Iron Studies: Ferritin, Transferrin and the Inflammation Trap

Someone is exhausted, pale and short of breath on stairs. Their haemoglobin is at the bottom of the range, their red cells are small, and their ferritin comes back comfortably normal. On the face of it the iron stores are fine and the anaemia must be something else. In a substantial number of these cases, the iron stores are not fine at all, and the normal ferritin is the problem rather than the reassurance.

Ferritin has a double life. It is the body’s iron storage protein, and its concentration in blood does broadly track how much iron is banked in tissue. It is also an acute-phase reactant, meaning its production is driven up by inflammation regardless of iron status. When both influences act at once, they push in opposite directions, and the result is a number that cannot be read at face value.

This is the central difficulty of the iron studies panel, and it is why the panel contains several tests rather than one. Each measures a different part of the iron economy, each fails in a different way, and reading them together is what recovers the answer.

What Ferritin Stores and What It Signals

Free iron is dangerous. It catalyses reactions that generate reactive oxygen species and damage cell components, so the body keeps almost none of it loose. Iron is either bound to a transport protein, incorporated into haemoglobin and other functional proteins, or locked inside ferritin.

Ferritin is a hollow protein shell, assembled from subunits, that sequesters thousands of iron atoms in a mineral core and releases them under controlled conditions. Most of it sits inside cells, particularly in the liver, spleen and bone marrow macrophages that recycle iron from ageing red cells. A small quantity leaks into plasma, and it is this circulating fraction that laboratories measure.

The relationship between circulating ferritin and total body iron stores is the basis of the test, and within limits it is a good one. A genuinely low serum ferritin is close to definitive for depleted stores, because there is no mechanism by which someone with adequate iron reserves would have very little ferritin in their blood. This is worth stating clearly, because it is the one direction in which the test is unambiguous: low ferritin means iron deficiency, full stop.

The trouble is entirely in the other direction. A high or normal ferritin has several possible explanations, only one of which is adequate iron. The circulating fraction is not purely a passive overflow from stores; it is actively regulated, and the regulator that matters most in practice is inflammation.

Ferritin as an Acute-Phase Reactant

The acute-phase response is a coordinated shift in liver protein production triggered by inflammatory signalling. The liver increases output of some proteins and decreases output of others, and the pattern is remarkably consistent regardless of what caused the inflammation.

Ferritin is among the proteins that rise. The rationale appears to be iron withholding: during infection, restricting the availability of iron to invading organisms is a defensive strategy, and sequestering iron inside ferritin while reducing its release into circulation serves that purpose. The same signalling raises hepcidin, the hormone that controls iron export from cells, which traps iron in macrophages and blocks absorption from the gut.

The practical consequence is that inflammation raises ferritin while simultaneously making iron less available to the marrow. Someone with an infection, a chronic inflammatory condition, recent surgery, obesity-related low-grade inflammation, liver disease or an active malignancy can therefore have a ferritin in the normal or raised range while their functional iron supply is genuinely inadequate.

The magnitude of the effect varies with the intensity of the inflammation, which is why a single correction factor does not work. Clinical guidance handles this by raising the ferritin threshold used to diagnose deficiency when inflammation is present, sometimes substantially, so that a value that would be reassuring in a healthy person is treated as low in someone with active inflammatory disease.

Liver damage adds a separate mechanism. Hepatocytes are full of ferritin, and when they are injured that intracellular ferritin spills into plasma. A raised ferritin in someone with hepatitis or alcohol-related liver disease may reflect cell leakage rather than either stores or inflammation.

Transferrin, TIBC and Saturation Percentage

Iron moving between sites travels bound to transferrin, a plasma protein with two iron-binding sites. Measuring the transport system gives a view of iron in transit rather than iron in storage, and it fails differently from ferritin, which is precisely why it is useful.

Three related quantities appear on reports and are frequently confused. Transferrin is the protein concentration itself. Total iron binding capacity is a functional measurement of how much iron the plasma could carry if every site were filled, and it tracks transferrin closely. Transferrin saturation is the proportion of that capacity currently occupied, calculated by dividing serum iron by the binding capacity.

Transferrin behaves as a negative acute-phase reactant, meaning inflammation lowers it. This is the key to the pattern differences in the table. In uncomplicated iron deficiency, the body raises transferrin production in an attempt to capture more iron, so binding capacity goes up while serum iron goes down, and saturation falls sharply. In anaemia of inflammation, transferrin falls rather than rises, so although serum iron is also low, saturation may be only modestly reduced.

Saturation has its own instability. It depends on serum iron, which varies through the day in a diurnal pattern and rises for hours after an iron-containing meal or supplement. A saturation calculated from a sample drawn shortly after a dose of iron can look normal in someone profoundly deficient.

Why Serum Iron Alone Is Nearly Useless

Serum iron measures the small quantity of iron bound to transferrin at the instant the blood was drawn. Almost everything about that quantity is unstable.

It follows a daily rhythm, generally higher in the morning and lower later in the day, with the size of the swing varying between individuals. It rises within hours of oral iron and after a meal containing meat. It falls during acute inflammation, sometimes dramatically and within a day, as hepcidin locks iron inside cells. It also rises in haemolysis, whether the red cells broke down inside the patient or inside the tube during a difficult venepuncture.

Because of all this, a single serum iron value cannot distinguish deficiency from inflammation from a badly timed sample. Its value on the panel is almost entirely as an input to the saturation calculation, where dividing by binding capacity removes some, though not all, of the noise.

The practical implication is that a report showing only serum iron should not be used to make a decision about iron status. It is also why laboratories that receive a request for “iron” often add binding capacity and ferritin automatically, and why a sample drawn in a fasting morning state gives a more comparable result than one taken opportunistically.

The same reasoning applies to anyone already taking iron supplements. Testing during treatment measures the tablet as much as the patient, and a meaningful assessment usually requires a gap of days rather than hours after the last dose.

Using CRP to Reinterpret a Ferritin Result

Since inflammation is the main thing that breaks ferritin, measuring inflammation directly is what repairs the interpretation. C-reactive protein is the usual choice because it is inexpensive, widely available and responds quickly.

The logic is straightforward. A low ferritin with a normal CRP is uncomplicated iron deficiency. A low ferritin with a raised CRP is still iron deficiency, and in fact more emphatically so, because inflammation was pushing ferritin up and it stayed low anyway. A normal ferritin with a normal CRP argues reasonably against deficiency. It is the fourth combination, a normal or raised ferritin with a raised CRP, that is genuinely ambiguous and requires the rest of the panel.

In that fourth case, transferrin saturation and the red cell indices become the deciding evidence. Small red cells with a raised distribution width and a low saturation, in someone whose ferritin sits in the middle of the range with active inflammation, describe iron deficiency that the ferritin is concealing. A therapeutic trial of iron, with a follow-up count to see whether the response occurs, is sometimes the most efficient way to settle it.

CRP is not a perfect proxy. It rises and falls faster than ferritin, so a sample taken while inflammation is resolving can show a normalising CRP alongside a ferritin still elevated from the preceding days. Some laboratories add a second, slower marker to bracket the timing. The general principle holds regardless: a ferritin without a concurrent inflammatory marker is a number missing its context.

Soluble Transferrin Receptor as a Workaround

Cells that need iron display transferrin receptors on their surface, and they display more of them when iron is scarce. A fragment of these receptors is shed into plasma in proportion to the total number expressed, and it can be measured.

This gives a marker with an unusual and valuable property: it reflects cellular iron demand and is largely indifferent to inflammation. Where ferritin is confounded by the acute-phase response and transferrin saturation is confounded by timing, soluble transferrin receptor rises in iron deficiency and stays put in anaemia of inflammation. It is the closest thing the panel has to a clean discriminator in the exact situation where the other tests fail.

It is not perfect either. The receptor is expressed heavily by developing red cells, so anything that expands the red cell precursor population raises the marker independently of iron status, including haemolysis and recovery from bleeding. Conversely, a marrow that is not producing much of anything will show a lower value than the iron status alone would predict. Some laboratories combine it with ferritin in a ratio or index intended to sharpen the separation.

The main practical limitation is availability and standardisation. Assays from different manufacturers have not historically produced interchangeable values, so reference ranges are method-specific and comparing results across laboratories is unsafe. It is not a routine first-line test, and it earns its place in specific cases where inflammation has made the standard panel uninterpretable.

Distinguishing Deficiency From Chronic Disease

The two conditions that most often need separating are absolute iron deficiency, where there is not enough iron in the body, and anaemia of inflammation, where there is enough iron but it is locked away from the marrow. They are treated differently, and the second does not respond to oral iron.

The distinguishing pattern rests on transferrin. Absolute deficiency raises transport capacity as the body tries to scavenge more iron; inflammation lowers it as part of the acute-phase response. Ferritin is low in the first and normal or raised in the second. Saturation is low in both, which is why it cannot separate them alone. Soluble transferrin receptor rises in the first and not the second.

The complication is that the two coexist frequently. Inflammatory bowel disease causes both chronic inflammation and blood loss. Chronic kidney disease combines inflammation with reduced erythropoietin and often with losses. Heart failure, rheumatoid disease and malignancy all produce mixed pictures. In these situations the question is not which of the two it is but how much of each, and the panel is being used to judge whether there is an iron component worth treating.

Two further habits improve interpretation. The first is to treat a low ferritin as actionable regardless of what else is on the panel, since nothing else lowers it. The second is to remember that finding iron deficiency is the beginning of the investigation rather than the end. Iron deficiency in an adult who is not menstruating heavily, pregnant or a regular blood donor implies loss from somewhere, most often the gastrointestinal tract, and correcting the number without asking where the iron went is the more consequential error on this panel.