impact of fair bowel preparation quality on adenoma and serrated polyp detection: data from the new...

ORIGINAL ARTICLE

CRC,SDR,

DISCNatioauthNatioother

Copy0016http:

Rece

www

Impact of fair bowel preparation quality on adenoma and serratedpolyp detection: data from the New Hampshire ColonoscopyRegistry by using a standardized preparation-quality rating

coloreserrate

LOSURnal Caors annal Cfinan

right ª-5107/$//dx.do

ived Ju

.giejo

Joseph C. Anderson, MD,1,2 Lynn F. Butterly, MD,2,3 Christina M. Robinson, MS,4 Martha Goodrich, MS,4

Julia E. Weiss, MS4

White River Junction, Vermont, USA

Background: The effect of colon preparation quality on adenoma detection rates (ADRs) is unclear, partlybecause of lack of uniform colon preparation ratings in prior studies. The New Hampshire Colonoscopy Registrycollects detailed data from colonoscopies statewide, by using a uniform preparation quality scale after the endo-scopist has cleaned the mucosa.

Objective: To compare the overall and proximal ADR and serrated polyp detection rates (SDR) in colonoscopieswith differing levels of colon preparation quality.

Design: Cross-sectional.

Setting: New Hampshire statewide registry.

Patients: Patients undergoing colonoscopy.

Interventions: We examined colon preparation quality for 13,022 colonoscopies, graded by using specific de-scriptions provided to endoscopists. ADR and SDR are the number of colonoscopies with at least 1 adenomaor serrated polyp (excluding those in the rectum and/or sigmoid colon) detected divided by the total numberof colonoscopies, for the preparation categories: optimal (excellent and/or good), fair, and poor.

Main Outcome Measurements: Overall/proximal ADR/SDR.

Results: The overall detection rates in examinations with fair colon preparation quality (SDR 8.9%; 95% confi-dence interval [CI], 7.4-10.7, ADR 27.1%; 95% CI, 24.6-30.0) were similar to rates observed in colonoscopieswith optimal preparation quality (SDR 8.8%; 95% CI, 8.3-9.4, ADR 26.3%; 95% CI, 25.6-27.2). This finding alsowas observed for rates in the proximal colon. A logistic regression model (including withdrawal time) foundthat proximal ADR was statistically lower in the poor preparation category (odds ratio 0.45; 95% CI, 0.24-0.84;P! .01) than in adequately prepared colons.

Limitations: Homogeneous population.

Conclusion: In our sample, there was no significant difference in overall or proximal ADR or SDR betweencolonoscopies with fair versus optimal colon preparation quality. Poor colon preparation quality may reducethe proximal ADR. (Gastrointest Endosc 2014;-:1-8.)

Abbreviations: ADR, adenoma detection rate; BMI, body mass index; Current affiliations: Department of Veterans Affairs Medical Center, White
ctal cancer; NHCR, New Hampshire Colonoscopy Registry;d polyp detection rate.
E: The project was supported by grant R01CA131141 from thencer Institute. The content is solely the responsibility of thed does not necessarily represent the official views of theancer Institute or the National Institutes of Health. Nocial relationships relevant to this article were disclosed.

2014 by the American Society for Gastrointestinal Endoscopy36.00i.org/10.1016/j.gie.2014.03.021

ne 20, 2013. Accepted March 15, 2014.

River Junction, Vermont (1), The Geisel School of Medicine at Dartmouth,Hanover, New Hampshire (2), Section of Gastroenterology, DartmouthHitchcock Medical Center, Lebanon, New Hampshire (3), Department ofCommunity and Family Medicine, The Geisel School of Medicine atDartmouth, Hanover, New Hampshire (4), USA.

Reprint requests: Joseph C Anderson, MD, Department of Veterans AffairsMedical Center, 215 North Main Street, White River Junction, VT 05009.

If you would like to chat with an author of this article, you may contactDr Anderson at [email protected].

urnal.org Volume -, No. - : 2014 GASTROINTESTINAL ENDOSCOPY 1

Delta:1_given name

Delta:1_surname

Delta:1_given name

Delta:1_surname

Delta:1_given name

http://dx.doi.org/10.1016/j.gie.2014.03.021

mailto:[email protected]

http://www.giejournal.org

Impact of fair bowel preparation on adenoma and serrated polyp detection Anderson et al

Colonoscopy is currently the most widely used screeningtest for colorectal cancer (CRC) prevention and early detec-tion in the United States and is a critical part of recom-mended screening guidelines.1,2 Prevention of CRC isaccomplished through removal of potentially precancerouspolyps, both adenomas and the more recently describedsessile serrated polyps, before those lesions can progressto CRC. Patients are instructed to prepare for colonoscopyby drinking colon-cleansing fluids and restricting their dietsfor 24 hours before the procedure. Variable compliancewith these instructions results in patients arriving for colo-noscopy with colons in varying stages of preparation,ranging from excellent to poor. It seems reasonable toexpect that detection of precancerous lesions during colo-noscopy could be affected by the quality of the colonpreparation.

However, little is known about outcomes based on thequality of colonoscopy preparation. For example, are morelesions detected in colonoscopies with optimal (excellentor good) preparation quality, or does suboptimal colonpreparation differentially affect findings in the right orleft side of the colon? A few studies have suggested that pa-tients with suboptimal preparations may have a high rate ofmissed advanced adenomas.3,4 However, lack of standard-ization for grading the quality of preparation has hinderedinvestigation of the impact of suboptimal preparation.5 Forexample, one study found similar adenoma detection rates(ADRs) in examinations with fair, good, and excellentbowel preparation, but there was no standardization inpreparation quality6 or in whether the preparation wasgraded before or after clearing of the colon. Another chal-lenge has been the lack of information regarding relatedvariables such as withdrawal time in studies examining co-lon preparation quality.6 As a result, there are no clear rec-ommendations regarding whether follow-up screening orsurveillance intervals should be modified for examinationswith suboptimal colon preparation. However, in practice,subsequent surveillance intervals are frequently shortenedfor patients with suboptimal colon preparation in orderto address the greater potential for missed lesions than ex-ists for patients with optimal (good or excellent) colonpreparation.7

Inadequate or suboptimal colon preparation in the rightside of the colon may partly explain the lack of protectionfrom advanced neoplasia in the proximal versus the distalsections of the colon provided by colonoscopy.8,9 It is un-clear whether suboptimal colon preparation may dispro-portionately affect detection of serrated as opposed toadenomatous lesions. This may be especially true becausesessile serrated adenomas, the more worrisome subset ofthese lesions, are often flat and proximally located.10 Thesefactors may play a role in the finding that interval cancersare more likely to be located proximally.11 Clarification ofthe impact of suboptimal colon preparation by location,incorporating patient risk factors, will allow more specificand targeted responses to the persistent question of

2 GASTROINTESTINAL ENDOSCOPY Volume -, No. - : 2014

Take-home Message

� By using a standardized preparation quality rating, weobserved that patients with fair colon preparation qualityhad proximal adenoma and serrated polyp detectionrates similar to those with excellent or good preparationquality.

when to repeat tests for which the preparation was subop-timal (neither good nor excellent).

The New Hampshire Colonoscopy Registry (NHCR) is apopulation-based, statewide registry that collects compre-hensive patient, procedure, and pathology information.Endoscopists complete a procedure form that provides adetailed description for each category of colon preparationquality and instructs endoscopists to grade the preparationaccording to the worst-prepared segment after clearing,providing consistent terminology among the diverse groupof participating endoscopists. The NHCR assesses colonos-copy quality measures, including ADR and serrated polyp(subset that does not include those in the rectum or sig-moid) detection rate (SDR). Our aim in this analysis wasto examine the overall as well as the proximal ADR andSDR for colonoscopies performed in patients with varyinglevels of colon preparation quality, particularly to comparedetection rates in procedures with fair and optimal prepa-ration quality, which previously have been reported to besimilar.6

METHODS

The design and development of the NHCR is describedin detail elsewhere.12-14 Nearly all endoscopy sites in NewHampshire currently contribute data to the NHCR, with afew sites currently undergoing human subjects reviewand implementation. The NHCR is a registry used togenerate evidence for multiple studies; therefore, thereare no specific criteria for endoscopists in New Hampshireto participate in the registry. Consenting patients completea self-administered patient questionnaire before colonos-copy, providing information on demographic characteris-tics, health history, and risk factors for CRC. On theNHCR procedure form, completed during or immediatelyafter colonoscopy, endoscopists or endoscopy nurses re-cord the indication for the colonoscopy (specific optionswithin screening, surveillance, or diagnostic categories),findings (location, size, and specific treatment, if any, ofpolyps, cancer, or other findings), quality of colon prepara-tion, sedation medication, colon region reached during theprocedure, withdrawal time, follow-up recommendations,and immediate adverse events. The NHCR requests pathol-ogy reports for all colonoscopies, with findings directlyfrom the pathology laboratory used by each participatingendoscopy facility. Trained NHCR staff abstract data from

www.giejournal.org


Anderson et al Impact of fair bowel preparation on adenoma and serrated polyp detection

these pathology reports, including location, size, and his-tology of all findings, and enter it into the NHCR database,linking it to information from the procedure form at thepolyp level.14 All data collection and study procedureswere approved by the Committee for the Protection ofHuman Subjects at Dartmouth College as well as by otherrelevant human subjects reviewing bodies at participatingsites.

CohortDuring the time period used for this analysis (April 6,

2009-March 22, 2011), 12 endoscopy facilities includinghospitals, ambulatory surgery centers, and communitypractices across New Hampshire were participating in theNHCR, and 16,574 colonoscopies were performed by 54endoscopists at these 12 facilities. Across these 12 facilities,pathology reports were received for a median of 92% ofcolonoscopies with polyps detected. Incomplete colonos-copies, those in which colon preparation was not indi-cated, evaluation of inflammatory bowel disease, andcolonoscopies in patients with familial syndromes (familialpolyposis or hereditary nonpolyposis colon cancer) notedon the procedure form were excluded. Our analysisincluded data from patients aged R40 years who hadeither a screening or surveillance (personal history ofcolorectal cancer or adenomatous polyps) colonoscopyperformed. Because colon preparation quality affects colo-noscopies of all indications, we also included examinationsin patients with diagnostic colonoscopies for indicationssuch as GI bleeding, anemia, and change in bowel habits.After excluding 3552 colonoscopies, 13,022 colonoscopiesfor 12,948 patients remained for this analysis.

Exposure measureThe NHCR procedure form provides the 4 options for

examination preparation quality (excellent, good, fair, orpoor) that were the basis of the exposure measure inthis analysis. On the procedure form, endoscopists wereinstructed to score quality of colon preparation, based onthe worst prepared segment of the colon, after clearingof the mucosa. The recommended colon preparation scoreoptions were created by NHCR staff after a review of themedical literature.15 We developed our preparation scoringsystem approximately 10 years ago, independently of theBoston16 or Ottawa preparation scores,17 and have beenusing it since that time. This preparation rating is describedin detail on every endoscopy form: (1) excellent, only scat-tered, tiny particles and/or clear liquidd100% visualizationpossible throughout colon; (2) good, easily removablesmall amounts of particles and/or liquiddvery unlikely toimpair visualization throughout colon; (3) fair, residual fecesand/or non-transparent fluiddpossibly impairing visualiza-tion; (4) poor, feces and/or non-transparent fluidddefinitely impairing visualization.

For this analysis, colon preparation was defined asoptimal if the endoscopist selected excellent or good.

www.giejournal.org

Because the registry was designed to measure real-worldpractice, outcomes were assessed based on the endoscop-ists’ usual practices. Therefore, we offered no guidance toendoscopists regarding when to terminate clearing efforts.

CofactorsPatient demographics, including age, sex, smoking sta-

tus, family history of CRC, and body mass index (BMI) aswell as the colonoscopy characteristics of indication for ex-amination (screening, surveillance, or diagnostic) and with-drawal time of the examination (dichotomized above andbelow 6 minutes) were examined in relation to colon prep-aration quality and outcome measures.

Outcome measuresWe calculated the overall colon ADR (the number of co-

lonoscopies with at least one adenoma detected [includingtubular or villous adenomas and adenomas with high-gradedysplasia or adenocarcinoma], divided by the total numberof colonoscopies), and 95% confidence interval (CI) foreach colon preparation category.

We calculated the proximal ADR in a similar manner tothat for the overall rate. Our primary definition of proximalcolon included the cecum, ascending colon, and thehepatic flexure. We also included the transverse colon ina secondary analysis to examine the potential effect ofdefining proximal lesions as those proximal to the splenicflexure and to account for potential misclassification of he-patic flexure polyps.

We defined serrated polyps to include sessile serratedpolyps and serrated adenomas as well as hyperplasticpolyps, as has been done in previous studies.18 Our sampleexcluded those hyperplastic polyps in the rectum or sig-moid colon. Although hyperplastic polyps are traditionallyconsidered benign lesions, their role in the serratedpathway to CRC remains unclear, and both interobserverreliability in histologic diagnosis and the logistic challengesof re-examining over 1000 histology slides argued for theirinclusion within our definition of serrated polyps for thisanalysis.19,20 Our calculation of SDR was similar to thatfor ADRdwe divided the number of colonoscopies withat least one serrated polyp by the total number of colonos-copies for each colon preparation category by using the 2definitions described before. In addition, we calculated theproximal SDR.

Statistical analysesADR, SDR, and 95% CI were computed by using a

nonparametric statistical method for proportions,21 andthe Fisher exact test was used to compare the rates. Uni-variate analyses were performed for all cofactors acrossthe 3 categories of colon preparation quality. The ADRand SDR odds ratios (OR) and 95% CIs are presented forthe colon preparation results from a logistic regressionmodel adjusted for all cofactors. Additionally, this modelincorporated clustering on the endoscopist because

Volume -, No. - : 2014 GASTROINTESTINAL ENDOSCOPY 3



patients were nested within endoscopist, to give more con-servative estimates of the standard errors. All analyses wereconducted in SAS 9.3 (SAS Institute Inc., Cary, NC), and aP value! .05 was considered significant.

RESULTS

Our cohort had a mean (� standard deviation) age of59 years (� 9 years), and was 53% female and 94% white,with 22% self-reporting a first-degree relative with CRC. Ofthe 13,022 colonoscopies, 11,620 (89%) were judged tohave an optimal preparation, 1201 (9%) had a fair prepara-tion, and 201 (2%) had a poor preparation. Participatingendoscopists had a median of 14 years of experience (in-terquartile range 7-21 years) and had performed a meannumber of 241 examinations in this dataset. Less than7% (n Z 894) of the examinations were performed by en-doscopists with !100 examinations in this dataset. Thenumber of examinations performed by each endoscopistthat are included in this analysis does not reflect theirentire colonoscopy caseload during this time period,because of exclusion criteria and the fact that not all endo-scopists were participating in the NHCR for the full timeperiod. With regard to variation in preparation, all endo-scopists had mean colon preparation scores of eithergood or excellent. Thus, the mean score was in theoptimal range for all endoscopists.

Table 1 presents ADR and SDR in the entire and prox-imal colon, calculated for all colonoscopies and by colonpreparation quality. There was a trend toward statistical sig-nificance in the difference in proximal ADR between theoptimal, fair, and poor colon preparation quality groups.For serrated polyps, there was a decrease in the proximalSDR across the 3 groups, with poor preparation havingthe lowest rate, but this was not statistically significant.We performed the same analysis, adding polyps from thetransverse colon to those from the original definition ofproximal colon, and observed similar nonsignificant resultsfor both ADR: optimal 17.3% (95% CI, 16.6-18.0); fair 18.2%(95% CI, 16.0-20.5); and poor 13.4% (95% CI, 9.0-18.9)(P Z .26; Fisher exact test) and SDR: optimal 7.0% (95%CI, 6.5-7.5); fair 6.5% (95% CI, 5.2-8.0); and poor 4.5(95% CI, 2.1-8.3) (P Z .34; Fisher exact test).

In addition, we conducted a subset analysis in surveil-lance and screening examinations (N Z 10,310), droppingdiagnostic examinations, and observed similar results. Spe-cifically, in this subset analysis, of 10,310 examinations(optimal Z 9263 examinations, 89.8%; fair Z 899 examina-tions, 8.7%; poor Z 148 examinations, 1.4%) the observedproximal ADRs for each group of colon preparation qualitywere as follows: optimal 13.9% (95% CI, 13.2-14.6); fair13.3% (95% CI, 11.2-15.7); and poor 10.1% (95% CI, 5.8-16.2) (P Z .42; Fisher exact test). Proximal SDRs were asfollows: optimal 5.0% (95% CI, 4.5-5.4); fair 4.1% (95%CI, 2.9-5.6); and poor 4.7 (95% CI, 1.9-9.5) (P Z .57; Fisher


exact test). We also performed subanalyses examining theADR and SDR for screening and surveillance examinationsseparately.22 These data are shown in Table 2.

In another subanalysis, we examined the ADR by the4 preparation categories, with the following results: excel-lent preparation 23.8 (95% CI, 22.5-25.1); good 28.0 (95%CI, 26.9-29.0); fair 27.1 (95% CI, 24.6-30.0); and poor20.9 (95% CI, 15.5-27.2). A similar analysis was done forSDR, with the following results: excellent preparationSDR 8.1 (95% CI, 8.3-8.9); good 9.3 (95% CI, 8.6-10.0);fair 8.9 (95% CI, 7.4-10.7); and poor 7.5 (95% CI, 4.2-12.0).

Table 3 shows the clinically important characteristics ofthe 3 preparation groups. Patients with poor preparationswere more likely to be younger, male, obese, and currentsmokers than patients in the optimal or fair groups. Alarger proportion of colonoscopies in the poor colon prep-aration group were performed for a diagnostic indicationas compared with the other preparation groups. Therewere no differences in family history of CRC or withdrawaltime between the 3 preparation groups. The examinationswith missing withdrawal times had a similar bowel prepara-tion quality distribution to those examinations for whichwithdrawal times were collected.

A multivariable logistic regression analysis, adjusting forage, sex, BMI, smoking, family history of CRC, indicationfor examination, withdrawal time, and endoscopist (Table 4),showed that the proximal ADR for colonoscopies in thepoor colon preparation group was lower than the ADR forthe optimal or fair preparation groups (OR 0.45; 95% CI,0.24-0.84). This difference was not observed for proximalSDR (OR 0.75; 95% CI, 0.31-1.80). When we included thetransverse colon, we obtained similar results in the poorpreparation group for both ADR (OR 0.48; 95% CI, 0.33-0.73) and SDR (OR 0.70; 95% CI, 0.37-1.35).

The proximal SDR varied among the 54 endoscopists,with a median of 4.5% and interquartile range of 2.4% to7.5%. To determine whether endoscopists with higherADRs gave lower scores for colon preparation quality (ashas been reported previously),23 we examined the correla-tion between the endoscopists’ ADRs and the percentageof colonoscopy preparations that they rated as optimal.We found no correlation between the endoscopists’mean proximal ADRs and mean quality of reported prepa-ration (Spearman correlation coefficient -0.09; P Z .50).

DISCUSSION

In data collected for the NHCR, we observed that theoverall ADR in colonoscopies with fair colon preparationwas similar to the ADR observed in examinations with anoptimal preparation (good or excellent). This relationshipalso was observed for the proximal ADR. As expected,the proximal ADR was lower in poor preparation groupsthan in the optimal or fair preparation groups. With regardto serrated polyps, we observed a decrease in the proximal

www.giejournal.org


TABLE 2. ADR and SDR for screening versus surveillance examinations

Rate % (95% CI) Total Excellent/good Fair Poor

ADR in screening cohort 24.1 (23.1-25.1) 24.1 (23.0-25.2) 24.0 (20.5-27.7) 24.5 (16.4-34.2)

ADR in surveillance cohort 36.4 (34.7-38.0) 36.5 (34.7-38.2) 36.6 (31.4-42.0) 28.0 (16.2-42.5)

SDR in screening cohort 8.4 (7.7-9.1) 8.4 (7.8-9.1) 8.2 (6.0-10.7) 6.1 (2.3-12.9)

SDR in surveillance cohort 10.7 (9.7-11.8) 10.6 (9.6-11.8) 10.4 (7.4-14.2) 8.0 (7.2-29.1)

ADR, Adenoma detection rate; SDR, serrated polyp detection rate; CI, confidence interval.

TABLE 1. Colon adenoma detection rate and serrated polyp detection rate by colon preparation quality

Rate, % (95% CI)Total

N [ 13,022

Optimal (excellent/good)N [ 11,620(89.5%)

FairN [ 1201

(9%)

PoorN [ 201(1.5%) P value*

ADR in entire colon 26.3 (25.6-27.1) 26.3 (25.6-27.2) 27.1 (24.6-30.0) 20.9 (15.5-27.2) .18

SDR in entire colon 8.8 (8.3-9.3) 8.8 (8.3-9.4) 8.9 (7.4-10.7) 7.5 (4.2-12.0) .84

ADR in proximal colon 12.8 (12.2-13.4) 12.9 (12.3-13.5) 12.4 (10.6-14.4) 8.0 (4.6-12.6) .10

SDR in proximal colon 4.7 (4.4-5.1) 4.8 (4.5-5.3) 3.9 (2.9-5.2) 3.5 (1.47.0) .24

CI, Confidence interval; ADR, adenoma detection rate; SDR, serrated polyp detection rate.*Fisher exact test.


SDR across the 3 preparation quality groups from optimalto poor, but this difference was not statistically significant.

CRC prevention with colonoscopy relies on the ability todetect and resect potentially precancerous polyps beforefurther development to adenocarcinoma. The ADR is aquality indicator that has been shown to be inversely asso-ciated with the interval cancer rates for endoscopists24;patients of endoscopists with lower ADRs have higher ratesof interval CRC. Thus, adequate detection of adenomas is acrucial component of colonoscopy practice.

An optimal colon preparation is vital to ensure completemucosal inspection for adenomas. An analysis of ClinicalOutcomes Research Initiative data demonstrated that anadequate colon preparation was achieved in approximatelythree fourths of all examinations.15 Patients with subopti-mal colon preparations have been observed to have highrates of missed advanced neoplasia; one recent studyobserved that patients with poor or fair colon preparationshad a miss rate of 27% for advanced lesions.3 Anotherstudy observed that a better colon preparation was corre-lated with a higher rate of detection for adenomas butnot serrated polyps.25 A recent study of Veterans Affairspatients observed that examinations in patients with fairpreparations had rates of adenomas and advanced lesionssimilar to those with optimal preparations.6 Although thisstudy provides intriguing data, it is limited by its inabilityto standardize preparation scores or to include withdrawal

www.giejournal.org

times as well as its reliance on data from a single site. Ourstudy, which incorporated data from 12 practices that usedstandardized preparation quality scores and factored inwithdrawal time, substantiated the findings of the VeteransAffairs study, finding similar detection rates in examina-tions with fair and optimal bowel preparation quality.

Poor colon preparation quality has been implicated inthe development of interval cancers, which are often prox-imal in location.11 The serrated pathway has been shown toshare molecular abnormalities with interval cancers,26 sug-gesting that detection of serrated lesions may be essentialto preventing interval cancers. A recent study found thatcolonoscopy conferred a statistically significant reductionin the future risk of proximal advanced adenomas butnot proximal sessile serrated polyps.27 It is unclear whetherinadequate colon preparation in the proximal colon mighthave played a role in this finding. To date, there have beenlittle data in the medical literature regarding ADR andSDR in the proximal colon of patients with inadequate orsuboptimal colon preparations, leading to our currentinvestigation.

We observed that there was a slight decrease in theproximal SDR as colon preparation quality moved fromoptimal to poor, but this trend was not statistically signifi-cant. de Wijkerslooth et al25 recently demonstrated no rela-tionship between detection of proximal serrated lesionsand colon preparation quality as graded by the Ottawa



TABLE 3. Characteristics and quality of colon preparation

Characteristics,* no. (%)

Quality of colon preparation

Optimal (excellent/good)*N [ 11,554

(89%)

FairN [ 1194

(9%)

PoorN [ 200(2%)

TotalN [ 12,948

(100%) P valueyAge, y .05

40-64 8499 (74) 840 (70) 150 (75) 9489 (73)

R65 3055 (26) 354 (30) 50 (25) 3459 (27)

Sex ! .001

Male 5122 (47) 585 (52) 103 (54) 5810 (47)

Female 5880 (53) 550 (48) 86 (46) 6516 (53)

BMI ! .0001

!25 (underweight/normal) 3127 (29) 286 (26) 35 (19) 3448 (28)

R25 to!30 (overweight) 4177 (38) 389 (35) 66 (36) 4632 (38)

R30 to!35 (obesity class I) 2287 (21) 252 (22) 45 (25) 2583 (21)

R35 (obesity classes II/III) 1333 (12) 194 (17) 37 (20) 1564 (13)

Smoking status ! .0001

Never 5491 (49) 465 (41) 68 (36) 6024 (48)

Past 4677 (42) 510 (44) 91 (48) 5278 (42)

Current 1079 (10) 173 (15) 30 (16) 1282 (10)

Family history of CRC .94

No 8968 (78) 920 (77) 155 (78) 10043 (78)

Yes 2568 (22) 270 (23) 44 (22) 2882 (22)

Indication for examination ! .001

Screening 6310 (56) 563 (50) 97 (50) 6970 (55)

Surveillance 2906 (26) 329 (29) 50 (26) 3285 (26)

Diagnostic 2024 (18) 235 (21) 47 (24) 2306 (18)

Withdrawal time, minz .47

%6 2136 (21) 219 (21) 41 (25) 2396 (21)

O6 8054 (79) 834 (79) 124 (75) 9012 (79)

BMI, Body mass index; CRC, colorectal cancer.*Missing (no., %): sex (622, 5), BMI (721, 6), smoking status (364, 3), family history of CRC (23,!1), indication for examination (387, 3).yChi-square test compares quality of bowel preparation categories for each characteristic.zWithdrawal time (no., %): (1540, 12). The examinations with missing withdrawal times had a bowel preparation quality distribution similar to thoseexaminations for which withdrawal times were collected.


Bowel Preparation Scale score.17 Given the flat nature ofproximal serrated polyps, the authors of this study weresurprised at this finding and postulated that perhapspoorer colon preparations may allow serrated polyps tobe detected more easily because of residual stool attachingto the mucous cap.

We were equally surprised that the difference in SDRsamong the 3 levels of preparation quality was not morepronounced in our study. It is possible, as de Wijkerslooth


et al25 suggested, that serrated polyps are made morevisible when residual stool is attached to their mucuscaps. Perhaps in the colons with poorer preparation, theadditional washing helped to focus the attention of the en-doscopist more closely to the mucosa, thus allowing forhigher detection of serrated polyps. Because of the lowerincidence of proximal serrated polyps, it is possible thatour study may have been underpowered with regard tothese lesions as compared with adenomas. It is unlikely

www.giejournal.org


TABLE 4. Logistic regression model* results for colonpreparation quality for proximal colon ADR and SDR

Colon preparationProximal ADROR (95% CI)

Proximal SDROR (95% CI)

Optimal(excellent/good)

Reference (1.0) Reference (1.0)

Fair 0.89 (0.71-1.14) 0.82 (0.58-1.15)

Poor 0.45 (0.24-0.84) 0.75 (0.31-1.80)

ADR, Adenoma detection rate; SDR, serrated polyp detection rate;OR, odds ratio; CI, confidence interval.*Adjusted for age, sex, body mass index, smoking, family history ofcolorectal cancer, indication for examination, withdrawal time, andendoscopist.


that the lack of a difference in SDR by colon preparationquality can be explained by anatomic location becausethe overall SDR for the entire colon had similar results.Although we cannot readily explain why suboptimal colonpreparation quality does not affect SDR at this time, westrongly agree with quality standards that assert that anadequate colon preparation is an important quality mea-sure for screening colonoscopy, especially in the proximalcolon.2

There are a few significant differences between our cur-rent study and that of de Wijkerslooth et al.25 Our analysiswas designed to examine the impact of colon preparation,whereas the other study examined withdrawal time.Furthermore, de Wijkerslooth et al examined the resultsof only 5 endoscopists, whereas our analysis includeddata from 54 endoscopists. Because SDR may vary widelyamong endoscopists,18 our results may provide more reli-able data. It has been suggested that endoscopists withhigher ADRs may rate the quality of colon preparationslower than do endoscopists with lower ADRs.23 We wereable to examine this potential confounding factor amongthe 54 endoscopists in this analysis, and we found no cor-relation between endoscopist’s ADRs and quality of re-ported colon preparation.

A limitation of our study is the potential lack of general-izability because of the limited racial diversity of the popu-lation of New Hampshire. In addition, we do not knowthe salvage efforts of the participating endoscopists whoachieved high rates of optimal colon preparations (89%).The ratings of the preparation before irrigation andclearing and the preparation ratings for individual colonsegments are unknown. This may be problematic if onlythe proximal colon, for example, had a fair-quality prepara-tion, whereas the rest of the bowel had a good-qualitypreparation. Thus, in our sample, the endoscopist wouldhave rated the overall preparation as fair (the preparationquality in the worst prepared segment) despite theremainder of the colon having a good quality preparation.It is also possible that the proximal colon may have been

www.giejournal.org

flooded in an effort to preferentially clean this particularsegment. Estimations of these efforts to clear the colonmucosa of residual stool and yield additional adenomasand serrated polyps during these salvage efforts or “pre-irrigation preparation ratings” might have provided morevaluable information than the absolute percentages ofADR and SDR in the various categories of colon prepara-tion. Although we use a uniform preparation rating, wedo not have data such as videos and do not examine avail-able photographs to provide feedback to the endoscopistsor perhaps to validate the rating. Thus, we may be limitedin assessing concordance of colon preparation among en-doscopists or whether some had veered away from thewritten definitions during the course of the study. Wealso do not provide data regarding the type of preparationused at different sites. This would allow us to determinethe optimal preparation for maximizing proximal polypdetection. In addition, we do not have data regarding thespecific morphology (flat vs sessile) or type (microvesicu-lar vs goblet cell) of hyperplastic polyps. Finally, weacknowledge the heterogeneity of pathologists’ interpreta-tions of serrated polyps at the different pathology labora-tories across New Hampshire.

Strengths of this study include the large numbers ofcolonoscopies and endoscopists. In addition, the NHCRcollects colonoscopy characteristics such as withdrawaltime, and information regarding CRC risk factors includingBMI, smoking, and family history of CRC, all of which havebeen shown to be important predictors of adenomas andserrated polyps.10,28,29 The linked pathology data, availablefor over 90% of examinations with findings, are a uniquestrength of the NHCR’s comprehensive data collection pro-tocol. A particular strength was the method for collectingdata regarding the quality of colon preparation. Endoscop-ists were provided with clear descriptions of each categoryof colon preparation quality on the endoscopy form andwere instructed to provide a preparation rating after theyhad cleaned the colon. Because preparation scored as fairin studies lacking this standardization might actually begood or excellent when the endoscopist finished clearingthe colon, the standardization of our preparation datastrengthens the validity of our findings. Although not vali-dated, our rating system enables consistency by using actualdescriptions, rather than just terminology, to describe whatwe mean by each term and because the same endoscopistshave been using those terms for almost 10 years.

Our goal was to measure and compare the overall andproximal ADR and SDR between colonoscopies with vary-ing colon preparation quality in the statewide, population-based NHCR. We observed no significant differencebetween fair and optimal preparations with regard to prox-imal ADR or SDR. The decision to recommend similar sur-veillance intervals for fair and optimal preparations shouldbe based on further research, including studies thatexamine the longitudinal risk of advanced neoplasia inpatients with varying qualities of colon preparation.




REFERENCES

1. Levin B, Lieberman DA, McFarland B, et al. Screening and surveillancefor the early detection of colorectal cancer and adenomatous polyps,2008: a joint guideline from the American Cancer Society, the US Multi-Society Task Force on Colorectal Cancer, and the American College ofRadiology. Gastroenterology 2008;134:1570-95.

2. Rex DK, Johnson DA, Anderson JC, et al. American College of Gastro-enterology guidelines for colorectal cancer screening 2009 [corrected].Am J Gastroenterol 2009;104:739-50.

3. Lebwohl B, Kastrinos F, Glick M, et al. The impact of suboptimal bowelpreparation on adenoma miss rates and the factors associated withearly repeat colonoscopy. Gastrointest Endosc 2011;73:1207-14.

4. Chokshi RV, Hovis CE, Hollander T, et al. Prevalence of missed ade-nomas in patients with inadequate bowel preparation on screeningcolonoscopy. Gastrointest Endosc 2012;75:1197-203.

5. Menees SB, Kim HM, Elliott EE, et al. The impact of fair colonoscopypreparation on colonoscopy use and adenoma miss rates in patientsundergoing outpatient colonoscopy. Gastrointest Endosc. Epub 2013May 2.

6. Sherer EA, Imler TD, Imperiale TF. The effect of colonoscopy prepara-tion quality on adenoma detection rates. Gastrointest Endosc2012;75:545-53.

7. Ben-Horin S, Bar-Meir S, Avidan B. The impact of colon cleanlinessassessment on endoscopists’ recommendations for follow-up colonos-copy. Am J Gastroenterol 2007;102:2680-5.

8. Baxter NN, Goldwasser MA, Paszat LF, et al. Association of colonoscopyand death from colorectal cancer. Ann Intern Med 2009;150:1-8.

9. Brenner H, Hoffmeister M, Arndt V, et al. Protection from right- andleft-sided colorectal neoplasms after colonoscopy: population-basedstudy. J Natl Cancer Inst 2010;102:89-95.

10. Anderson JC, Rangasamy P, Rustagi T, et al. Risk factors for sessileserrated adenomas. J Clin Gastroenterol 2011;45:694-9.

11. Bressler B, Paszat LF, Chen Z, et al. Rates of new or missed colorectalcancers after colonoscopy and their risk factors: a population-basedanalysis. Gastroenterology 2007;132:96-102.

12. Weaver DL, Rosenberg RD, Barlow WE, et al. Pathologic findings fromthe Breast Cancer Surveillance Consortium: population-based out-comes in women undergoing biopsy after screening mammography.Cancer 2006;106:732-42.

13. Butterly LF, Goodrich M, Onega T, et al. Improving the quality of colo-rectal cancer screening: assessment of familial risk. Dig Dis Sci 2010;55:754-60.

14. Greene MA, Butterly LF, Goodrich M, et al. Matching colonoscopy andpathology data in population-based registries: development of a novel


algorithm and the initial experience of the New Hampshire Colonos-copy Registry. Gastrointest Endosc 2011;74:334-40.

15. Harewood GC, Sharma VK, de Garmo P. Impact of colonoscopy prepa-ration quality on detection of suspected colonic neoplasia. Gastroint-est Endosc 2003;58:76-9.

16. Lai EJ, Calderwood AH, Doros G, et al. The Boston Bowel PreparationScale: a valid and reliable instrument for colonoscopy-orientedresearch. Gastrointest Endosc 2009;69:620-5.

17. Rostom A, Jolicoeur E. Validation of a new scale for the assessment ofbowel preparation quality. Gastrointest Endosc 2004;59:482-6.

18. Kahi CJ, Hewett DG, Norton DL, et al. Prevalence and variable detectionof proximal colon serrated polyps during screening colonoscopy. ClinGastroenterol Hepatol 2011;9:42-6.

19. Singh H, Bay D, Ip S, et al. Pathological reassessment of hyperplasticcolon polyps in a city-wide pathology practice: implications for polypsurveillance recommendations. Gastrointest Endosc 2012;76:1003-8.

20. Anderson JC, Robertson DJ. Serrated polyps misclassified: the value oflooking back and the impact of looking forward. Gastroenterology2013;144:1564-6; discussion 1566.

21. Conover WJ. Practical nonparametric statistics. 2nd ed. New York: JohnWiley & Sons, Inc; 1980.

22. Anderson JC, Butterly LF, Goodrich M, et al. Differences in detectionrates of adenomas and serrated polyps in screening versus surveil-lance colonoscopies, based on the New Hampshire Colonoscopy Reg-istry. Clin Gastroenterol Hepatol 2013;11:1308-12.

23. Thomas-Gibson S, Rogers P, Cooper S, et al. Judgement of the qualityof bowel preparation at screening flexible sigmoidoscopy is associatedwith variability in adenoma detection rates. Endoscopy 2006;38:456-60.

24. Kaminski MF, Regula J, Kraszewska E, et al. Quality indicators for colo-noscopy and the risk of interval cancer. N Engl J Med 2010;362:1795-803.

25. de Wijkerslooth TR, Stoop EM, Bossuyt PM, et al. Differences in proximalserrated polyp detection among endoscopists are associated with vari-ability in withdrawal time. Gastrointest Endosc 2013;77:617-23.

26. Arain MA, Sawhney M, Sheikh S, et al. CIMP status of interval coloncancers: another piece to the puzzle. Am J Gastroenterol 2010;105:1189-95.

27. Burnett-Hartman AN, Newcomb PA, Phipps AI, et al. Colorectal endos-copy, advanced adenomas, and sessile serrated polyps: implicationsfor proximal colon cancer. Am J Gastroenterol 2012;107:1213-9.

28. Anderson JC, Attam R, Alpern Z, et al. Prevalence of colorectalneoplasia in smokers. Am J Gastroenterol 2003;98:2777-83.

29. Anderson JC, Messina CR, Dakhllalah F, et al. Body mass index: amarker for significant colorectal neoplasia in a screening population.J Clin Gastroenterol 2007;41:285-90.

www.giejournal.org

http://refhub.elsevier.com/S0016-5107(14)01289-9/sref1





















































































impact of fair bowel preparation quality on adenoma and serrated polyp detection: data from the new...

Documents