M Mendall

Advanced age influences the dynamic changes in circulating C-reactive protein following injury

Background The concentration of C-reactive protein (CRP), a biomarker of systemic inflammation, is determined by genetic, clinical and demographic factors including gender, smoking and body mass index (BMI). The influence of age on CRP dynamic changes following insult has, however, been poorly characterised. Methods We used unilateral hernia repair as a model of standardised insult to investigate the influence of baseline demographic and clinico-pathological factors affecting the dynamic changes in CRP, interleukin (IL) 6 and tumour necrosis factor-α over a time course of 48 h following injury. Results We derived CRP negativisation kinetics on 100 prospectively enrolled male subjects with mean age of 60.6 years (range 24–90 years) and mean BMI of 25.7 kg/m2 (range 17.9–37 kg/m2). Patients who failed to normalise CRP to<10 mg/l at 48 h (n=74) were significantly older (p<0.001), had longer surgical times (p=0.05), higher waist/hip ratio (p=0.02). Multiple regression analysis confirmed age as the only independent predictor of delayed CRP normalisation (p=0.03). Persistent CRP elevation was associated with higher peak CRP values (p<0.001), higher IL-6 concentrations at 24 (p=0.01) and 48 h (p=0.03). Conclusions CRP decline following insult is delayed in elderly patients as a result of unopposed IL-6 release. Age should be routinely incorporated in the assessment of CRP response to avoid misinterpretation of age-related delay in CRP clearance with ongoing systemic inflammation.

Faecal calprotectin: factors affecting levels and its potential role as a surrogate marker for risk of development of Crohn's Disease.

BackgroundFaecal calprotectin (FC) is one of the most widely used non-invasive tests for the diagnosis and assessment of Crohn’s disease (CD) activity. Despite this, factors other than disease activity which affect levels have not been extensively reviewed. This is of importance when using FC in the diagnostic setting but also may be of utility in studying the aetiology of disease.ObjectivesOur review outlines environmental risk factors that affect FC levels influencing diagnostic accuracy and how these may be associated with risk of developing CD. FC as a surrogate marker could be used to validate risk factors established in case control studies where prospective studies are not feasible. Proof of this concept is provided by our identification of obesity as being associated with elevated FC, our subsequent confirmation of obesity as risk factor for CD and the subsequent verification in prospective studies, as well as associations of lack of physical activity and dietary fibre intake with elevated FC levels and their subsequent confirmation as risk factors in prospective studies.ConclusionWe believe that FC is likely to prove a useful surrogate marker for risk of developing CD. This review has given a theoretical basis for considering the epidemiological determinants of CD which to date has been missing.

Childhood body mass index and risk of inflammatory bowel disease in adulthood: a population-based cohort study

BACKGROUND: The increasing incidence of inflammatory bowel disease (IBD) in western countries has led to the hypothesis that obesity‐related inflammation could play a role in the etiology of IBD. However, this hypothesis lacks confirmation in studies of individuals prior to the typical onset of IBD in young adulthood. METHODS: In a cohort of 316,799 individuals from the Copenhagen School Health Records Register (CSHRR), we examined whether BMI at ages 7 through 13 years was associated with later IBD. Linking the CSHRR to the Danish National Patient Register, we identified cases of Crohns disease (CD) and ulcerative colitis (UC) diagnosed during follow‐up. Cox regression was used to estimate the hazard ratios (HR) with 95% confidence intervals. RESULTS: During 10 million person‐years of follow‐up, 1500 individuals were diagnosed with CD and 2732 with UC. At all examined ages, a 1 unit increase in BMI z‐score was associated with a significantly decreased risk of UC (HRs = 0.9) and with a significantly increased risk of CD when diagnosed before age 30 (HRs = 1.2). We observed no associations between changes in BMI z‐score between 7 and 13 years and later risk of CD or UC. CONCLUSION: We found a direct association between childhood BMI and CD diagnosed before 30 years of age, and an inverse association between childhood BMI and UC irrespective of age. Our results support the previous hypotheses of obesity being a risk factor for CD, and suggest that childhood underweight might be a risk factor for UC.

PWE-015 Inflammatory Bowel Disease in The Elderly: Characteristics and Risk Factors for Disease

Introduction The number of patients presenting with IBD at older ages is increasing and the proportion of patients over the age of 60 is increasing as the IBD epidemic matures. There are only a few case series in literature concerning IBD presenting at older age ages. It is unknown whether the risk factors for disease differ compared to younger age groups. Methods A database was established at CUH in 2002 and currently has 1326 subjects registered. Information was obtained pertaining to the subjects’ age at diagnosis, exposure to established risk factors (smoking history and appendectomy), and disease profile. Their requirement for progression therapy was assessed, in terms of the need for immunosuppression, biological agents, and surgery. Patients who had required surgery for peri-anal disease only were not counted as having required surgery. Results The characteristics of the subjects are shown in the table. A higher proportion of patients diagnosed over the age of 60 had UC compared to other age groups. In all in December 2015 303/1326 were aged over 60 giving a prevalence of 22.9% of all subjects with IBD versus an incidence in the over 60 age group of 11.5%. Male predominance is present at older ages of diagnosis of CD as opposed to female preponderance at younger ages. The female predominance in UC did not change across the age groups. Smoking although more common in older CD did not reach significance unlike at younger ages. Likewise appendectomy was not a risk factor in the over 60 s for CD unlike at younger ages. Older subjects were less likely to have had surgery or be on immunomodulators or biologics. For CD colonic disease was common in older subjects.Abstract PWE-015 Table 1 Conclusion CD diagnosed in later years is phenotypically different in terms of the associated risk factors and disease trajectory. It appears to run a more indolent course, with fewer patients requiring immunosuppressive or biologic therapy, and fewer surgical procedures. This raises the question as to whether old age CD disease is a different disease to that observed in younger subjects. The behaviour of UC does not demonstrate such marked differences with age of diagnosis. Disclosure of Interest None Declared

PTH-058 Retrospective Analysis of Crohn’s Disease Risks for Further Surgery

Introduction Surgery plays an important role in the treatment of Crohn’s disease. Only a few large surgical series have been published in the literature. We describe our own experience and factors that determine risk of reoperation over a ten-year period. Methods All Crohn’s disease surgeries at St George’s Hospital from 1 st January 2004 to 31 st December 2013 were identified. A random sample of just over 200 patients was selected; not including patients less than 16 years of age at time of surgery or those who had surgery for perianal disease. Information was collected for weight, height, gender, smoking status as well as their medical and surgical management. Patients whose records were incomplete were contacted by telephone for further information. Results There were a total of 154 patients selected who had completed histories. This accounted to just over 10,000 months of follow-up. Surgeries were for bowel resections, strictureplasties or fistula repairs. 80 were male and 74 were female. Mean age of diagnosis was 23 years and 10 months. Mean time from diagnosis to first ever surgery was 5 years and 5 months. Mean time to further surgery was 31 months. 69% were non-smokers, defined as never having smoked more than 100 cigarettes in their lifetime. At the time of surgery 18% had a low BMI, 60.1% had a normal BMI and 21.9% had a high BMI. 72.72% were receiving treatment with the immunomodulators: azathioprine, 6 mercaptopurine or methotrexate. 7.14% were receiving biologic therapy at the time of surgery. 46.7% had albumin levels <35 at time of surgery and mean CRP was 27.15. Cox regression analysis was used to assess for risk for further surgery and covariates of: age of diagnosis, number of previous surgeries, time from diagnosis to first ever surgery for Crohn’s disease, treatment with immunomodulators, albumin at time of surgery, change in albumin at 1 year follow up, sex, BMI and smoking status. Significant risk for further surgery was found in those that had ever smoked and low BMI (hazard ratio 4.775 and 2.147 respectively). Importantly previous surgery, albumin and age of diagnosis were not a risk factor for recurrent surgery (hazard ratios 1.016, 0.962 and 1.198 respectively). Conclusion We review over 10,000 months of follow-up in a selection of Crohn’s disease patients. We have identified significant risks for further surgery of smoking and low BMI (probably a reflection of more severe disease). It continues to be important to address these risk factors as well as continue to try and identify other cofactors and markers that can be useful in predicting course of disease. Disclosure of Interest None Declared

PTH-183 Upper gastrointestinal cancer red flag symptoms alone except dysphagia do not predict the presence of upper gastrointestinal tract cancer: implications for practice

Introduction The epidemiology of upper gastrointestinal (UGI) cancer is rapidly changing. Current guidelines for red flag symptoms for UGI cancer were drawn up using evidence from a past age. With the exception of dysphagia, many of the symptoms are non-specific for UGI cancer in particular dyspepsia, particularly with the rapidly declining prevalence of H. pylori and its associated gastric cancers. Aims To determine the value of individual UGI cancer red flag symptoms alone or in combination in predicting the presence of UGI and non-UGI cancers. Method Our routine practice is to investigate all patients referred under the urgent suspected cancer (USC) pathway with an oesophagogastroduodenoscopy (OGD) and a computed tomography (CT) scan if over 50 years of age and symptoms are not purely oesophageal. A retrospective case review between October 2013 to March 2014 of all referrals on the USC pathway to our district general hospital with a case review of all the UGI cancers detected by our service over a one year period (October 2013–2014). Clinical follow-up after referral was a minimum of 11 months. Results 391 patients (20–96 age range) were seen under the USC pathway between October 2013–March 2014. In total 372 and 177 patients were investigated with an OGD and a CT scan respectively; 289 and 161 respectively were patients over the age of 50. 19 cancers were detected giving a detection rate of 4.75%. Of those only nine were UGI cancers (three gastric, six oesophageal). Other cancers found included three colorectal, one pancreatic, three lung, one cholangiocarcinoma, one ovarian and one peritoneal. The table shows the symptoms distribution for both UGI and non-UGI cancers. Isolated red flag symptoms had poor predictive power. Weight loss and dysphagia in combination with other symptoms had the best predictive value. There was a similar number of UGI cancers (23) and non UGI cancers (19). Analysis of all UGI cancers n = 47 over the course of one year not necessarily presenting through the USC route (October 2013–2014) revealed that isolated dyspepsia was found in only two of 47 patients compared to iron deficiency anaemia, weight loss and dysphagia which were the sole presenting symptoms in seven, four and three patients respectively, reinforcing the findings from the USC referrals.Abstract PTH-183 Figure 1 Conclusion In modern times, isolated UGI cancer red flag symptoms particularly dyspepsia are poorly predictive of UGI cancer, and equally well predict the presence of cancer outside the UGI tract. CT scanning should be a routine part of evaluation of UGI cancer referrals in the over 50’s. Disclosure of interest None Declared.

PTU-092 Crohn’s disease: strictureplasty and risks for further surgery

Introduction Surgery plays an important role in the treatment of Crohn’s disease. Strictureplasty in particular is a practical and attractive treatment for Crohn’s disease, conserving bowel length. Only a few large surgical series have been published in the literature. We describe our own experience and factors that determine risk of reoperation or death over a ten year period. Method All the Crohn’s disease strictureplasties for a specialist IBD surgeon at St George’s Hospital from 1st January 2005 to 31st December 2014 were identified. Information was collected for weight, height, gender, smoking status as well as their medical and surgical management. Patients whose records were incomplete were contacted for further information. Results There were a total of 53 cases of strictureplasty, with just under 3,400 months of follow-up. 22 had concurrent bowel resection. 32 were male. Mean age of diagnosis was 24 years and 10 months. Mean time from diagnosis to first ever surgery was 6 years and 3 months. Mean number of prior surgeries was 1. Total number of further surgery in the follow-up period was 11, with mean time to reoperation 42 months. No patients died during in the follow-up period. 86.54% were non smokers. At the time of surgery 30.25% had a low BMI, 60.45% had a normal BMI and 9.3% had a high BMI. 72.55% were receiving treatment with the immunomodulators: azathioprine, 6-mercaptopurine or methotrexate. Kaplan Meier survival analysis was used to compare those that had strictureplasties and those that had strictureplasty and concurrent bowel resection, with further surgery or death up to 31st December 2014 as the endpoint. Log rank test for trend did not find that there was any significant difference between the two groups and probability of further surgery. Cox regression analysis was used to assess for risk for further surgery: age of diagnosis, number of previous surgeries, time from diagnosis to first ever surgery, treatment with immunomodulators, concurrent bowel resection, sex, BMI and smoking status. Significant risk for further surgery was found in those that had ever smoked (hazard ratio 4.775, p = 0.003) and those patients with a low BMI who had concurrent bowel resection as well as strictureplasty (hazard ratio 6.31, p = 0.002). Importantly previous surgery and age of diagnosis were not a risk factor for recurrent surgery (hazard ratios 1.016 and 1.198 respectively). Conclusion We review a decade of strictureplasties in the management of Crohn’s disease and have identified significant risks for further surgery of smoking and low BMI as well as concurrent bowel resection. We show that strictureplasty remains a viable and attractive management for stricturing Crohn’s disease with no deaths resulting from the surgery and a low number of reoperations. Disclosure of interest None Declared.

PWE-122 Determinants Of Weight Loss Prior To Diagnosis In Inflammatory Bowel Disease: A Retrospective Observational Study

Introduction Weight loss is a recognised presenting feature of Inflammatory Bowel Disease (IBD) and considered as marker of malnutrition and disease severity. This is well established in children with IBD, however in adults, evidence based data is lacking. There is little in the literature characterising weight loss prior to formal diagnosis, or defining the disease factors associated with it. Methods Patients with IBD attending outpatient clinic were provided with a questionnaire enquiring into various aspects of their disease. Percentage Body Mass Index loss prior to diagnosis was calculated for each subject. Patients were sub-categorised into groups according to severity of % BMI loss (insignificant <5%, moderate 5–10%, severe 10–20%, extreme >20%), and disease and patient factors associated with weight loss were determined. Results 494 subjects were recruited. 237 had Crohn’s Disease (CD); 257 had Ulcerative Colitis (UC). Mean%BMI loss prior to diagnosis was greater in CD (CD 9.76% vs UC 7.63, p = 0.02). Increasing age at diagnosis was inversely associated with weight loss prior to diagnosis in UC (–0.1 per year of age, 95% CI -0.17 – -0.03, p = 0.004), and CD (-0.15 per year of age, 95% CI -0.23 – -0.06, p = 0.003). In CD, patients with prior appendectomy had reduced risk of weight loss (HR 0.38, p = 0.014). Ileal disease was a risk factor in patients with extreme weight loss. Conclusion Weight loss is a significant problem for many IBD patients at presentation, especially in younger age and Crohn’s disease with ileal involvement. Improved awareness of the presenting features of IBD should encourage wider use of malnutrition screening tools and earlier investigative tests to uncover patients at risk. Appendectomy confers a protective effect in CD. Disclosure of Interest None Declared.

PWE-162 Concurrent Computerised Tomography Can Optimise the Detection of Cancer in Patients Presenting with Unexplained Anaemia

Introduction Active investigation for gastrointestinal (GI) cancers is often triggered by “alarm symptoms”; features in the clinical presentation that may predict malignancy and warrant urgent referral. Unexplained anaemia (UA) is a highly prevalent presentation. The BSG guidelines recommend only upper GI endoscopy (OGD) and colonoscopy (COL). We investigated the additional diagnostic value of concurrent contrast enhanced computerised tomography of the chest, abdomen and pelvis (CT) in the investigation of patients (pts) aged > 50 referred to the urgent suspected cancer (USC) pathway for GI malignancies. We evaluated its accuracy in detecting upper GI, lower GI and extraluminal malignancies in a cohort of consecutive pts presenting with and without UA. Methods We retrospectively analysed characteristics and outcomes of 350 consecutive GI USC referrals (07/2010–07/2012): 200 (Group A) presented with UA and were investigated with OGD (178, 89%), COL (70, 39%) and CT (157, 78%, with 138, 87% aged > 50 years). The diagnostic outcomes were compared with a second group of 150 pts (Group B) referred with alarm symptoms (unintentional weight loss, abdominal pain, progressive dysphagia) who underwent OGD (91, 60%), COL (32, 21%) and CT (139, 93%, with 121, 89% aged > 50 years). Results Group A had a mean age of 70 years (range 22–96), 51% males, mean haemoglobin (Hb) of 10.2 (5–13.8) g/dL. Pts in Group B had a mean age of 67 years (range 20–92), 60% males, mean Hb 13.9 (11.5–17.5) g/dL. Malignancy was diagnosed in 38 (19%) Group A and 17 (12%) Group B patients (p = 0.07). The proportion of malignant cases diagnosed endoscopically was not different across the studied groups (4/176, 2% and 7/86, 8% for OGD and COL in Group A; 5/91, 6% and 3/32, 9% in Group B, p = 0.3). Conversely, the rate of incident cancers identified by CT favoured Group A (33/157, 21% vs. 10/139, 7%, p < 0.001), where 71% of the incident cancers were extraluminal and diagnosed in pts > 50 (29/33, 87%). Conclusion Concurrent CT can optimise the detection of malignancy in pts over the age of 50 referred under the GI USC pathway, with subjects presenting with unexplained anaemia achieving the greatest diagnostic benefit. We therefore propose that CT should be incorporated within the routine investigation pathway of anaemia in the over 50s. Disclosure of Interest None Declared.