A prospective study of the gastroenterological causes of iron deficiency anaemia in a general hospital.

Current practice is to investigate routinely both upper and lower gastrointestinal tracts in patients with unexplained iron deficiency anaemia. To determine the efficacy of this approach and whether the use of more stringent biochemical criteria for iron deficiency, symptoms, or a positive immunochemical faecal human haemoglobin (FHH) influenced the findings of the investigations and could help target investigations more efficiently. Eighty patients were studied prospectively, 51 who had "definite' iron deficiency anaemia (low ferritin and transferrin saturation) and 29 with "probable' iron deficiency anaemia (either low ferritin or transferrin saturation). Patients underwent a standardised symptom assessment and testing for FHH, upper endoscopy with small bowel biopsy and colonoscopy, and a small bowel series if upper endoscopy and colonoscopy were negative. Lesions potentially causative for iron deficiency anemia were found in 54/80 (60%) of patients. Five patients (7%) had lesions in both upper and lower tracts. Small bowel biopsy was abnormal in one of 80 patients and small bowel series one of 25 patients. Significant lesions in either the upper or lower gastrointestinal tract were found in 14/20 patients with positive FHH and 25/47 with negative FHH. Symptoms, use of non-steroidal anti-inflammatory drugs and classification of patients into "definite' and "probable' iron deficiency did not influence yield of investigations or site of lesions found. Gastrointestinal lesions are common in patients with unexplained iron deficiency anaemia. Neither symptoms nor presence of FHH predict the presence of site of detectable lesions and neither testing for FHH nor more stringent biochemical criteria for iron deficiency alters clinical decision making. The findings support the routine performance of both upper endoscopy and colonoscopy in the investigation of patients with unexplained iron deficiency anaemia, however routine investigation of the small bowel is of questionable value.