludes the formation of any strong conclusions. Few modifiable risk factors were identified, limiting the potential impact of currently available published literature on clinical practice.

2.  What Are the Published Recommendations for Controlling Modifiable Risk Factors?

Four publications, comprised of 2 literature reviews and 2 clinical practice guidelines, reported 17 recommendations to reduce perforation rate associated with colonoscopy (see Table 4). The evidence used to support these recommendations was weak. The types of evidence that the recommendations were based on included observational studies, animal studies, and nonclinical (laboratory) studies. At least two recommendations were based upon anecdotal reports. For many recommendations, the supporting evidence was unclear but may have been based in medical opinion.

Table 4. Published Recommendations to Reduce the Risk of Perforation

Publication

 

Issuing Party

 

Year
Published

Recommendation

 

Supporting Information

Before Colonoscopy Begins

​ ​

Reickert and Beck

None reported

2001

"Assure that a thorough and adequate bowel preparation has been completed whenever possible."

Unclear

Reickert and Beck

None reported

 

2001

 

"Provide adequate sedation with experienced personnel and monitoring equipment."

Unclear

 

Reickert and Beck

None reported

2001

"Check endoscopic and electrocautery equipment settings prior to use."

Unclear

Mechanical

Rex et al.  "Quality in the technical . . . "

The U.S. Multi-Society Task Force on Colorectal Cancer

2002

 

"The most important rule to avoid mechanical perforation is not to push against the sensation of fixed resistance."

 

Unclear

 

Reickert and Beck

None reported

2001

 

"Use proper insertion technique, maintaining visualization of the lumen at all times."

Unclear

 

Reickert and Beck

None reported

2001

"Maintain one-to-one motion with the scope as much as possible."

Unclear

Reickert and Beck

None reported

2001

"Limit looping/bowing of the endoscope, with position changes or external counterpressure."

Unclear

 

Barotrauma-Related

Rex et al.  "Quality indicators . . .  "

 

The American Society for Gastrointestinal Endoscopy (ASGE) and American College of Gastroenterology (ACG) Task Force on Endoscopy

2006

 

"Barotrauma perforations can probably best be avoided in most cases by judicious use of air during insufflation, particularly after passing strictures, perhaps by insufflation of carbon dioxide rather than air, and by ensuring that the air pump and light source will not continue to insufflate air when intraluminal pressures exceed the bursting pressure of the colon."

 

Observational studies

 

Reickert and Beck

None reported

 

2001

 

"Minimize colonic distension to minimize discomfort and the risk of barotraumas."

Unclear

 

Polypectomy/Biopsy-Related

Fatima and Rex

None reported

 

2007

 

"Because thermal injury from electrocautery is the cause of essentially all polypectomy-related perforations, using cold techniques (i.e. techniques that do not involve electrocautery) seems very appropriate for small polyps in which the risk of developing invasive cancer is extremely low."

Unclear

 

Fatima and Rex

 

None reported

 

2007

 

"Ancillary techniques to prevent perforation during resection of large sessile colorectal polyps include the use of submucosal injection. . . . It is unclear exactly which polyps are most appropriate for submucosal injection but in general the larger the sessile polyp and the closer it is to the cecum, the more appropriate would be submucosal injection techniques to prevent perforation."

Animal studies

 

Fatima and Rex

 

None reported

 

2007

 

"The risk [of perforation following removal of pedunculated polyps with electrocautery] is reduced by optimal location of the snare on the stalk, which is often said to be approximately one third of the distance from the base of the polyp to the colon wall."

Unclear

 

Rex et al.  "Quality indicators . . .  "

 

The ASGE and ACG Task Force on Endoscopy

 

2006

 

"Perforations may also result from polypectomy. In virtually every case, they are the result of the electrocautery burn. The risk of perforation is greatest with large polyps in the proximal colon. Submucosal saline solution injection is now frequently used by gastroenterologists, although no standardized guidelines regarding the size and location of polyps that require submucosal saline solution injection have been developed."(sic)

Observational studies

 

Rex et al.  "Quality indicators . . .  "

The ASGE and ACG Task Force on Endoscopy

 

2006

 

"Anecdotal reports have suggested an increased risk of complications associated with the use of hot biopsy forceps, and forceps removal of small polyps reduces the chance of complete removal. Cold snaring is attractive for the removal of small polyps because it effectively reduces small polyps and has been associated with exceedingly low risks of complications."

Anecdotal reports

 

Rex et al.  "Quality indicators . . .  "

The ASGE and ACG Task Force on Endoscopy

 

2002

 

"Injection of submucosal saline before piecemeal polypectomy of large sessile polyps reduces injury to the deep wall layers in experimental models but has not been convincingly shown to reduce perforation rates in clinical practice. . . However, the technique facilitates removal of some sessile polyps and probably reduces perforation."

An experimental model but no clinical study

Rex et al.  "Quality in the technical . . . "

The U.S. Multi-Society Task Force on Colorectal Cancer

2002

 

"Anecdotal data suggest that perforations and bleeding are more likely with hot forceps, but definite proof of increased risk is lacking. Cold snaring is particularly attractive for polyps of <7-8mm in size, as anecdotal suggest no risk of perforation and a very low risk of post-polypectomy bleeding."

Anecdotal reports

Reickert and Beck

None reported

 

2001

 

"Minimize current exposure to the colon wall during therapeutic procedures."

 

Unclear

Sources: Fatima H, Rex DK. Minimizing endoscopic complications: colonoscopic polypectomy. Gastrointest Endosc Clin N Am 2007 Jan;17(1):145-56;  Reickert CA, Beck DE. Complications of colonoscopy. Clin Colon Rectal Surg 2001;14(4):379-85;  Rex DK, Bond JH, Winawer S, et al. Quality in the technical performance of colonoscopy and the continuous quality improvement process for colonoscopy: Recommendations of the U.S. Multi-Society Task Force on Colorectal Cancer. Am J Gastroenterol 2002 Jun;97(6):1296-1308;  Rex DK, Petrini JL, Baron TH, et al. Quality indicators for colonoscopy. Am J Gastroenterol 2006 Apr;101(4):873-85.

  

Preparation. Before colonoscopy begins, Reickert and Beck recommend assuring adequate bowel preparation, providing adequate sedation, and checking all equipment settings before using it.6

Technique. To reduce the risk of mechanical perforation during colonoscopy, the U.S. Multi-Society Task Force on Colorectal Cancer recommend against pushing against fixed resistance.7 Reickert and Beck offered several recommendations, including proper insertion technique, maintenance of visualization, maintenance of “one to one motion with the scope,” and minimizing endoscope looping or bowing.6 To reduce the risk of barotraumatic perforation during colonoscopy, the American Society for Gastrointestinal Endoscopy (ASGE) and American College of Gastroenterology (ACG) Task Force on Endoscopy recommend using air judiciously, especially after passing the endoscope by strictures, replacing air with carbon dioxide, and ensuring proper equipment performance.8 Reickert and Beck also recommended minimizing colonic distension.6

Polypectomy and biopsy. Several authors recommended strategies to reduce the risk of polypectomy or biopsy-associated perforation. Fatima and Rex,9 the ASGE and ACG Task Force on Endoscopy,8 and the U.S. Multi-Society Task Force on Colorectal Cancer7 suggest cold techniques for small polyps with low risk of cancer. Submucosal saline injection is recommended by the same three groups. Reickert and Beck also recommend exposing the colon wall to the smallest amount of current possible.6 Fatima and Rex recommend placing the snare on the stalk approximately “one third of the distance from the base of the polyp to the colon wall” to remove pedunculated polyps with electrocautery.9

Summary of Recommendations for Controlling Modifiable Risk Factors

Very few recommendations for best clinical practices to minimize the risk of perforation during colonoscopy are available in the peer-reviewed literature. The small amount of literature that is available is based upon weak types of evidence. The recommendations reported in this systematic qualitative review of the medical literature and clinical practice guidelines have not been definitively shown to reduce the risk of colon perforation.

Discussion

 Identifying modifiable risk factors associated with perforation is desirable for improving the overall safety of colonoscopy. Although the risk of perforation to any individual is not high, minimizing the perforation rate is important given the large number of patients getting colonoscopy for screening, diagnostic, and therapeutic purposes. Prospective comparisons may be particularly helpful for identifying factors not typically available in reports of perforations that have occurred.

Conclusion

Colonoscopists should review these comprehensive lists of modifiable risk factors for perforation and clinical practice guidelines for prevention of this complication. The lack of good evidence supporting these conclusions is sobering. The PA-PSRS analysts will continue to review reports for stronger evidence that this information makes patients safer. Endoscopists and Pennsylvania Patient Safety Officers can help by ensuring that all perforations associated with colonoscopy are known and reported.

Notes

  1. Jemal A, Murray T, Ward E, et al. Cancer statistics, 2005. CA Cancer J Clin 2005 Jan-Feb;55(1):10-30.
  2. Pennsylvania Patient Safety Reporting System. Perforations of the colon during colonoscopy. PA PSRS Patient Saf Advis [online] 2006 Dec [cited 2008 Jun 2]. Available from Internet: http://patientsafetyauthority.org/ADVISORIES/AdvisoryLibrary/2006/Dec3(4)/Pages/10.aspx.
  3. Cobb WS, Heniford BT, Sigmon LB, et al. Colonoscopic perforations: incidence, management, and outcomes. Am Surg 2004 Sep;70 (9):750-7; discussion 757-8.
  4. Vokurka J. Iatrogenic perforation during an endoscopic examination of the gastrointestinal tract. Bratisl Lek Listy 2004;105(10-11):387-9.
  5. Wexner SD, Garbus JE, Singh JJ; SAGES Colonoscopy Study Outcomes Group. A prospective analysis of 13,580 colonoscopies. Reevaluation of credentialing guidelines. Surg Endosc 2001 Mar;15(3):251-61.
  6. Reickert CA, Beck DE. Complications of colonoscopy. Clin Colon Rectal Surg 2001;14(4):379-85.
  7. Rex DK, Bond JH, Winawer S, et al. Quality in the technical performance of colonoscopy and the continuous quality improvement process for colonoscopy: Recommendations of the U.S. Multi-Society Task Force on Colorectal Cancer. Am J Gastroenterol 2002 Jun;97(6):1296-1308.
  8. Rex DK, Petrini JL, Baron TH, et al. Quality indicators for colonoscopy. Am J Gastroenterol 2006 Apr;101(4):873-85.
  9. Fatima H, Rex DK. Minimizing endoscopic complications: colonoscopic polypectomy. Gastrointest Endosc Clin N Am 2007 Jan;17(1):145-56.

Supplemental Material

Advisory Panel to Minimize Colonoscopy Perforations in Pennsylvania

The members of the Advisory Panel are providing domain expertise for efforts to minimize perforations associated with colonoscopies.

The members of the Advisory Panel are as follows:

James C. Reynolds, MD,
Chairman of Medicine,
Drexel University College of Medicine,
Philadelphia (Chair of the Advisory Panel)

Nancy Bacci, RN,
Division of Gastroenterology,
Hershey Medical Center, Hershey

R. Bradley Hayward, MD,
Tri-State Surgical Associates, Bridgewater

Ann Ouyang, MD,
Division of Gastroenterology,
Hershey Medical Center, Hershey

Robert E. (Rocky) Schoen, MD,
Professor of Medicine,
UPMC Presbyterian University Hospital, Pittsburgh

Robert J. Sinnott, MD,
Chief of Colorectal Surgery,
Lehigh Valley Medical Center, Allentown

Joel L. Weissfeld, MD, MPH,
Associate Professor of Epidemiology,
University of Pittsburgh, Pittsburgh

The PA-PSRS analysts are indebted to them for their guidance and contributions.

PSA LOGO

The Pennsylvania Patient Safety Advisory may be reprinted and distributed without restriction, provided it is printed or distributed in its entirety and without alteration. Individual articles may be reprinted in their entirety and without alteration, provided the source is clearly attributed.

Current and previous issues are available online at http://patientsafety.pa.gov.

©2019 Pennsylvania Patient Safety Authority