Introduction Between 2000 and 2002, 639 consecutive patients referred to the Iron Deficiency Anaemia (IDA) pathway were prospectively collated on a database. We present a case series of potentially missed malignancies.

Methods 639 patients presenting to the IDA pathway were prospectively entered onto a database. Initial assessment included OGD and colonic imaging. A 5-year review of outcomes was undertaken using patient medical records and hospital histopathology databases to identify important missed diagnoses.

Results 126 (20%) patients were initially found to have a GI luminal tumour; 55 (9%) of these being malignant. These comprised 48 colorectal and seven upper GI malignancies. 12 non-GI, or GI non-luminal malignancies were found. Complete medical records were available for 595 (93%) of patients. Of the 584 patients in whom a malignancy was not initially found, 15 (3%) were subsequently found to found to have one. Six were colorectal, meaning 11% of the total 54 colorectal cancers presenting initially had been missed. The most common cause was a falsely negative barium enema. Three were upper GI, meaning a surprisingly high 30% of the total of 10 upper GI malignancies presenting had been missed. The most common cause was falsely reassuring OGD biopsy histology. Five non-luminal or non-GI tumours were subsequently found meaning 29% of the total 17 had been initially missed. When the malignancy was found on initial presentation 43% of the patients survived for 5 years after diagnosis, whereas none of the patients whose diagnosis was delayed survived 5 years, with a median survival of 10.5 months from diagnosis.

Conclusion This is an effective pathway with a very high yield of tumours: 161 (25%), of which 131 were luminal GI tumours and 30 were non-luminal or non-GI tumours. However the pathway still misses significant numbers of malignancies, with barium enemas more prone than colonoscopy to missing tumours, and upper GI, non-luminal and non-GI tumours being more difficult to find. Missed malignancies had poor survival outcomes. For these reasons a high index of suspicion for malignancy in patients with IDA should be maintained.