Introduction Anaemia is a common indication for referral to GI endoscopy services. Appropriate triage of these referrals may be aided by further characterisation of the anaemia and review of iron studies. However, value of the MCV in predicting a positive endoscopic finding has been questioned. Our aim was to determine predictors of positive endoscopic findings in patients referred to a single centre for investigation of anaemia in a 6-month period, in relation to full blood count (FBC), iron studies, and patient age.
Methods A retrospective cohort study examining endoscopic procedures performed where anaemia without an obvious cause was the primary indication. Patients with overt GI haemorrhage were excluded. Data were extracted from an electronic database and specific parameters included demographic variables, indication(s) for endoscopy, haematological values, iron studies, and endoscopic findings. Positive endoscopy was defined as a finding considered responsible for the anaemia. Analysis was performed using GraphPad Prism and Microsoft Excel.
Results A total of 359 endoscopic procedures (210 gastroscopies, 137 full colonoscopies, 12 left colonoscopies) were performed in 243 patients. In 48/243 (20%) patients, a cause for anaemia was found; 10/243 (4%) had malignancy; [3/243 (1%) gastric ca, 7/243 (3%) colorectal ca]. 38/243 (16%) had a non-malignant cause of anaemia, including gastric ulcer 15/243 (6%), angiodysplasia 13/243 (5%), duodenal ulcer 4/243 (2%), and coeliac disease 2/243 (1%). Endoscopy was normal or revealed incidental findings in the remaining 195/243 (80%) patients. Older age and higher RDW were significantly associated with positive endoscopy (p=0.006, p=0009 respectively). While the association with lower serum Hb trended towards significance (p=0.07), no association between MCV and positive endoscopy was observed (p=0.87). Low serum ferritin and MCV were significantly associated with malignancy (p=0.05, p=0.05), as was a higher RDW (p=0.03).
Conclusion While a low MCV was significantly associated with malignancy on GI endoscopy, it was not significantly associated with a positive endoscopy overall. RDW is a good predictor of positive endoscopy with regard to both malignant and non-malignant causes of anaemia. Hence, the performance of GI endoscopy can be considered in the context of a normal MCV.
Competing interests None declared.