Clinical Practice - Wiley Online Library

bs_bs_banner

Journal of Gastroenterology and Hepatology (2015) 30 (Suppl. 3)

Clinical Practice Investigation of comorbities associated with helicobacter pylori infection in the kingdom of Saudi Arabia DR. KHALED AL DOSSARI,1 DR. SANAA KAMAL1 1 College of Medicine Salman Bin Abdulaziz University KSA Background/Aims: Helicobacter pylori has worldwide distribution In various developing countries, more than 80% of the population is H. pylori positive, even at young ages. Pylori s detected in the majority of patients with duodenal and gastric ulcer. H. pylori has been linked to a variety of disorders such as coronary heart disease, thyroid disease and anemia. No studies in Saudi Arabia assessed the relation of Pylori to other diseases. Therefore, the aim of the current study is to investigate comorbidities associated with H. pylori infection in the Kingdom of Saudi Arabia. Patients and Methods: The current study enrolled patients presenting with dyspeptic symptoms at Salman Bin Abdel Aziz University Hospital, Al Kharj, Saudi Arabia and National Hospital in Riyadh between 2012 and 2015. Patients responded to a questionnaire to investigate possible dyspeptic symptoms. All patients had complete blood picture, liver, lipid, thyroid profile. Diagnosis of Pylori was based on serology, urea breath test, H. pylori stool ELISA test, upper endoscopy and gastric biopsy. Results: In the current study, 323 patients (174 men and 149 women) with dyspepsia and symptoms suggestive of gastritis were enrolled. Of those, 166 (51.4%) patients were Pylori positive. Dyspepsia, epigastric pain, nausea, vomiting and bloating were the frequent symptoms among Pylori positive patients. Anemia, dyslipidemia, hypothyroidism, diabetes, and hyperuricemia were more frequent among H. pylori patients. Multivariate analysis showed an association between Pylori infection and each of baseline cholesterol level, hemoglobin, diabetes and baseline TSH levels. Conclusions: H. pylori infection is frequent in Saudi Arabia. Pylori is associated with comorbidities particularly diabetes, dyslipidemia and hypothyroidism. In patients with such disorders, H.Pylori screening and management is recommended Key words: Helicobacter pylori, serology, urea breath test, stool antigen test, endoscopy Contact information: [email protected]

Acute upper gastrointestinal haemorrhage: audit of experience in a regional tertiary referral hospital S BRIENESSE, R FOSTER Gastroenterology Department, John Hunter Hospital, NSW Background: Acute upper gastrointestinal haemorrhage (AUGIH) is common and carries a high risk of morbidity and mortality. The GlasgowBlatchford score (GBS) predicts risk of requiring endoscopic intervention with increasing score. A score of 0 has previously been shown to be associated with a low likelihood of need for urgent endoscopic intervention. Management requires coordination of resuscitation and clinical assessment to facilitate early referral for endoscopy. Aim: To describe the characteristics, management and outcomes of patients presenting with acute upper gastrointestinal haemorrhage in a regional tertiary referral hospital. Methods: We performed a retrospective analysis of all patients admitted to the John Hunter Hospital with AUGIH over a 6 month period. Patient characteristics, Glasgow-Blatchford score, length of stay, use of blood products, medications, time to endoscopy and details of endoscopy (if performed), and mortality rates were reviewed.

Results: There were 198 admissions from 193 unique patients, 66% male with an average age of 67 (range 19–94). 31 (16%) patients had a history of liver cirrhosis and 159 (81%) had more than 2 comorbid diseases. The most common causes of AUGIH were ulcers 64 (32%), Mallory-Weiss tears 20 (10%) and oesophageal varices 18 (9%). Analysis of medication use revealed 72 (37%) were on aspirin, 30 (15%) were on clopidogrel or other anti-platelet agents, 29 (15%) were on NSAIDs and 28 (14%) of patients were on oral anticoagulation. 9 (5%) were identified as low risk with a GBS score of 0. Of these patients, 3 (33%) had endoscopy during the admission with findings including an oesophageal erosion, Mallory-Weiss tear, gastritis and duodenal erosions. There was no requirement for endoscopic intervention or blood transfusion in this group. 189 (95%) were identified as high risk with a GBS score >1, 101 (53%) required blood transfusions. Overall mortality was 9 (5%), the primary risk factors for death were older age (>60 years) and multiple comorbidities. Conclusion: Patients presenting with acute upper gastrointestinal haemorrhage to our facility are elderly, frequently take ulcerogenic medications and have multiple comorbidities. Ulcers are the most common cause of bleeding. Risk stratification into high (GBS score >1) and low (GBS score 0) risk groups predicts need for endoscopic intervention and blood transfusion. Our overall mortality rate is lower than in previously published clinical audits.

Hospital admission for acute diarrhoea: better off under a gastroenterologist or not? G CHEN,1 MC GRIMM1,2 1 Department of Gastroenterology and Hepatology, St George Hospital, Sydney, 2St George Clinical School, University of New South Wales, Sydney Background: Acute diarrhoea (70 year and chronic kidney disease were associated with a lower haemoglobin at presentation (p < 0.05). 31/74 (42%) required packed red cell transfusion, with a mean of 2.6 units given. 33/74 (45%) of patients had endoscopy within 24 hours of presentation. The median time to endoscopy in our cohort was 24 hours, with a mean of 36 hours. A specific cause for upper gastrointestinal bleeding was found on endoscopy in 59/74 (80%). Repeat endoscopy was required in 7/74 (9%) due to rebleed. Only one patient required surgery and the cause of bleeding was determined to be from jejunal diverticula. There was no mortality recorded. The mean length of stay in our study was 10 days. Discussion: A number of potential areas for improvement were identified. The time from presentation to endoscopic investigation was below established benchmarks [1, 2]. An explanation for this delay may be due to the lack of a proper upper gastrointestinal bleeding protocol in our emergency department to facilitate management and prompt referrals. Another reason may have been the absence of an established on-call emergency endoscopy roster.This study also identified a long length of stay in our study cohort when compared to the minimum ACG recommendation of 3 days [2]. Factors that may explain this longer length of stay include a predominantly elderly population, with majority of patients requiring extensive discharge planning; and limited access to non-emergency endoscopy lists for stable patients. Conclusion: Our regional population of patients presenting with upper gastrointestinal bleeding are predominantly elderly, and this contributes to the observed longer length of stay. Access to endoscopy lists for nonemergent patients may also play a role. The number of patients who received endoscopy within 24 hours is considerably lower than current guidelines. This identified an important area of improvement for our department, in terms of expediting the time from presentation to endoscopy. This will involve establishing an upper gastrointestinal bleed protocol and pathway for emergency department, in addition to a structured urgent endoscopy roster. Our needs for surgery and mortality rates are not significantly different from established figures. References 1. ASGE guideline: The role of endoscopy in the management of acute non-variceal upper GI bleeding. GastrointestEndosc 2012. 2. ACG practice guideline: Management of patients with ulcer bleeding. Am J Gastroenterology 2011.

Journal of Gastroenterology and Hepatology 2015; 30 (Suppl. 3): 13–26 © 2015 The Authors. Journal of Gastroenterology and Hepatology © 2015 Journal of Gastroenterology and Hepatology Foundation and Wiley Publishing Asia Pty Ltd

Clinical Practice

What keeps gastroenterologists in the public sector busy? A prospective one month snapshot audit A KHEIR,1,2,3 N KOLOSKI,4,5,6 G HOLTMANN,5,6 M WALKER,4 M VEYSEY,1,2,3 N TALLEY4,7 1 Department of Gastroenterology, Central Coast Local Health District, NSW, Australia, 2Teaching & Research Unit, Central Coast Local Health District, NSW, Australia, 3 School of Medicine & Public Health, University of Newcastle, NSW, Australia, 4Faculty of Health & Medicine, University of Newcastle, NSW, Australia, 5 Department of Gastroenterology, Princess Alexandra Hospital, QLD, Australia, 6School of Medicine, University of Queensland, TRI, QLD, Australia, 7Australian Gastrointestinal Research Alliance Background: Very little data are available about the service demand in the field of gastroenterology. However this information is important for service and workforce planning and to meet patients’ needs. We aimed to characterise patients and their presentation of disease in three public gastroenterology outpatient facilities in Australia. Methods: We prospectively assessed all patients presenting to three major public hospital gastroenterology outpatient clinics in Australia. All sites had gastroenterology and hepatology patients. Facility 1 is a tertiary referral centre, facility 2 a tertiary referral centre with specialised IBS, motility, hepatology and transplant hepatology services whereas centre 3 was a public gastroenterology clinic. Data was collected over a one month period in 2013. The number of new patients with their main presenting symptoms and the number of follow-up patients and their primary diagnosis were recorded. Results: The total number of patients seen at three sites was 1562 (274, 919, and 369, respectively), with an approximately equal distribution of males and females. Table 1 shows the number of new patients presenting to each facility. The main presenting symptoms were abdominal pain, constipation, diarrhea and gastrointestinal bleeding (Table 1). The majority of the workload at these clinics is seeing follow-up patients (Table 1). Chronic organic conditions including inflammatory bowel disease (IBD) and reflux were the main primary diagnoses seen at these clinics but iron deficiency anaemia was also common. Contrasting the high prevalence of functional GI symptoms in the population, only a relatively small proportion of these patients are seen in the tertiary setting. Conclusions: New presentations account for only 20 to 40% of all gastroenterology outpatient presentations. Pain, diarrhea and constipation are the main drivers for gastroenterology referrals followed by referrals for positive fecal occult blood test (FOBT) or rectal bleeding. In follow-up consultations, IBD requires up to 25% of review consultations across the various sites even though there is considerable variability, likely driven by differences in service profiles. Lower proportions of new patients (and increased follow-up presentations) are associated with highly specialised services (e.g. IBD/motility services or hepatology and transplant hepatology services). Whilst it is well recognised that the majority of patients with pain, diarrhea or constipation are suffering from functional gastrointestinal disorders they are not managed routinely in a specialised GI outpatient setting after initial diagnostic assessment.

Table 1. Facility Total number of consultations:

1 (n = 274)

2 (n = 919)

3 (n = 369)

New Patients (n, %) Main presenting symptom – n (%) Abdominal pain Epigastric pain Diarrhea Constipation Dysphagia Rectal bleed + Positive FOBT Other (including liver disease) Follow-up Patients (n, %) Primary GI diagnosis – n (%) Liver disease Inflammatory Bowel Disease Irritable bowel Syndrome Reflux/GORD Iron Deficiency Anaemia Dysphagia Other

110,40%

193, 21%

150,40%

6 (5.4) 6 (5.4) 3 (2.7) 3 (2.7) 10 (9.0) 21 (19.0) 61 (55.4) 164,60%

19 (9.9) 10 (5.2) 10 (5.2) 8 (4.2) 3 (1.5) 16 (8.3) 127 (65.8) 726,80%

12 (8.0) 4 (2.6) 7 (4.6) 4 (2.6) 10 (6.6) 27 (18.0) 86 (57.3) 219, 60%

42 (25.6) 19 (11.5) 4 (2.4) 10 (6.0) 1 (0.6) 2 (1.2) 86 (52.4)

281 (38.7) 164 (22.6) 10 (1.4) 31 (4.3) 61 (8.4) 22 (3.0) 157 (21.6)

42 (19.1) 54 (24.6) 9 (4.1) 8 (3.6) 0 (0) 0 (0) 106 (48.4)

Diagnostic uncertainty signalled by specialists and ongoing investigations may contribute to patient insecurity in functional gastrointestinal disorders E LINEDALE,1 A CHUR-HANSEN,1 A MIKOCKA-WALUS,2 P GIBSON,3 J ANDREWS4 1 The University of Adelaide, 2University of York, 3Monash University, 4The Royal Adelaide Hospital Background: Functional Gastrointestinal Disorders (FGIDs) affect up to 40% of the population yet are poorly handled in the healthcare system, creating frustration in patients and doctors alike. Amongst patients, fear of a missed organic diagnosis and a lack of “ownership” of a functional disorder appear to lead to repeat consultation. Amongst doctors, repeat consultation is a major factor driving the overuse of investigations. This is not an issue with “organic” gastrointestinal disease (OGID). We hypothesised that this difference between how patients with “functional” and “organic” GIDs perceive their diagnosis may stem from differences in the language used in specialist communication. Aim: The aim of this research was to explore and describe any differences in specialist communication in “organic” versus “functional” gastrointestinal disorders, as well as any differences in specialist approach to these two types of disorders. Methods: Content analysis was carried out on consecutive diagnostic letters written by treating gastroenterologists (in the RAH Gastroenterology Unit) back to the referring GP. Letters were collected until there were ∼100 letters for each patient category. In total 207 letters were analysed, 119 FGID and 108 OGID. Content analysis was carried out on consecutive diagnostic letters written by treating gastroenterologists (in the RAH Gastroenterology Unit) back to the referring GP. Letters were recruited until there were ∼100 letters for each patient category. Results: It was noted that language used was either clear/definite or “qualified” – appearing to signal a level of certainty (or lack thereof) on the part of the doctor writing the letter. One example of the difference between clear and ‘qualified’ diagnostic statements is “the patient is diagnosed with. . . .” vs. “it is possible that this patient might have. . . .”. Qualified diagnostic language was used significantly more often in letters regarding patients with FGID as compared to those with OGID (60.5% vs. 14.8%;


19

Clinical Practice

p < .001). In addition, patients with FGIDs were found to be investigated by endoscopy and colonoscopy significantly more often than those with OGIDs (79% vs. 63%; p < .05). Conclusion: There is a considerable amount of diagnostic uncertainty conveyed by the language specialist’s use in correspondence to referring doctors of patients with functional gastrointestinal disorders. This may be a contributing factor to the suboptimal patient acceptance of a functional diagnosis and be a driving factor in the use of further unwarranted endoscopic investigations.

Could a structured screening approach be the answer to the avalanche of functional gastrointestinal disorder referrals? (An interim report) E LINEDALE,1 A MIKOCKA-WALUS,2 P GIBSON,3 J ANDREWS4 1 The University of Adelaide, 2University of York, 3Monash University, 4The Royal Adelaide Hospital Background: GPs appear to have difficulty making a positive diagnosis of a functional gastrointestinal disorder (FGID). This results in a large numbers of referral to gastroenterologists due to the concern of needing investigations to avoid missing something. Whilst guidelines suggest minimal tests are required these recommendations do not appear to be generally adopted. Aim: To examine the yield of abnormal screening test results in a cohort of patients referred for GE review, previously judged likely to have a FGID and classified as non-urgent. Methods: Referrals of adults triaged as ‘non-urgent’ with suspected FGIDs were invited from the ‘routine waitlist’ of the Gastroenterology Department of an Australian public hospital between July 2013 and April 2015. Patients were screened with the following panel of blood and stool tests: full blood count, C-reactive protein, biochemistry, thyroid function tests, iron studies, coeliac serology, +/- H. pylori serology (upper GI symptoms only) and/or faecal calprotectin and elastase (lower GI symptoms). A structured, alarms-based medical history questionnaire was completed and patient symptoms classified according to the Rome III criteria for FGIDs. Information from patients with any alarm symptom and/or abnormal test was reviewed by a senior gastroenterologist (GE) and, where judged appropriate, a prompt GE appointment offered. Results: In 62 patients screened to date (36 female, aged 21–68 yrs, mean 38 yrs) 21 (324%) warranted GE review. Abnormal screening tests were detected in the majority of these cases with 7 patients returning abnormal blood tests and 11 abnormal stool tests. The most common reason for GE review was elevated faecal calprotectin (n = 9) (7 with levels >100 ug/g and 2 with intermediate levels 50–100 ug/g) and iron deficiency (n = 7), followed by abnormal faecal elastase (n = 2) and positive coeliac serology (n = 1). Only one patient had more than one abnormal test, with both iron deficiency and raised C-reactive protein. Two patients had normal screening test results but warranted GE review based on reported alarm features alone. Of note, no alarm features were identified in GP referral for these 21 patients, yet all patients reported alarms (6 with one alarm, 15 with multiple). To date, 12 of the 21 patients redirected to GE review have undergone assessment, with further investigations ordered in 10 and a diagnosis given in 4. Functional disorders were diagnosed in 3 patients (alcohol induced gastritis ± functional dyspepsia, IBS, functional abdominal pain + dietary iron deficiency), and inflammatory bowel disease in 1. Two patients were discharged from clinic without a diagnosis explicitly recorded (1 with reassurance, 1 with indefinite high fibre diet), 1 diagnosis could not be verified due to incomplete medical records and 4 have test results pending. One patient was diagnosed with ulcerative colitis whilst awaiting GE review.

20

Conclusion: A significant proportion of patients triaged as non-urgent with suspected FGIDs were subsequently found to warrant GE review after undergoing a structured panel of blood and stool tests and questionnaire. Whilst more definitive conclusions cannot be made until all patients are assessed for alternative serious diagnoses at 12 month follow up, it would seem to be worth trialling this structured screening process in primary care prior to referral.

Effectiveness of rifaximin compared with ciprofloxacin in the treatment of Traveler’s Diarrhea (TD): a meta-analysis JA MALALA,1 JA GOZUN,1 C DE CASTRO1 1 Deparment of Medicine, St. Luke’s Medical Center, Quezon City, Philippines Background: Traveler’s Diarrhea (TD) is the second leading cause of morbidity in developing countries, incapacitating travellers and disrupting valuable business and leisure excursions. Empiric treatment of TD with ciprofloxacin is common and effective. However, due to emerging issues on bacterial resistance and its potential side-effects, it is imperative to look for other drugs that may be used for TD. Aim: Here, we look at rifaximin, a drug which has no reported antibiotic resistance and has low potential for systemic side effects, as an alternative or better treatment drug to ciprofloxacin. Methodology: Prospective, randomized, double-blind studies that have a quality scale of A-B were retrieved from the NCBI PubMed (http:// www.ncbi.nlm.nih.gov/pubmed), Cochrane (http://www.cochrane.org/), Elsevier – Embase (http://www.elsevier.com/solutions/embase), Google Scholar (https://scholar.google.com.ph/), Ovid (http://www.ovid.com/site/ index.jsp) and some unpublished articles. Two reviewers independently reviewed the studies and agreed on eligibility of the trials. 2/3 trials (Dupont 2001 & Taylor 2006, n = 486) were included. Trials were combined using a random-effects model. Analyses were performed using Cochrane Review Manager version 5.0. Results: There is treatment of no effect with rifaximin compared to ciprofloxacin with evidence of significant heterogeneity in Time of Last Unformed Stools (TLUS) [2 trials, n = 485; MD 2.04, 95% CI (−0.41 to 4.48)] and Total # of Unformed Stools (TUS) [(2 trials, n = 485; MD 1.27, 95% CI (−1.38 to 3.91)]. Possible source of heterogeneity is attrition bias. Subgroup analysis for Bacteriologic Cure showed treatment favoring rifaximin [2 trials, n = 324; OR 0.39, 95% CI (0.20 to 0.76)] with no evidence of heterogeneity. Rifaximin has no effect of treatment in Wellness [2 trials, n = 485; OR 0.91, 95% CI (0.56 to 1.47)] and in Adverse Experience [2 trials, n = 488; OR 1.02, 95% CI (0.68 to 1.53)], both without evidence of heterogeneity. Conclusions: These findings suggest that rifaximin and ciprofloxacin may be equally efficacious based on TLUS and TUS. However, the results may still be inconclusive due to the significant heterogeneity. Wellness may be equally efficacious between rifaximin and ciprofloxacin due to the use of standard treatment in both groups such as hydration. However, Bacteriologic Cure is superior in patients who took rifaximin, which may be explained by its low incidence of bacterial resistance.


Clinical Practice

Does a low fodmap diet reduce symptoms associated with functional gastrointestinal disorders? – A meta-analysis A MARSH,1 E ESLICK,1 G ESLICK1 1 The Whiteley-Martin Research Centre, Discipline of Surgery, The Sydney Medical School Nepean, Penrith, New South Wales, Australia Introduction: Functional gastrointestinal symptoms such as abdominal pain, bloating, constipation, diarrhea and flatulence are common in patients with Irritable Bowel Syndrome (IBS) or Inflammatory Bowel Disease (IBD). The diversity of symptoms presenting has meant finding an effective treatment challenging with most treatments alleviating only the primary symptom. A novel treatment option for IBS and IBD currently generating much excitement is the low FODMAP (Fermentable, Oligo-, Di-, Mono-saccharides And Polyols) diet. Objective: To determine the evidence of the efficacy of a low FODMAP diet in the treatment of such symptoms. Methods: Electronic databases were searched. Pooled odds ratios (ORs) and 95% confidence intervals were calculated for the effect of a low FODMAP diet in the reduction of IBS Symptom Severity Score (SSS) and increase in IBS Quality of Life (QOL) score for both randomized clinical trials (RCTs) and non-randomized interventions using a random-effects model. Results: Six RCTs and 16 non-randomized interventions were included in the analysis. There was a significant decrease in IBS SSS scores for those individuals on a low FODMAP diet in both the RCTs (OR: 0.44, 95% CI: 0.25–0.76) and non-randomized interventions (OR: 0.03, 95% CI: 0.01– 0.20). In addition, there was a significant improvement in the IBS QOL score for RCTs (OR: 1.84, 95% CI: 1.12–3.03) and for non-randomized interventions (OR: 3.18, 95% CI: 1.60–6.31). Moreover, following a low FODMAP diet was found to significantly reduce symptom severity for abdominal pain (OR: 1.81, 95% CI: 1.13–2.88), bloating (OR: 1.75, 95% CI: 1.07–2.87) and overall symptoms (OR: 1.81, 95% CI: 1.11–2.95) in the RCTs. In the non-randomized interventions similar findings were observed. Conclusion: This meta-analysis supports the efficacy of a low FODMAP diet in the treatment of functional gastrointestinal symptoms. Further research should ensure studies include an assessment of dietary adherence as part of a low FODMAP diet.

Gastroenterologist perceptions of faecal microbiota transplantation S PARAMSOTHY,1,3,4,6 A WALSH,1 T BORODY,2 D SAMUEL,3 J VAN DEN BOGAERDE,4 R LEONG,3 S CONNOR,5 W NG,5 H MITCHELL,6 N KAAKOUSH,6 M KAMM7 1 Gastroenterology and Hepatology Department, St Vincent’s Hospital, Sydney, Australia, 2Centre for Digestive Diseases, Sydney, Australia, 3Gastroenterology and Hepatology Department, Bankstown-Lidcombe Hospital, Australia, 4Gastroenterology and Hepatology Department, Nambour General Hospital, Nambour, Australia, 5 Gastroenterology and Hepatology Department, Liverpool Hospital, Sydney, Australia, 6School of Biotechnology and Biomolecular Sciences, University of New South Wales, Australia, 7Gastroenterology and Hepatology Department, St Vincent’s Hospital Melbourne, Australia Background: While interest in faecal microbiota transplantation (FMT) has grown markedly, its use in clinical practice remains contentious. Controversy exists regarding potential indications for FMT, efficacy and safety,

and the current evidence base supporting FMT use. We thus decided to conduct a pilot study to gauge gastroenterologist perceptions towards FMT. Methods: A pilot survey of 17 questions was anonymously distributed amongst Australian gastroenterologists between October 2013 and April 2014 to assess their experiences and opinions towards FMT. Results: Fifty two clinicians participated, consisting of 20 general gastroenterologists (38.5%), 12 IBD subspecialists (23%), 10 trainee gastroenterologists (19%), 4 general physicians with a gastroenterology interest (7.5%), 3 hepatology subspecialists (6%) and 3 endoscopy subspecialists (6%). 21% had previously referred a patient for FMT, only 7.5% more than once. 90.5% would refer patients with Clostridium difficile infection (CDI) for FMT if easily available. 36.5% would refer patients with ulcerative colitis for FMT if easily available, 13.5% for Crohn’s disease and 6% for irritable bowel syndrome. 6% would not refer for FMT for any indication, including recurrent CDI. 86.5% would enroll their patients in clinical trials assessing FMT. 36.5% thought the optimal mode of FMT administration was transcolonoscopic, 17.5% nasoduodenal, 13.5% enema and 7.5% oral capsule while 21% did not have an opinion. When asked about their greatest concern regarding FMT, 42% said lack of evidence, 11.5% infection risk from donor stool despite screening, 9.5% non infectious adverse reaction and lack of safety data, 9.5% “yuck” factor, 9.5% lack of efficacy, 6% had no concerns regarding FMT, 4% said disease exacerbation while 2% stated inappropriate usage. 32% did not think FMT should be available for routine clinical use. 77% believed there is a lack of availability / accessibility to FMT while 52% had an interest in learning how to process and administer FMT. 79% felt only a few select centres that satisfy appropriate regulatory requirements should be available in any region to offer FMT. Only 6% of respondents were already offering FMT as a therapeutic option at their practice / institution. Conclusions: Despite general enthusiasm, currently the majority of gastroenterologists have limited or no experience with FMT. Accessibility to FMT appears to be a significant issue. The greatest concerns regarding FMT were a lack of evidence, followed by safety issues due to infectious and non infectious adverse events. Almost one third felt FMT should not be available for routine clinical use even for recurrent CDI. Conversely, a surprising proportion of gastroenterologists reported they would refer patients for FMT for indications other than CDI despite insufficient evidence to recommend routine clinical use.

Gastrointestinal outpatients referral quality: safe to use? MA SHAHZAD,1 E LINEDALE,2 JM ANDREWS,1 AM WALUS3 1 Royal Adelaide Hospital, 2University of Adelaide, 3 University of York Background: Up to a quarter of all gastroenterology ambulatory referrals are for Functional Gastrointestinal Disorders (FGIDs) despite the fact that these can be positively diagnosed without invasive investigations and managed in primary care in most instances. The safety of referral triage relies on complete and accurate referrals. Aim: To assess the quality of General Practitioner (GP) referrals to a large public teaching hospital for patients with probable FGIDs as verified by structured patient questionnaires. Methods: As part of an ongoing study to evaluate a new model of care in patients with likely FGID, referrals received between July 2013 and April 2015 were examined. Discrepancies between patient responses to a structured intake questionnaire and GP referrals became apparent and an interim review of the quality of GP referrals and their concordance with patient reports was conducted. Results: To date questionnaire responses from 68 patients (42 females) and the corresponding GP referrals have been analysed. Communication of


21

Clinical Practice

basic patient information in GP referrals was poor with 5/68 referrals being illegible, and age and gender not explicitly stated in 36 and 25 respectively. Smoking/alcohol history was specified in only 3 referrals and current medications not provided in 14. The majority of referrals did not provide a clear reason for the referral (n = 49), and less than a third mentioned a provisional diagnosis (n = 22). Symptom duration was mentioned in less than half of referrals (n = 32) with only 14 agreeing with the symptom duration from patient questionnaires. Alarm features were poorly addressed in the majority of referrals. The presence or absence of alarm features was not mentioned in 42 referrals, 6 stated ‘no alarm features’ without providing further detail and 7 stated ‘no rectal bleeding and no alarms’. The most commonly reported alarm feature was rectal bleeding (n = 17, present in 6) followed by weight loss (n = 11, present in 1), nocturnal symptoms (n = 3, present in 1), fever (n = 2, present in 1) and haematemesis (n = 1, absent). However, in the structured patient questionnaire the majority of patients (n = 61) reported the presence of alarm features. ‘Nocturnal symptoms’ was most commonly reported alarm (n = 51) followed by rectal bleeding (n = 20), unexplained weight loss of >5 kgs (n = 15), unexplained fever (n = 12) and haematemesis (n = 3). Recent onset of symptoms in patients over 50 years of age was also poorly recorded with 9/17 referrals failing to report on symptom onset and only half of those stating recent symptom onset in agreement with the patient questionnaire (2/4). Of concern, was the large number of GP referrals that under-reported relevant family history. Only 5 referrals identified positive relevant family history as compared with 20 patient questionnaires. Conclusion: The quality of these GP referrals is well below that required to accurately triage into urgency categories. It is proposed that the routine use of a “customised referral form” be evaluated to ensure quality of patient care.

Bariatric Surgery and the Gastroenterologist ML TALBOT UNSW Department of Upper GI Surgery. St George Hospital, Sydney Background: Bariatric Surgery changes foregut anatomy in order to alter the physiology of eating. While weight-loss may lead to health improvements in some domains, this disruption of normal anatomy and physiology can lead to the development of gastrointestinal disease. Gastroenterologists managing these patients need use a range of diagnostic modalities to best characterise the effects the altered anatomy is having on a patient’s function. Methods: Description of the anatomy and physiology of current and historic bariatric procedures with a focus not just on endoscopy and barium studies, but also newer modalities such as 3D CT and oesophageal physiology testing. Common endoscopic findings and interventions required when patients present with symptoms will be discussed. More complex phenomena such as hypergastrinaemia and hyperinsulinaemia and their management will be reviewed. Results: Diagnostic endoscopy has great utility in patients with symptoms after bariatric surgery, but a keen understanding of surgical anatomy is required. Therapeutic endoscopy techniques including dilation and management of leaks and fistulae need to be anticipated. 3D CT and oesophageal physiology testing are required for foregut symptoms that fail to resolve after endoscopic evaluation and some patients will also require evaluation of humoral factors in order make an appropriate diagnosis. Discussion: Bariatric surgery leads to predictable patterns of altered physiology that can lead to development of disease requiring gastroenterological diagnosis and therapy. Managing the patient requires understanding of the chronic nature of their obesity condition and obesity therapy. Apart from for post-surgical stricture, endoscopy alone is almost never sufficient to provide enough information to guide therapy.

22

Fecal microbiota transplantation for treatment of slow transit constipation: a prospective open-label study HL TIAN, C DING, XL GE, JF GONG, N LI Jinling Hospital, Research Institute of General Surgery, Nanjing University School of Medicine, Nanjing, Jiangsu, China Background: There were some cases reported bacteriotherapy for constipation in 1990s1. There are no systematic large studies to evaluate the efficacy of this treatment. In the present study, we investigated the use of fecal microbiota transplantation (FMT) to provide evidence for slowtransit constipation (STC) patients. Methods: Fifteen patients with STC, aged 22 to 72 were enrolled in this prospective open-label study. Patients received FMT on three consecutive days via nasojejunal tube with follow-up at 1,2,4 and 8 weeks after treatment. Rate of clinical improvement and remission, wexner constipation scale, and bowel movement per week were evaluated at each study visit. Results: The rate of clinical improvement and remission based on clinical activity at 8 weeks was 53.33% (8/15) and 40% (6/15) respectively. The patient’s stool frequency increased from a mean of 1.23 (SD, 1.76) per week pre-FMT to 4.58 (SD, 3.25) at 8 weeks postoperatively without laxative-use (p < 0.001). Meanwhile, comparison of pre-FMT and postFMT Wexner constipation scores 2 demonstrated a significant reduction between baseline and 1 week (8.60 ± 3.73), 2 weeks (7.48 ± 2.43), 4 weeks (7.35 ± 4.52) which was maintained up to the 8 weeks follow-up (7.32 ± 2.35; p < 0.001). Compared with baseline, significant overall improvements were also seen in PAC-QOL satisfaction scores 3 and gastrointestinal quality of life 4 at 1,2,4, and 8 weeks of follow-up (p < 0.01). There were no severe adverse events during the whole FMT procedure follow-up except for venting(6/15), abdominal pain(3/15), bloating(2/15) and diarrhea(2/15). Conclusions: This is a pilot study with the largest sample of patients to demonstrate that FMT has the potential to be somehow of help in managing patients with STC but considerable further efforts are necessary to maintain a long-term effect. Despite the small number of patients studied, the marked clinical improvement following fecal bacteriotherapy warrants further study to elucidate the contribution of the fecal flora in the etiology of constipation. References 1. Andrews PJ, Barnes P, Borody TJ. Chronic constipation reversed by restoration of the bowel flora.A case and a hypothesis. European J Gastroenterol Hepatol 1992;4:245–7. 2. Agachan F, Chen T, Pfeifer J, et al. A constipation scoring system to simplify evaluation and management of constipated patients. Dis Colon Rectum 1996;39:681–5. 3. Marquis P, De La Loge C, Dubois D, et al. Development and validation of the Patient Assessment of Constipation Quality of Life questionnaire. Scand J Gastroenterol 2005; 40:540–51. 4. Eypasch E, Williams JI, Wood-Dauphinee S, et al. Gastrointestinal Quality of Life Index: development, validation and application of a new instrument. Br J Surg 1995;82:216–222.

Impact of introduction of Fibroscan® on the change in management of patients with Chronic Hepatitis C (CHC) infection R UCHILA,1 N KONTORINIS,1 J KONG,1 S NAZARETH,1 R TUMA,1 SLCHEN,1 W CHENG1 1 Royal Perth Hospital, Perth, WA Background: Liver stiffness measurements (LSM) using Fibroscan has been used to measure liver fibrosis and is highly reproducible and allows


Clinical Practice

evaluation of liver stiffness, which is strongly correlated with liver fibrosis stages, assessed by simultaneous biopsies. Aims: To examine the impact of Fibroscan on the change in management of chronic Hepatitis C infection since its recent introduction at our institute. Methods: Retrospective analysis of 270 CHC patients who underwent Fibroscan between June 2013 and February 2015. LSM were used to prioritize anti-viral treatment, hepatocellular carcinoma (HCC) surveillance and clinical follow up. Advanced fibrosis and cirrhosis had been based on clinical, radiological and hepascore (>0.84) assessments. The distribution of patients is depicted in Fig 1.

Fig1.

Distribution of patient by fibrosis and treatment status

Results: The majority of the patients, 162 of the 270 (60%) patients were treatment naïve. In 11 of the 56 (20%) untreated F3-F4 patients, advanced fibrosis was not known prior and these patients were prioritized to commence treatment. Thirteen out of the 96 (13%) patients with fibrosis ≥F3 in whom surveillance for HCC was commenced was not detected prior to fibroscan. LSM was performed in 32 patients with F0-F2 fibrosis who had achieved SVR following treatment. Of these patients 31% who had high hepascore could be discharged without the need for HCC surveillance. Conclusion: 1) Fibroscan is a useful tool in detecting advanced fibrosis to prioritize treatment for patients with CHC (2) High LSM allowed the detection of additional 13% of patients with advanced fibrosis and cirrhosis for HCC surveillance (3) Patients who had achieved SVR and previously considered to have advanced fibrosis for hepatocellular carcinoma surveillance may be discharged following fibroscan. (4) Fibroscan plays an important role in the clinical practice to both help prioritize and initiate anti-viral therapy as well as to guide follows up.

An exploratory study into the unmet supportive needs of people diagnosed with cirrhosis in Queensland, Australia PC VALERY,1,2 PJ CLARK,1,2,3 SM MCPHAIL,4,5 J MARTIN,6,7 L HORSFALL,3,8 ML VOLK,9 E POWELL3,8 1 QIMR Berghofer Medical Research Institute, Brisbane, Qld, Australia, 2School of Medicine, University of Queensland, Brisbane, Qld, Australia, 3Princess Alexandra Hospital, Brisbane, Qld, Australia, 4Institute of Health and Biomedical Innovation and School of Public Health & Social Work, Queensland University of Technology, Brisbane, Qld, Australia, 5Centre for Functioning and Health Research, Metro South Health, Brisbane, Qld, Australia, 6School of Medicine and Public Health, University of Newcastle, New Castle, NSW, Australia, 7 Diamantina Institute, University of Queensland, Brisbane, Qld, Australia, 8The Centre for Liver Disease Research, University of Queensland, Brisbane, Qld, Australia, 9Loma Linda University, Loma Linda, CA, USA Introduction: People with chronic liver disease, particularly those with decompensated cirrhosis, must follow a complex and variable regimen of dietary restrictions, medications, laboratory testing, and clinic visits. In addition they frequently suffer debilitating complications that impact on quality of life and activities of daily living. These impairments combined with complex treatment means that patients may be faced with high levels of supportive care needs. Methods: This study explores the concerns and challenges of people diagnosed with cirrhosis attending a tertiary hospital hepatology clinic. Self-completed structured interview including close-ended and openended questions was used to collect data for this investigation. Patients were initially asked in an open-ended manner about ‘what is the most significant challenge or problem for you now that you have liver disease’ (‘volunteered concern’). Patients were then provided with a list of 10 potential concerns, asked to prioritize five, and to identify those that were of utmost importance to them (ranking items from 1 = most important to 5 = least important). Results: Fifty three consecutive English speaking patients attending the hepatology clinic who had been identified as having cirrhosis were invited to participate; 50 (94%) were interviewed, and 3 declined participation. Participants were aged 39–90 years (mean 58 years, SD = 10.2). The majority were male (78%), Caucasian (86%), were Australian born (72%), spoke English at home (96%), had a partner (50%), and were confident with written English (96%). Fifty-six percent had compensated cirrhosis, 36% had decompensated cirrhosis and 8% had hepatocellular carcinoma. The most common concerns that were volunteered by patients included: fatigue, alcohol/control drinking, eating/losing weight, curing their liver disease, and reclusive/not socialising. Of the 33 who answered the rank questions correctly, the most common prioritized concerns were: ‘developing liver cancer’ (79%), followed by ‘Losing ability to do daily tasks for yourself’ (76%), ‘Fear of dying’ (64%), ‘Fear of the unknown – what the future holds’ (64%), and ‘Lack of treatments for cirrhosis’ (55%). When the results were analysed by mean ranking scores, the highest priority scores were for ‘developing cancer’ [mean 2.5, SD = 1.3] followed by ‘fear of dying’ (2.6, SD = 1.6). Regarding use of community support services, 24% reported accessing a dietician, 20% a pharmacist, and 18% a psychologist. Of 21 patients ranking ‘fear of dying’ as one of their top five concerns, 7 reported having accessed a psychologist. Of 5 patients who volunteered a concern about eating/losing weight, one reported having accessed a dietician. Most patients reported having someone they can depend on for assistance if they really need it, 90% have someone they can talk to about important decisions in life, 86% reported they would see a GP if concerned about


23

Clinical Practice

their cirrhosis, and 78% feel confident that they can control their symptoms of cirrhosis. One participant (2%) had an advance health directive in place, 9 (18%) did not have one, and 40 (80%) didn’t know what that was, and 26% have an enduring power of attorney in place. Thirty two percent were in the workforce, 60% were unemployed (28% on Centrelink support and 32% on disability support), and 8% were either on home duties or retired. Of those working (n = 16), 81% need to take time off work for treatment or review of their cirrhosis. Sixty nine percent have to travel more than 30 minutes to get treatment. Overall, 56% percent of patients worry about their health; 73% of those with decompensated cirrhosis worry about their health compared to 43% of those with compensated cirrhosis (p = 0.035). Conclusions: Our results provide insight about the many concerns and challenges of people diagnosed with cirrhosis, and their use of available support services. Our pilot data suggest that cirrhosis patients not only worry about health concerns, but also about the impact of their disease on societal functioning and interaction – concerns which the medical system is poorly equipped to address. Further research into the supportive care needs of this patient group, particularly the small subgroup unable to rank potential concerns or unwilling to complete the survey, is essential to inform service development, particularly around multidisciplinary care to address the clinical and psychosocial needs of this patient group.

Retrospective audit of time delay from positive FOBT to colonoscopy in a teaching hospital in South West Sydney G WARK, S AL-SOHAILY, S FULHAM, I TURNER Campbelltown Hospital, Sydney, Australia University of Western Sydney, Sydney, Australia Background: Colorectal cancer is the second leading cause of cancer deaths in Australia. According to NHMRC data a person with a positive FOBT has a 30–45% chance of having an adenoma and a 3–10% chance of colorectal cancer. When a patient has a positive FOBT result, prompt colonoscopy enables timely diagnosis and thus treatment of disease at the earliest possible stage giving the patient the best prognosis. The benchmark for colonoscopy after positive FOBT is within thirty days (NSW Ministry of Health National Definition for Elective Surgery Categories, 2012). We are concerned that waiting times far exceed this in the public hospital setting, with possible adverse consequences and prolonged patient anxiety about the result. Aims: To assess the current public hospital waiting time from positive FOBT to colonoscopy in a busy outer suburban public hospital. Methods: This retrospective audit was conducted in a 350 bed public hospital in South West Sydney. Patients were included if they had undergone colonoscopy for the indication of positive FOBT between 01/12/2013 and 31/03/15 as indicated on the electronic colonoscopy report using ProVation endoscopy reporting software. This search identified 240 patients however 14 patients were then excluded as on review of the clinical file the true indication was for PR bleeding rather than positive FOBT. There were a total of 226 patients with a median age of 61 (range 26–87), 54% were male. Patient files were then reviewed in the hospital or in the endoscopists outpatient rooms for date of positive FOBT test, date of referral from GP, date seen by endoscopist and date of colonoscopy. Results: The median waiting time from positive FOBT result to colonoscopy was 120 days. There was a median time from positive FOBT to GP consultation of 8 days, GP referral to review by endoscopist of 22 days and time from review by endoscopist to date of colonoscopy of 78 days. No data about the waiting time was available for 41 patients. 18% of the total cohort was from a non-English speaking background and the median wait time for this group was 127 days which was not statistically longer than for people of English speaking backgrounds.

24

Discussion: The main delays to colonoscopy after positive FOBT appear to be the time from GP referral to review by endoscopist and especially time from review by endoscopist to the colonoscopy being performed. The time delay from positive FOBT to colonoscopy far exceeds the benchmark target of 30 days. Reasons for the delays may include the financial burden of consultation fee, delay in submitting paperwork, specialists’ high workload, excessive endoscopy of patients with functional symptoms and insufficient public endoscopy lists to serve the demand. The delay may be improved by the implementation of an open access public hospital colonoscopy program to streamline colonoscopy booking and to reduce the financial burden of socioeconomic disadvantage.

What is the utility of performing gastroscopy in patients with positive FOBT without upper GI symptoms, iron deficiency or anaemia? G WARK, I TURNER, S AL-SOHAILY Campbelltown Hospital, Sydney, Australia University of Western Sydney, Sydney, Australia Background: A large number of patients who undergo colonoscopy for FOBT do not have significant pathology to explain the positive result. Gastroscopy is often performed as well as colonoscopy to exclude significant upper GI pathology as a cause of the positive FOBT. The yield of performing gastroscopy in a patient with a positive FOBT in the absence of upper GI symptoms or iron deficiency anaemia is uncertain. Aims: To assess the frequency of clinically significant upper GI lesions in patients who undergo gastroscopy for the indication of positive FOBT compared with those who undergo gastroscopy for anaemia or upper GI symptoms in addition to the finding of positive FOBT. Methods: This retrospective audit was conducted in a 350 bed public teaching hospital in South West Sydney. Patients were included if they had undergone colonoscopy for the indication of positive FOBT between 01/12/2013 and 31/03/15 as indicated on the electronic report using ProVation endoscopy reporting software. This search initially identified 240 patients however 14 patients were then excluded as, on review of the clinical file the true indication was not for positive FOBT. There were 226 patients with mean age of 61 (Range 26–87), 54% were male. Of this cohort one patient’s file was unable to be accessed leaving 225 patients. Of this group 44% (100/225) of patients underwent gastroscopy at the same time of colonoscopy. 45% (45/100) of these were for positive FOBT only, 21% (21/100) were performed to investigate iron deficiency and/or anaemia and 34% (34/100) were indicated in the basis of upper GI symptoms. Results: Of the patients who underwent gastroscopy for positive FOBT, 24% (11/45) were found to have significant pathology (severe gastritis, ≥ grade B reflux oesophagitis, gastric ulcers, Barrett’s oesophagus, upper GI adenomas or gastric cancer). Significant pathology was identified in 24% (5/21) of patients who underwent the procedure to investigate iron deficiency and/or anaemia and in 32% (11/34) patients who had upper GI symptoms. The mean age for patients in whom significant pathology was detected was 62 compared to the mean age in which no significant pathology was detected being 63 which was not statistically significant. Conclusion: If colonoscopy is normal in an asymptomatic patient in the absence of iron deficiency anaemia then consideration of a gastroscopy during the same anaesthetic is worthwhile as it would add little to the total cost and significant pathology may be detected in up to a quarter of patients, which is a similar incidence to patients undergoing gastroscopy to also investigate iron deficiency, anaemia and upper GI symptoms. This is an area that requires further investigation in the form of prospective trials to further assess the utility of gastroscopy in this setting.


Clinical Practice

A retrospective audit of quality performance measures and identified pathology in colonoscopy for positive FOBT at an outer suburban hospital in South West Sydney G WARK, S AL-SOHAILY, S FULHAM, I TURNER Campbelltown Hospital, Sydney, Australia University of Western Sydney, Sydney, Australia Background: The effectiveness of colonoscopy as a screening tool for advanced adenoma and colorectal cancer in patients with positive FOBT depends on the quality performance of the procedure. This is of particular importance in detection and removal of adenomas from the colon to reduce future cancer risk. The American Society of Gastroenterology (ASGE) Guidelines for quality in performance of colonoscopy suggest the benchmarks of; documentation of adequacy of bowel preparation in >98%, adequate bowel preparation in >85% cases, minimum caecal intubation rates with photo documentation of 90%, recording of withdrawal time in 98% of cases and adenoma detection rates of >25%. Aims: Firstly, to identify the incidence and location of significant pathology and to measure the impact of bowel preparation on the identification of significant pathology in patients who underwent colonoscopy after a positive FOBT. Secondly, to compare our results with the available ASGE benchmarks for quality performance in colonoscopy. Method: A retrospective audit was conducted in a 350 bed public hospital in South West Sydney. Patients were included if they had undergone colonoscopy for the indication of positive FOBT between 01/12/2013 and 31/03/15 as indicated on the electronic report using ProVation endoscopy reporting software. This search initially identified 240 patients however 14 patients were excluded as review of the clinical file showed the true indication was not for positive FOBT. There were a total of 226 patients with a median age of 61(range 26–87), 54% were male. The reports were audited for presence of record of bowel preparation, quality of bowel preparation, caecal intubation rate, photo documentation of completion and documentation of withdrawal time. Pathology and its location were also recorded. Three medical records could not be accessed so the cohort comprised 223 patients. Results: Of the 223 patients audited the quality of bowel preparation was reported 99% (220/223) of the time, with adequate bowel preparation in 94% (207/220). The caecum or terminal ileum was reached in 99% (220/ 223), however photo documentation was recorded in only 72% (160/223) of cases and withdrawal time was not recorded in any cases. The ADR was 53 % (119/223) with multiple ADR (ADR2+ number of patients with multiple adenomas/number of patients with adenomas) of 54% (64/119). Colonoscopy found no significant pathology (normal, small or hyperplasic polyps, diverticulosis, angioectasia or haemorrhoids) in 49% (110/223) of patients, low risk adenoma in 16%(36/223) of patients, high risk adenoma (polyp ≥1 cm, villous change, high grade dysplasia or ≥3 adenomas) in 31% (68/223) of patients and carcinoma in 4% (9/223) of patients. Males had a greater number of high risk adenomas or carcinomas than females (43% vs 25%) This reached statistical significance (p = 0.001). An index lesion (polyp ≥1 cm, villous change, high grade dysplasia or carcinoma) was found in 27% (61/223) of cases. In 62% (38/61) of cases the index lesion was in the left colon and the index lesion was in the right colon in 38% (23/61). This difference did not reach statistical significance. Patients with fair bowel preparation had comparable ADR’s compared with patients with good bowel preparation (62% vs 46%). Poor bowel preparation reduced the detection rate of significant pathology. Detection rates of high risk adenomas or carcinoma with poor bowel prep was 38% (5/13) compared to adequate bowel prep of 51% (105/207). This was not statistically significant. Conclusion: Although ADR is well above the ASGE guideline of 25% minimum, probably related to the obesity prevalence, the quality performance of colonoscopy could further be optimised by increased photo documentation of completion and recording of withdrawal time. The lower

ADR after poor bowel preparation confirms that patients with poor bowel preparation should undergo repeat colonoscopy within 1 year. The method with which this audit was conducted highlights the utility of electronic reporting tools in facilitating the performance of clinical audits in order to improve quality of practice.

Ten-days bismuth-containing quadruple therapy for Helicobacter pylori eradication in diabetic patients: a single center pilot study CT WU,1,2 YJ YANG,3 HC CHENG,2 WI CHANG,1,2 HY KUO,1,2 HY OU,2 CT WU,1 HB YANG,4 BS SHEU1,2 1 Institute of Clinical Medicine, 2Department of Internal Medicine, 3Department of Pediatrics, National Cheng Kung University Hospital, College of Medicine, National Cheng Kung University, Tainan, Taiwan, 4Department of Pathology, Ton-Yen General Hospital, Hsin-Chu, Taiwan Background and aims: International consensus recommended 80% as the goal of Helicobacter pylori eradication rate. Clarithromycin-based therapy is prescribed universally, but its efficacy is limited when clarithromycin resistance reaches more than 15–20%. Current clarithromycin-based eradication rate of 50–62% in diabetic patient is still unsatisfactory. We aimed to investigate the efficacy of 10-day bismuthcontaining quadruple therapy for Helicobacter pylori eradication in diabetic patients. Methods: Patients with positive Helicobacter pylori IgG serology and C13 urea breath test (C13-UBT) were assumed to have current infection. After upper endoscopic gastric tissue biopsy and bacterial culture, all patients received 10-days bismuth-containing quadruple therapy (pantoprazole 40 mg, bismuth subcitrate 600 mg, metronidazole 500 mg all twice daily, and tetracycline 500 mg 4 times daily), followed by C13-UBT again to confirm eradication result. Non-diabetic H. pyloriinfected patients previously been treated by clarithromycin-based regimen were selected retrospectively as comparison group, which could stand for current therapeutic effect in general non-diabetic population. The primary outcome was to compare eradication rate between the two groups. Results: There were 34 and 100 patients enrolled in diabetic and nondiabetic groups respectively. It was not significantly different between the two groups in demographic and pathological characteristics. Clarithromycin had marginally higher resistance in diabetic group (28.6%) than it was in non-diabetics (12.1%), despite no statistical significance (p = .11). Thirty patients completed follow-up C13-UBT. H. pylori eradication efficacy was non-inferior in the bismuth-containing quadruple therapy diabetic group than it was in the clarithromycin-based non-diabetic group in both the intention-to-treat analysis (82.4% V.S. 93.0%; odds ratio 0.35 [95% CI 0.11–1.13]; p = .09) and per-protocol analysis (93.3% V.S. 93.0%; odds ratio 1.05 [95% CI 0.21–5.36]; p = 1.00). Conclusions: Ten-days bismuth-containing quadruple therapy may serve as an effective regimen to achieve the ideal H. pylori eradication rate in diabetic patients. Table 1

Antimicrobial resistance

Levofloxacin (%) Metronidazole (%) Clarithromycin (%) Tetracycline (%) Amoxicillin (%) Chi-square test

Diabetic n = 28

Non-diabetic n = 33

p

6 4 8 0 0

5 9 4 0 0

.53 .22 .11 – –

(21.4) (14.3) (28.6) (0.0) (0.0)

(15.2) (27.3) (12.1) (0.0) (0.0)


25

Clinical Practice

Table 2

Outcomes

Eradication rate Intention to treat (%) Per protocol (%) Fisher’s exact test

26

Diabetic n = 34

Non-diabetic n = 100

Odds ratio

p

28/34 (82.4)

93/100 (93.0)

0.35 (0.11–1.13)

.09

28/30 (93.3)

93/100 (93.0)

1.05 (0.21–5.36)

1.00
