No ironclad values for iron deficiency

What lab values constitute iron deficiency and when people might benefit from iron replacement have been evolving. Thirty percent to 40% of healthy women younger than age 50 years have no iron stores on bone marrow biopsy, so choosing a cutoff based on distribution of healthy people may give a ferritin level that is too low when looking at lab reports. Gut absorption of iron remains high until the ferritin level reaches 50 mcg/L, and several studies have found an improvement in fatigue and small improvements in hemoglobin level in women treated with iron when their ferritin level was less than 50 mcg/L, so a ferritin level of 50 mcg/L might be a physiologic cut-point for iron deficiency.

The American Gastroenterological Association chose a ferritin level of 45 mcg/L to signify iron deficiency, as described in a technical review published September 2020 in Gastroenterology, noting that based on confirmation with bone marrow biopsy this cutoff had a sensitivity of 0.85 (95% CI, 0.82 to 0.87) and a specificity of 0.92 (95% CI, 0.91 to 0.94). The CDC and the World Health Organization define iron deficiency as a ferritin level of 15 mcg/L, which is 99% specific for iron deficiency compared to bone marrow biopsy but not sensitive and misses many iron deficiency cases.

When iron stores are adequate, hepcidin is released, which decreases absorption of iron and signals the body to store iron as ferritin. Inflammation also releases hepcidin, so there is a theory that people with chronic inflammation who are not iron deficient based on ferritin don't have access to iron to make heme and experience a functional iron deficiency. The 2025 draft KDIGO guideline is recommending changing the term “functional iron deficiency” to “iron-restricted hematopoiesis” to better represent this process.

A ferritin level of 100 mcg/L was chosen as a cutoff for iron deficiency for people with chronic kidney disease (CKD) and heart failure with reduced ejection fraction (HFrEF), but compared to bone marrow biopsy in patients with heart failure, this has a sensitivity of 82% and specificity of 72%, which aren't great. Using just the iron saturation of less than 20% had a sensitivity of 94% and a specificity of 84%, according to a scientific statement from the American Heart Association published in July 2023 in the Journal of Cardiac Failure.

The definition of functional iron deficiency for CKD stages 2 through 4 is a ferritin level of 100 to 299 mcg/L with an iron saturation below 25%. For HFrEF, it's a ferritin level of 100 mcg/L to 299 mcg/L with an iron saturation below 20%. Oral or IV iron is recommended by CKD guidelines, while heart failure guidelines recommend IV iron. Studies have shown that treatment is expected to improve heart failure symptoms for patients with HFrEF, but reductions in repeat hospitalizations weren't clear and depended on how trials handled data during the COVID-19 pandemic. For patients with CKD, hemoglobin level improved by 0.6 g/dL to 1 g/dL, but there was no improvement in quality of life and improvements in mortality or hospitalizations weren't clear. A meta-analysis of the heart failure trials published in September 2024 in Circulation found that a low iron saturation (<19.8%) was a much better indicator of who would benefit from IV iron than ferritin.

IV iron is generally safe, with death from anaphylaxis occurring in about 1 in 200,000 infusions. Ferric carboxymaltose has caused severe, prolonged, and untreatable cases of hypophosphatemia. Older studies have not found an increase in infections from iron supplementation, but the most recent and largest meta-analysis, published in JAMA Network Open in November 2021, did find a statistically significant increased risk (relative risk, 1.16; 95% CI, 1.03 to 1.29).

A ferritin level less than 50 mcg/L should make you think about diagnosing and treating patients for iron deficiency. Using a cutoff of 100 mcg/L for people with inflammation will cause overdiagnosis. Excluding patients receiving erythropoietin-stimulating agents, what constitutes a functional iron deficiency or iron-restricted erythropoiesis will likely change over time, but currently it should be a ferritin level less than 300 mcg/L and an iron saturation less than 20% to 25%. However, the iron saturation is looking like a better value to find patients who might benefit from iron replacement.