P354 A multi-centre audit of lower gastrointestinal bleeding management

Introduction Two national reviews identified unwanted variation in the care of patients with gastrointestinal haemorrhage. British Society of Gastroenterology (BSG) acute lower gastrointestinal bleeding (LGIB) guidelines define a specific management approach including risk stratification, investigation and intervention for patients presenting with LGIB. Guideline adoption, impact on ‘real-world’ clinical outcomes and resource implications are not known. The Mersey Gastroenterology Network conducted a region-wide, multi-centre, pilot audit of LGIB management in relation to BSG guidance to assess this. Methods Patients aged ≥16 years presenting with LGIB to five hospital trusts from June 1st-July 1st 2019 were included. Data on presentation, management and outcomes of patients were recorded. These were audited against BSG guidelines. Results 104 patients were included. Mean age at presentation was 60 years (range 23–96). A Shock Index (SI) >1 was rare (7.7%), with a mean SI of 0.69 on presentation. 67% of patients haemodynamically unstable despite initial resuscitation underwent computed tomogram angiography (CTA). 21 patients (20.2%) had a low risk Oakland Score (OS) (≤8 points); 29% were admitted whilst 71% were discharged for outpatient management. 75 patients (78.9%) had a high-risk OS (>8 points); 61% were admitted and 39% were discharged. In patients discharged with a high-risk OS (median 11), 6.9% were readmitted within 30 days. Admitted patients underwent CTA (10.2%), colonoscopy (8.5%), flexible sigmoidoscopy (3.4%) or OGD (22%) as initial investigation. 55.9% were observed without diagnostic investigations for an average of 3 days before discharge. Rebleeding within 7 days, readmission within 30 days and inpatient mortality was 9.3%, 15.8% and 3.4% respectively. The median length of admission was 3 days (range 0–60). 38.5% of patients were managed in line with BSG guidance. The most common deviation was patients with an OS >8 being monitored without investigation during admission (31.7%). Conclusions LGIB management does not follow current guidelines. We report low mortality and readmission rates, even in patients discharged with a ‘high-risk’ OS. Admitted patients rarely receive an urgent inpatient colonoscopy. We aim to collect data over an extended period to provide further insights and analyse economic and resource feasibility of guideline adherence.