ORIGINAL CONTRIBUTION
Achieving a Complete Colonic Evaluation in Patients With Incomplete Colonoscopy Is Worth the Effort Timothy J. Ridolfi, M.D. • Michael A. Valente, D.O. • James M. Church, M.D. Department of Colorectal Surgery, Digestive Diseases Institute, Cleveland Clinic Foundation, Cleveland, Ohio
BACKGROUND: Patients with an incomplete colonoscopy are potentially at risk for missed lesions. OBJECTIVE: The purpose of this work was to identify the percentage of patients completing colonic evaluation after incomplete colonoscopy, the manner in which the evaluation was completed, and the incidence of significant pathology. DESIGN: This was a retrospective analysis of prospectively collected data. SETTINGS: The study was conducted in an outpatient colonoscopy clinic in the colorectal surgery department of a tertiary referral center. PATIENTS: Patients included those undergoing incomplete colonoscopy from a database of 25,645 colonoscopies performed from 1982 to 2009. INTERVENTIONS: Procedures aimed at completing
colorectal evaluation were included in the study. MAIN OUTCOME MEASURES: Reason for incompletion, secondary study, its success, and findings were measured. RESULTS: A total of 242 patients with incomplete colonoscopies were identified; 166 (69%) were women. The average age of patients was 59 years. Most frequent causes for incomplete colonoscopy were inadequate preparation (34%), pain (30%), and tortuosity (20%). The scope could not pass the splenic flexure in 165 patients (71%). A total of 218 patients (90%) were offered completion studies, and 179 patients (82%) complied. Financial Disclosure: None reported. Podium presentation at the meeting of The American Society of Colon and Rectal Surgeons, Phoenix, AZ, April 27 to May 1, 2013. Correspondence: James M. Church, M.D., Desk A 30, 9500 Euclid Ave, Cleveland, OH, 44195. E-mail: [email protected] Dis Colon Rectum 2014; 57: 383–387 DOI: 10.1097/DCR.0000000000000072 © The ASCRS 2014 Diseases of the Colon & Rectum Volume 57: 3 (2014)
Seventy-three of 82 patients who had a surveillance colonoscopy had a follow-up (89%), compared with 72 (87%) of 83 with symptoms and 40 (74%) of 54 who had a screening. Barium enema (BE) was performed in 74 (41%), repeat colonoscopy in 71 (40%), CT colonography in 17 (9%), and colonoscopy under general anesthesia in 9 patients (5%). Resection with intraoperative/perioperative colonoscopy was required in 8 patients (4%). Repeat colonoscopy found 32 lesions (24 tubular adenomas, 4 tubulovillous adenomas, and 4 sessile serrated polyps) in 17 patients (24%). Radiology demonstrated new abnormalities in 11 (12%) of 91 patients, prompting 7 colonoscopies. In 3 patients, colonoscopy showed an inverted appendix, a tubulovillous adenoma, and a sigmoid stricture. Overall, clinically significant lesions were found in 19 patients (10%). LIMITATIONS: This study was limited by an incomplete colonoscopy subjectively determined at the time of colonoscopy, as well as by a lack of comparison group. CONCLUSIONS: Complete colonic evaluation in patients with an incomplete colonoscopy is important. Repeat colonoscopy may be the most efficient way to achieve this. KEY WORDS: Incomplete colonoscopy; Yield of
colonoscopy.
C
olonoscopy is an important technique for examining the large intestine. It is used to investigate symptoms such as bleeding, change in bowel function, and abdominal pain. It is also used to detect and treat colorectal neoplasia, where patients may present with a variety of indications and risks from average risk screening to a family history-based screening and follow-up of cancer or adenomas. The slight decrease in incidence rates of colorectal cancer over the last 2 decades has been attributed in part to increases in the use of colorectal cancer screening tests that allow for the early detection and removal of colorectal polyps before they progress to cancer.1 383
384
Each year, ≈14 million screening colonoscopy procedures are performed, ranking colonoscopy as one of the most widely used endoscopic procedures in the United States.2 Efficient colonoscopy depends on examination of the entire colon, but even patients undergoing complete colonoscopy are at risk of missed lesions. Adenoma miss rates vary according to adenoma size,3 and recent population-based data indicate that, although colonoscopy is associated with fewer deaths from colon and rectal cancer, this association is primarily limited to deaths from cancer developing in the left side of the colon.4 The authors suggest that the reason colonoscopy may be less effective in preventing death from right-sided colon cancer is multifactorial, including worse bowel preparation in the right colon, biological and genetic differences in right and left colon lesions, and the possibility that some complete colonoscopies do not evaluate the entire colon. Rates of incomplete colonoscopic examinations range from 4% to 25%.5–8 There are several possible reasons for incompletion, but whatever the reason, patients with incomplete colonoscopy are at risk of missed lesions. This matters if the aim of the colonoscopy was to diagnose and treat premalignant polyps or lesions. Because it matters, attempts are usually made to complete the colonic examination, using a variety of strategies that depend on the reason for colonoscopy failure. However, if a patient’s experience with the incomplete colonoscopy is unpleasant, the patient may be reluctant to submit to further examination of the large bowel. The reported percentage of patients who undergo additional studies after incomplete colonoscopy is broad and varies from 29% to 79%.8,9 Data on the diagnostic yield of completion examinations are limited. This study aims to identify the percentage of patients who undergo complete colonic evaluation after incomplete colonoscopy, the manner in which the evaluation was completed, and the frequency with which significant pathology is found.
METHODS The institutional review board of the Cleveland Clinic Foundation approved this study. A prospectively maintained departmental colonoscopy database, with data collected from 1982 to 2009, was queried for a subset of incomplete colonoscopies in which “incomplete” was listed in the findings of the study, as well as the reason for the incomplete study. Incomplete colonoscopy was determined by the endoscopist at the time of the procedure. All of the colonoscopies were performed by a colorectal surgeon, sometimes with a fellow. These colonoscopies were typically performed under intravenous sedation using 1 drug or a combination of midazolam, diazepam, meperidine, and/or fentanyl. A chart review was completed for patients
RIDOLFI ET AL: COMPLETING COLON EVALUATION
with incomplete colonoscopy, and data pertaining to secondary studies offered, their success, and findings of the secondary study were abstracted. Only patients with the incompletion described in the colonoscopy report were included. Diverticulosis and tortuosity were not considered new findings on secondary studies. Adenomas, advanced adenomas, and sessile serrated adenomas/polyps were considered clinically significant. Hyperplastic polyps were excluded from analysis.
RESULTS Between August 26, 1982, and November 6, 2009, 25,645 patients were recorded as having undergone colonoscopy in the database. Patients underwent colonoscopy for 1 of 16 indications. These were grouped into 4 major areas: 1) IBD (IBD survey and IBD status), 2) screening (screening study or family history), 3) surveillance (personal history of polyp or personal history of cancer), and 4) symptoms (abdominal pain, change in bowel habits, constipation, diarrhea, hemorrhage, outlet rectal bleeding, suspect rectal bleeding, anemia, and weight loss). A total of 242 patients were recorded as having an incomplete colonoscopy, 166 women (69%) and 76 men (31%). Causes for incomplete colonoscopy were inadequate bowel preparation in 80 patients (34%), pain in 69 patients (30%), tortuosity in 46 patients (20%), stricture in 17 patients (7%), and perforation in 1 patient (0.004%). The cause was not recorded in 29 patients (12%). The scope was not able to pass the splenic flexure in 165 patients (71%; Table 1). The most common cause for termination of the colonoscopy at the level of the rectum and sigmoid and ascending colons was inadequate bowel preparation. Pain was the most common cause for termination of the scope at the level of the splenic flexure, whereas pain, tortuosity, and inadequate preparation contributed in similar proportions to reasons for incomplete colonoscopy at the hepatic flexure (Table 1). A total of 218 patients (90%) were offered completion studies, and 179 patients (82%) complied. Patients having surveillance colonoscopy were most compliant (73/82; 89%), followed by those being scoped for symptoms (72/83) (87%) and those having screening (40/54; 74%). BE was performed in 74 patients (41%), colonoscopy was repeated under conscious sedation in 71 (40%), general anesthetic in 9 (5%), and computed tomographic colonography (CTC) was performed in 17 (9%). Colonic resection with intraoperative/perioperative colonoscopy was performed in 8 patients (4%). Of the 8 intraoperative/perioperative colonoscopies performed, 2 tubular adenomas were detected, as well as 1 villous adenoma, which changed operative management in the patient by indicating a separate resection (Table 2).
385
Diseases of the Colon & Rectum Volume 57: 3 (2014)
TABLE 1. Reason for incomplete study by extent of index colonoscopy
Rectum Sigmoid Splenic flex Hepatic flex Ascending All
Pain
Stool
Tort
Strict
Other
Perf
Total
0 27 24 13 5 69 (30)
9 31 12 17 11 80 (34)
0 16 11 16 3 46 (20)
1 8 7 1 0 17 (7)
0 13 5 1 0 19 (8)
0 1 0 0 0 1 (0)
10 96 59 48 19 232
Data are n or n (%). Tort = tortuosity; Strict = stricture; Perf = perforation; flex = flexure.
Radiologic studies, including both BE and CTC, demonstrated new abnormalities in 11 (12%) of 91 patients, but only 3 abnormalities were confirmed. Five (7%) of 74 patients undergoing BEs had a positive finding, which prompted follow-up colonoscopy in 2 and 1 resection for stricture. One of the follow-up colonoscopies demonstrated an inverted appendix, whereas the other was normal. Two patients with positive findings on BE did not undergo further evaluation at our institution. Six of the 17 patients undergoing CTCs had abnormalities, prompting 5 follow-up colonoscopies. No abnormality could be identified on 3 of the follow-up colonoscopies, 1 colonoscopy confirmed a tubulovillous adenoma, and 1 colonoscopy was again unable to be completed. One patient with positive findings on CTC did not undergo further evaluation at our institution. Repeat colonoscopy found 32 lesions (24 tubular adenomas, 4 tubulovillous adenomas, and 4 sessile serrated polyps) in 17 patients (19%). Thirteen of the 32 polyps in 11 patients were considered advanced adenomas. Twelve polyps were found in those undergoing a repeat study for poor prep, whereas 9, 2, 1, and 8 polyps were found in those who had an index procedure terminated for pain, stricture, tortuosity, or other reasons. Twenty-one of the polyps were beyond the level of the index colonoscopy, 8 of which were considered high-risk polyps. Five high-risk polyps were found in those undergoing repeat colonoscopy secondary to poor prep. Of all of the patients with positive findings on repeat colonoscopy, patients with a personal history of colon cancer or polyps were the most frequently represented, composing two thirds of the group (Table 3).
DISCUSSION Our study examines a cohort of 242 patients who were identified as having an incomplete colonoscopy. The most significant finding of the study is that 10% of those undergoing a completion study were ultimately found to have a clinically significant lesion, with 6% having an advanced adenoma. Other important findings include that 82% of patients with incomplete colonoscopy eventually had the entire colon screened, with those patients having an index colonoscopy for surveillance being the most compliant (89%) and those with an index colonoscopy for screening the least compliant (74%) with recommendations. Additionally, 19% of those undergoing repeat colonoscopy as a secondary study were found to have a lesion, whereas only 7% of those undergoing BE were found to have a lesion. Finally, in the time period of the study of the 6 CTCs with positive findings (35%), half were shown to be false positives. The strengths of the study are that it examines a large group of incomplete colonoscopies over a 27-year time period. Also we were able to report the findings of the completion study performed and not simply the end point percentage of people having had a completion study. Additionally, it was possible to validate abnormal findings on radiologic completion studies in a subset of patients. One weakness in our study is that it is unlikely that all of the incomplete colonoscopies were captured within the database. The percentage of incomplete colonoscopies was 0.9%. This percentage is less that that reported previously at our institution, at 6.4% within a cohort of 2907 patients.10 Our entry criteria required the colonoscopy
TABLE 2. Patients undergoing resection after index colonoscopy Indication for resection
Operation
Intraoperative/perioperative colonoscopy finding
Action
Sigmoid stricture Sigmoid stricture Sigmoid stricture Sigmoid cancer Sigmoid cancer Sigmoid cancer Descending cancer Crohn’s disease
Sigmoid resection Sigmoid resection Sigmoid resection Sigmoid resection Sigmoid resection Sigmoid resection Left hemicolectomy Ileocecal resection
Normal 3-mm descending TA 7-mm transverse TA 6-cm villous adenoma Normal Normal Normal Normal
None Polypectomy Polypectomy Right hemicolectomy None None None None
TA = tubular adenoma.
386
RIDOLFI ET AL: COMPLETING COLON EVALUATION
TABLE 3. Findings of secondary study by indication for initial colonoscopy
Family history Personal history Symptoms Screening
Patients
TA
TVA
SSA
Advanced adenoma
2 12 3 1
3 15 2 7
0 1 1 1
0 4 0 0
2 8 2 1
Polyps beyond first colonoscopy 2 18 2 0
TA = tubular adenoma; TVA = tubulovillous adenoma; SSA = sessile serrated adenoma.
report to explicitly state that the colonoscopy was incomplete. Undoubtedly there are incomplete colonoscopies within the database that did not contain the word “incomplete” in the findings and would therefore not be included in this study. However, the objectives of our study were to identify the percentage of patients completing colonic evaluation after incomplete colonoscopy, the manner in which the evaluation was completed, and the incidence of significant pathology, not the absolute incompletion rate itself. An additional weakness of the study is its retrospective nature. We are unable to ascertain why a particular modality of completion study was chosen. Finally, the limited number of CTCs performed during the time period of the study, as well as the use of previous-generation CTC technology, does not allow us to make recommendations on its use in contemporary practice. Our rate of complete colonoscopic evaluation is consistent with other reports. Complete colonic evaluation was achieved in 74% of patients. Other studies have reported that 30% to 75% of patients with incomplete colonoscopies underwent a successful second diagnostic study.8,11,12 Because of the retrospective nature of this study and the lack of a suitable field in the database, it is unclear why only 218 of 242 patients were offered secondary examinations. Only patients with a documented secondary study or with documentation within the index colonoscopy r eport of the secondary study being offered were considered to have been offered a secondary study. BE was most frequently used as the primary method of follow-up after incomplete colonoscopy (41%), although the diagnostic yield of this investigation was disappointing. Only 5 of 75 BEs demonstrated a lesion other than diverticulosis or tortuosity and, furthermore, only 2 were verified. A recent large, population-based study showed a cancer miss rate of 22% for BE, which makes the examination a very poor second best to colonoscopy, either optical or CTC.13 However, BE does show the anatomy of the colon, which can help when a second attempt at colonoscopy is made. In addition, the limited availability of colonoscopic expertise in some parts of the country means that BEs should still be considered as a follow-up to incomplete colonoscopy. Repeat colonoscopy was the second most frequent study ordered. A total of 19% of patients were found to have a clinically significant lesion. This compares favor-
ably with previous studies in which clinically significant lesions were found in 13% to 29% of patients who underwent repeat colonoscopies for incomplete colonoscopies.9 As an alternative, CTC has a reported accuracy, at least for cancers and large- and medium-sized polyps, that can equal or exceed that of optical colonoscopy. It depends on the skill and experience of the radiologist. In our study however, 6 (35%) of 17 CT colonographies had positive findings, half of which were ultimately shown to be false positives. Our ability to comment on CTC is limited by the small number of CTCs ordered, which is likely a reflection of the time period over which the database was collected. Because our study is not a prospective, randomized controlled trial, we are unable to state with certainty which is the best study to perform after incomplete colonoscopy. However, in our experience, the most efficient way to complete a colonic examination in patients with an incomplete index examination is to try colonoscopy again. The advantages are an accurate examination and the ability to biopsy or remove lesions. The suitability of repeat colonoscopy depends on the reason that the index examination failed. If it was because of poor preparation, the reasons for the poor preparation must be examined. Noncompliance with instructions should be correctable, and a poor preparation because of chronic constipation can be managed by a more extensive or prolonged lavage. If there is an obstructing lesion or a diseased segment of bowel then the lesion can be resected with an on-table or perioperative colonoscopy to clear the rest of the colon. This was the case in 8 of our patients. If there was too much pain to complete the initial colonoscopy or if the colon was too tortuous, the examination can be repeated under general anesthesia (9 of our patients) or by another, more expert, colonoscopist. We hope that the findings of our study underscore the importance of completing a colonic evaluation after incomplete colonoscopy. The patients most likely to be noncompliant with recommendations are those undergoing an index colonoscopy for screening, and they may be a good initial target for improving education on the importance of follow-through with such exams. Also, the data presented within, particularly those related to the percentage of new findings on repeat studies, may be helpful in generating a cost-effective decision analysis on the study of choice after incomplete colonoscopy when combined with contemporary data on CTC.
Diseases of the Colon & Rectum Volume 57: 3 (2014)
CONCLUSION Complete colonic evaluation in patients with an incomplete colonoscopy is important. Repeat colonoscopy may be the most efficient way of achieving this. REFERENCES 1. American Cancer Society. Cancer facts and figures 2013. www.cancer.org/research/cancerfactsstatistics/cancerfactsfigures2013/index. Accessed December 5, 2013. 2. Seeff LC, Richards TB, Shapiro JA, et al. How many endoscopies are performed for colorectal cancer screening? Results from CDC’s survey of endoscopic capacity. Gastroenterology. 2004;127:1670–1677. 3. Rex DK, Cutler CS, Lemmel GT, et al. Colonoscopic miss rates of adenomas determined by back-to-back colonoscopies. Gastroenterology. 1997;112:24–28. 4. Baxter NN, Goldwasser MA, Paszat LF, Saskin R, Urbach DR, Rabeneck L. Association of colonoscopy and death from colorectal cancer. Ann Intern Med. 2009;150:1–8. 5. Brown AL, Skehan SJ, Greaney T, Rawlinson J, Somers S, Stevenson GW. Value of double-contrast barium enema performed immediately after incomplete colonoscopy. AJR Am J Roentgenol. 2001;176:943–945.
387
6. Chong A, Shah JN, Levine MS, et al. Diagnostic yield of barium enema examination after incomplete colonoscopy. Radiology. 2002;223:620–624. 7. Gollub MJ, Flaherty F. Barium enema following incomplete colonoscopy. Clin Imaging. 1999;23:367–374. 8. Rizek R, Paszat LF, Stukel TA, Saskin R, Li C, Rabeneck L. Rates of complete colonic evaluation after incomplete colonoscopy and their associated factors: a population-based study. Med Care. 2009;47:48–52. 9. Kao KT, Tam M, Sekhon H, Wijeratne R, Haigh PI, Abbas MA. Should barium enema be the next step following an incomplete colonoscopy? Int J Colorectal Dis. 2010;25: 1353–1357. 10. Church JM. Complete colonoscopy: how often? And if not, why not? Am J Gastroenterol. 1994;89:556–560. 11. Martinez F, Kondylis P, Reilly J. Limitations of barium enema performed as an adjunct to incomplete colonoscopy. Dis Colon Rectum. 2005;48:1951–1954. 12. Neerincx M, Terhaar sive Droste JS, Mulder CJ, et al. Colonic work-up after incomplete colonoscopy: significant new findings during follow-up. Endoscopy. 2010;42:730–735. 13. Toma J, Paszat LF, Gunraj N, Rabeneck L. Rates of new or missed colorectal cancer after barium enema and their risk factors: a population-based study. Am J Gastroenterol. 2008;103:3142–3148.
Achieving a Complete Colonic Evaluation in Patients With Incomplete Colonoscopy Is Worth the Effort Timothy J. Ridolfi, M.D. • Michael A. Valente, D.O. • James M. Church, M.D. Department of Colorectal Surgery, Digestive Diseases Institute, Cleveland Clinic Foundation, Cleveland, Ohio
BACKGROUND: Patients with an incomplete colonoscopy are potentially at risk for missed lesions. OBJECTIVE: The purpose of this work was to identify the percentage of patients completing colonic evaluation after incomplete colonoscopy, the manner in which the evaluation was completed, and the incidence of significant pathology. DESIGN: This was a retrospective analysis of prospectively collected data. SETTINGS: The study was conducted in an outpatient colonoscopy clinic in the colorectal surgery department of a tertiary referral center. PATIENTS: Patients included those undergoing incomplete colonoscopy from a database of 25,645 colonoscopies performed from 1982 to 2009. INTERVENTIONS: Procedures aimed at completing
colorectal evaluation were included in the study. MAIN OUTCOME MEASURES: Reason for incompletion, secondary study, its success, and findings were measured. RESULTS: A total of 242 patients with incomplete colonoscopies were identified; 166 (69%) were women. The average age of patients was 59 years. Most frequent causes for incomplete colonoscopy were inadequate preparation (34%), pain (30%), and tortuosity (20%). The scope could not pass the splenic flexure in 165 patients (71%). A total of 218 patients (90%) were offered completion studies, and 179 patients (82%) complied. Financial Disclosure: None reported. Podium presentation at the meeting of The American Society of Colon and Rectal Surgeons, Phoenix, AZ, April 27 to May 1, 2013. Correspondence: James M. Church, M.D., Desk A 30, 9500 Euclid Ave, Cleveland, OH, 44195. E-mail: [email protected] Dis Colon Rectum 2014; 57: 383–387 DOI: 10.1097/DCR.0000000000000072 © The ASCRS 2014 Diseases of the Colon & Rectum Volume 57: 3 (2014)
Seventy-three of 82 patients who had a surveillance colonoscopy had a follow-up (89%), compared with 72 (87%) of 83 with symptoms and 40 (74%) of 54 who had a screening. Barium enema (BE) was performed in 74 (41%), repeat colonoscopy in 71 (40%), CT colonography in 17 (9%), and colonoscopy under general anesthesia in 9 patients (5%). Resection with intraoperative/perioperative colonoscopy was required in 8 patients (4%). Repeat colonoscopy found 32 lesions (24 tubular adenomas, 4 tubulovillous adenomas, and 4 sessile serrated polyps) in 17 patients (24%). Radiology demonstrated new abnormalities in 11 (12%) of 91 patients, prompting 7 colonoscopies. In 3 patients, colonoscopy showed an inverted appendix, a tubulovillous adenoma, and a sigmoid stricture. Overall, clinically significant lesions were found in 19 patients (10%). LIMITATIONS: This study was limited by an incomplete colonoscopy subjectively determined at the time of colonoscopy, as well as by a lack of comparison group. CONCLUSIONS: Complete colonic evaluation in patients with an incomplete colonoscopy is important. Repeat colonoscopy may be the most efficient way to achieve this. KEY WORDS: Incomplete colonoscopy; Yield of
colonoscopy.
C
olonoscopy is an important technique for examining the large intestine. It is used to investigate symptoms such as bleeding, change in bowel function, and abdominal pain. It is also used to detect and treat colorectal neoplasia, where patients may present with a variety of indications and risks from average risk screening to a family history-based screening and follow-up of cancer or adenomas. The slight decrease in incidence rates of colorectal cancer over the last 2 decades has been attributed in part to increases in the use of colorectal cancer screening tests that allow for the early detection and removal of colorectal polyps before they progress to cancer.1 383
384
Each year, ≈14 million screening colonoscopy procedures are performed, ranking colonoscopy as one of the most widely used endoscopic procedures in the United States.2 Efficient colonoscopy depends on examination of the entire colon, but even patients undergoing complete colonoscopy are at risk of missed lesions. Adenoma miss rates vary according to adenoma size,3 and recent population-based data indicate that, although colonoscopy is associated with fewer deaths from colon and rectal cancer, this association is primarily limited to deaths from cancer developing in the left side of the colon.4 The authors suggest that the reason colonoscopy may be less effective in preventing death from right-sided colon cancer is multifactorial, including worse bowel preparation in the right colon, biological and genetic differences in right and left colon lesions, and the possibility that some complete colonoscopies do not evaluate the entire colon. Rates of incomplete colonoscopic examinations range from 4% to 25%.5–8 There are several possible reasons for incompletion, but whatever the reason, patients with incomplete colonoscopy are at risk of missed lesions. This matters if the aim of the colonoscopy was to diagnose and treat premalignant polyps or lesions. Because it matters, attempts are usually made to complete the colonic examination, using a variety of strategies that depend on the reason for colonoscopy failure. However, if a patient’s experience with the incomplete colonoscopy is unpleasant, the patient may be reluctant to submit to further examination of the large bowel. The reported percentage of patients who undergo additional studies after incomplete colonoscopy is broad and varies from 29% to 79%.8,9 Data on the diagnostic yield of completion examinations are limited. This study aims to identify the percentage of patients who undergo complete colonic evaluation after incomplete colonoscopy, the manner in which the evaluation was completed, and the frequency with which significant pathology is found.
METHODS The institutional review board of the Cleveland Clinic Foundation approved this study. A prospectively maintained departmental colonoscopy database, with data collected from 1982 to 2009, was queried for a subset of incomplete colonoscopies in which “incomplete” was listed in the findings of the study, as well as the reason for the incomplete study. Incomplete colonoscopy was determined by the endoscopist at the time of the procedure. All of the colonoscopies were performed by a colorectal surgeon, sometimes with a fellow. These colonoscopies were typically performed under intravenous sedation using 1 drug or a combination of midazolam, diazepam, meperidine, and/or fentanyl. A chart review was completed for patients
RIDOLFI ET AL: COMPLETING COLON EVALUATION
with incomplete colonoscopy, and data pertaining to secondary studies offered, their success, and findings of the secondary study were abstracted. Only patients with the incompletion described in the colonoscopy report were included. Diverticulosis and tortuosity were not considered new findings on secondary studies. Adenomas, advanced adenomas, and sessile serrated adenomas/polyps were considered clinically significant. Hyperplastic polyps were excluded from analysis.
RESULTS Between August 26, 1982, and November 6, 2009, 25,645 patients were recorded as having undergone colonoscopy in the database. Patients underwent colonoscopy for 1 of 16 indications. These were grouped into 4 major areas: 1) IBD (IBD survey and IBD status), 2) screening (screening study or family history), 3) surveillance (personal history of polyp or personal history of cancer), and 4) symptoms (abdominal pain, change in bowel habits, constipation, diarrhea, hemorrhage, outlet rectal bleeding, suspect rectal bleeding, anemia, and weight loss). A total of 242 patients were recorded as having an incomplete colonoscopy, 166 women (69%) and 76 men (31%). Causes for incomplete colonoscopy were inadequate bowel preparation in 80 patients (34%), pain in 69 patients (30%), tortuosity in 46 patients (20%), stricture in 17 patients (7%), and perforation in 1 patient (0.004%). The cause was not recorded in 29 patients (12%). The scope was not able to pass the splenic flexure in 165 patients (71%; Table 1). The most common cause for termination of the colonoscopy at the level of the rectum and sigmoid and ascending colons was inadequate bowel preparation. Pain was the most common cause for termination of the scope at the level of the splenic flexure, whereas pain, tortuosity, and inadequate preparation contributed in similar proportions to reasons for incomplete colonoscopy at the hepatic flexure (Table 1). A total of 218 patients (90%) were offered completion studies, and 179 patients (82%) complied. Patients having surveillance colonoscopy were most compliant (73/82; 89%), followed by those being scoped for symptoms (72/83) (87%) and those having screening (40/54; 74%). BE was performed in 74 patients (41%), colonoscopy was repeated under conscious sedation in 71 (40%), general anesthetic in 9 (5%), and computed tomographic colonography (CTC) was performed in 17 (9%). Colonic resection with intraoperative/perioperative colonoscopy was performed in 8 patients (4%). Of the 8 intraoperative/perioperative colonoscopies performed, 2 tubular adenomas were detected, as well as 1 villous adenoma, which changed operative management in the patient by indicating a separate resection (Table 2).
385
Diseases of the Colon & Rectum Volume 57: 3 (2014)
TABLE 1. Reason for incomplete study by extent of index colonoscopy
Rectum Sigmoid Splenic flex Hepatic flex Ascending All
Pain
Stool
Tort
Strict
Other
Perf
Total
0 27 24 13 5 69 (30)
9 31 12 17 11 80 (34)
0 16 11 16 3 46 (20)
1 8 7 1 0 17 (7)
0 13 5 1 0 19 (8)
0 1 0 0 0 1 (0)
10 96 59 48 19 232
Data are n or n (%). Tort = tortuosity; Strict = stricture; Perf = perforation; flex = flexure.
Radiologic studies, including both BE and CTC, demonstrated new abnormalities in 11 (12%) of 91 patients, but only 3 abnormalities were confirmed. Five (7%) of 74 patients undergoing BEs had a positive finding, which prompted follow-up colonoscopy in 2 and 1 resection for stricture. One of the follow-up colonoscopies demonstrated an inverted appendix, whereas the other was normal. Two patients with positive findings on BE did not undergo further evaluation at our institution. Six of the 17 patients undergoing CTCs had abnormalities, prompting 5 follow-up colonoscopies. No abnormality could be identified on 3 of the follow-up colonoscopies, 1 colonoscopy confirmed a tubulovillous adenoma, and 1 colonoscopy was again unable to be completed. One patient with positive findings on CTC did not undergo further evaluation at our institution. Repeat colonoscopy found 32 lesions (24 tubular adenomas, 4 tubulovillous adenomas, and 4 sessile serrated polyps) in 17 patients (19%). Thirteen of the 32 polyps in 11 patients were considered advanced adenomas. Twelve polyps were found in those undergoing a repeat study for poor prep, whereas 9, 2, 1, and 8 polyps were found in those who had an index procedure terminated for pain, stricture, tortuosity, or other reasons. Twenty-one of the polyps were beyond the level of the index colonoscopy, 8 of which were considered high-risk polyps. Five high-risk polyps were found in those undergoing repeat colonoscopy secondary to poor prep. Of all of the patients with positive findings on repeat colonoscopy, patients with a personal history of colon cancer or polyps were the most frequently represented, composing two thirds of the group (Table 3).
DISCUSSION Our study examines a cohort of 242 patients who were identified as having an incomplete colonoscopy. The most significant finding of the study is that 10% of those undergoing a completion study were ultimately found to have a clinically significant lesion, with 6% having an advanced adenoma. Other important findings include that 82% of patients with incomplete colonoscopy eventually had the entire colon screened, with those patients having an index colonoscopy for surveillance being the most compliant (89%) and those with an index colonoscopy for screening the least compliant (74%) with recommendations. Additionally, 19% of those undergoing repeat colonoscopy as a secondary study were found to have a lesion, whereas only 7% of those undergoing BE were found to have a lesion. Finally, in the time period of the study of the 6 CTCs with positive findings (35%), half were shown to be false positives. The strengths of the study are that it examines a large group of incomplete colonoscopies over a 27-year time period. Also we were able to report the findings of the completion study performed and not simply the end point percentage of people having had a completion study. Additionally, it was possible to validate abnormal findings on radiologic completion studies in a subset of patients. One weakness in our study is that it is unlikely that all of the incomplete colonoscopies were captured within the database. The percentage of incomplete colonoscopies was 0.9%. This percentage is less that that reported previously at our institution, at 6.4% within a cohort of 2907 patients.10 Our entry criteria required the colonoscopy
TABLE 2. Patients undergoing resection after index colonoscopy Indication for resection
Operation
Intraoperative/perioperative colonoscopy finding
Action
Sigmoid stricture Sigmoid stricture Sigmoid stricture Sigmoid cancer Sigmoid cancer Sigmoid cancer Descending cancer Crohn’s disease
Sigmoid resection Sigmoid resection Sigmoid resection Sigmoid resection Sigmoid resection Sigmoid resection Left hemicolectomy Ileocecal resection
Normal 3-mm descending TA 7-mm transverse TA 6-cm villous adenoma Normal Normal Normal Normal
None Polypectomy Polypectomy Right hemicolectomy None None None None
TA = tubular adenoma.
386
RIDOLFI ET AL: COMPLETING COLON EVALUATION
TABLE 3. Findings of secondary study by indication for initial colonoscopy
Family history Personal history Symptoms Screening
Patients
TA
TVA
SSA
Advanced adenoma
2 12 3 1
3 15 2 7
0 1 1 1
0 4 0 0
2 8 2 1
Polyps beyond first colonoscopy 2 18 2 0
TA = tubular adenoma; TVA = tubulovillous adenoma; SSA = sessile serrated adenoma.
report to explicitly state that the colonoscopy was incomplete. Undoubtedly there are incomplete colonoscopies within the database that did not contain the word “incomplete” in the findings and would therefore not be included in this study. However, the objectives of our study were to identify the percentage of patients completing colonic evaluation after incomplete colonoscopy, the manner in which the evaluation was completed, and the incidence of significant pathology, not the absolute incompletion rate itself. An additional weakness of the study is its retrospective nature. We are unable to ascertain why a particular modality of completion study was chosen. Finally, the limited number of CTCs performed during the time period of the study, as well as the use of previous-generation CTC technology, does not allow us to make recommendations on its use in contemporary practice. Our rate of complete colonoscopic evaluation is consistent with other reports. Complete colonic evaluation was achieved in 74% of patients. Other studies have reported that 30% to 75% of patients with incomplete colonoscopies underwent a successful second diagnostic study.8,11,12 Because of the retrospective nature of this study and the lack of a suitable field in the database, it is unclear why only 218 of 242 patients were offered secondary examinations. Only patients with a documented secondary study or with documentation within the index colonoscopy r eport of the secondary study being offered were considered to have been offered a secondary study. BE was most frequently used as the primary method of follow-up after incomplete colonoscopy (41%), although the diagnostic yield of this investigation was disappointing. Only 5 of 75 BEs demonstrated a lesion other than diverticulosis or tortuosity and, furthermore, only 2 were verified. A recent large, population-based study showed a cancer miss rate of 22% for BE, which makes the examination a very poor second best to colonoscopy, either optical or CTC.13 However, BE does show the anatomy of the colon, which can help when a second attempt at colonoscopy is made. In addition, the limited availability of colonoscopic expertise in some parts of the country means that BEs should still be considered as a follow-up to incomplete colonoscopy. Repeat colonoscopy was the second most frequent study ordered. A total of 19% of patients were found to have a clinically significant lesion. This compares favor-
ably with previous studies in which clinically significant lesions were found in 13% to 29% of patients who underwent repeat colonoscopies for incomplete colonoscopies.9 As an alternative, CTC has a reported accuracy, at least for cancers and large- and medium-sized polyps, that can equal or exceed that of optical colonoscopy. It depends on the skill and experience of the radiologist. In our study however, 6 (35%) of 17 CT colonographies had positive findings, half of which were ultimately shown to be false positives. Our ability to comment on CTC is limited by the small number of CTCs ordered, which is likely a reflection of the time period over which the database was collected. Because our study is not a prospective, randomized controlled trial, we are unable to state with certainty which is the best study to perform after incomplete colonoscopy. However, in our experience, the most efficient way to complete a colonic examination in patients with an incomplete index examination is to try colonoscopy again. The advantages are an accurate examination and the ability to biopsy or remove lesions. The suitability of repeat colonoscopy depends on the reason that the index examination failed. If it was because of poor preparation, the reasons for the poor preparation must be examined. Noncompliance with instructions should be correctable, and a poor preparation because of chronic constipation can be managed by a more extensive or prolonged lavage. If there is an obstructing lesion or a diseased segment of bowel then the lesion can be resected with an on-table or perioperative colonoscopy to clear the rest of the colon. This was the case in 8 of our patients. If there was too much pain to complete the initial colonoscopy or if the colon was too tortuous, the examination can be repeated under general anesthesia (9 of our patients) or by another, more expert, colonoscopist. We hope that the findings of our study underscore the importance of completing a colonic evaluation after incomplete colonoscopy. The patients most likely to be noncompliant with recommendations are those undergoing an index colonoscopy for screening, and they may be a good initial target for improving education on the importance of follow-through with such exams. Also, the data presented within, particularly those related to the percentage of new findings on repeat studies, may be helpful in generating a cost-effective decision analysis on the study of choice after incomplete colonoscopy when combined with contemporary data on CTC.
Diseases of the Colon & Rectum Volume 57: 3 (2014)
CONCLUSION Complete colonic evaluation in patients with an incomplete colonoscopy is important. Repeat colonoscopy may be the most efficient way of achieving this. REFERENCES 1. American Cancer Society. Cancer facts and figures 2013. www.cancer.org/research/cancerfactsstatistics/cancerfactsfigures2013/index. Accessed December 5, 2013. 2. Seeff LC, Richards TB, Shapiro JA, et al. How many endoscopies are performed for colorectal cancer screening? Results from CDC’s survey of endoscopic capacity. Gastroenterology. 2004;127:1670–1677. 3. Rex DK, Cutler CS, Lemmel GT, et al. Colonoscopic miss rates of adenomas determined by back-to-back colonoscopies. Gastroenterology. 1997;112:24–28. 4. Baxter NN, Goldwasser MA, Paszat LF, Saskin R, Urbach DR, Rabeneck L. Association of colonoscopy and death from colorectal cancer. Ann Intern Med. 2009;150:1–8. 5. Brown AL, Skehan SJ, Greaney T, Rawlinson J, Somers S, Stevenson GW. Value of double-contrast barium enema performed immediately after incomplete colonoscopy. AJR Am J Roentgenol. 2001;176:943–945.
387
6. Chong A, Shah JN, Levine MS, et al. Diagnostic yield of barium enema examination after incomplete colonoscopy. Radiology. 2002;223:620–624. 7. Gollub MJ, Flaherty F. Barium enema following incomplete colonoscopy. Clin Imaging. 1999;23:367–374. 8. Rizek R, Paszat LF, Stukel TA, Saskin R, Li C, Rabeneck L. Rates of complete colonic evaluation after incomplete colonoscopy and their associated factors: a population-based study. Med Care. 2009;47:48–52. 9. Kao KT, Tam M, Sekhon H, Wijeratne R, Haigh PI, Abbas MA. Should barium enema be the next step following an incomplete colonoscopy? Int J Colorectal Dis. 2010;25: 1353–1357. 10. Church JM. Complete colonoscopy: how often? And if not, why not? Am J Gastroenterol. 1994;89:556–560. 11. Martinez F, Kondylis P, Reilly J. Limitations of barium enema performed as an adjunct to incomplete colonoscopy. Dis Colon Rectum. 2005;48:1951–1954. 12. Neerincx M, Terhaar sive Droste JS, Mulder CJ, et al. Colonic work-up after incomplete colonoscopy: significant new findings during follow-up. Endoscopy. 2010;42:730–735. 13. Toma J, Paszat LF, Gunraj N, Rabeneck L. Rates of new or missed colorectal cancer after barium enema and their risk factors: a population-based study. Am J Gastroenterol. 2008;103:3142–3148.
