H-L Ching

PWE-135 Pushing Propofol-Assisted Double Balloon Enteroscopy Out of Theatre and Into The Endoscopy Suite: Does The Pound Have Enough Pull?

Introduction Double balloon enteroscopy (DBE) has poorer tolerability compared to other forms of endoscopy due to procedure duration. Anaesthetic led propofol sedation in theatre is increasingly used for complex endoscopy however has cost implications. We studied the feasibility and cost effectiveness of providing an anaesthetic led propofol-based DBE service outside theatre. Methods All patients undergoing DBE with propofol between March 2013 and December 2015 were prospectively recruited. The propofol lists were carried out in an endoscopy room with anaesthetists and fully equipped with anaesthetic equipment. Patient data including hospital anxiety and depression scores (HADS) were performed prior to DBE. Pain, discomfort and distress scores were measured post-DBE (range 0–10 for each category). Costs including consumables, staffing, ward and theatre expenses were compared. Results A total of 82 patients (44% male, 79% oral DBE) were included (mean age 54±13 years). Indications for DBE included iron deficiency anaemia (37%), overt gastrointestinal bleeding (8.5%), Peutz-Jeghers syndrome (18%), abnormal radiology (2.4%), suspected Crohn’s disease (22%), suspected coeliac complications (4.8%) and small bowel tumours/masses (6.1%). The mean dose of propofol was 1089 mg±422.9 and the mean procedural time 77 mins±25. The diagnostic yield was 63%. The findings included polyps/tumours (18%), angioectasias (22%), ulcers (12%), strictures (7%), and abnormal mucosa (6%). DBE was normal in 36.5%. DBE was well tolerated with median scores for pain, discomfort and distress all being 0.2. There were no correlations between pre-DBE HADS and post-DBE pain, discomfort or distress scores (p > 0.05 for all categories). Therapeutic intervention was performed in 40% which included argon plasma coagulation for angioectasias (13.4%), polypectomy (22%) and dilation of strictures (3.7%). One patient experienced transient bradycardia and required glycopyrrolate while another patient became wheezy and required nebuliser therapy post-procedure. Neither cases necessitated abandonment of DBE and were deemed non-serious adverse events. The cost of running a single DBE list with propofol is £2391 at our unit compared to a national range of £3174 - £4730 if run in theatre. Extrapolating this information would imply cost-savings of between £40,700 - £110,000 per year. Conclusion We have demonstrated that propofol-assisted DBE in an endoscopy suite is safe and feasible. It provides excellent tolerability, irrespective of pre-DBE HADS and is cost effective for the department. This allows better utilisation of theatre resources and should be encouraged nationally for other forms of complex endoscopy. Disclosure of Interest None Declared

The Changing Platform of Small Bowel Endoscopy: a 15-Year Experience of Demand and Outcomes

Dear Editor, We have read with interest the article by Green et al. on the Approach to the Diagnostic Workup and Management of Small Bowel Lesions at a Tertiary Care Center. This study offers insights into the diagnosis and surgical management of small bowel (SB) lesions from a surgical perspective where all patients ultimately underwent surgery. We would like to share our experience on the management of SB pathology from a tertiary SB centre. A total of 5065 small bowel capsule endoscopies (SBCEs), 569 double balloon enteroscopies (DBEs) and 22 intraoperative enteroscopies (IOEs) have been performed in total over the past 15 years (January 2002 and December 2016). Obscure gastrointestinal bleeding (OGB) and anaemia were the most common indications for SBCE (37.8%), DBE (30%) and IOE (55%). The majority of patients had a SBCE prior to DBE (83.4%) and IOE (81.8%) (p = 0.0001). Fifty percent of patients had a DBE prior to IOE. The introduction of SBCE in 2010 at our centre has made the investigation of the SB more accessible. The demand of SBCE has risen steadily until 2014. Its request has plateaued since then. The demand for DBE is still rising unlike the need for IOE that has fallen dramatically (p = 0.0001). The use of IOE is reserved only for lesions beyond limits of DBE. In 2016, for every 11 SBCE performed, 1 patient underwent a DBE locally. The diagnostic yield for SBCE, DBE and IOE were 29, 53.8 and 89.5% respectively (p = 0.0001). The diagnostic yield for SBCE has fallen over the years (p = 0.0001) possibly due to the lower threshold for performing SBCE by the referring clinicians. It was highest for the indication of occult gastrointestinal bleeding (OGB) (36%; p = 0.001) similar to findings in this study where the diagnostic rates of SBCE in patients presenting with anaemia/OGB were higher than in those presenting with obstructive symptoms. This contrasts with the diagnostic yield of DBE that has risen gradually (p = 0.0001) possibly because of better planning of the procedures with prior SBCE. The yield for DBE in those with a negative SBCE was only 4.3% (p = 0.0001). We report a higher diagnostic yield of IOE than that reported by the authors of this paper. The rate of therapeutics for DBE and IOE was 29 and 77% respectively. The majority of therapeutic procedures performed were argon plasma coagulation (APC) to angioectasias and polypectomies. The complication rates were 2% for DBE and 13.6% for IOE. (p = 0.004) IOE carries a high procedure-related mortality and postoperative complication rate of 5 and 21% respectively. Its use is limited to when the probability of finding SB pathology is still high despite negative noninvasive SB endoscopy, to guide therapeutic surgery or as a therapeutic procedure itself when SB bleeding lesions have been managed unsuccessfully by angiographic embolisation or endoscopic treatment. The complication rate of DBE is much less (0.9–1.2% reported in the literature). The lower complication rate and the less invasive nature of DBE should make it a preferable option over IOE. Only 8 (1.41%) of patients who underwent DBEwent on to have surgery for strictures/small bowel lesions. This is a considerably low rate and emphasises that patients can be managed more safely with less invasive procedures such as DBE even when therapeutic interventions are needed. Patients with a suspicion of SB pathology should initially undergo SBCE and/or imaging. CE offers a non-invasive approach of visualising the entire length of the SB, making it an attractive first-line approach for patients with suspected SB pathology with persistent symptoms following negative bidirectional endoscopies. It provides mucosal views of flat lesions such as ulcers and angioectasias that otherwise would * S. Chetcuti Zammit stf_che@yahoo.com

PTU-018 Endoscopy: the effect of hospital anxiety and depression scale on patient experience

Introduction The effect a patients baseline anxiety and depression has on their experience of oesophagogastroduodenoscopy (OGD) is unclear, as is how this affects their view of the service. We present a questionnaire study of patients undergoing OGD in our endoscopy department. Method A closed ended questionnaire was given to adult patients attending outpatient OGD between November 2016 and January 2017. The patients hospital anxiety and depression scale (HADS), sedation requirement, expectations, experience of OGD and service evaluation were assessed in the questionnaire. Patients were asked to rank questions focused on unpleasant sensations (gagging, vomiting, choking and bloating), discomfort and pain. All patients were offered sedation with fentanyl and midazolam. Results Questionnaires were completed by 100 patients: 46 female (med: 62 years±13.5 years), 54 male (med: 57±13 years). There was a significant difference in anxiety score (AS) between females and males (med: 7±3.15 vs 4.5±3.5 respectively p=0.001) and depression score (DS) (med: 5±3.5 vs 2±2.75 respectively p=0.026). Patients with a high AS or DS had statistically significantly greater concerns and expectations prior to OGD but no statistically significant difference in their experiences of unpleasant sensations or discomfort (Table 1). Patients with a higher AS and DS experienced statistically significantly more pain than those with lower scores. Sedated patients experienced less unpleasant sensations and discomfort. Conclusion Patients with higher AS or DS have greater concerns prior to OGD. However, their actual experiences are no different to those with lower scores with regards to unpleasant sensations and discomfort. Sedated patients reported a better experience suggesting a beneficial role of sedation in reducing awareness of unpleasant sensations and discomfort. Disclosure of Interest None Declared Abstract PTU-018 Table 1 Influence of HADS and sedation on patient expectations and experience of OGD Anxiety score Depression score Sedation given Low High P value Low High P value No Yes P value Concerns of unpleasant sensations (out of 40) 136.9 268.9 <0.05 149 2010.8 0.05 168.50 2010.50 0.32 Expectation of discomfort (out of 30) 7±3.9 15±5.5 <0.05 7±5.4 13.5±5.4 <0.05 9±4.00 15±6.00 0.12 Expectation of pain(out of 30) 6±4 14.5±5.4 <0.05 7±6 12.5±5.9 <0.05 9±5.50 12±5.50 0.24 Experienced unpleasant sensations (out of 40) 8±7 11.5±7.65 0.23 9.5±6.75 11±9.4 0.47 12±7.00 7±5.50 <0.05 Experienced discomfort(out of 30) 5±3 5±2.5 0.88 5±2.9 5±2.9 0.99 5±3.00 4±30 0.05 Experienced pain(out of 30) 2±1 3.5±1.9 <0.05 2±1 3.5±2 <0.05 2±1.25 2±1.50 0.94

PTU-132 The changing platform of small bowel endoscopy: a 15-year experience of demand and outcomes

Introduction The development of capsule endoscopy (CE) and enteroscopy have increased the ability to investigate and treat pathology in the small bowel. We have been providing a comprehensive small bowel endoscopy service and we present our data from the last 15 years evaluating demand, diagnostic yield and impact on patient management. Method A retrospective analysis was conducted on patients who underwent CE, push enteroscopy (PE), double balloon enteroscopy (DBE) and intraoperative enteroscopy (IOE) between January 2002 and December 2016. Results A total of 5065 CEs, 569 DBEs, 327 PEs and 22 IOEs were performed over 168 months. Obscure gastrointestinal bleeding (OGB) and anaemia were the most common indications for CE (37.8%), DBE (30%) and IOE (55%). Exclusion of coeliac disease complications (26.6%) was the most common indication for PE. The majority of patients had a CE prior to DBE (83.4%) and IOE (81.8%) (p=0.0001). Fifty percent of patients had a DBE prior to IOE. The demand for CE has plateaued over the past few years compared to the demand for DBE which is rising. The demand for PE has fallen dramatically (p=0.0001). This is also true for IOE and its use is reserved only for lesions beyond the limits of DBE (p=0.0001). In 2016, for every 11 CEs performed, 1 patient underwent a DBE locally. The diagnostic yield for CE, PE, DBE and IOE were 29%, 43.9%, 53.8% and 89.5% respectively (p=0.0001). The diagnostic yield for CE has fallen over the years (p=0.0001). This is perhaps due to the lower threshold for performing CE by the referring clinicians. The diagnostic yield for CE was highest for the indication of OGB (36%) compared to Crohn’s disease (30.0%, p=0.001). This contrasts with the diagnostic yield of DBE which has risen gradually (p=0.0001). Fifteen percent of patients underwent a DBE despite a negative CE. The yield for DBE in this group of patients was only 4.3% (p=0.0001). The rate of therapeutics for DBE, PE and IOE were 29%, 20% and 77% respectively. The majority of therapeutic procedures performed were argon plasma coagulation to angioectasias and polypectomies. The complication rates were 2% for DBE, 0.9% for PE and 13.6% for IOE (p=0.004). Complications following DBE included cardiovascular events (0.2%), respiratory compromise (0.9%), pancreatitis (0.2%) and gastrointestinal bleeding following polypectomy in (0.2%). Conclusion This is one of the largest series to date comparing the clinical utility and yield of all 4 small bowel endoscopic modalities. Enteroscopy has an important role in providing further management directed by CE. In patients with a negative CE, the yield of DBE is low, hence careful patient selection is required to maximise the use of resources. Disclosure of Interest None Declared

PTU-015 Patient perspective on the tolerance of gastroscopy

Introduction The degree of tolerance patients have for gastroscopy may be less than the endoscopist believes. We therefore present a questionnaire study of patients undergoing oesophagogastroduodenoscopy (OGD) in an endoscopy centre to determine what this difference may be. Method A close ended questionnaire was given to patients attending adult outpatient OGD between November 2016 and January 2017. The patients’ Gloucester Comfort Score (GCS), was recorded as was the endoscopist rating score. Doctors were not aware of the GCS given by patients nor that the data was to be collected for comparison. All patients were offered the choice of sedation with fentanyl and midazolam. Results Questionnaires were completed by 65 patients: 30 female (median: 60 years±13.5 years), 35 male (median: 60±16.5 years). 18.5% reported no pain, 36.9% minimal discomfort, 23.1% mild discomfort, 18.5% moderate discomfort and 3.1% severe discomfort. Sedation was administered in 39% patients.There was no significant difference in reported scores based on sex (p=0.283), age (p=0.202) or whether they were sedated (p=0.107). Endoscopists rated 43% of procedures good, 29.2% acceptable, 3% poor and 0% not tolerated. There was a moderate correlation (rs=0.413, p=0.001) between the endoscopist rating and the patients GCS. Conclusion Although there was some correlation, a fifth of patients still reported moderate to severe pain during OGD which was not identified by endoscopists. This difference highlights how the tolerability of this routine procedure may be perceived differently by patients and endoscopists. Disclosure of Interest None Declared

PTH-021 Reassessing The Value of Gastroscopy for The Investigation of Dyspepsia

Introduction The National Institute of Health and Clinical Excellence (NICE) first recommended a policy of H. pylori test and treat or empirical full dose proton pump inhibitor therapy for uninvestigated dyspepsia in 2004.1 Following an initial fall in demand for gastroscopy (OGD), there has been a 30% increase in the last five years.2 Whilst regarded as the gold standard, it is uncomfortable, costly, may require sedation and carries the risks of intubation. In an era when non-invasive tests are emerging, the role of OGD requires re-evaluation. Methods Data from consecutive OGDs performed between September 2015 and January 2016 to investigate dyspepsia was analysed. We determined the percentage of patients in whom OGD ± biopsy changed management as defined by an approach other than the non-invasive NICE recommendations.1 Results 500 patients (39.8% male; mean age 58±16.1) underwent OGD for dyspepsia, some of whom also had dysphagia (6%), anaemia (4%), vomiting (4.2%) or suspected gastrointestinal (GI) bleeding (0.6%). 145 (29%) were sedated (midazolam (mean±SEM) 2.0 mg±1.0; fentanyl (when used) 50 mcg±23). 381 patients (76.2%) had abnormal endoscopy; 417 (83.4%) had biopsies taken (15.8% for histological assessment, 27.4% for rapid urease tests, 40.2% for both). Findings of uncertain relevance, or which could have been managed with empirical therapies, were seen in 299 patients (59.8%; including oesophagitis (n = 122), hiatus hernias (n = 178), gastric polyps (n = 34), gastritis (n = 236), gastric ulcers (n = 9), gastric erosions (n = 32), duodenitis (n = 40), duodenal ulcer (n = 1), duodenal erosions (n = 15) and a duodenal polyp (n = 1)). Diagnoses which would not have been appropriately managed by empirical therapies numbered 82 (16.4%). These included 71 (14.2%) patients with Barrett’s oesophagus (n = 39), oesophageal stricture (n = 2), oesophageal cancer (n = 1) and gastric cancer (n = 4) diagnosed at the time of endoscopy. An additional 11 (2.2%) diagnoses were made solely by histology, which included eosinophilic oesophagitis (n = 1), eosinophilic gastritis (n = 1), intestinal metaplasia (n = 3) and coeliac disease (n = 6). Conclusion Diagnoses which alter management are made by endoscopy in only 14.2% of patients with dyspepsia. Although the majority of patients have biopsies taken, the added value increases the yield to only 16.4%. Non-invasive, cost-effective diagnostic strategies are needed to better guide patient management and select the minority of patients who need endoscopic biopsy or therapy. References 1 www.nice.org.uk/guidance/CG184 (2014 update) 2 https://www.england.nhs.uk/statistics Disclosure of Interest None Declared

PTH-022 A Novel Experience of The Pillcam® ESO 3 Capsule Endoscope In Patients Unwilling To Undergo Conventional Upper GI Endoscopy

Introduction Gastroscopy (OGD) is the gold standard test for investigating dyspepsia, but is uncomfortable, incurs the risk of intubation and sedation and has a diagnostic yield in our practice in 16.4% (data submitted). Capsule endoscopy is safe, non-invasive and well-tolerated. Furthermore, capsules to image the oesophagus are now used in routine clinical practice. We test the feasibility and safety of a novel protocol using the ESO3 upper gastrointestinal (GI) capsule (Given Imaging Ltd) which boasts an improved field of view and battery life compared to earlier models. Methods All consecutive patients undergoing investigation with the ESO3, between December 2015 and February 2016, were prospectively recruited. All patients required upper GI investigation but had declined conventional OGD ± barium studies. Patients were asked to drink 1 L of water containing simethicone prior to the procedure. They were then asked to swallow the ESO3 and adopt lying positions in 3 different directions at 3 angles (30° head down/up and horizontal) for 3 minutes each. Data on patient demographics, procedural data and patient tolerance was reviewed. Results 10 patients (mean age 49±17.3 years, 50% male) were included. Indications for the ESO3 included investigation for dyspepsia (n = 2), iron deficiency anaemia (n = 2), variceal screening (n = 3), suspected gastric Crohn’s disease (n = 1), subjective dysphagia (n = 1) and assessment of oesophageal ulcer healing (n = 1). Pathologies detected at CE included oesophagitis (n = 2), hiatus hernias (n = 3), oesophageal ulcers and erosions (n = 3), oesophageal varices (n = 3) and gastric and Cameron ulcers (n = 1, each). The mean time of the ESO3 transit in the oesophagus, stomach and small bowel was 12±9 secs, 58±27 mins, 23±22 mins respectively. The mean reading time for the capsule video was faster with the accelerated reading mode (39±18.4 mins) than with standard mode (79±14.6 mins) using the RAPID® software (p < 0.05). Mean post-procedural anxiety, discomfort and pain scores were excellent (scores from worst-best =0-10; 0, 0.7, 0 respectively). Visualisation (grade from worst-best =1-5) of all areas of the upper GI tract was also excellent; oesophagus (5±0), gastric cardia (5±03), fundus (4±1.1), body (5±0.6), antrum (5±0.3) and the first and second part of the duodenum (both 5±0). Complete examination to D2 was achieved in 80%. No complications were seen. Conclusion We report the first study of a novel protocol demonstrating the feasibility and safety of the ESO3. Further protocol amendments are necessary to improve reading time and visualisation while randomised control trials are needed to compare diagnostic yield to conventional OGD. However, the excellent patient tolerance affords an alternative upper GI investigative tool worth pursuing. Disclosure of Interest None Declared

PWE-053 Multicentre European Study of Double Balloon Enteroscopy in Patients with Cardiovascular Disease: A Relationship Without Heartbreak?

Introduction Double balloon enteroscopy (DBE) is a relatively invasive and lengthy procedure necessitating careful consideration of patients’ comorbidities. We aim to assess the safety of DBE in patients with cardiovascular disease (CVD). Methods Between June 2006 and January 2016, 568 consecutive patients undergoing DBE were reviewed across 3 teaching hospitals in the UK and Italy. Demographic and clinical data were collected and patients were categorised by age (elderly: ≥70 years and young: <70 years) and the presence or absence of CVD. Comparisons were made of diagnostic and therapeutic yields and complications rates. Results CVD was present in 185 patients (mean age 70±9.7, 51% male) who underwent DBE for iron deficiency anaemia (54%) overt gastrointestinal bleeding (25%), suspected Crohn’s disease (10%), small bowel strictures (6%) and suspected coeliac disease complications (4%). CVD (elderly vs young) included ischaemic heart disease (59% vs 68%, p = 0.2), valve replacement (23% vs 16%, p = 0.3), atrial fibrillation (31% vs 11%, p < 0.05) and congestive cardiac failure (21% vs 8.5%, p < 0.05). The 2 groups (elderly vs young) had similar propofol requirements (1320 mg vs 999 mg, p = 0.5), but midazolam (4 mg vs 5.5 mg, p < 0.05) and fentanyl (36.5 mcg vs 75 mcg, p < 0.05) use was less in the elderly. The most common abnormalities (elderly vs young) were ulcers (5.5% vs 10%, p = 0.4), strictures (3% vs 3%, p = 1.0), tumours (2% vs 7%, p = 0.2) and angioectasias (43% vs 27%, p < 0.05). Diagnostic yield (67% vs 64%, p = 0.8) and complication rates (5.5% vs 2%, p = 0.3) were comparable. However, therapeutic yield was higher in the elderly (50.5% vs 33%, p < 0.05). All CVD patients were compared to 383 patients without CVD (mean age 50±14.1, 44% male). Diagnostic yield was higher in those with CVD compared to those without (65% vs 50%, p < 0.05), as was therapeutic yield (42% vs 17%, p < 0.05). Irrespective of age, angioectasias were commoner in patients with CVD compared to those without (34.6% vs. 8.4%, p < 0.05). The difference seen in complications between both groups was not significant (2.7% vs 0.8%, p = 0.12). Complications seen in the CVD group included 2 cases of systemic infections, 2 cases of respiratory compromise and 1 case of myocardial infarction. Conclusion We report the first multicentre study attesting the safety of patients with CVD undergoing DBE. Moreover, patients with CVD have higher diagnostic and therapeutic yield at DBE and thus with careful selection, these patients are most likely to benefit from the procedure. Disclosure of Interest None Declared

PWE-199 Diagnostic Radiation Exposure in Patients with Crohn’S Disease Treated with Infliximab

Introduction Exposure to radiation from diagnostic imaging is thought to be associated with an increase in cancer risk. Patients with Crohn’s Disease (CD) frequently require x-ray exposure throughout the course of their illness. Studies have quantified cumulative radiation exposure in patients with CD1. However, as far as we know, there has only been one study that has identified infliximab as an independent risk factor for increased radiation exposure2. We aimed to quantify all imaging and associated radiation exposure for this subset of patients in our institution. Methods Our unit is in a district hospital serving a population of 500,000 across 2 sites. All patients with CD who have received infliximab from January 1997 to January 2013 were identified from our hospital databases. Diagnostic imaging records were also retrieved from the databases. Using local protocols, a mean effective dose for each imaging modality was calculated. The cumulative effective dose (CED) per patient was calculated as the sum of the mean effective dose for all imaging modalities. The mean diagnostic radiation exposure per year = CED/years of follow-up. Results 131 patients (52% female, mean age 46 years) were included in the study. Total number of imaging procedures was 624 (plain abdominal film = 351, CT abdomen/pelvis = 168, barium study = 105). Other imaging modalities included MRI abdomen/pelvis = 72 and abdominal ultrasound = 65. Average total number of imaging per patient was 5 (plain abdominal film = 3, CT abdomen/pelvis = 1, barium study = 1). The total CED for all imaging was 1828.1 mSv (85% attributed to CT imaging). The mean CED per patient was 14.0mSv (range 0–171.9mSv) with the mean diagnostic radiation exposure per year being 2.4mSv/year (range 0–28.7mSv/year). Conclusion Imaging requests in patients with CD is high, particularly when they have received infliximab. In spite of concerns about radiation risk, 93% of our patients have CED less than 50mSv exposure, a level considered to be associated with a low cancer risk. However, most patients are young and are likely to accumulate more radiation exposure over time. The move towards more specialised care should lead to reduction in use of CT scans and increased use of MRI and ultrasound studies. Disclosure of Interest None Declared. Reference Peloquin JM, et al. Diagnostic ionising radiation exposure in a population-based cohort of patients with inflammatory bowel disease. Am J Gastro 2008; 103:2015–22 2. Desmond AN, et al. Crohn’s disease: factors associated with exposure to high levels of diagnostic radiation. Gut 2008; 57:1524–1529