Alex J. Ball

Position change during colonoscope withdrawal increases polyp and adenoma detection in the right but not in the left side of the colon: results of a randomized controlled trial.

BACKGROUND It has been suggested that changing patient position during colonoscope withdrawal increases adenoma detection. The results of previous studies have been conflicting. OBJECTIVE To evaluate whether routine position change during colonoscope withdrawal improves polyp detection. DESIGN Randomized, 2-way, crossover study. SETTING Teaching hospital. PATIENTS A total of 130 patients attending for diagnostic colonoscopy. INTERVENTIONS Patients undergoing colonoscopy had each colon segment examined twice: the right side of the colon (cecum to hepatic flexure) in the supine and left lateral position and the left side of the colon (splenic flexure and descending colon) in the supine and right lateral position. The transverse colon was examined twice in the supine position. MAIN OUTCOME MEASUREMENTS The primary outcome measure was the polyp detection rate (≥1 polyp) per colon segment. Secondary outcome measures included the number and proportion of patients with ≥1 adenoma in each segment and adequacy of luminal distension (1 = total collapse and 5 = no collapse). RESULTS Examination of the right side of the colon in the left lateral position significantly improved polyp detection (26.2% vs 17.7%; P = .01) and luminal distension (mean = 4.0 vs 3.5; P < .0001). Position change did not improve polyp detection in the left side of the colon (5.4% vs 4.6%; P = .99). There was no significant correlation between luminal distension and polyp detection in the right side of the colon (r = .03). LIMITATIONS Single center and open study design. CONCLUSION Examining the right side of the colon in the left lateral position increased polyp detection compared with examination in the supine position. Polyp detection in the left side of the colon was similar in the right lateral and supine positions. ( CLINICAL TRIAL REGISTRATION NUMBER NCT01554098.).

Is gluten sensitivity a "No Man's Land" or a "Fertile Crescent" for research?

To the Editor: We are pleased with the interest expressed by Dr Montagnese et al. in our previous observation (1) as shown in their thoughtful commentary (2) . Th ey provide data that show that high levels of bilirubin can cause interference with the assay of melatonin and they propose that this could be the case in our patient. We did not perform chloroform extraction, which is of unquestionable value in cases of high bilirubin, but we do not believe that this is the reason for the high values of melatonin and for the lack of discernible rhythm that returned to normal aft er liver transplantation. At the time of the study before liver transplantation, our patient had a Child – Pugh score of 10 and a serum bilirubin of 2.4 mg / dl that decreased to 1.2 mg / dl aft er undergoing the liver transplant. Th e upper limit of sensitivity of our assay was 0.3 pg / ml, and the intra-assay coeffi cient of variation in our hands was 2.3 % . Dr Montagnese et al. reported the absence of melatonin rhythm in two healthy volunteers and in two patients. Including our patient, we observed a lack of melatonin rhythm in 3 of 20 patients with cirrhosis who were studied in the Clinical Research Center at the Northwestern Memorial Hospital. Th e reasons for the lack of plasma melatonin are complex. Melatonin secretion that may be aff ected by disorders that aff ect the central circadian clock and melatonin metabolism, which undergoes both phase I (hydroxylation) and phase II (sulfation) reactions, could be prolonged in patients with cirrhosis. As in the report by Montagnese et al. , we were unable to associate melatonin rhythms with parameters of liver function (such as biochemical tests, Child – Pugh score). In addition, we conducted neuropsychological tests and administered treatment with lactulose to 10 patients, but we did not fi nd any association to support a disruption of central circadian control secondary to minimal hepatic encephalopathy. Studies on circadian rhythms in humans are diffi cult to conduct and the data provided by Montagnese et al. are a step forward to understanding the complex abnormalities present in cirrhosis. We believe that our observation following liver transplant support a role of hepatic metabolism in determining circadian abnormalities, but we acknowledge that further studies are necessary to explain melatonin arrhythmia.

Sedation practice and comfort during colonoscopy: lessons learnt from a national screening programme.

Aim Medication may be used to manage discomfort during colonoscopy but practice varies. The relationship between medication use and comfort during colonoscopy was examined in the English Bowel Cancer Screening Programme. Methods Data related to patient comfort and medication use from all 113 316 examinations performed within the English Bowel Cancer Screening Programme between 1 January 2010 and 31 December 2012 were analysed. Comfort was rated on the five-point Modified Gloucester Comfort Scale: 1, no discomfort; 5, severe discomfort. Scores of 4 and 5 were considered to indicate significant discomfort. Correlations between the proportion of examinations associated with significant discomfort and the amounts of medication used by colonoscopists were assessed using Spearman’s &rgr;. Logistic regression modelling examined the independent predictors of significant discomfort. Results Patients had a mean age of 65.7 years, and 58% were male. Examinations were performed by 290 endoscopists. In 91% of examinations, there was no significant discomfort reported during examination; however, there was considerable variation between individual colonoscopists (range 76.1–99.2%). Intravenous sedation and opiate analgesia were used during most examinations, but there was wide variation between colonoscopists, with a median (range) usage of 95.1% (4.1–100%) and 97.3% (5.6–100%), respectively. There was no association between the amount of sedation and analgesia used and significant discomfort (&rgr;<0.2). On multivariate analysis, significant discomfort was found to be more common among female individuals [odds ratio (OR)=2.0], on incomplete examinations (OR=6.7), and among patients with diverticulosis (OR=1.4). Conclusion There was wide variation in medication practice among English screening colonoscopists, but this was unrelated to the occurrence of significant discomfort.

PTU-019 A Comparison of Two Colonoscope withdrawal Techniques: Interim Analysis of a Randomised Cross over Study

Introduction Many endoscopists withdraw the colonoscope with the patient in a single position (left lateral or supine), while others advocate position change. A previous study in a small group of patients suggested position change is beneficial in the transverse and left colon. We have compared colonoscope withdrawal in the supine position with position change. Methods A randomised cross-over study compared colonoscope withdrawal in the supine position with position change (caecum to hepatic flexure in the left lateral position, transverse colon in the supine position and the hepatic flexure and descending colon in the right lateral position). Colonic segments were precisely defined using biopsy sites as markers, aided by a Scopeguide imager. Segments were cleansed and examined for at least 2 minutes during which air was insufflated to distend the colon. After each segment was examined the colonoscope was reinserted and the same segment was re-examined in the alternative position. Luminal distension was rated on a validated 5 point scale, ranging from 1 = completely collapsed to 5 = maximal distension. Ordinal and categorical data were compared with the Mann Whitney U test and Fisher’s exact test respectively. Results This is an interim analysis of 65 patients (mean age 62, 38 male). 30 patients were initially examined in the supine position and 35 patients with position change. Distension scores were higher in the right colon when examined in the left lateral position (mean = 3.9 vs. 3.5 p = < 0.001), and in the left colon when examined in the right lateral position (4.4 vs. 3.6, p = < 0.001) (see table 1). The proportion of patients with scores of 4–5 (adequate) were higher in the right colon when examined in the left lateral position (42/65 vs. 27/65, p = 0.014) and in the left colon when examined in the right lateral position (59/65 vs. 34/65, p = < 0.001). There was no significant carry-over effect in any of the examined segments. Abstract PTU-019 Table 1 Distribution of distension scores with each colonoscope withdrawal strategy Distension score Withdrawal position 1 2 3 4 5 Right colon Supine 0 5 33 19 8 Left Lateral 0 0 22 26 16 Descending colon Supine 0 2 29 29 5 Right lateral 0 0 6 26 33 Conclusion Position change improves suboptimal distension of the right and left colon. Further analysis of this study will clarify whether position change also improves polyp detection. Disclosure of Interest None Declared References East JE et al. Position changes improve visibility during colonoscope withdrawal: a randomised, blinded, crossover trial. Gastrointest Endosc. Feb 2007; 65(2):263–269. East JE et al. Dynamic patient position changes during colonoscope withdrawal increase adenoma detection: a randomised, crossover trial. Gastrointest Endosc. Mar 2011; 73(3):456–463.

Position change during colonoscope withdrawal: is it worth the effort?

We read the recent article by Kim et al with some incredulity regarding the manner in which informed consent was obtained. The authors state that in those patients undergoing an unsedated colonoscopy in whom the sigmoid colon was not transversed after 10 minutes, the instrument was withdrawn. The patients were then given an explanation of the proposed study with a cap and, if they agreed, were randomized to an E-cap or a C-cap. We think that informed consent should have been obtained well before the colonoscopic procedure was started. The recommendations of good clinical practice suggest at least 3 days before the scheduled appointment. It is clearly problematic to explain a research protocol to obtain informed consent after spending 10 minutes in an unsuccessful attempt to transverse the sigmoid colon. The patient is likely to be in some discomfort and fearful, and there is a possibility that the patient will see a refusal to consent as having implications for the next colonoscopy. If the patients did not consent, then they would presumably have to be rescheduled for an examination under sedation and another bowel preparation. It would clearly have been preferable to explain the research proposal before scheduling the colonoscopy. In our institutions, such a protocol would not have been passed by the institutional review board, and we suspect that this is the case also in the United States. The requirements for informed consent have rightly become more stringent in recent years and are universal.

Nitrous oxide use during colonoscopy: a national survey of English screening colonoscopists

Introduction Nitrous oxide can improve patient experience during colonoscopy, and its rapid elimination minimises after effects and inconvenience. Despite its advantages, nitrous oxide is used infrequently in the UK. We sought to understand the reasons for its low use. Methods Colonoscopists within the English Bowel Cancer Screening Programme (BCSP) were invited to participate in a web-based survey assessing the availability, current practices and perceptions towards nitrous oxide. Respondents were able to select predefined answers or offer written responses. Free text responses were assessed using thematic analysis. Results The survey was completed by 68% of the English BCSP colonoscopists. Nitrous oxide was available to 73% of respondents but with considerable regional variation. Most colonoscopists rated the properties of nitrous oxide favourably and would use it if they had a colonoscopy themselves. Despite this, nearly half used it in less than 20% of examinations. 80% instruct patients to use nitrous oxide as required, and differences in how it was used in combination with intravenous sedation and analgesia were reported. Written responses suggest nitrous oxide is often used in the patients who are expected to have the least discomfort. Conclusions Most colonoscopists perceive that nitrous oxide is effective and reduces inconvenience and would use it themselves if they required a colonoscopy. Studies to improve patient selection and optimise the use of nitrous oxide would be of value.

A randomized controlled trial comparing continuous and as-required nitrous oxide use during screening colonoscopy.

Background and study aims Entonox is a 50 : 50 combination of nitrous oxide and oxygen, which may be used to manage pain during colonoscopy. The optimal mode of Entonox administration is unknown. The aim of this study was therefore to compare continuous and as-required Entonox use. Patients and methods Patients attending for screening colonoscopy at a single centre were randomized to continuous or as-required Entonox use. The primary outcome measure was the patient’s overall pain rating at the time of discharge (verbally administered numerical ratings scale, 0=no pain and 10=extreme pain). Secondary outcome measures included the patients’ experience of pain during the colonoscopy (rated every 2 min), side effects and the need for rescue intravenous medications. Results A total of 108 patients were randomized, and 100 completed the study (46 continuous, 54 as required). The overall pain scores at discharge did not differ between those who used Entonox continuously and as required (mean=2.4 vs. 3.2, P=0.08). There were also no differences in the experience of pain during colonoscopy (mean=1.8 vs. 2.2, P=0.28; peak=4.2 vs. 4.8, P=0.26; and area under curve=23 vs. 30, P=0.24). Patients with high anxiety had greater overall pain scores (mean=3.7 vs. 2.4, P=0.03). Light headedness occurred more often with continuous Entonox use (48 vs. 21%, P=0.009). Conclusion Among patients attending for screening colonoscopy, comfort ratings were similar in those using Entonox continuously and as required, but light headedness was more common with continuous use (NCT identifier: 01865721).

PWE-029 Position Changes Among English Bcsp Colonoscopists: A Survey Of Practices

Introduction Studies suggest that modifying a patient’s position during colonoscope withdrawal may improve luminal distension and polyp detection. It is unclear whether this practice is widely adopted by endoscopists. Methods Colonoscopists within the English Bowel Cancer Screening Programme (BCSP) were invited to participate in a web-based survey assessing the use of position change during colonoscope withdrawal. Free text responses were assessed using thematic analysis. Results The survey was completed by 204/298 (68%) of English BCSP colonoscopists. 64.7% of respondents indicated that they almost always change a patient’s position, 16.7% usually, 13.7% sometimes, 3.4% occasionally and 1.5% rarely do so. 77% of those who almost always or usually changed a patient’s position did so as part of their routine, but 75.3% were less likely to change position in those with poor mobility and 75.3% would not change position if luminal distension was adequate. 93% of these respondents most often positioned patients supine while examining the transverse colon and nearly half examined the right and descending colon in a sub-optimal position (Table 1). Of those respondents who sometimes, occasionally or rarely changed a patient’s position, 42% were unconvinced that routine position change was beneficial. A further 21.1% felt it took too long, 7.8% felt it was inconvenient for the patient and 7.8% felt it was inconvenient for the endoscopist. These respondents were most likely to examine segments without changing patient position. Free text responses revealed that some endoscopists position patients differently during insertion and withdrawal and also use position change to optimise access during therapy. Abstract PWE-029 Table 1 Patient position most often used by endoscopists who almost always or usually change position and those who sometimes, occasionally or rarely change position Position change usage Segment Right lateral Supine Left lateral In which ever position they arrive Almost always or usually Caecum to hepatic flexure 7.8% 25.3% 60.2% 7.8% Transverse colon 1.2% 93.4% 5.4% 0.6% Splenic flexure and descending colon 51.2% 34.4% 11.4% 3.6% Sometimes, occasionally or rarely Caecum to hepatic flexure 0% 31.6% 34.2% 34.2% Transverse colon 0% 34.2% 28.9% 36.8% Splenic flexure and descending colon 7.9% 31.6% 26.3% 34.2% Conclusion Most BCSP colonoscopists change patients’ position during most colonoscope withdrawals, but the patient position is often sub-optimal. Increased awareness of the existing literature and further research assessing positioning strategy is warranted. Reference East JE et al. Gastrointest Endosc. 2011 Mar;73(3):456–63 Disclosure of Interest None Declared.

Sa1533 Patient Comfort and Sedation and Analgesic Practices During Colonoscopy in the English Bowel Cancer Screening Programme

of endoscopy and its potential complications. The aim of this study is to perform a systematic review on the safety of endoscopic procedures following ACS. Methods: An initial search of predefined keywords on the EMBASE, Medline, and ISI Web of Science databases yielded 1152 abstracts for possible inclusion. Two independent reviewers analyzed abstracts, and reviewed manuscript content. Data from all relevant papers, including demographic information, type of endoscopy, indications, complication rates, and ACS subtypes were compiled. We determined weighted data for timing of endoscopy, and rates of endoscopic complications and all-cause mortality. Results: Nineteen retrospective cohorts were included. All patients suffered an ACS (53.6% NSTEMI, 20.1% STEMI, 0.6% unstable angina, and nonspecified in 25.6%); 33% developed congestive heart failure, and 19.4% arrhythmias secondary to ACS before endoscopy. The cumulative incidence of endoscopy following ACS was 0.41% (data on 316/76,941). Overall, 964 patients (mean age 71.1 5.2 years, 60.3% male) underwent 985 endoscopies (1.02 per patient, with therapeutic goal in 20.1%). Procedures included 693 EGDs, (70.3%), 133 colonoscopies (13.5%), 92 sigmoidoscopies (9.3%), 34 PEGs (3.5%), 26 ERCPs (2.6%), and 5 enteroscopies (0.5%). The primary indications for endoscopy were unspecified symptoms of GI bleeding (33.9%), hematemesis (17.5%), melena (13.5%), occult blood loss (13.3%), hematochezia (6.4%), and other indications (15.4%). The average timing to endoscopy was 9.2 5.5 days after ACS. The most common endoscopic findings were peptic ulcer disease (24.7%, 95% CI 21.8-27.8%), followed by normal endoscopic findings (22.0%, 95% CI 19.2-25.0%). Twenty-one repeat endoscopies were performed (2.1%, 95% CI 1.4-3.2%). 3.5% of patients required surgery or angiography (3.6, 95% CI 2.2-5.7%). Including deaths, 91 complications occurred (9.4%, 95% CI 7.8-11.5%). Of those, hypotension (27.5%, 95% CI 19.437.4%), arrhythmias (12.1%, 95% CI 6.9-20.4%), and repeat ACS (7.7%, 95% CI 3.815.0%) were the most frequent following endoscopy. Six deaths were attributed to endoscopy (!48 hours post-ACS, 6.6% of all complications, 95% CI 3.1-13.7 %). Allcause mortality was 8.6% (63/733, 95% CI 6.8%-10.9%). Conclusion: This systematic review identified a significant degree of possible endoscopy-related negative outcomes following ACS, highlighting the need for adequate cardiac and hemodynamic monitoring during endoscopy. Further studies are required to better characterize indication and patient selection, as well as the appropriate timing of endoscopy in this high-risk cohort.

PWE-060 Polypectomy Practices In The English Bowel Cancer Screening Programme

Introduction Most polyps are <10 mm in size and a range of polypectomy techniques are available with wide variations in practice. We aimed to examine the techniques employed for removal of <10 mm polyps in relation to polyp characteristics, completeness of excision, safety and changes over time. Methods Data relating to removal of polyps <10 mm between Jan 2010 and Dec 2012 were retrieved from the national Bowel Cancer Screening Programme (BCSP) database. Categorical data was compared using x2 . Results 147174 polyps were removed during 62679 colonoscopies. A range of techniques was used (cold biopsy forceps (CBF) 19.7%, cold snare (CS) 22.1%, hot biopsy forceps (HBF) 12.2%, hot snare (HS) 35.1%, EMR 10.9%). EMR was used more frequently in the right colon compared to the left (14.3 vs. 8.3%, OR = 1.84, 95% CI: 1.78–1.90). Most pedunculated polyps were removed using HS; this was lower in the right vs. left colon (69.6 vs. 88.3%, OR = 0.30, CI: 0.28–0.33). CS was most common for non-pedunculated polyps in the right colon (29.8 vs. 19.0% in left, OR = 1.81 CI: 1.76–1.85); whereas most common in the left colon was HS (34.8 vs. 22.5% in right, OR = 1.84 CI: 1.79–1.88). Surgeons were more likely than physicians to use diathermy irrespective of site or morphology (65.6 vs. 56.5%, OR = 1.46 CI: 1.43–1.5). In 60% of polyps removed completeness of excision was not histologically assessable. 21.2% were completely excised, 5.8% incomplete and 13% not stated. For non-pedunculated polyps, histologically-confirmed complete excision was more common after EMR (23.4 vs. 6.2%, OR = 1.16, CI: 1.08–1.25) compared to other techniques (CBF 17.7%, CS 15.1%, HBF 19.1%, HS 21.5%); for pedunculated polyps it was more common after EMR (42.3%) and HS (42.0%). Complications were rare for colonoscopies (45227) where only polyps <10 mm were removed. 12 (0.03%) bleeding episodes required transfusion; rates for single and multiple polypectomy cases were 0.01 and 0.04% respectively (OR = 5.01, CI: 1.10–22.8). The HS technique was most commonly used. There were 16 (0.04%) perforations; 0.02% for single vs. 0.05% for multiple polypectomies (OR = 2.20, CI: 0.77–6.34, p = 0.13). No technique dominated for single compared with HS for multiple polypectomies. Between 2010 and 2012, use of CBF, CS and EMR increased, whereas HBF and HS decreased (p < 0.01) Abstract PWE-060 Table 1 % 2010 2012 CBF 15.2 23.0 OR = 1.67, CI: 1.61–1.72 CS 21.3 23.3 OR = 1.12, CI: 1.09–1.16 HBF 14.1 10.1 OR = 0.68, CI: 0.66–0.71 HS 41.0 31.1 OR = 0.65, CI: 0.63–0.67 EMR 8.5 12.5 OR = 1.55, CI: 1.48–1.62 Conclusion The removal of polyps <10 mm within the BCSP is safe, but histological evidence of completeness of excision is poor with all techniques. Wide variations in practice reflect the lack of evidence guiding these decisions, although use of cold resection techniques has increased over time Disclosure of Interest None Declared.