- Email: [email protected]
Prolonged cecal insertion time is associated with decreased adenoma detection
ORIGINAL ARTICLE: Clinical Endoscopy
Prolonged cecal insertion time is associated with decreased adenoma detection Daniel von Renteln, MD,1 Douglas J. Robertson, MD,2,3 Steve Bensen, MD,3 Heiko Pohl, MD2,3 Montreal, Quebec, Canada; White River Junction, Vermont; Lebanon, New Hampshire, USA
Background and Aims: The adenoma detection rate (ADR) is an important colonoscopy quality parameter. A longer withdrawal time is associated with increased adenoma detection; however, the effect of cecal insertion time on adenoma detection is unclear. The aim of this study was to evaluate if cecal insertion time is associated with adenoma detection. Methods: The study included 50- to 89-year-old subjects undergoing an elective outpatient colonoscopy as part of a randomized trial on cap-assisted colonoscopy. The primary outcome was the adjusted mean number of adenomas per patient across quartiles of lengthening patient insertion times. Secondary outcomes included the mean number of advanced adenomas, ADR, and advanced ADR. Adjusted regression analysis was applied to assess for a possible trend. Results: Among 1043 included study subjects, the median cecal insertion time was 5.3 minutes (interquartile range [IQR], 3.5-8.1) and the median withdrawal time was 8.6 minutes (IQR, 7.1-10.7). Cecal insertion time was not associated with withdrawal time (P Z .950). The mean number of adenomas per patient decreased across increasing insertion time quartiles from 1.1 to .7, corresponding to a 11% decline per quartile (P Z .031). Similarly, the number of advanced adenomas decreased from .15 to .06, corresponding to a 7% decline per quartile (P Z .013). The observed decrease in adenoma detection was similar for proximal and distal adenomas. Endoscopists’ individual insertion times were not associated with adenoma detection. Conclusions: A longer cecal insertion time was associated with a decreased detection of adenomas and advanced adenomas. A long insertion time may be a marker for a more difficult examination that requires a longer withdrawal time to assure adequate examination and adenoma detection. (Clinical trial registration number: NCT01935180.) (Gastrointest Endosc 2017;85:574-80.)
The adenoma detection rate (ADR) is considered an important quality metric for screening colonoscopy.1 A low ADR is inversely associated with a higher risk for the patient to develop postcolonoscopy colorectal cancer.2 Although many studies have examined procedure characteristics and interventions to improve adenoma detection, few factors have been found to be
effective.3 A withdrawal time of at least 6 minutes is associated with a higher ADR compared with a shorter withdrawal time.4 However, lengthening withdrawal time alone does not ensure optimal polyp detection,5 and a study suggests that technical ability and diligence of the examiner are similarly important.6
Abbreviations: ADR, adenoma detection rate; IQR, interquartile range.
Current affiliations: Department of Medicine, Division of Gastroenterology, Montreal University Hospital (CHUM), Montreal University Hospital Research Center (CRCHUM), Montreal, Quebec, Canada (1), Section of Gastroenterology, Veterans Affairs Medical Center, White River Junction, Vermont, USA (2), Section of Gastroenterology, Dartmouth-Hitchcock Medical Center, Lebanon, New Hampshire, USA (3).
DISCLOSURE: All authors disclosed no financial relationships relevant to this publication. Research support for this study was provided by The Veterans Health Administration. The views expressed in this article are those of the authors and do not necessarily represent the views of the Department of Veterans Affairs. Copyright ª 2017 by the American Society for Gastrointestinal Endoscopy 0016-5107/$36.00 http://dx.doi.org/10.1016/j.gie.2016.08.021
Reprint requests: Daniel von Renteln, MD, Department of Medicine, Division of Gastroenterology, Montreal University Hospital (CHUM), Montreal University Hospital Research Center (CRCHUM), 900 Rue SaintDenis, Montréal, QC H2X 0A9, Montreal, Canada.
Received March 8, 2016. Accepted August 17, 2016.
574 GASTROINTESTINAL ENDOSCOPY Volume 85, No. 3 : 2017
www.giejournal.org
von Renteln et al
There is reason to believe that insertion time may also be associated with adenoma detection. At the patient level, cecal insertion time is influenced by patient and procedural factors, including bowel cleansing, age, history of constipation, prior abdominal surgical, sex, and body mass index.7-8 A longer insertion time may reflect a more difficult examination, which may hinder adenoma detection during withdrawal. At the endoscopist level, technical skill is required for advancing the endoscope to the cecum. Cecal intubation times and cecal intubation rates > 90% have been identified as competency metrics.7 A short insertion time may indicate technical skill, which may also promote a meticulous examination with increased adenoma detection during withdrawal. Further, after a prolonged insertion time, the endoscopist may feel fatigued or under time pressure and accelerate endoscope withdrawal with the potential to miss polyps. The primary aim of this study was to assess whether cecal insertion time is associated with adenoma detection. We were further interested in understanding whether a possible association is related to patient or endoscopist factors.
Cecal insertion time and adenoma detection
Outcome measures The primary outcome was the mean number of adenomas per patient, which was chosen because it is a more valid representation of adenoma detection as opposed to the ADR.10 Secondary outcomes included the mean number of advanced adenomas, ADR, and advanced ADR. The ADR represents the proportion of patients with at least 1 adenoma. Advanced adenomas were defined as adenomas 10 mm in size or containing villous features, high-grade dysplasia, or cancer. The advanced ADR included all patients with at least 1 advanced adenoma. Procedure characteristics were assessed as part of the initial trial. Cecal insertion time was defined as the time elapsed from introducing the colonoscope into the rectum until intubation of the cecum. Withdrawal time was defined as the time to withdraw the endoscope from the cecum to the rectum among all patients without a polypectomy. Procedure data were recorded during the colonoscopy by trained personnel (sedation nurse or research assistant).
Statistical analysis
The study included subjects who were 50 to 89 years of age undergoing an elective outpatient colonoscopy at the Dartmouth-Hitchcock Medical Center, Lebanon, New Hampshire, or the VA Medical Center, White River Junction, Vermont. All participants were part of a randomized trial on cap-assisted colonoscopy (clintrials.gov NCT01935180).9 The cap used was a 4-mm transparent distal attachment cap (Disposable Distal Attachment, Models D-201-15004, D-201-12704; Olympus America Inc Center Valley, PA, USA). Patients with known inflammatory bowel disease, active colitis, coagulopathy, polyposis syndrome, or poor general health as defined by American Society of Anesthesiologists class greater than 3 and those undergoing an emergency colonoscopy were excluded from the study. For the current analysis, we additionally excluded patients with prior colon resection, incomplete colonoscopy, poor bowel preparation, and incomplete documentation of insertion or withdrawal times. All study subjects underwent a standard bowel cleansing preparation, typically a 4-L polyethylene glycol solution the day before the colonoscopy. Each study patient underwent only 1 study colonoscopy. Study procedures were performed by 10 experienced board-certified gastroenterologists using standard endoscopes (H-CF/H-PCF 180; Olympus). All study participants provided informed written consent for the initial trial and use of the data for additional analysis. The study was approved by the local institutional review board.
To examine a possible association between cecal insertion time and adenoma detection, we performed patientand endoscopist-level analyses. For the patient-level analysis, we categorized insertion times of all patients into quartiles of insertion time. We then assessed the measures of adenoma detection and withdrawal time across lengthening quartiles of insertion times. In addition, we assessed the effect of insertion time on proximal (proximal to, and not including, the splenic flexure) and distal adenoma detection. For the endoscopist-level analysis, we assessed adenoma detection and withdrawal times by individual endoscopist’s median insertion times. Proportions are presented as percentages. Continuous variables are expressed as means with standard deviations if normally distributed and as medians with interquartile ranges (IQR) if not normally distributed. Poisson regression analysis was applied to examine an association between insertion time and number of detected adenomas or advanced adenomas. Unit changes across quartiles of insertion times are presented as relative change expressed as percentage with 95% confidence interval. To examine an association between insertion time and ADR and advanced ADR, we used adjusted logistic regression analysis expressed as odds ratio with 95% confidence interval across quartiles of insertion time and to calculate a test of trend. To examine the effect of endoscopists on associations between insertion time adenoma detection, we applied the factor endoscopist as a fixed effect. We calculated test of trend using Poisson regression (withdrawal time, mean number of adenomas, and advanced adenomas) and logistic regression (ADR and advanced ADR). All regression models considered possible confounders, including age, sex, quality of colon
www.giejournal.org
Volume 85, No. 3 : 2017 GASTROINTESTINAL ENDOSCOPY 575
METHODS Study population and procedures
Cecal insertion time and adenoma detection
von Renteln et al
TABLE 1. Study subjects and procedure characteristics Insertion time quartiles Total (n [ 1043)
<3.5 min (n [ 256)
3.5-5.3 min (n [ 257)
5.3-8.1 min (n [ 268)
>8.1 min (n [ 262)
P value
Mean age, y (SD)
61.7 (8.3)
61.2 (8.4)
61.2 (8.9)
61.7 (8.1)
62.6 (7.9)
.208
Sex, male
667 (64.0)
220 (85.4)
176 (68.5)
152 (56.7)
119 (45.4)
<.001
ASA I
371 (33.6)
72 (28.1)
98 (38.1)
105 (39.2)
96 (36.6)
ASA II
561 (53.8)
152 (59.4)
138 (53.7)
134 (50.0)
137 (52.3)
ASA III
111 (10.6)
31 (12.5)
21 (8.2)
29 (10.8)
29 (11.1)
ASA class
.117
Indication
.108
Screening
554 (53.1)
133 (52.0)
144 (56.0)
149 (55.6)
128 (48.9)
Surveillance
317 (30.4)
70 (27.3)
83 (32.3)
77 (28.7)
87 (33.2)
Diagnostic
172 (16.5)
53 (20.7)
30 (11.7)
42 (15.7)
47 (17.9)
Procedure Preparation
.002
Excellent
391 (37.5)
89 (34.8)
109 (42.4)
104 (38.8)
89 (34.0)
Good
571 (54.7)
159 (62.1)
129 (50.2)
142 (53.0)
141 (53.8)
Fair
81 (7.8)
8 (3.1)
19 (7.4)
22 (8.2)
32 (12.2)
Cap use
532 (51.0)
157 (61.3)
134 (52.1)
122 (45.5)
119 (45.4)
.001
Median insertion time, min (IQR)
5.3 (4.6)
2.7 (1.0)
4.3 (.9)
6.4 (1.3)
10.6 (4.7)
<.001
Median withdrawal time,* min (IQR)
8.6 (3.6)
9.0 (3.3)
8.9 (3.4)
8.4 (3.7)
8.4 (3.9)
.725
Values are number of cases with percents in parentheses, unless otherwise noted. SD, Standard deviation; ASA, American Society of Anesthesiologists; IQR, interquartile range. *Among all patients without a polypectomy.
preparation, indication, and use of cap-assisted colonoscopy. We used stepwise forward logistic regression analysis to identify covariates to adjust for, and included variables with, a P .2 in univariate analysis in the final model. Considering that cap-assisted colonoscopy may decrease insertion time, we assessed for possible interaction between insertion time and cap use. Results were rounded to the nearest percentage point to facilitate text presentation; the tables provide more precision. All authors had access to the study data and have reviewed and approved the final manuscript.
polypectomies was 8.6 minutes (IQR, 7.1-10.7). Male sex, better quality of colonoscopy preparation, and use of a cap were associated with faster insertion times, whereas indication, American Society of Anesthesiologists class, or age was not.
Insertion time and adenoma detection
Among 1113 participants of the original trial on capassisted colonoscopy, 1043 subjects met inclusion criteria for this study. Sixty-four percent of subjects were men, and the mean age was 62 years (standard deviation, 8.3). Patient and procedure characteristics are provided in Table 1. Colon preparation was good or excellent in 92% of cases. For 84% of subjects the indication for colonoscopy was screening or surveillance. The median cecal insertion time was 5.3 minutes (IQR, 3.5-8.1), and the median withdrawal time for all patients without
Patient-level analysis. The mean number of adenomas per patient was .84 (standard deviation, 1.5), and the mean number of advanced adenomas per patient was .11 (standard deviation, .39) The ADR was 40% and the advanced ADR was 8.6%. Withdrawal time remained similar across all patient insertion time quartiles and ranged from 9.0 to 8.4 minutes from the fastest to the slowest insertion time quartile (P Z .950). The mean number of adenomas per patient significantly decreased from the fastest to the slowest insertion time quartile from 1.1 to .7 (P Z .031, Fig. 1). Similarly, the number of advanced adenomas decreased from .15 to .06 per patient across insertion time quartiles (P Z .013, Fig. 2). Although there was a nonsignificant trend toward a decreased ADR from 44% to 38% (P Z .543), the advanced ADR decreased from 13% to 5.0% (P Z .001). A lower adenoma detection was seen for proximal and distal adenomas across quartiles (proximal P Z .022, distal P Z .009). Similarly, the number of proximal advanced adenomas decreased from
576 GASTROINTESTINAL ENDOSCOPY Volume 85, No. 3 : 2017
www.giejournal.org
RESULTS Baseline characteristics
von Renteln et al
Cecal insertion time and adenoma detection
Adenomas per patient, mean n
1.2
Distal Proximal
1 0.8 0.6 0.4 0.2 0 Q1 < 3.5 min (n= 256)
Q2 3.5 - 5.3 min (n= 257)
Q3 5.3 - 8.1 min (n= 268)
Q4 > 8.1 min (n= 262)
P trend
Adenomas per patient, mean
1.06
0.93
0.70
0.69
.031
ADR
44.1%
41.2%
38.1%
38.2%
Insertion Time Quartile
Adjusted OR (95% CI)
1.00 (reference)
0.98 (0.69-1.41)
0.89 (0.62-1.28)
0.91 (0.63-1.33)
.543
Figure 1. Association between insertion time and adenoma detection among all study participants. Subjects were categorized into quartiles of insertion time. Proximal to splenic flexure. Adjusted analysis included adjustment for age, sex, cap use, and quality of preparation. OR, odds ratio adjusted for age, gender, and quality of colonoscopy preparation; CI, confidence interval.
Advanced adenomas per patient, mean n
0.16
Distal Proximal
0.14 0.12 0.1 0.08 0.06 0.04 0.02 0 Q1 < 3.5 min (n= 256)
Q2 3.5 - 5.3 min (n= 257)
Q3 5.3 - 8.1 min (n= 268)
Q4 > 8.1 min (n= 262)
P trend
Adv. adenomas per patient, mean
0.15
0.13
0.08
0.06
.013
Adv. ADR
12.9%
10.9%
6.0%
5.0%
-
1.00 (reference)
0.90 (0.52-1.56)
0.48 (0.26-0.92)
0.40 (0.20-0.82)
.001
Insertion Time Quartile
Adjusted OR (95% CI)
Figure 2. Association between insertion time and advanced adenoma detection among all study participants. Study subjects were categorized into quartiles of insertion time. Proximal to splenic flexure. Adjusted analysis included adjustment for age, sex, cap use, and quality of preparation. Adv., Advanced; OR, odds ratio adjusted for age, gender, and quality of colonoscopy preparation; CI, confidence interval.
www.giejournal.org
Volume 85, No. 3 : 2017 GASTROINTESTINAL ENDOSCOPY 577
Cecal insertion time and adenoma detection
von Renteln et al
TABLE 2. Association between endoscopists’ insertion time and adenoma detection among all endoscopists Insertion time categories
Withdrawal time
Adenomas
Advanced adenomas
ADR
Advanced ADR
No. of subjects
Median minutes (IQR)
Mean n (SD)
Mean n (SD)
% (95% CI)
% (95% CI)
5.3 (4.6)
1043
8.6 (3.6)
.84 (1.5)
.11 (.39)
40.4 (37.4-43.4)
8.6 (7.0-10.5)
3.8
184
8.3
1.02
.16
48.9
15.2
B
4.6
111
8.5
1.55
.16
63.1
12.6
C
4.7
233
9.1
.47
.05
24.9
4.3
D
5.4
104
10.6
.88
.12
40.4
9.6
E
5.6
53
7.3
.51
.04
32.1
1.9
F
6.3
72
10.3
.78
.15
40.3
9.7
G
6.4
88
5.6
.41
.03
28.4
3.4
H
7.2
76
14.0
1.08
.13
52.6
11.8
I
7.6
34
5.8
1.82
.24
41.2
14.7
J
8.2
88
7.9
.65
.05
40.9
3.4
P trend
d
d
.367
.883
.585
.857
.202
Median minutes (IQR)
All A
Endoscopists
Presented are withdrawal times, adenoma detection, and advanced adenoma detection for individual endoscopists ranked by insertion time. P values represent results of multivariate analysis adjusted for age, sex, cap use, and quality of preparation. SD, Standard deviation; IQR, interquartile range; ADR, adenoma detection rate; CI, confidence interval; d, not applicable. Any adenoma Advanced adenoma
Mean number of adenomas per patient
2
I B
1,6 1,2
adenoma or advanced adenoma detection (Fig. 3, Table 2).
DISCUSSION H
A
.10 to .05 (P Z .196) and of distal advanced adenomas detected from .05 to .01 across insertion time quartiles (P Z .003). In adjusted regression analysis the number of detected adenomas declined by 11% (P Z .031) and of advanced adenomas by 7% (P Z .013) from a faster to a slower insertion time quartile. This was independent of colonoscopy indication, cap use, quality of colonoscopy preparation, sex, or age. Further, we did not find an interaction between cap use and insertion time. Endoscopist-level analysis. Median endoscopists’ withdrawal time was 8.6 minutes (IRQ, 7.1-10.7), ranging from 5.6 to 14 minutes (Table 2). There was no association between individual endoscopist’s insertion times and withdrawal times. Further, individual endoscopist’s insertion time was not associated with
In this study a longer cecal insertion time was associated with a decreased detection of adenomas and advanced adenomas. The mean number of adenomas per patient decreased from 1.1 to .7 from the fastest to the slowest insertion time quartile, which represents an almost 40% decline. Furthermore, the mean number of advanced adenomas decreased from .15 to .05, an even more pronounced decline of 66%. This association was independent of individual endoscopists’ insertion times and therefore appears to be solely related to patient factors. Interestingly, longer insertion did not lead to a faster or slower withdrawal time. Considering that insertion time may reflect technical skill, there are surprisingly little data on a possible association between cecal insertion time and adenoma detection. A large study from South Korea examined more than 12,000 patients undergoing screening colonoscopy.11 Similar to our study, adenoma detection decreased with longer cecal insertion time independent of the endoscopist individual insertion times. In contrast to our results, this association was limited to diminutive (5 mm) adenomas. These, and our findings, support the assertion that patient factors, and not endoscopist skills, account for the association between insertion time and adenoma detection. A second smaller, retrospective study of 550 patients undergoing a screening colonoscopy found significantly more adenomas, if the ratio of insertion time to withdrawal time was less than 1.12 This
578 GASTROINTESTINAL ENDOSCOPY Volume 85, No. 3 : 2017
www.giejournal.org
D
F
0,8
J C
0,4 0
E
G
0 1 2 3 4 5 6 7 8 9 Endoscopists’ median cecal insertion time, minutes
Figure 3. Individual endoscopists’ insertion time plus individual adenoma or advanced adenoma detection.
von Renteln et al
Cecal insertion time and adenoma detection
study did not adjust for withdrawal time and applied total withdrawal time (including polyp removal) as an outcome measure, therefore limiting the generalizability of the results. In contrast to these 2 studies, we did not find that longer insertion time led to shorter withdrawal times. In our study, withdrawal time remained similar across insertion time quartiles, whereas adenoma detection decreased. Two possible explanations may explain this observation. First, endoscopists with a faster insertion time may have technical skills that allow a more meticulous examination that enhance adenoma detection during withdrawal. However, we did not find that individual endoscopists’ insertion times were associated with adenoma detection. A second plausible explanation may be related to patient factors that led to a longer insertion time. Female sex, older age, waist-to-hip ratio, previous abdominal surgery, inadequate bowel preparation, and chronic constipation are independently associated with a longer cecal insertion time.8,13-15 In our study we found that female sex was associated with prolonged insertion. However, we did not find that sex affected the association between insertion time and adenoma detection in adjusted analysis. A long insertion time may therefore reflect a more challenging colonoscopy that may translate into a higher miss rate during endoscope withdrawal independent of the underlying factors like age or sex. For instance, reaching the cecum in a looped position may hinder examination of the proximal colon, and a difficult insertion through the sigmoid colon may portend challenges in mucosal inspection during withdrawal. The observed similar decrease in adenoma detection in the proximal and distal colon suggests that after a long insertion time visualization may be more difficult across the entire colon. Another possible explanation for the outcomes might be that a longer and more difficult insertion results in endoscopist fatigue. Withdrawal times may remain the same in such cases but fatigue might explain lower adenoma detection. Although we did not find a difference in overall ADR with increased insertion time, we detected a decrease in the number of adenomas. We chose the mean number of adenomas per patient as the main outcome measure, because it has been described as a more valid representation of adenoma detection than ADR.10 Although ADR is used as the main quality indicator, it leaves room for a “one-and-done” approach. This might explain part of the observed difference between ADR and mean number of adenomas. However, the study also may have been underpowered to show the observed difference in ADR as significant. It is interesting to note that withdrawal times among all endoscopists were greater than the recommended minimum of 6 minutes. Even when maintaining long withdrawal times, fewer adenomas were found the longer it took to reach the cecum. This was independent of the
endoscopists’ individual insertion times. Technical skills that enable a fast cecal insertion may therefore be different from those that are required for adequate adenoma detection. Rather than technical skill, the observed findings appear to be related to patient factors. It can be speculated that a longer insertion time may be a surrogate marker for individuals with a difficult anatomy. In patients with a longer insertion time, greater care during withdrawal may then be required to compensate for a challenging examination. Whether a longer withdrawal time that is adjusted for a longer insertion time will in fact result in increased adenoma detection should be subject to further research. Further, if studies confirm an association between insertion time and adenoma detection, insertion time should be included in the endoscopy report. The observed significant decline in the detection of advanced adenomas among patients with long insertion times raises concerns of whether a subset of these patients may be at higher risk for interval colorectal cancer. Strengths of the study include its size, prospective data collection, and participation of multiple experienced endoscopists without trainee involvement. Several limitations should be noted. Colonoscopies were performed by endoscopists at 2 academic centers, which limits the generalizability of the results. All endoscopists surpassed the recommended minimum withdrawal time. Findings might have been different with a broader variation. Further, we included patients with different indications and hence different adenoma risk. However, we did not find that indication modified the association between insertion time and adenoma detection. Because it is known that cap-assisted colonoscopy decreases insertion times, we considered a possible interaction between insertion time and cap use.9 We did not find such interaction or an effect of cap use on the main results. Finally, we did not record whether polyps were removed during insertion. We believe this occurred infrequently, because it is general practice at both centers to remove all polyps during withdrawal. In conclusion, we found a decrease in the detection of adenomas and particularly of advanced adenomas with increased cecal insertion time despite an adequate withdrawal time. This effect was independent of the endoscopist. One possible explanation of our findings is that insertion time represents a surrogate marker for a difficult examination that hinders an adequate inspection of the colon during withdrawal. Whether a purposefully longer withdrawal time after a long insertion time may compensate for a lower adenoma detection should be further studied.
www.giejournal.org
Volume 85, No. 3 : 2017 GASTROINTESTINAL ENDOSCOPY 579
REFERENCES 1. Kaminski MF, Regula J, Kraszewska E, et al. Quality indicators for colonoscopy and the risk of interval cancer. N Engl J Med 2010;362: 1795-803.
Cecal insertion time and adenoma detection 2. Corley DA, Jensen CD, Marks AR, et al. Adenoma detection rate and risk of colorectal cancer and death. N Engl J Med 2014;370: 1298-306. 3. Fayad NF, Kahi CJ. Quality measures for colonoscopy: a critical evaluation. Clin Gastroenterol Hepatol 2014;12:1973-80. 4. Barclay RL, Vicari JJ, Doughty AS, et al. Colonoscopic withdrawal times and adenoma detection during screening colonoscopy. N Engl J Med 2006;355:2533-41. 5. Corley DA, Jensen CD, Marks AR. Can we improve adenoma detection rates? A systematic review of intervention studies. Gastrointest Endosc 2011;74:656-65. 6. Lee RH, Tang RS, Muthusamy VR, et al. Quality of colonoscopy withdrawal technique and variability in adenoma detection rates (with videos). Gastrointest Endosc 2011;74:128-34. 7. Sedlack RE, Coyle WJ, Sedlack RE, et al. Assessment of competency in endoscopy: establishing and validating generalizable competency benchmarks for colonoscopy. Gastrointest Endosc 2016;83: 516-23. 8. Lee HL, Eun CS, Lee OY, et al. Significance of colonoscope length in cecal insertion time. Gastrointest Endosc 2009;69:503-8.
von Renteln et al 9. Pohl H, Bensen SP, Toor A, et al. Cap-assisted colonoscopy and detection of Adenomatous Polyps (CAP) study: a randomized trial. Endoscopy 2015;47:891-7. 10. Wang HS, Pisegna J, Modi R, et al. Adenoma detection rate is necessary but insufficient for distinguishing high versus low endoscopist performance. Gastrointest Endosc 2013;77:71-8. 11. Yang MH, Cho J, Rampal S, et al. The association between cecal insertion time and colorectal neoplasm detection. BMC Gastroenterol 2013;13:124. 12. Benson ME, Reichelderfer M, Said A, et al. Variation in colonoscopic technique and adenoma detection rates at an academic gastroenterology unit. Dig Dis Sci 2010;55:166-71. 13. Hsieh YH, Kuo CS, Tseng KC, et al. Factors that predict cecal insertion time during sedated colonoscopy: the role of waist circumference. J Gastroenterol Hepatol 2008;23:215-7. 14. Krishnan P, Sofi AA, Dempsey R, et al. Body mass index predicts cecal insertion time: the higher, the better. Dig Endosc 2012;24:439-42. 15. Nagata N, Sakamoto K, Arai T, et al. Predictors for cecal insertion time: the impact of abdominal visceral fat measured by computed tomography. Dis Colon Rectum 2014;57:1213-9.
GIE on LinkedIn Follow GIE on LinkedIn. Followers will receive news, updates, and links to author interviews, podcasts, articles, and tables of contents. Search on LinkedIn for “GIE: Gastrointestinal Endoscopy with Editor Michael B. Wallace” and follow us today.
580 GASTROINTESTINAL ENDOSCOPY Volume 85, No. 3 : 2017
www.giejournal.org
Prolonged cecal insertion time is associated with decreased adenoma detection Daniel von Renteln, MD,1 Douglas J. Robertson, MD,2,3 Steve Bensen, MD,3 Heiko Pohl, MD2,3 Montreal, Quebec, Canada; White River Junction, Vermont; Lebanon, New Hampshire, USA
Background and Aims: The adenoma detection rate (ADR) is an important colonoscopy quality parameter. A longer withdrawal time is associated with increased adenoma detection; however, the effect of cecal insertion time on adenoma detection is unclear. The aim of this study was to evaluate if cecal insertion time is associated with adenoma detection. Methods: The study included 50- to 89-year-old subjects undergoing an elective outpatient colonoscopy as part of a randomized trial on cap-assisted colonoscopy. The primary outcome was the adjusted mean number of adenomas per patient across quartiles of lengthening patient insertion times. Secondary outcomes included the mean number of advanced adenomas, ADR, and advanced ADR. Adjusted regression analysis was applied to assess for a possible trend. Results: Among 1043 included study subjects, the median cecal insertion time was 5.3 minutes (interquartile range [IQR], 3.5-8.1) and the median withdrawal time was 8.6 minutes (IQR, 7.1-10.7). Cecal insertion time was not associated with withdrawal time (P Z .950). The mean number of adenomas per patient decreased across increasing insertion time quartiles from 1.1 to .7, corresponding to a 11% decline per quartile (P Z .031). Similarly, the number of advanced adenomas decreased from .15 to .06, corresponding to a 7% decline per quartile (P Z .013). The observed decrease in adenoma detection was similar for proximal and distal adenomas. Endoscopists’ individual insertion times were not associated with adenoma detection. Conclusions: A longer cecal insertion time was associated with a decreased detection of adenomas and advanced adenomas. A long insertion time may be a marker for a more difficult examination that requires a longer withdrawal time to assure adequate examination and adenoma detection. (Clinical trial registration number: NCT01935180.) (Gastrointest Endosc 2017;85:574-80.)
The adenoma detection rate (ADR) is considered an important quality metric for screening colonoscopy.1 A low ADR is inversely associated with a higher risk for the patient to develop postcolonoscopy colorectal cancer.2 Although many studies have examined procedure characteristics and interventions to improve adenoma detection, few factors have been found to be
effective.3 A withdrawal time of at least 6 minutes is associated with a higher ADR compared with a shorter withdrawal time.4 However, lengthening withdrawal time alone does not ensure optimal polyp detection,5 and a study suggests that technical ability and diligence of the examiner are similarly important.6
Abbreviations: ADR, adenoma detection rate; IQR, interquartile range.
Current affiliations: Department of Medicine, Division of Gastroenterology, Montreal University Hospital (CHUM), Montreal University Hospital Research Center (CRCHUM), Montreal, Quebec, Canada (1), Section of Gastroenterology, Veterans Affairs Medical Center, White River Junction, Vermont, USA (2), Section of Gastroenterology, Dartmouth-Hitchcock Medical Center, Lebanon, New Hampshire, USA (3).
DISCLOSURE: All authors disclosed no financial relationships relevant to this publication. Research support for this study was provided by The Veterans Health Administration. The views expressed in this article are those of the authors and do not necessarily represent the views of the Department of Veterans Affairs. Copyright ª 2017 by the American Society for Gastrointestinal Endoscopy 0016-5107/$36.00 http://dx.doi.org/10.1016/j.gie.2016.08.021
Reprint requests: Daniel von Renteln, MD, Department of Medicine, Division of Gastroenterology, Montreal University Hospital (CHUM), Montreal University Hospital Research Center (CRCHUM), 900 Rue SaintDenis, Montréal, QC H2X 0A9, Montreal, Canada.
Received March 8, 2016. Accepted August 17, 2016.
574 GASTROINTESTINAL ENDOSCOPY Volume 85, No. 3 : 2017
www.giejournal.org
von Renteln et al
There is reason to believe that insertion time may also be associated with adenoma detection. At the patient level, cecal insertion time is influenced by patient and procedural factors, including bowel cleansing, age, history of constipation, prior abdominal surgical, sex, and body mass index.7-8 A longer insertion time may reflect a more difficult examination, which may hinder adenoma detection during withdrawal. At the endoscopist level, technical skill is required for advancing the endoscope to the cecum. Cecal intubation times and cecal intubation rates > 90% have been identified as competency metrics.7 A short insertion time may indicate technical skill, which may also promote a meticulous examination with increased adenoma detection during withdrawal. Further, after a prolonged insertion time, the endoscopist may feel fatigued or under time pressure and accelerate endoscope withdrawal with the potential to miss polyps. The primary aim of this study was to assess whether cecal insertion time is associated with adenoma detection. We were further interested in understanding whether a possible association is related to patient or endoscopist factors.
Cecal insertion time and adenoma detection
Outcome measures The primary outcome was the mean number of adenomas per patient, which was chosen because it is a more valid representation of adenoma detection as opposed to the ADR.10 Secondary outcomes included the mean number of advanced adenomas, ADR, and advanced ADR. The ADR represents the proportion of patients with at least 1 adenoma. Advanced adenomas were defined as adenomas 10 mm in size or containing villous features, high-grade dysplasia, or cancer. The advanced ADR included all patients with at least 1 advanced adenoma. Procedure characteristics were assessed as part of the initial trial. Cecal insertion time was defined as the time elapsed from introducing the colonoscope into the rectum until intubation of the cecum. Withdrawal time was defined as the time to withdraw the endoscope from the cecum to the rectum among all patients without a polypectomy. Procedure data were recorded during the colonoscopy by trained personnel (sedation nurse or research assistant).
Statistical analysis
The study included subjects who were 50 to 89 years of age undergoing an elective outpatient colonoscopy at the Dartmouth-Hitchcock Medical Center, Lebanon, New Hampshire, or the VA Medical Center, White River Junction, Vermont. All participants were part of a randomized trial on cap-assisted colonoscopy (clintrials.gov NCT01935180).9 The cap used was a 4-mm transparent distal attachment cap (Disposable Distal Attachment, Models D-201-15004, D-201-12704; Olympus America Inc Center Valley, PA, USA). Patients with known inflammatory bowel disease, active colitis, coagulopathy, polyposis syndrome, or poor general health as defined by American Society of Anesthesiologists class greater than 3 and those undergoing an emergency colonoscopy were excluded from the study. For the current analysis, we additionally excluded patients with prior colon resection, incomplete colonoscopy, poor bowel preparation, and incomplete documentation of insertion or withdrawal times. All study subjects underwent a standard bowel cleansing preparation, typically a 4-L polyethylene glycol solution the day before the colonoscopy. Each study patient underwent only 1 study colonoscopy. Study procedures were performed by 10 experienced board-certified gastroenterologists using standard endoscopes (H-CF/H-PCF 180; Olympus). All study participants provided informed written consent for the initial trial and use of the data for additional analysis. The study was approved by the local institutional review board.
To examine a possible association between cecal insertion time and adenoma detection, we performed patientand endoscopist-level analyses. For the patient-level analysis, we categorized insertion times of all patients into quartiles of insertion time. We then assessed the measures of adenoma detection and withdrawal time across lengthening quartiles of insertion times. In addition, we assessed the effect of insertion time on proximal (proximal to, and not including, the splenic flexure) and distal adenoma detection. For the endoscopist-level analysis, we assessed adenoma detection and withdrawal times by individual endoscopist’s median insertion times. Proportions are presented as percentages. Continuous variables are expressed as means with standard deviations if normally distributed and as medians with interquartile ranges (IQR) if not normally distributed. Poisson regression analysis was applied to examine an association between insertion time and number of detected adenomas or advanced adenomas. Unit changes across quartiles of insertion times are presented as relative change expressed as percentage with 95% confidence interval. To examine an association between insertion time and ADR and advanced ADR, we used adjusted logistic regression analysis expressed as odds ratio with 95% confidence interval across quartiles of insertion time and to calculate a test of trend. To examine the effect of endoscopists on associations between insertion time adenoma detection, we applied the factor endoscopist as a fixed effect. We calculated test of trend using Poisson regression (withdrawal time, mean number of adenomas, and advanced adenomas) and logistic regression (ADR and advanced ADR). All regression models considered possible confounders, including age, sex, quality of colon
www.giejournal.org
Volume 85, No. 3 : 2017 GASTROINTESTINAL ENDOSCOPY 575
METHODS Study population and procedures
Cecal insertion time and adenoma detection
von Renteln et al
TABLE 1. Study subjects and procedure characteristics Insertion time quartiles Total (n [ 1043)
<3.5 min (n [ 256)
3.5-5.3 min (n [ 257)
5.3-8.1 min (n [ 268)
>8.1 min (n [ 262)
P value
Mean age, y (SD)
61.7 (8.3)
61.2 (8.4)
61.2 (8.9)
61.7 (8.1)
62.6 (7.9)
.208
Sex, male
667 (64.0)
220 (85.4)
176 (68.5)
152 (56.7)
119 (45.4)
<.001
ASA I
371 (33.6)
72 (28.1)
98 (38.1)
105 (39.2)
96 (36.6)
ASA II
561 (53.8)
152 (59.4)
138 (53.7)
134 (50.0)
137 (52.3)
ASA III
111 (10.6)
31 (12.5)
21 (8.2)
29 (10.8)
29 (11.1)
ASA class
.117
Indication
.108
Screening
554 (53.1)
133 (52.0)
144 (56.0)
149 (55.6)
128 (48.9)
Surveillance
317 (30.4)
70 (27.3)
83 (32.3)
77 (28.7)
87 (33.2)
Diagnostic
172 (16.5)
53 (20.7)
30 (11.7)
42 (15.7)
47 (17.9)
Procedure Preparation
.002
Excellent
391 (37.5)
89 (34.8)
109 (42.4)
104 (38.8)
89 (34.0)
Good
571 (54.7)
159 (62.1)
129 (50.2)
142 (53.0)
141 (53.8)
Fair
81 (7.8)
8 (3.1)
19 (7.4)
22 (8.2)
32 (12.2)
Cap use
532 (51.0)
157 (61.3)
134 (52.1)
122 (45.5)
119 (45.4)
.001
Median insertion time, min (IQR)
5.3 (4.6)
2.7 (1.0)
4.3 (.9)
6.4 (1.3)
10.6 (4.7)
<.001
Median withdrawal time,* min (IQR)
8.6 (3.6)
9.0 (3.3)
8.9 (3.4)
8.4 (3.7)
8.4 (3.9)
.725
Values are number of cases with percents in parentheses, unless otherwise noted. SD, Standard deviation; ASA, American Society of Anesthesiologists; IQR, interquartile range. *Among all patients without a polypectomy.
preparation, indication, and use of cap-assisted colonoscopy. We used stepwise forward logistic regression analysis to identify covariates to adjust for, and included variables with, a P .2 in univariate analysis in the final model. Considering that cap-assisted colonoscopy may decrease insertion time, we assessed for possible interaction between insertion time and cap use. Results were rounded to the nearest percentage point to facilitate text presentation; the tables provide more precision. All authors had access to the study data and have reviewed and approved the final manuscript.
polypectomies was 8.6 minutes (IQR, 7.1-10.7). Male sex, better quality of colonoscopy preparation, and use of a cap were associated with faster insertion times, whereas indication, American Society of Anesthesiologists class, or age was not.
Insertion time and adenoma detection
Among 1113 participants of the original trial on capassisted colonoscopy, 1043 subjects met inclusion criteria for this study. Sixty-four percent of subjects were men, and the mean age was 62 years (standard deviation, 8.3). Patient and procedure characteristics are provided in Table 1. Colon preparation was good or excellent in 92% of cases. For 84% of subjects the indication for colonoscopy was screening or surveillance. The median cecal insertion time was 5.3 minutes (IQR, 3.5-8.1), and the median withdrawal time for all patients without
Patient-level analysis. The mean number of adenomas per patient was .84 (standard deviation, 1.5), and the mean number of advanced adenomas per patient was .11 (standard deviation, .39) The ADR was 40% and the advanced ADR was 8.6%. Withdrawal time remained similar across all patient insertion time quartiles and ranged from 9.0 to 8.4 minutes from the fastest to the slowest insertion time quartile (P Z .950). The mean number of adenomas per patient significantly decreased from the fastest to the slowest insertion time quartile from 1.1 to .7 (P Z .031, Fig. 1). Similarly, the number of advanced adenomas decreased from .15 to .06 per patient across insertion time quartiles (P Z .013, Fig. 2). Although there was a nonsignificant trend toward a decreased ADR from 44% to 38% (P Z .543), the advanced ADR decreased from 13% to 5.0% (P Z .001). A lower adenoma detection was seen for proximal and distal adenomas across quartiles (proximal P Z .022, distal P Z .009). Similarly, the number of proximal advanced adenomas decreased from
576 GASTROINTESTINAL ENDOSCOPY Volume 85, No. 3 : 2017
www.giejournal.org
RESULTS Baseline characteristics
von Renteln et al
Cecal insertion time and adenoma detection
Adenomas per patient, mean n
1.2
Distal Proximal
1 0.8 0.6 0.4 0.2 0 Q1 < 3.5 min (n= 256)
Q2 3.5 - 5.3 min (n= 257)
Q3 5.3 - 8.1 min (n= 268)
Q4 > 8.1 min (n= 262)
P trend
Adenomas per patient, mean
1.06
0.93
0.70
0.69
.031
ADR
44.1%
41.2%
38.1%
38.2%
Insertion Time Quartile
Adjusted OR (95% CI)
1.00 (reference)
0.98 (0.69-1.41)
0.89 (0.62-1.28)
0.91 (0.63-1.33)
.543
Figure 1. Association between insertion time and adenoma detection among all study participants. Subjects were categorized into quartiles of insertion time. Proximal to splenic flexure. Adjusted analysis included adjustment for age, sex, cap use, and quality of preparation. OR, odds ratio adjusted for age, gender, and quality of colonoscopy preparation; CI, confidence interval.
Advanced adenomas per patient, mean n
0.16
Distal Proximal
0.14 0.12 0.1 0.08 0.06 0.04 0.02 0 Q1 < 3.5 min (n= 256)
Q2 3.5 - 5.3 min (n= 257)
Q3 5.3 - 8.1 min (n= 268)
Q4 > 8.1 min (n= 262)
P trend
Adv. adenomas per patient, mean
0.15
0.13
0.08
0.06
.013
Adv. ADR
12.9%
10.9%
6.0%
5.0%
-
1.00 (reference)
0.90 (0.52-1.56)
0.48 (0.26-0.92)
0.40 (0.20-0.82)
.001
Insertion Time Quartile
Adjusted OR (95% CI)
Figure 2. Association between insertion time and advanced adenoma detection among all study participants. Study subjects were categorized into quartiles of insertion time. Proximal to splenic flexure. Adjusted analysis included adjustment for age, sex, cap use, and quality of preparation. Adv., Advanced; OR, odds ratio adjusted for age, gender, and quality of colonoscopy preparation; CI, confidence interval.
www.giejournal.org
Volume 85, No. 3 : 2017 GASTROINTESTINAL ENDOSCOPY 577
Cecal insertion time and adenoma detection
von Renteln et al
TABLE 2. Association between endoscopists’ insertion time and adenoma detection among all endoscopists Insertion time categories
Withdrawal time
Adenomas
Advanced adenomas
ADR
Advanced ADR
No. of subjects
Median minutes (IQR)
Mean n (SD)
Mean n (SD)
% (95% CI)
% (95% CI)
5.3 (4.6)
1043
8.6 (3.6)
.84 (1.5)
.11 (.39)
40.4 (37.4-43.4)
8.6 (7.0-10.5)
3.8
184
8.3
1.02
.16
48.9
15.2
B
4.6
111
8.5
1.55
.16
63.1
12.6
C
4.7
233
9.1
.47
.05
24.9
4.3
D
5.4
104
10.6
.88
.12
40.4
9.6
E
5.6
53
7.3
.51
.04
32.1
1.9
F
6.3
72
10.3
.78
.15
40.3
9.7
G
6.4
88
5.6
.41
.03
28.4
3.4
H
7.2
76
14.0
1.08
.13
52.6
11.8
I
7.6
34
5.8
1.82
.24
41.2
14.7
J
8.2
88
7.9
.65
.05
40.9
3.4
P trend
d
d
.367
.883
.585
.857
.202
Median minutes (IQR)
All A
Endoscopists
Presented are withdrawal times, adenoma detection, and advanced adenoma detection for individual endoscopists ranked by insertion time. P values represent results of multivariate analysis adjusted for age, sex, cap use, and quality of preparation. SD, Standard deviation; IQR, interquartile range; ADR, adenoma detection rate; CI, confidence interval; d, not applicable. Any adenoma Advanced adenoma
Mean number of adenomas per patient
2
I B
1,6 1,2
adenoma or advanced adenoma detection (Fig. 3, Table 2).
DISCUSSION H
A
.10 to .05 (P Z .196) and of distal advanced adenomas detected from .05 to .01 across insertion time quartiles (P Z .003). In adjusted regression analysis the number of detected adenomas declined by 11% (P Z .031) and of advanced adenomas by 7% (P Z .013) from a faster to a slower insertion time quartile. This was independent of colonoscopy indication, cap use, quality of colonoscopy preparation, sex, or age. Further, we did not find an interaction between cap use and insertion time. Endoscopist-level analysis. Median endoscopists’ withdrawal time was 8.6 minutes (IRQ, 7.1-10.7), ranging from 5.6 to 14 minutes (Table 2). There was no association between individual endoscopist’s insertion times and withdrawal times. Further, individual endoscopist’s insertion time was not associated with
In this study a longer cecal insertion time was associated with a decreased detection of adenomas and advanced adenomas. The mean number of adenomas per patient decreased from 1.1 to .7 from the fastest to the slowest insertion time quartile, which represents an almost 40% decline. Furthermore, the mean number of advanced adenomas decreased from .15 to .05, an even more pronounced decline of 66%. This association was independent of individual endoscopists’ insertion times and therefore appears to be solely related to patient factors. Interestingly, longer insertion did not lead to a faster or slower withdrawal time. Considering that insertion time may reflect technical skill, there are surprisingly little data on a possible association between cecal insertion time and adenoma detection. A large study from South Korea examined more than 12,000 patients undergoing screening colonoscopy.11 Similar to our study, adenoma detection decreased with longer cecal insertion time independent of the endoscopist individual insertion times. In contrast to our results, this association was limited to diminutive (5 mm) adenomas. These, and our findings, support the assertion that patient factors, and not endoscopist skills, account for the association between insertion time and adenoma detection. A second smaller, retrospective study of 550 patients undergoing a screening colonoscopy found significantly more adenomas, if the ratio of insertion time to withdrawal time was less than 1.12 This
578 GASTROINTESTINAL ENDOSCOPY Volume 85, No. 3 : 2017
www.giejournal.org
D
F
0,8
J C
0,4 0
E
G
0 1 2 3 4 5 6 7 8 9 Endoscopists’ median cecal insertion time, minutes
Figure 3. Individual endoscopists’ insertion time plus individual adenoma or advanced adenoma detection.
von Renteln et al
Cecal insertion time and adenoma detection
study did not adjust for withdrawal time and applied total withdrawal time (including polyp removal) as an outcome measure, therefore limiting the generalizability of the results. In contrast to these 2 studies, we did not find that longer insertion time led to shorter withdrawal times. In our study, withdrawal time remained similar across insertion time quartiles, whereas adenoma detection decreased. Two possible explanations may explain this observation. First, endoscopists with a faster insertion time may have technical skills that allow a more meticulous examination that enhance adenoma detection during withdrawal. However, we did not find that individual endoscopists’ insertion times were associated with adenoma detection. A second plausible explanation may be related to patient factors that led to a longer insertion time. Female sex, older age, waist-to-hip ratio, previous abdominal surgery, inadequate bowel preparation, and chronic constipation are independently associated with a longer cecal insertion time.8,13-15 In our study we found that female sex was associated with prolonged insertion. However, we did not find that sex affected the association between insertion time and adenoma detection in adjusted analysis. A long insertion time may therefore reflect a more challenging colonoscopy that may translate into a higher miss rate during endoscope withdrawal independent of the underlying factors like age or sex. For instance, reaching the cecum in a looped position may hinder examination of the proximal colon, and a difficult insertion through the sigmoid colon may portend challenges in mucosal inspection during withdrawal. The observed similar decrease in adenoma detection in the proximal and distal colon suggests that after a long insertion time visualization may be more difficult across the entire colon. Another possible explanation for the outcomes might be that a longer and more difficult insertion results in endoscopist fatigue. Withdrawal times may remain the same in such cases but fatigue might explain lower adenoma detection. Although we did not find a difference in overall ADR with increased insertion time, we detected a decrease in the number of adenomas. We chose the mean number of adenomas per patient as the main outcome measure, because it has been described as a more valid representation of adenoma detection than ADR.10 Although ADR is used as the main quality indicator, it leaves room for a “one-and-done” approach. This might explain part of the observed difference between ADR and mean number of adenomas. However, the study also may have been underpowered to show the observed difference in ADR as significant. It is interesting to note that withdrawal times among all endoscopists were greater than the recommended minimum of 6 minutes. Even when maintaining long withdrawal times, fewer adenomas were found the longer it took to reach the cecum. This was independent of the
endoscopists’ individual insertion times. Technical skills that enable a fast cecal insertion may therefore be different from those that are required for adequate adenoma detection. Rather than technical skill, the observed findings appear to be related to patient factors. It can be speculated that a longer insertion time may be a surrogate marker for individuals with a difficult anatomy. In patients with a longer insertion time, greater care during withdrawal may then be required to compensate for a challenging examination. Whether a longer withdrawal time that is adjusted for a longer insertion time will in fact result in increased adenoma detection should be subject to further research. Further, if studies confirm an association between insertion time and adenoma detection, insertion time should be included in the endoscopy report. The observed significant decline in the detection of advanced adenomas among patients with long insertion times raises concerns of whether a subset of these patients may be at higher risk for interval colorectal cancer. Strengths of the study include its size, prospective data collection, and participation of multiple experienced endoscopists without trainee involvement. Several limitations should be noted. Colonoscopies were performed by endoscopists at 2 academic centers, which limits the generalizability of the results. All endoscopists surpassed the recommended minimum withdrawal time. Findings might have been different with a broader variation. Further, we included patients with different indications and hence different adenoma risk. However, we did not find that indication modified the association between insertion time and adenoma detection. Because it is known that cap-assisted colonoscopy decreases insertion times, we considered a possible interaction between insertion time and cap use.9 We did not find such interaction or an effect of cap use on the main results. Finally, we did not record whether polyps were removed during insertion. We believe this occurred infrequently, because it is general practice at both centers to remove all polyps during withdrawal. In conclusion, we found a decrease in the detection of adenomas and particularly of advanced adenomas with increased cecal insertion time despite an adequate withdrawal time. This effect was independent of the endoscopist. One possible explanation of our findings is that insertion time represents a surrogate marker for a difficult examination that hinders an adequate inspection of the colon during withdrawal. Whether a purposefully longer withdrawal time after a long insertion time may compensate for a lower adenoma detection should be further studied.
www.giejournal.org
Volume 85, No. 3 : 2017 GASTROINTESTINAL ENDOSCOPY 579
REFERENCES 1. Kaminski MF, Regula J, Kraszewska E, et al. Quality indicators for colonoscopy and the risk of interval cancer. N Engl J Med 2010;362: 1795-803.
Cecal insertion time and adenoma detection 2. Corley DA, Jensen CD, Marks AR, et al. Adenoma detection rate and risk of colorectal cancer and death. N Engl J Med 2014;370: 1298-306. 3. Fayad NF, Kahi CJ. Quality measures for colonoscopy: a critical evaluation. Clin Gastroenterol Hepatol 2014;12:1973-80. 4. Barclay RL, Vicari JJ, Doughty AS, et al. Colonoscopic withdrawal times and adenoma detection during screening colonoscopy. N Engl J Med 2006;355:2533-41. 5. Corley DA, Jensen CD, Marks AR. Can we improve adenoma detection rates? A systematic review of intervention studies. Gastrointest Endosc 2011;74:656-65. 6. Lee RH, Tang RS, Muthusamy VR, et al. Quality of colonoscopy withdrawal technique and variability in adenoma detection rates (with videos). Gastrointest Endosc 2011;74:128-34. 7. Sedlack RE, Coyle WJ, Sedlack RE, et al. Assessment of competency in endoscopy: establishing and validating generalizable competency benchmarks for colonoscopy. Gastrointest Endosc 2016;83: 516-23. 8. Lee HL, Eun CS, Lee OY, et al. Significance of colonoscope length in cecal insertion time. Gastrointest Endosc 2009;69:503-8.
von Renteln et al 9. Pohl H, Bensen SP, Toor A, et al. Cap-assisted colonoscopy and detection of Adenomatous Polyps (CAP) study: a randomized trial. Endoscopy 2015;47:891-7. 10. Wang HS, Pisegna J, Modi R, et al. Adenoma detection rate is necessary but insufficient for distinguishing high versus low endoscopist performance. Gastrointest Endosc 2013;77:71-8. 11. Yang MH, Cho J, Rampal S, et al. The association between cecal insertion time and colorectal neoplasm detection. BMC Gastroenterol 2013;13:124. 12. Benson ME, Reichelderfer M, Said A, et al. Variation in colonoscopic technique and adenoma detection rates at an academic gastroenterology unit. Dig Dis Sci 2010;55:166-71. 13. Hsieh YH, Kuo CS, Tseng KC, et al. Factors that predict cecal insertion time during sedated colonoscopy: the role of waist circumference. J Gastroenterol Hepatol 2008;23:215-7. 14. Krishnan P, Sofi AA, Dempsey R, et al. Body mass index predicts cecal insertion time: the higher, the better. Dig Endosc 2012;24:439-42. 15. Nagata N, Sakamoto K, Arai T, et al. Predictors for cecal insertion time: the impact of abdominal visceral fat measured by computed tomography. Dis Colon Rectum 2014;57:1213-9.
GIE on LinkedIn Follow GIE on LinkedIn. Followers will receive news, updates, and links to author interviews, podcasts, articles, and tables of contents. Search on LinkedIn for “GIE: Gastrointestinal Endoscopy with Editor Michael B. Wallace” and follow us today.
580 GASTROINTESTINAL ENDOSCOPY Volume 85, No. 3 : 2017
www.giejournal.org