BSH Guidelines on Investigation,Management of raised serum ferritin

British Society for Haematology (BSH) has released its latest guideline on Investigation and management of a raised serum ferritin. Serum ferritin level is a commonly requested investigation by clinicians in both primary and secondary care. But the clinical and laboratory management of patients with raised SF values is not at all well recognized and therefore this guideline has been developed. The guideline appears in British Journal of Haematology.

Low serum ferritin levels usually indicate reduced iron stores whereas raised serum ferritin levels can be due to multiple aetiologies, including iron overload, inflammation, liver or renal disease, malignancy and metabolic syndrome.

A sensitive immunoradiometric assay was developed in 1972 and since then measurement of serum ferritin (SF) has virtually replaced laboratory assays of serum iron and transferrin or total iron binding capacity as a surrogate measure of body iron stores in clinical practice. It is valuable to the clinician because SF is directly proportional to the level of iron stores. In fact, reduced SF levels are only found in patients with reduced body iron stores.

• A study of quantitative phlebotomy in normal volunteers showed a correlation between storage iron and SF concentration with 1 μg/l of SF equivalent to approximately 8 mg of storage iron (Walters et al, 1973).


• There is no other cause and guidelines for the management of patients with low SF and iron deficiency anemia are well established in medical practice (Goddard et al, 2011).


• In some circumstances like in patients with co‐existent inflammatory disorders, SF may be within the normal or elevated range even when iron stores are absent and anemia is due to iron deficiency.

Key Recommendations:

• The normal ranges for serum ferritin in an individual patient should take into account the variation due to age, gender and possibly ethnic origin (Grade 2A).


• Reactive causes of raised serum ferritin levels, including malignancy, inflammatory disorders, renal failure, liver disease and metabolic syndrome, should always be considered as they are all considerably more common than true iron overload (Grade 1B).


• Markedly elevated serum ferritin levels (>10 000 μg/l) should prompt consideration of rare conditions, such as adult-onset Still disease or haemophagocytic lymphohistiocytosis, but may also be seen in commoner conditions, such as renal or liver disease, infections, and malignancies (Grade 2B).


• Patients found to have raised serum ferritin should be questioned about alcohol intake and other risk factors for liver disease, transfusion history, family history of iron overload and the presence or absence of type 2 diabetes mellitus, obesity and hypertension, as well as for symptoms and signs that may point to an underlying inflammatory or malignant disorder (Grade 1C).


• In patients with a finding of elevated serum ferritin levels, first line investigations should include full blood count and film, repeat serum ferritin, transferrin saturation, inflammatory markers (C‐reactive protein, erythrocyte sedimentation rate or plasma viscosity) to detect occult inflammatory disorders, serum creatinine and electrolytes for renal function, liver function tests with consideration of viral hepatitis screening and abdominal ultrasonography (if abnormal liver function), and blood glucose and lipid studies (Grade 1C).


• In otherwise well patients with unexplained and moderately elevated serum ferritin levels (<1000 μg/l) and normal transferrin saturation, a period of observation, with lifestyle adjustment if appropriate, may be reasonable with repeat assessment after 3–6 months (Grade 2C).


• Patients with unexplained persistent hyperferritinemia (especially >1000 μg/l) require referral to a hepatologist (Grade 2C).


• There is no evidence to support venesection therapy to reduce serum ferritin levels in patients with non‐alcoholic fatty liver disease (Grade 1B).

To Read the full guideline click on the following link:

https://doi.org/10.1111/bjh.15166