Gastrointestinal evaluation of anaemic patients without evidence of iron deficiency

Abstract

Bleeding lesions are commonly identified in the gastrointestinal (GI) tract of patients with iron deficiency anaemia (IDA). Gastroenterologists, however, are frequently referred patients with anaemia without supporting evidence of iron deficiency, where the merit of GI investigation is unknown. We conducted a retrospective, case-control study to determine the prevalence of bleeding GI lesions in 100 anaemic patients without evidence of iron deficiency [noniron deficiency anemia (NIDA) group] and 271 patients with confirmed IDA (IDA group). The prevalence of bleeding lesions in the upper GI tract was significantly lower in NIDA patients (8%) compared with IDA patients (22.9%) (odds ratio 0.29, 95% confidence interval 0.15-0.55, P<0.0009). The prevalence of lower GI bleeding lesions was also significantly lower in NIDA patients (6.9%) compared with IDA patients (20.2%) (odds ratio 0.26, 95% confidence interval 0.12-0.57, P<0.025). No cases of GI malignancy were identified in the NIDA group, but in the IDA group 5.5% had an upper GI cancer and 10.7% colorectal cancer (P<0.002 and P<0.0002, respectively). Subgroup analysis of patients aged more than or equal to 70 years stratified according to serum ferritin concentration (low, intermediate and high), revealed that the prevalence of bleeding lesions in both the upper (26.4% vs. 23.5% vs. 2.9%) and lower (26.8% vs. 16.7% vs. 0%) GI tract did not differ significantly between patients with a low serum ferritin and intermediate serum ferritin concentrations, but was significantly lower in the low ferritin group compared with high ferritin group (P<0.003 and P<0.0001, respectively). NIDA patients were also significantly more likely to have chronic kidney disease (33%), compared with IDA patients (12%), P<0.0001, and to subsequently undergo bone marrow sampling (P<0.02) with a potential cause for their anaemia being found. GI bleeding lesions are infrequently found in anaemic patients without evidence of iron deficiency and alternative causes should first be sought. Elderly patients, however, with an intermediate ferritin concentration (30-100 microg/l) also had a high yield of bleeding lesions and should be considered for GI evaluation.