Abstracts

Of 12,798 colonoscopies (4278 for screening) and excluding those due to poorcolonic prep, there were 483 (3.8%) that were incomplete; 137 of which were forscreening (3.2%). There were 64 colonoscopies where active colitis (15),obstructing tumors (43) and strictures (6) prevented colonoscopic completion.Trainees were involved in 35% (169 colonoscopies) of all incompletecolonoscopies. Individual attending’s incompletion rates ranged from 1.7% to 7.3%.In average the duration of the incomplete procedures was 27 minutes (3-90). Themean BMI of patients with incomplete colonoscopy was 28.9. Diverticulosis wasfound in 220 (45.5%) of procedures and 21% of proceduers were in patients witha history of pelvic/abdominal surgery. Discomfort and/or intolerance was reportedas the cause of incompletion in 134 colonoscopies (27.7%) and was significantlymore common in younger patients and associatted with shorter total colonoscopytime than those with reason(s) other than intolerance/discomfort (Table). Therewere 140 colonoscopies that either had a prior and/or subsequent attempts,representing 28.7% of incomplete colonoscopies. Of these, 36 (25.7%) wereattempted by the same endoscopist and in 26/36 (72.2%) were successful. Likewise,110 of the 140 repeated attempts (78.6%) had a complete previous and/orsubsequent repeat exam. Conclusions: At our endoscopy unit the completion ratesfor all (and screening) colonoscopies was above 96%. Trainees were involved in 1/3of incomplete procedures. Almost 30% of incomplete procedures had a repeatedattempt at a prior or subsequent date, and the completion rate in these repeatedattempts was 78.6%, suggesting that a repeat attempt by the same or anotherexperienced endoscopist may improve the completion rates even further.

Patinets with incomplete colonoscopy due to Discomfort/intolerance vs. Otherreason

Discomfort/Intolerance (n) Other (n))

www.giejournal.org

Mean Age

59.5 � 0.9 (134)) 65.1 � 0.8 (357) Mean Colonoscopy time 21.6 � 1.0 (132) 29.1 � 0.9 (350) Mean BMI 28.5 � 0.6 (131) 29.1 � 0.4 (351)

)p ! 0.05 Discomfort/Intolerance vs. Other (Student’s t-test)

T1534

Prophylactic Octreotide for Prevention of Post-ERCP

Pancreatitis: A Meta-AnalysisAbhishek Choudhary, Matthew L. Bechtold, Srinivas R. Puli,Mohamed O. Othman, Wilson P. Pais, Mainor R. Antillon, Praveen K. RoyBackground: Acute pancreatitis is a common complication of ERCP. Over the years,attempts have been made to identify an agent to prevent this complication.Octreotide, an inhibitor of exocrine secretion with anti-inflammatory andcytoprotective effects, has been studied in randomized clinical trials (RCTs) to preventpost-ERCP pancreatitis with mixed results. We conducted a meta-analysis to evaluatethe efficacy and safety of prophylactic octreotide for prevention of post-ERCPpancreatitis. Methods: MEDLINE, Cochrane Central Register of Controlled Trials &Database of Systematic Reviews, PubMed, and recent abstracts from major conferenceproceedings were searched (through 10/07). RCTs comparing prophylacticoctreotide to placebo for prevention of post-ERCP pancreatitis were included.Standard forms were used to extract data by two independent reviewers. The effectsof octreotide were analyzed by calculating pooled estimates of post-ERCPpancreatitis, hyperamylasemia, hospital stay, and severity of pancreatitis. Separateanalyses were performed for each outcome by using odds ratio (OR) or weightedmean difference (WMD) by random effects model. Publication bias was assessed byfunnel plots. All studies were graded by Jadad score. Heterogeneity among studieswas assessed by calculating I2 measure of inconsistency. Results: Sixteen RCTs (N Z3,575) met the inclusion criteria. In 13 trials, octreotide was administeredsubcutaneously and in 3 trials, it was administered as IV infusion and subcutaneously.The dose of octreotide ranged from 0.1 mg to 0.5 mg in different studies. Octreotidewas administered at different times prior to ERCP: 4 studies administered octreotideon the night before to 2 days before ERCP and 12 studies used octreotide on the day ofthe procedure. Octreotide infusion decreased the odds of hyperamylasemia (OR 0.79,95% CI: 0.67 - 0.96, p Z 0.01), the length of hospital stay (WMD -0.72, 95% CI: -1.12 - -0.32, p ! 0.01), and the severity of pancreatitis (OR 1.97; 95% CI: 1.02 - 3.80, p Z 0.04).Octreotide infusion did not decrease the odds of post-ERCP pancreatitis (OR 0.79,95% CI: 0.61 - 1.03, p Z 0.08). However, on sub-group analysis of studies thatadministered octreotide at least a night before ERCP, a protective effect against post-ERCP pancreatitis was noted (OR 0.47, 95% CI: 0.23 - 0.97, p Z 0.04, NNT 17). Nosignificant publication bias was present. Conclusions: Octreotide administration thenight before ERCP prevents post-ERCP pancreatitis. Octreotide administration wasalso efficacious in reducing hyperamylasemia, shortening hospital stay, anddecreasing the severity of pancreatitis.

T1535

How Accurate Is Endoscopic Ultrasound in Staging Perirectal

Nodal Invasion By Rectal Cancers? A Meta-Analysis and

Systematic ReviewSrinivas R. Puli, Jyotsna Bk Reddy, Matthew L. Bechtold, Praveen K. Roy,Jack D. Bragg, Priya D. Ravindran, Jamal A. Ibdah, Mainor R. Antillon

Vo

Background: Nodal staging in a patient with rectal cancer predicts prognosis anddirects therapy. Endoscopic Ultrasound (EUS) may be the best imaging modality toevaluate for nodal metastasis. Published data on accuracy of endoscopic ultrasound(EUS) for diagnosing nodal invasion in patients with rectal cancer has beeninconsistent. Aim: To evaluate the accuracy of EUS in diagnosing nodal metastasis ofrectal cancers. Method: Study Selection Criteria: Only EUS studies confirmed bysurgery were selected. Only studies from which a 2 � 2 table could be constructedfor true positive, false negative, false positive, and true negative values wereincluded. Data collection & extraction: Articles were searched in Medline, Pubmed,Ovid journals, CINAHL, International pharmaceutical abstracts, old Medline,Medline non-indexed citations, and Cochrane control trial registry. Two reviewersindependently searched and extracted data. The differences were resolved bymutual agreement. 2 � 2 tables were constructed with the data extracted from eachstudy. Statistical Method: Meta-analysis for the accuracy of EUS was analyzed bycalculating pooled estimates of sensitivity, specificity, likelihood ratios, anddiagnostic odds ratios. Pooling was conducted by both Mantel-Haenszel method(fixed effects model) and DerSimonian Laird method (random effects model). Theheterogeneity among studies was tested using Cochran’s Q test based upon inversevariance weights. Results: Initial search identified 3,610 reference articles, in which,352 relevant articles were selected and reviewed. Data was extracted from 34studies (N Z 2,732) that met the inclusion criteria. Pooled sensitivity of EUS indiagnosing nodal involvement by rectal cancers was 73.2% (95% CI: 70.6�75.6).EUS had a pooled specificity of 75.8% (95% CI: 73.5 - 78.0). The positive likelihoodratio of EUS was 2.84 (95% CI: 2.16 - 3.72) and negative likelihood ratio was 0.42(95% CI: 33.3 - 52.2). The diagnostic odds ratio, the odds of having nodal metastasisin positive as compared to negative EUS studies, was 7.9 (95% CI: 5.3�11.7). All thepooled estimates, calculated by fixed and random effect models, were similar. SROCcurves showed an area under the curve of 0.79. The p for chi-squared heterogeneityfor all the pooled accuracy estimates was O 0.10. Conclusion: EUS is an importantand accurate diagnostic tool to evaluate nodal metastasis of rectal cancers. Thismeta-analysis shows that the sensitivity and specificity of EUS moderate. Furtherrefinement in EUS technologies and diagnostic criteria are needed to improve thediagnostic accuracy.

T1536

Recommendation for the Current Endoscopic Technique in the

Surveillance of HNPCC Gene CarriersRobert HuNeburg, Tilman Sauerbruch, Christof LambertiIntroduction: Due to the high risk of colorectal cancer in carriers of hereditarynonpolyposis colorectal cancer (HNPCC), and the shortened adenoma-carcinomasequence, good colonoscopic surveillance is obligatory. Up to 25% of small andflat adenomas, which are typically for HNPCC, are missed by standardcolonoscopy (SC), according to tandem colonoscopies. Therefore an endoscopicapproach is needed which is able to detect small adenomas. Methods: Allendoscopic studies that met the following criteria were included: 1. Screening ofHNPCC patients2. Comparison of different endoscopic techniques. Results: So farthere have been four studies concerning new endoscopic techniques, 3 alreadypublished, 1 publication in progress. In two studies SC was compared tochromocolonoscopy (CC), in one SC was compared to narrow-band-imagingcolonoscopy (NBI), in the last one, a two-armed study, SC was compared to CC aswell as CC to NBI. A total number of 105 patients were screened by comparing SCto CC. In all CC studies indigo carmine dye was used. In 62 patients CC wascompared to NBI. In one study NBI was compared to CC which included 62patients as well. The percentage of HNPCC gene mutation carriers differed a lot(84%/50%/13%/90%) as well as the rate of patients who had a colonoscopy beenperformed before inclusion into the study (32%/56%/100%/92%). In one CC andthe NBI study only the proximal colon was inspected, the other two included thewhole colon. In patients who were already embedded in a surveillance programthe interval to their last colonoscopy differed considerably (2/3/2/1 year). ThreeStudies showed a significant increase of adenoma detection while using CC. Onestudy showed a significant increase of adenomas while using NBI. One studyrevealed that CC detected significantly more adenomas than NBI. Conclusion: CCis superior to standard and narrow-band-imaging colonoscopy in detection ofpolypoid lesions in patients with HNPCC. Therefore we recommend CC as thecurrent gold standard of endoscopic surveillance.

Adenomas Adenomas AdenomasGenecarriers

miss rateof 1st odds

lume 6

n

7,

in SC

No. 5 :

in CC

2008 G

in NBI

ASTROINT

(%)

ESTINAL

p

END

examination

OSCOPY

ratio

Hurlstone

25 11 32 - 84 0.001 74 5.8 Lecomte 33 7 11 - 50 0.045 61 3.2 East 62 25 - 21 13 0.001 45 1.9 Huneburg 47 7 13 - 90 0.001 46 2.5 Huneburg 62 - 38 10 90 0.001 73 3.2

AB241