Anaemia investigation in practice
This article has a correction. Please see:
Editor – Mankodi and colleagues (Clin Med April 2010 pp 115–8) are to be congratulated for tackling the important issue of optimising investigational strategy in iron-deficiency anaemia (IDA). However, we would urge caution before extrapolating from their results and excluding patients from urgent gastrointestinal investigations if they do not meet strict criteria for IDA.
IDA is a continuum and, as Makondi et al's results nicely illustrate, an increase in specificity is inevitably at the expense of decreased sensitivity resulting in missed cancers (false negatives). In the IDA group 4/14 (29%) patients with cancer had a ferritin .15 ng/ml, while in the non-IDA group at least 2/4 (50%) were in all probability iron deficient. This is alluded to in the discussion where they state that a ferritin threshold of, 50 ng/ml increases the sensitivity for cancer detection to 94.4%. However, this is at odds with the central tenet of the paper demonstrating that applying strict thresholds for diagnosing IDA results in a reduction in investigations, and by extension, costs.
We have recently looked at the prevalence of anaemia in a sequential series of 87 patients diagnosed with right-sided colon cancer (caecal and ascending colon) at our institution between 2005 and 2008. At presentation, 72% of patients were anaemic according to the British Society of Gastroenterology criteria used by Mankodi et al. However, only 66% of these cases would have been classified as iron deficient using a ferritin level of, 15 ng/ml, whereas this rose to 91% using a ferritin cut-off of <50 ng/ml. Therefore, approximately 25% of patients with anaemia secondary to a right-sided colorectal cancer have a ferritin of between 15–50 ng/ml, and would be denied urgent investigation using strict criteria. Furthermore, the mean cell volume is of limited value as, though a microcytosis is useful in suggesting the presence of IDA, 51% of our cohort had a normocytic anaemia. Of these, 58% had IDA using a ferritin threshold of < 15 ng/ml, which rose to 83% using a threshold of < 50 ng/ml.
Based on these and Mankodi's results we therefore advocate that rather than enforcing strict criteria for the diagnosis of IDA, using a ferritin threshold of < 50 ng/ml significantly reduces false-negatives resulting in a higher cancer detection rate that outweighs the burden of increased investigations.
Footnotes
Please submit letters for the editor's consideration within three weeks of receipt of the Journal. Letters should ideally be limited to 350 words, and sent by email to: clinicalmedicine{at}rcplondon.ac.uk
- Royal College of Physicians
Article Tools
Citation Manager Formats
Jump to section
Related Articles
Cited By...
- No citing articles found.