Multicentre prospective study of panenteric endoscopy to localise bleeding lesions causing iron deficiency anaemia in the gastrointestinal tract (BLITGIT)

Aims Guidelines recommend upper and lower GI endoscopy to investigate the cause of iron deficiency anaemia (IDA). This practice is based on small, retrospective studies of select populations in which small bowel pathology was rarely identified. This study aimed to localise and grade all lesions with bleeding potential by performing panenteric endoscopy.

Methods Patients with IDA able to attend for capsule endoscopy in the week prior to upper and lower GI endoscopy and who had no risk factors for capsule retention were recruited from Sheffield (UK), Székesfehérvár (Hungary) and Hong Kong. Images of all upper and lower GI and small bowel lesions were assessed by a three member expert consensus panel, described using terminology from a pre-defined diagnostic register and the Saurin classification (P0: bleeding unlikely; P1: bleeding possible; P2: bleeding likely). The true prevalence of small bowel lesions in patients with IDA was assumed to be 10%, such that a sample size required to have 80% power of getting a 95% confidence interval for the prevalence no wider than 10 percentage points was 160 patients.

Results All gastroscopy, capsule endoscopy and colonoscopy (completion rates 99.4%, 88.2% and 95.2% respectively) data from 170 patients (median age 61 years (IQR 46- 71); 44% female) was analysed. Per-patient diagnostic yields of P1 and P2 lesions by gastroscopy, small bowel capsule endoscopy and colonoscopy were 27.6%, 49.4% and 20% (P<0.0001) and for P2 lesions alone, 6.5%, 10% and 9.4% respectively (P=0.48). P1 and P2 lesions were localised to oesophagus, stomach, small bowel, and colon in 5.5%, 18.3%, 63.0% and 13.2% respectively. P1 or P2 lesions occurred in more than one location in 27.6%.

Oesophageal lesions were erosions (2.7%), ulcers (1.8%) and varices (0.9%); gastric lesions were erosions (10.5%), ulcers (2.3%), eroded polyp (2.3%), gastric antral vascular ectasia (1.4%), neoplasia (0.9%: one cancer, one a secondary lesion from lymphoma diagnosed by breast biopsy), angioectasia (0.5%), portal hypertensive gastropathy (0.5%); small bowel lesions were angioectasia (30.6%), erosions (16.4%), ulcers (10.5%), ulcerated stenosis (1.8%), fresh blood (1.8%), eroded polyp (1.4%), vascular lesion (0.9%); colonic lesions were cancer (4.1%), haemorrhoids (4.1%), angioectasia (1.8%), ulcers (1.8%) and eroded polyps (1.4%).

Conclusions This is the first prosepctive study to perform panenteric endoscopy to identify and localise lesions causing IDA. The commonest location was the small bowel and commonest malignant cause, colon cancer. A low threshold to exclude patients with possible obstructive symptoms may partly explain the absence of small bowel neoplasia. Nonetheless, important small bowel pathologies likely to cause recurrent bleeding were common and should be sought routinely in patients with IDA.

Publikationsverlauf

Artikel online veröffentlicht:
27. März 2025