Adenoma detection and retroscopy
Article Outline
Abbreviation: TER, Third Eye Retroscope
Four interdependent technical variables govern polyp detection rates: (1) instrument angle of view, (2) withdrawal time, (3) bowel preparation quality, and (4) tip deflection technique.
During withdrawal of the colonoscope from the cecum, there are 4 interdependent technical variables that govern polyp detection rates: (1) instrument angle of view, (2) withdrawal time, (3) bowel preparation quality, and (4) tip deflection technique. Optimization of these 4 interdependent variables improves polyp detection rates through one single mechanism, that of maximizing the visualization of the colonic mucosal surface area.
Human colonic surface area visualization studies using CT colography have provided many insights into the impact of the enhancement of the angle of view on mucosal surface examination during colonoscope withdrawal. Pickhardt et al1 demonstrated that the majority of polyps missed during colonoscopy were on the proximal aspect of colonic folds. By using modified CT colography software that enables the conduct of endoluminal flythroughs at varying angles of view, East et al2 have estimated the nonvisualized colonic mucosal surface area for colonoscopes with different angles of view used in conjunction with and without a 135° angle of view retroscope (Table 1). Their study predicts an absolute visualization gain of 5.6% for the 170° angle of view colonoscope compared with the 140° angle of view colonoscope and an additional gain of 6.7% when the Third Eye Retroscope (TER) (Avantis Medical Systems, Inc, Sunnyvale, Calif) is used with the 170° angle of view colonoscope. Simulated addition of the TER to the 140° angle of view colonoscope resulted in a 12.1% absolute visualization gain. The predicted percent colonic surface area visualized with either instrument combined with the TER is 98.7% or more. The gain in the amount of surface area visualized by increasing the instrument angle of view from 140° and 170° has not translated into increased polyp detection3; however, more rapid insertion and withdrawal times without deterioration in polyp detection have been demonstrated.4, 5
Table 1. Angle of view–dependent nonvisualized colonic surface area
| Angle of view | 140° | 170° | 140° + 135° R∗ | 170° + 135° TER |
| Percent of colon surface area not visualized | 13.4% | 7.8% | 1.3% | 1.1% |
| Absolute percent visualization gain | 12.1% | 6.7% |
∗135° angle of view retroscope |
Colonoscopists with mean withdrawal times of <6 minutes detect significantly fewer polyps than those with mean withdrawal times of >6 minutes.6 The 6-minute standard7, 8, 9 is derived from a combination of observational data and expert opinion suggesting that optimal withdrawal examination requires 6 to 10 minutes.10 Barclay et al6 studied a Rockford, Illinois–based group of colonoscopists using 140° angle of view instruments and observed that, in a screened population of 2053 patients, 0.47 any size adenomas per patient were detected. In the same population, 0.075 any size adenomas were detected for every minute of mean withdrawal time (in patients with no adenomas). The relationship was linear between 3 and 16 minutes across all participating colonoscopists, regardless of individual mean withdrawal times. A subsequent investigation by Barclay et al11 with the same group of colonoscopists demonstrated that educational intervention coupled with procedural time cues targeting a 2-minute or more increase in mean withdrawal time significantly increased the observed any size adenoma per patient detection rate to 0.65 within the screened population. The predicted any per patient detection rate was 0.73, as calculated by multiplying the observed postintervention mean withdrawal time of 9.8 minutes (in patients with no adenomas) by the any size adenoma per mean withdrawal minute detection rate of 0.075. The difference between the observed and predicted any per patient detection rates suggests that enforced slowing of withdrawal time does not improve operator-dependent technical variables equally across all colonoscopists.
Colon preparation adequacy primarily impacts detection of polyps of <9 mm in size; detection of large polyps and malignant masses are marginally affected. Approximately 25% of patients undergoing colonoscopy present with inadequate bowel preparation.12, 13 Recently, Lai et al14 observed that assessment of colon preparation by section upon withdrawal and after cleansing maneuvers is of greater relevance to lesion detection rates than preparation assessment on insertion. The thoroughness of cleansing maneuvers performed by the colonoscopist during withdrawal may be an important factor influencing the duration of withdrawal among colonoscopists with high polyp detection rates.
Tip deflection during withdrawal can be considered as a time-dependent mechanical method of increasing the instrument angle of view. Tip deflection is believed to have an important influence on polyp detection rates,10 and it may be another technique factor that contributes to the increased withdrawal times observed among colonoscopists with high polyp detection rates. However, no studies attempting to independently quantify tip deflection with respect to polyp detection rates have been reported.
Waye et al15 in this issue of Gastrointestinal Endoscopy report that the use of the TER improved mean colonoscopy polyp detection rates by 13.2% and mean adenoma detection rates by 11%. Detection gains were greater for larger polyps. The mean withdrawal time during colonoscopy was 10.9 minutes. It is assumed that the reported mean withdrawal time includes polyp removal time because the authors do not specify censoring. For the combination of the colonoscope and TER, the any size adenoma per patient detection rate was 0.61, and the any size adenoma per mean withdrawal minute detection rate was 0.056 across the participating colonoscopists.
The any size adenoma per patient detection rate and the any size adenoma per mean withdrawal minute detection rate represent practical efficacy and efficiency standards for comparing the impact of different colonoscopy procedure modifications on adenoma detection, when applied across similar cohorts of colonoscopists and screened patients. These summary statistics are not easily confounded by the peculiarities of study design. For the Waye et al15 study, the any size adenoma per patient detection rate was 0.55 for the colonoscope alone and 0.61 for the colonoscope/TER combination. These per patient detection rates are similar to those for the Barclay et al studies6, 11 (Table 2) and suggest that an educational intervention and the use of procedural timing cues that produce a reduction in the mean colonoscope withdrawal time by approximately 3.5 minutes will achieve results comparable or superior to those obtained by use of the TER.
Table 2. Adenoma detection rate efficacy and efficiency comparison
| No. | Cecal intubation | Angle of view | Adenoma per patient rate | Uncorrected withdrawal time | Adenoma per minute rate (UWT) | Adenoma per minute rate (corrected) | Good-to-excellent bowel preparation | |
|---|---|---|---|---|---|---|---|---|
| Barclay et al | ||||||||
 Base 2003-2004 | 2053 | 99% | 140° | 0.47 | 7.4 | 0.064 | 0.075∗ | 93% |
| Post 2005-2006 | 2325 | 99% | 140° | 0.64 | 10.9 | 0.059 | 0.065∗ | 89% |
| Waye et al | ||||||||
| Colonoscope | 249 | 100% | 170° | 0.55 | ??† | ??† | ??‡ | |
| Colonoscope + TER | 249 | 100% | 170° + 135° R | 0.61 | 10.9 | 0.056 | ??‡ |
∗Surrogate mean withdrawal time for negative examinations. |
†Study design precluded estimation of actual withdrawal time or surrogate withdrawal time for colonoscope alone. |
‡Quality of bowel preparation not reported. |
The postintervention any size adenoma per mean withdrawal minute detection rate calculated with derived total withdrawal times from the Barclay studies6, 11 are superior to the per mean withdrawal minute detection rate reported by Waye et al15 for the colonoscope/TER combination (Table 2). Although there may be multiple explanations for this finding relating to differences in study cohorts, differences in bowel preparation quality, and withdrawal time calculation, the most obvious explanation is that fiddling with the TER during polyp removal and cleansing maneuvers adds dead time to the withdrawal. This is because the TER must be removed from the instrument accessory channel for polyp removal and, in some cases, for aspiration of fluid.
The 7.8% of colon surface area not visualized after examination with modern 170° angle of view colonoscopes is the last frontier for colonoscopists. However, there is no reason to suspect that the distribution of adenoma sizes should be different for this nonvisualized region, represented primarily by the proximal aspect of folds, than the distribution that exists for the 92.2% of colon surface area now routinely examined. In the Waye et al15 study, 23.53% of all polyps detected with the colonoscope alone were ≥6 mm, and 53.33% of all polyps detected with the TER alone were ≥6 mm. These two distributions should be approximately the same. Either the TER detects too few polyps of <6 mm (a problem that might be solved by improving the optical characteristics of the device), or the cohort studied had an unusually large number of large polyps hidden on the proximal aspect of folds. If the latter is true, then the reported adenoma detection gains of 25.0% and 33.3% for adenomas ≥6 mm and ≥10 mm, respectively, attributed to the TER may be substantially reduced when the device is studied (or used) in patient populations with normal topographic size distributions of adenomas.
It appears that colonoscopists using unassisted 140° angle of view colonoscopes who are willing to spend a few more minutes withdrawing the instrument and pay a bit more attention to procedural technique, particularly tip deflection and cleansing maneuvers, can produce efficacy and efficiency results equivalent or superior to those reported by Waye et al15 for colonoscopists using TER-assisted 140° to 170° angle of view colonoscopes. Therefore, the role of retroscopy remains unclear. Will the retroscope serve only as a crutch for the adenoma detection–challenged colonoscopist? Will it become a niche device useful only in searching for polyps detected or suspected by other means, such as CT colography? Or, will it be widely adapted as an essential adjunct for screening and surveillance colonoscopy? The challenges for the advocates of TER-assisted colonoscopy are to identify situations in which TER-related gains in adenoma detection are present and then to generate data that will convincingly demonstrate that such gains are unique and not easily achieved by simple procedural modifications.
Disclosure
The author disclosed no financial relationships relevant to this publication.
References
- Location of adenomas missed by optical colonoscopy. Ann Intern Med. 2004;141:352–359
- Surface visualization at CT colonography simulated colonoscopy: effect of varying field of view and retrograde view. Am J Gastroenterol. 2007;102:2529–2535
- Impact of wide-angle, high-definition endoscopy in the diagnosis of colorectal neoplasia: a randomized controlled trial. Gastroenterology. 2008;135:1062–1068
- . 170 degrees wide-angle colonoscope: effect on efficiency and miss rates. Am J Gastroenterol. 2004;99:2138–2142
- . Wide angle colonoscopy with a prototype instrument: impact on miss rates and efficiency as determined by back-to-back colonoscopies. Am J Gastroenterol. 2003;98:2000–2005
- Colonoscopic withdrawal times and adenoma detection during screening colonoscopy. N Engl J Med. 2006;355:2533–2541
- Quality in the technical performance of colonoscopy and the continuous quality improvement process for colonoscopy: recommendations of the U.S. Multi-Society Task Force on Colorectal Cancer. Am J Gastroenterol. 2002;97:1296–1308
- Quality indicators for colonoscopy. Gastrointest Endosc. 2006;63:S16–S28
- Quality indicators for colonoscopy. Am J Gastroenterol. 2006;101:873–885
- . Colonoscopic withdrawal technique is associated with adenoma miss rates. Gastrointest Endosc. 2000;51:33–36
- . Effect of a time-dependent colonoscopic withdrawal protocol on adenoma detection during screening colonoscopy. Clin Gastroenterol Hepatol. 2008;6:1091–1098
- Impact of colonic cleansing on quality and diagnostic yield of colonoscopy: the European Panel of Appropriateness of Gastrointestinal Endoscopy European multicenter study. Gastrointest Endosc. 2005;61:378–384
- . Impact of colonoscopy preparation quality on detection of suspected colonic neoplasia. Gastrointest Endosc. 2003;58:76–79
- The Boston bowel preparation scale: a valid and reliable instrument for colonoscopy-oriented research. Gastrointest Endosc. 2009;69:620–625
- A retrograde-viewing device improves detection of adenomas in the colon: a prospective efficacy evaluation (with videos). Gastrointest Endosc. 2010;551–556
PII: S0016-5107(09)02737-0
doi:10.1016/j.gie.2009.11.013
© 2010 American Society for Gastrointestinal Endoscopy. Published by Elsevier Inc. All rights reserved.
