Search

Introduction Tofacitinib is an oral partially selective Janus kinase inhibitor approved for the treatment of refractory moderate to severe ulcerative colitis (UC). Real-world experience of patients with UC treated with tofacitinib is however limited, and safety concerns over the risk of venous thromboembolism (VTE) have recently emerged. Further, factors linked to primary non-response remain poorly defined. We therefore sought to define the effectiveness and safety of tofacitinib in a real-world cohort. Methods We conducted a retrospective observational cohort study of 134 patients with UC (64% male; median age 37 years [range 16–81]; 83% patients had previously received at least one biologic) treated with tofacitinib from October 2018 to October 2019 in four UK centres. Disease activity was assessed using the Simple Clinical Colitis Activity Index (SCCAI) or Partial Mayo Score (PMS) depending on study site. Response and remission were defined as a reduction in SCCAI or PMS of ≥3, and SCCAI ≤2 or PMS ≤1, respectively. Results Overall, 74% (88/119; 95% CI 65–81%) patients responded to tofacitinib at week 8 and steroid free remission was observed in 44% (47/108; 95% CI 34%-53%) patients at week 26 (figure 1). Endoscopy was undertaken in 90 patients (67%) at baseline with routine follow-up endoscopy in 11 patients at week 8 and 34 patients at week 14. Median baseline UCEIS was 5 (IQR 4–6) falling to 2 (1–6) at week 8 and 2 (1–4) at week 14. Primary non-response was independently associated with younger age (p=0.014) and higher baseline CRP (p=0.004). Prior biologic exposure did not influence response or remission rates. Continuing tofacitinib in the setting of primary non-response was rarely helpful. Dose escalation recaptured response in 9/19 patients who lost response. Dyslipidaemia was observed in 20% (27/134; 95% CI 14%-28%) of patients but no major adverse cardiovascular events occurred. Seven patients had serious infections, with herpes zoster in 3 patients. Overall, adverse events that curtailed treatment were uncommon and no VTE occurred. Conclusions In this multi-centre real-world cohort, tofacitinib was well tolerated and clinically effective in a treatment refractory UC population.

Introduction Lack of comparative trial data on dosing regimens of infliximab rescue therapy in acute severe ulcerative colitis (ASUC) has resulted in variable use of rescue regimes in ASUC with potential impact on clinical outcomes. We aimed to (i) study variability in practice in management of ASUC and (ii) evaluate physician perspectives in decision making in rescue therapy. Methodology An internet-based survey of members of the IBD section of BSG was conducted. The non-vignette section evaluated provider characteristics. The vignette-based section with linked questions detailing progress of an ASUC patient failing intravenous corticosteroids (IVCS) evaluated diagnostic and therapeutic practices. Data was analysed by descriptive statistics and using Fisher’s exact test for categorical variables. Results The response rate of the survey was 31% (209/682 IBD section members). The provider characteristics as in table 1. In the scenario of IVCS failure, 82.9% will use Infliximab rescue with only 1.5% favouring colectomy. Of those who chose infliximab rescue therapy, 78.4% favoured standard dose (5 mg/kg) while 37 (21.6%) favoured a higher dose (10 mg/kg) with reasons cited including low albumin and high CRP. IBD specialists chose higher front-loading dose more often compared to other gastroenterologists (p=0.01) Accelerated Induction was favoured by 51.5% of the respondents while 25% preferred standard induction (SI) regime and 19% favoured colectomy. IBD specialists more often favoured AI compared to other gastroenterologists(p=0.03). The main factor for favouring AI was the presence of predictors of low infliximab trough levels (73.9%). The reasons cited by those favouring SI (n=57) included lack of evidence for AI (18), their usual practice (11), unlicensed regime (7), and safety concerns (4). 100% of the respondents who favoured colectomy cited safety concerns as their main reason for deciding against continuing medical therapy. Conclusions There is significant variation in practice in the use of infliximab rescue therapy in ASUC. There is an urgent need for development of care pathways to standardise practice.

Introduction Gastro-oesophageal Reflux Disease (GORD) is a common condition in which acid from the stomach leaks to the oesophagus causing heartburn and unpleasant taste in mouth, potentially leading to Barrett’s Oesophagus and Oesophageal Adenocarcinoma. Observationally, it has been reported that low body mass index (BMI), high BMI, past smoking, alcohol and caffeine consumption increase risk of GORD. Recently, it was reported that waist hip ratio (WHR) was a better measure of obesity than BMI for predicting GORD. However, observational relationships may be affected by bias and confounding. Methods We used data from 379,713 unrelated European participants in the UK Biobank, including 23,123 GORD cases to firstly explore the observational associations between GORD and adiposity measures, smoking status and smoking frequency, alcohol and caffeine consumption. We then used one- and two-sample Mendelian Randomisation (MR) techniques to test the causal relationship between the predictors and GORD. MR is a genetic technique which uses genetic variation to examine the causal effect of a risk factor on an outcome through the use of a genetic instrument for the risk factor. Results Observationally male gender, age, past smoking, current smoking, higher BMI, higher WHR, higher body fat percentage and waist circumference associated with higher odds of GORD, whilst higher caffeinated coffee consumption was associated with lower risk of GORD. There was some tentative evidence of a J-shaped relationship between BMI and GORD. Alcohol consumption was not associated with GORD. MR provided strong evidence of a causal role for WHR, with a one-SD higher WHR causing a 1.34 higher odds of GORD (95%CI: 1.16–1.55, p=5e-5). Similarly, a one-SD higher WHR adjusted for BMI was associated with 1.21 higher odds of GORD (95%CI: 1.14–1.29, p=2e-9). However, there was no evidence of a causal relationship for BMI (OR: 1.00 (95%CI: 0.09–1.12), p=0.97). There was no evidence of a causal role for body fat percentage, smoking and caffeine. All results were robust to 2 sample MR approaches. Conclusions The results show robust causal evidence for the link between obesity and reflux being primarily down to WHR, not BMI. These results emphasise the importance of measuring WHR when studying gastrointestinal disorders, and the importance of weight loss in reducing the risk of reflux.

Gastrointestinal symptoms constitute about 10% of all presenting complaints reported in new primary care appointments. We have recently introduced a calprotectin-based referral pathway for young adults presenting to primary care with GI symptoms: general practitioners can refer individuals with normal coeliac serology and negative stool calprotectin test straight to our dietician-led community clinic, which was previously only accessible following secondary care referral. We sought to define the caseload referred to- and the effectiveness of, our primary care dietician service. Methods We conducted a prospective observational cohort service of patients referred to our new clinic between April 2018 and Jan 2019. Patients with coeliac disease and possible inflammatory bowel disease were excluded using serological tests and stool calprotectin, respectively. Symptoms were assessed using a symptom scale (SS) modified from the Gastrointestinal Symptom Rating Scale for irritable bowel syndrome. Patients also completed the Patient Health Questionnaire-12 Somatic Symptom (PHQ-12) scale. Personalised dietary treatments were instituted and patients were followed up after 8 weeks. Response was assessed using the symptom score and the dietitian assessment. Treatment failure was defined by a static symptom score and dieticians global assessment. Comparisons were made between referrals made to the service from primary compared with secondary care. Results During the 9-month inclusion period our dietician treated 152 patients: 43% (66/152) were referred from primary care, and 78% (119/152) female. Referred primary care patients were younger (median (IQR) 30 (2–9) years vs 46 (3–3) years, p Conclusions Overall, dietary therapies were effective in the management of about two-thirds of patients referred with gastrointestinal disorders. Patients referred direct from primary care were typically younger than those referred from secondary care. We plan to develop the service further by signposting referred patients with high PHQ-12 scores to our integrated depression and anxiety and psychology services.

Introduction Iron deficiency anaemia (IDA) commonly complicates gastrointestinal disease and impairs quality of life (QoL). Intravenous iron therapy is widely used in IDA when oral iron is poorly tolerated or ineffective, and normalisation of haemoglobin improves QoL scores. We sought to define the dosing regimens of, durability of haemoglobin responses and the prevalence of adverse events to intravenous iron (III) isomaltoside (Monofer). Methods We undertook a service evaluation of intravenous iron use in 508 outpatients (40% male) with gastrointestinal disease treated with 648 Monofer infusions between 2014 and 2017. Demographic, diagnosis, and treatment factors including dose of Monofer used and the number of patients treated with repeat infusions were recorded form the medical record. Anaemia was defined by WHO criteria. Iron deficiency was defined as transferrin saturation 5 mg/L). We extracted laboratory results from the electronic record for baseline, 12 weeks and 52 weeks. We sought factors associated with treatment failure, defined as ongoing anaemia at 12 weeks, using logistic regression. Results Overall, 93% (568/613) were anaemic at baseline with median (IQR) haemoglobin of 100 g/L (IQR 8–12). 73% (476/648) had haematinics tested and 91% [431/476] had proven IDA. Inflamatory bowel disease was the most common indication at 30% (193/648) infusions. A variety of dosing regimens were used: 26% (167/648) received fixed dosing of 1 g, 25% (163/648) were dosed according to the Ganzoni formula, 14% (91/648) had the dose calculated by the Ganzoni formula but limited to a single 20 mg/kg infusion, 8% (54/648) were dosed by the simplified dosing table and 27% (173/648) other dosing strategies. 74% (479/648) infusions had follow-up haemoglobin measured 6 to 18 weeks post-infusion. The median change in haemoglobin between baseline and –8 weeks was 18 g/L (IQR –9). 42% (186/438) of previously anaemic patients had normalised their haemoglobin by this time. Factors associated with failure to normalise haemoglobin were male sex (odds ratio (OR) 2.9 [95%CI 1.–.3]), age ≥65 years (OR 3.4 [95%CI 2.–.0]), higher comorbidity (OR 3.7 [95%CI 2.–.5]) and under-dosing versus Ganzoni-calculated dose (OR 4.8 per gram underdosed [95%CI 2.–0]). Only, 31% (44/143) patients whose haemoglobin normalised at week 12 had recurrent anaemia at 1 year. Adverse events were rare: only one patient had a probable complement activation-related pseudoallergy that was mitigated by slowing infusions, and there were no anaphylactic reactions. Discussion A wide-variety of dosing strategies are used in our trust. Treatment failure was associated with under-dosing , age, sex and comorbidity. Adverse events were rare.

Background International cohort studies have previously identified Crohn’s disease (CD), ileal disease, smoking, and age ( Methods In total, 163 patients between the age of 18 and 46 years who first presented to their general practitioner (GP) with gastrointestinal symptoms from January 2014 were included in this study. Patients above the age of 46 were excluded due to the increased risk of colorectal cancer with increasing age. This was also the upper age limit recommended for faecal calprotectin use in the investigation of suspected IBD. In addition to baseline demographic data, our main outcome measure was time to overall diagnosis including time from onset of symptoms to GP presentation (patient delay), time of GP presentation to referral (primary care delay), and time of referral to diagnosis (secondary care delay). Results The median time to diagnosis was 6.7 months [IQR 3.3–14.1], with no significant difference in time to diagnosis for IBD sub-types [CD, 9.8 months [IQR 5.5–18.5]; IBD-Unclassified, 7.0 months [IQR 4.5–8.5] and ulcerative colitis (UC), 5.2 months [IQR 2.9 −12.3] (p = 0.56 )]. The median time it took patients to present to their GP was 3.0 months [IQR 1.4–6.0]; median time for GP to refer to a gastroenterologist was 0.6 months [IQR 0.2–1.7]; and the median time from GP referral to diagnosis was 1.5 months [IQR 0.8–2.5]. On multivariable analysis, rectal bleeding (OR 0.33, 95% CI 0.15–0.71, p = 0.005) and abdominal pain (OR 2.49; 95% CI 1.13–5.89, p = 0.029) was negatively and positively associated with being in the upper quartile of patient delay. Urgent GP referrals (OR 0.14; 95% CI 0.05–0.36, p Conclusion Referrals triaged urgently and by a gastroenterologist were associated with a reduction in secondary care diagnostic delay. Adopting a combination of primary care faecal calprotectin testing and secondary care straight-to-test may impact diagnostic delays.