Article Text
Abstract
Introduction Iron deficiency anaemia (IDA) is common, affecting 2–5% of people in the developed world. The BSG guidelines recommend bidirectional (upper and lower gastrointestinal [GI] tracts) investigations for patients with IDA in men, and in women who are postmenopausal, age > 50 or with a family history of colorectal cancer (CRC). We aimed to evaluate the yield of bidirectional investigations for GI malignant and non malignant causes.
Methods We prospectively collected data on patients attending a single consultant led IDA clinic from 2013–2015. IDA was defined as anaemia with microcytosis, low ferritin, raised zinc protoporphyrin or a compatible iron profile. Gastroscopy and colonoscopy were requested unless frailty, comorbidity, or patient choice dictated otherwise. Alternatives included barium studies, CT pneumocolon or CT abdomen. Coeliac and intrinsic factor (IF) antibodies, urine microscopy and renal tract ultrasound (unless patient had a CT) were also requested. Polyps or ulcers <1 cm, non-haemorrhagic or non-friable oesophagitis, gastritis or Barrett’s oesophagus were not considered to be the cause for the iron deficiency.
Results A total of 282 IDA and 11 patients with isolated hypoferritinaemia attended the clinic. 21 patients were excluded as they declined investigations, leaving 272 for analysis. 264 (97.1%) patients had UGI tests (240 gastroscopy/20 Barium swallow/4 CT) and 266 (97.8%) had LGI tests (198 colonoscopy/11 barium enema/57 CT). There were 170 symptomatic and 91 asymptomatic IDA patients, and 7 symptomatic and 4 asymptomatic hypoferritinaemic patients. CRC was found in 17 (6.4%) patients, with 10 and 7 in the symptomatic and asymptomatic IDA groups respectively. No malignancies were identified by any gastroscopies. An identifiable non-malignant cause of IDA was found in 8 (3%) gastroscopies. CT identified thickening of the gastric wall in an elderly patient who declined further investigations. In the isolated hypoferritinaemia group, there was no positive yield apart from 1 bleeding haemorrhoids. 1.5% (4/262) patients had pernicious anaemia and 0.7% (2/273) had coeliac disease.
Conclusion The yield of UGI and LGI cancer was 0.38% (1/264) and 6.4% (17/266). There was no yield for malignancy from all 240 gastroscopies undertaken. This suggests that gastroscopies provide very limited diagnostic value as a first line investigation in patients with IDA without UGI symptoms or suspected coeliac disease.
Disclosure of Interest None Declared