P233 Rectal bleeding in young adults: factors associated with significant bowel disease. (19th June 2022)

• Record Type: Journal Article
• Authors: Ip, Chak Lam; Kalla, Rahul; Din, Shahida; Shandro, Ben; Watson, Eleanor; Noble, Colin; Spyros, Siakavellas; Gail, Masterton; Dethier, Anne
• Is Part Of: Gut. Volume 71(2022)Supplement 1
• Page Start: A154
• Page End: A155
• Publication Date: 2022-06-19
• Subjects: Gastroenterology -- Periodicals; 616.33
• DOI: 10.1136/gutjnl-2022-BSG.287
• Languages: English
• ISSNs: 0017-5749

Abstract

Abstract : Introduction: There are no unified criteria to risk stratify patients under 50 with lower gastrointestinal bleed (LGIB) for endoscopic investigations. We evaluated this cohort to identify predictors for significant bowel disease (SBD), defined as malignancy, high risk adenoma (≥3 adenomas or any adenoma ≥1 cm) or new diagnosis of Inflammatory bowel disease (IBD). Methods: All patients under 50 in NHS Lothian with no colorectal cancer or IBD history referred for endoscopy for LGIB in 2019 were identified through the endoscopy reporting system. Electronic patient records were reviewed to extract variables including LGIB history, diarrhoea, abdominal pain, weight loss, and diagnoses. Univariate analysis, with Chi-square or t-test/Wilcoxon test as appropriate, was used to identify significant associations with SBD. Normality was tested with Shapiro-Wilk test. Complete case analysis was used where data were missing. Results: 532 patients were included in the analysis, 223 and 309 patients underwent flexible sigmoidoscopy and colonoscopy respectively. Mean age was 38.0 years (IQR 32.0 to 45.6) and 57.4% were females. SBD were present in 17.3% of all cases (1.3% malignancy; 1.3% advanced adenoma; 14.7% IBD). Normal assessment was reported in 42.7% of endoscopies and haemorrhoids were the only finding in 29.5% of cases. A significant proportion of GP referrals contained missing data for duration (37.4%) and frequency (35.9%) of LGIB. In those with data for frequency (n=341), referrals that reported 'frequent' LGIB were more likely to have SBD than those with infrequent LGIB (28.0% vs 11.2%; p=0.002), but the duration was not associated with SBD (p=0.19). All other recorded symptoms in clinical history were insignificant, including diarrhoea (p=0.28), which was present in 58.5% of cases. 83% of patients had full blood count measured before endoscopy. Faecal calprotectin (FC) was missing in over half of patients. Iron studies were performed in 12.4% of cases. Patients with SBD were more likely to be anaemic (24.7% vs 10.7%; p=0.004) or have raised FC (89.4% vs 44.3%; p<0.001) than those without SBD. Age and gender were not associated with SBD. LGIB frequency remained the only significant predictor of SBD (OR 2.34; IQR: 1.36-4.13) after multivariate stepwise regression using age, gender, and haemoglobin as covariates. Conclusions: Determining appropriate investigations for LGIB in under 50 can be hampered by limited referral details. A dedicated GP referral proforma may improve risk stratification. In a limited dataset, our study identifies LGIB frequency, anaemia and raised FC as factors that could be used to prioritise endoscopy referrals.