Iron studies: ferritin, transferrin saturation and TIBC

Iron is essential for life, but the human body has no regulated way of excreting it. Every extra red cell, every iron-containing enzyme and every store is built from iron that came through the gut, and the whole system is controlled at one point: absorption. An iron panel exists because no single test can tell you whether a person has too little iron, too much iron, or iron that is locked away and unavailable. The panel is not a diagnosis by itself; it is a way of asking four separate questions — how much iron is in the blood today, how much capacity is there to carry it, how full are the stores, and how tightly the body is holding on to it. ^{[1] [2]}

The iron panel is one of the most misread groups of tests in medicine. A normal ferritin in an inflamed patient can hide iron deficiency, a raised ferritin with a normal transferrin saturation is not overload, and a low transferrin saturation with a normal ferritin can be the first sign of anaemia of chronic disease. The purpose of this page is to work through what each test measures, the reference ranges, the causes of high and low values, the named patterns, and how to interpret a set of results step by step.

By the end you should be able to look at a panel of serum iron, TIBC, transferrin saturation and ferritin, decide which pattern it represents, and state the next investigation or treatment.

Serum iron is the amount of iron circulating bound to transferrin. It falls in iron deficiency and rises in iron overload, but it is the least useful single value because it changes with the time of day, meals and recent iron intake, and it is low whenever reticuloendothelial iron is being held back. ^[1][8]

Transferrin is the main iron transport protein, made by the liver. TIBC (total iron-binding capacity) measures how many iron-binding sites transferrin has available. When iron stores are low, the liver makes more transferrin, so TIBC rises; when iron is plentiful or inflammation is present, transferrin falls. ^[9]

Transferrin saturation (TSAT) is the proportion of transferrin actually carrying iron, calculated as serum iron divided by TIBC, expressed as a percentage. It is more stable than serum iron alone and is the best quick guide to whether iron is available for new red cell production. A low TSAT means iron restriction; a high TSAT suggests iron overload or release of iron from damaged cells. ^[1][3]

Ferritin is the intracellular iron storage protein, and a small fraction leaks into plasma in proportion to body iron stores. In the absence of inflammation it is the single most useful marker of iron deficiency: a low ferritin is highly specific for depleted stores. ^[1] But ferritin is also an acute-phase reactant released by macrophages and damaged cells, so it can be raised by inflammation, infection, liver disease, malignancy and alcohol even when iron stores are normal or low. ^[3]

Soluble transferrin receptor (sTfR) is a cleaved fragment of the transferrin membrane receptor on erythroid precursors. When iron is scarce inside the developing red cell, more receptor is made and sTfR rises. Its big advantage is that it is not an acute-phase protein, so it stays informative when ferritin is falsely raised by inflammation. ^[4] The main limitation is that assays are not standardised between manufacturers, so the sTfR alone is hard to compare across hospitals; the sTfR divided by the logarithm of ferritin (the sTfR/log ferritin index) is used for this, with values above 2 suggesting iron deficiency anaemia and below 1 suggesting anaemia of chronic disease. ^[4]

Hepcidin is the master regulator, a peptide made chiefly by hepatocytes. Low iron stores, hypoxia and increased erythropoiesis (for example after bleeding or with erythropoietin) suppress hepcidin, which opens the ferroportin channel on gut enterocytes and macrophages so iron can enter the plasma. Inflammation, especially interleukin-6, raises hepcidin within hours; ferroportin is then broken down, iron is trapped inside macrophages and absorption stops, producing functional iron deficiency even though total body iron is normal or high. ^[4][3]

Because hepcidin sits at the top of the control loop, it is also used as a teaching tool for the panel's patterns: every pattern can be explained by asking whether hepcidin is high or low. (Routine clinical hepcidin testing is still limited and is not part of the standard panel in most hospitals.)

Ranges are adult values. They differ by laboratory, age, sex and method, and the laboratory's own range always wins; the numbers below are common values, not universal cut-offs. ^[8][9][10]

Two teaching cut-offs are widely used and are quoted here from open guidance. A ferritin under 15 microg/L is highly specific for iron deficiency (specificity 0.99), and a cut-off of 45 microg/L gives a respectable specificity of about 0.92, so levels below this may warrant consideration of further investigation, particularly with a chronic inflammatory process and anaemia. ^[1] Iron stores are effectively exhausted by the time ferritin falls to 15, even though anaemia may not have appeared. ^[5]

The ferritin row needs emphasis on direction of change: low ferritin virtually always means depleted stores (in the absence of a massive recent transfusion), while a high ferritin is non-specific and is often an inflammatory signal rather than overload. ^[3] In routine practice only about 10% of cases of high ferritin are associated with iron overload; the rest are explained by inflammation, metabolic syndrome, alcohol, liver disease or other conditions. ^[3]

Always start by asking whether the patient is deficient, inflamed, overloaded, or mixed. Hepcidin is high in the inflammatory and overloaded states and low in true deficiency. ^[4]

Mixed pictures are common and clinically important. The most frequent is a patient with rheumatoid arthritis or active inflammatory bowel disease who also has occult gastrointestinal bleeding causing absolute iron deficiency. A ferritin that looks normal (for example 60) in an inflamed patient may actually conceal depleted stores, because the same inflammation has pushed the ferritin up. In that situation the sTfR and the sTfR/log ferritin index, or a two-to-four-week trial of oral iron with a rise in haemoglobin of 10 g/L, help settle the question. ^[1][4]

Thalassaemia trait is the classic mimic of iron deficiency: a low MCV with a ferritin that is not low. The MCV is typically reduced out of proportion to the degree of anaemia, and high-performance liquid chromatography of haemoglobin shows raised HbA2. Giving iron in thalassaemia trait does not correct the microcytosis and only risks overload. ^[1][6]

Sideroblastic anaemia is the opposite problem: iron is present (stores and TSAT are normal or high, and ferritin is raised), but it cannot be incorporated into haem because of a defect in haem synthesis. The ring sideroblast on the bone marrow Perls' stain shows iron trapped in mitochondria. ^[6]

CKD has its own thresholds, taken from the CKD anaemia literature: absolute iron deficiency is defined as transferrin saturation of 20% or less with ferritin of 100 microg/L or less in pre-dialysis and peritoneal dialysis patients, or 200 microg/L or less in haemodialysis patients. ^[1][7] In chronic heart failure, iron deficiency is taken as a ferritin under 100 microg/L, or a ferritin of 100-299 microg/L with a transferrin saturation under 20%. ^[1]

Haemochromatosis (the most common kind, HFE-related) presents with a high transferrin saturation and a rising ferritin. A normal transferrin saturation usually excludes pathologically increased iron absorption, although it does not necessarily exclude the presence of an iron overload; interpretation requires the clinical context. ^[3] High ferritin with a normal or low saturation is far more likely to be inflammation, metabolic syndrome, alcohol or liver disease than overload.

All values below are invented for teaching and are not real patient data. Each sheet represents a typical panel, with the patient's sex and context stated, so that every value can be read against the appropriate range.

Context: 28-year-old woman, no anaemia, ferritin checked as part of a general health screen. Fasted morning sample.

Interpretation, step by step:

1. Check the red cell side first if a full blood count is available; here there is no anaemia, so we are looking for iron deficiency without anaemia. ^[1]
2. Serum iron is normal and the TIBC is not raised, so there is no systemic signal of iron shortage. ^[9]
3. Transferrin saturation of 27% sits comfortably inside the 15-45% female range. ^[8]
4. Ferritin of 46 ng/mL is well above the 15 microg/L specificity cut-off and the 30 level at which deficiency is still possible. ^[1][5]
5. There is no inflammation mentioned, so the ferritin can be taken at face value. ^[3]

Conclusion: the iron panel is normal; there is no evidence of iron deficiency or overload. The next step is simply to continue routine care, and to reassess only if symptoms or a repeat full blood count change.

Context: 44-year-old woman, tired, with a microcytic anaemia on the full blood count. No known inflammatory disease.

Interpretation, step by step:

1. The low MCV and MCH on the accompanying full blood count point to a microcytic anaemia; iron deficiency is the most common cause. ^[1]
2. Serum iron is low and TIBC is raised, the classic direction-of-change pair for iron deficiency: transferrin is being up-regulated by the liver. ^[9]
3. Transferrin saturation of 4% is far below the female range of 15-45%, confirming that iron is not reaching the red cell. ^[8]
4. Ferritin of 9 ng/mL is below the lower limit of the female range and below the 15 microg/L level that is highly specific for iron deficiency. ^[1][10]
5. There is no chronic inflammation in the history, so a falsely normal ferritin is not a concern here. ^[1]
6. The cause of the iron deficiency still has to be found: menstrual loss, pregnancy, malabsorption, or gastrointestinal blood loss. ^[1]

Conclusion: absolute iron deficiency anaemia. The next step is to treat with oral iron replacement and, in this adult, to investigate for a source of blood loss, including the gastrointestinal tract.

Context: 62-year-old man with active rheumatoid arthritis and a normocytic anaemia. Morning sample.

Interpretation, step by step:

1. The anaemia is normocytic, which in a patient with active rheumatoid arthritis raises anaemia of chronic disease. ^[4]
2. Serum iron is low, which in ACD reflects iron being trapped inside macrophages rather than available iron being absent. ^[4]
3. TIBC is low rather than raised, the opposite direction from absolute iron deficiency; inflammation suppresses transferrin synthesis. ^[9][4]
4. Transferrin saturation of 16% is below the male range of 20-50%, consistent with iron-restricted erythropoiesis. ^[8]
5. Ferritin of 248 ng/mL is inside the male range and is being pushed up by the inflammatory state; it does not exclude concurrent iron deficiency. ^[3][1]
6. The high CRP confirms an active inflammatory process, which also raises hepcidin and locks iron away. ^[4][3]
7. Because the stores cannot be read directly, an sTfR and the sTfR/log ferritin index (or a monitored trial of iron) is needed before deciding iron is absent. ^[4]

Conclusion: anaemia of chronic disease with functional iron deficiency, with iron deficiency not yet excluded. The next step is to treat the underlying arthritis, correct other causes of anaemia, and test iron reserve (sTfR/log ferritin index) before giving routine iron; intravenous iron is reserved for symptomatic iron deficiency or where targets must be met.

• Do not call a normal ferritin "reassuring" in an inflamed patient. Ferritin is an acute-phase reactant and can look normal even when stores are empty; correlation with the clinical picture and C-reactive protein is essential. ^[1][3]
• Do not treat thalassaemia trait with iron. A microcytic, hypochromic anaemia with a normal ferritin is thalassaemia trait until proven otherwise; check haemoglobin electrophoresis first. ^[1][6]
• Do not diagnose overload from ferritin alone. A raised ferritin with a normal or low transferrin saturation is inflammation, alcohol or liver disease far more often than haemochromatosis. ^[3]
• Beware recent iron intake and timing. Serum iron varies through the day and rises after an iron dose; an early-morning fasted sample is standard. ^[8]
• Beware haemolysis of the sample, which releases iron and can falsely raise serum iron (TIBC and ferritin are less affected in the same way).
• Beware very high ferritin without overload in HLH, malignancy, liver disease, adult Still's disease and severe infections; ferritin above 1000 microg/L is a marker of serious pathology whatever the cause. ^[3]
• In CKD the numbers are different: a ferritin of 80 may be adequate in a patient on haemodialysis, while under 100 with a TSAT under 20% indicates absolute deficiency in pre-dialysis disease. Always use the CKD thresholds, not the general ones. ^[1][7]
• Do not forget that iron deficiency is a symptom, not a disease. Once confirmed, the cause (gastrointestinal blood loss, menorrhagia, coeliac disease, malabsorption) still has to be found, particularly in adults. ^[1]

• An iron panel asks four things: how much iron is in the blood (serum iron), how much capacity there is to carry it (TIBC), how full the stores are (ferritin), and how tightly iron is being held (hepcidin-driven through TSAT, sTfR and ferritin together). ^[1]
• A low ferritin is highly specific for iron deficiency; a ferritin under 15 microg/L has specificity of 0.99. ^[1]
• Ferritin is also an acute-phase reactant, so a normal value can hide iron deficiency during inflammation. ^[3]
• In inflammation, serum iron and transferrin saturation fall, TIBC falls and ferritin rises or stays normal — the opposite pattern to absolute deficiency. ^[4]
• A high ferritin with a normal transferrin saturation is usually inflammation, alcohol, liver disease or metabolic syndrome, not iron overload. ^[3]
• sTfR is not an acute-phase protein and the sTfR/log ferritin index (>2 for iron deficiency anaemia, under 1 for anaemia of chronic disease) helps separate mixed cases. ^[4]
• Hepcidin is high when inflammation traps iron and low when stores are depleted; it is the mechanism behind the patterns. ^[4][3]
• CKD and heart failure each have their own iron thresholds, which differ from standard ones. ^[1][7]
• Thalassaemia trait and sideroblastic anaemia are the two important non-deficiency microcytic pictures, and neither is treated with iron. ^[1][6]
• Reference ranges differ by laboratory, age and sex; always use the laboratory's own range. ^[8][9][10]