How to Read an Anemia Workup Without Jumping to Conclusions

An anemia workup is best read as a set of imperfect tests that constrain one another, not as a single number that hands you a diagnosis. Ferritin, transferrin saturation, and the mean corpuscular volume (MCV) each measure something different, each carries measurement error, and one of them, ferritin, can climb even when the body's iron stores are empty. The reason guidelines interpret these results together is that no single test performs well enough alone across the range of real patients. What follows appraises how those tests behave; it is educational and not medical advice.

Three tests, three different questions

Ferritin is an intracellular storage protein whose serum concentration tracks total body iron, but only when nothing else is disturbing it. It answers, roughly, how much iron is in the warehouse. Transferrin saturation (TSAT) is the percentage of iron-carrying transferrin actually loaded with iron, and it speaks more to iron in transit, the supply reaching the bone marrow right now. MCV describes the average size of red cells, which tends to fall when iron is scarce over time and rise in other deficiency states such as low vitamin B12 or folate. Because these three describe storage, transport, and the downstream product, a coherent story usually shows all three pointing the same direction. When they disagree, the disagreement is the finding, and that mismatch is a reason to slow down rather than force a label.

Why the ferritin cutoff is contested

The number that defines a low ferritin has moved, and the movement is instructive. The World Health Organization's guideline on using ferritin to assess iron status sets the threshold for deficiency in apparently healthy adults at less than 15 micrograms per liter, a value that maximizes specificity, meaning a result below it is very likely to be true deficiency. The tradeoff is sensitivity: a strict cutoff misses people who are genuinely iron deficient but whose ferritin has not yet fallen that far.

The American Gastroenterological Association took the opposite side of that tradeoff for its clinical question. Its guideline on the gastrointestinal evaluation of iron deficiency anemia recommends defining iron deficiency at a ferritin below 45 nanograms per milliliter rather than 15. In the pooled data the AGA reviewed, the higher cutoff carried roughly 85 percent sensitivity and 92 percent specificity, while the traditional lower cutoff sat near 59 percent sensitivity and 99 percent specificity. Neither threshold is wrong. They are tuned for different goals: WHO is describing populations and wants to avoid overcalling deficiency, while the AGA is trying not to miss an individual whose iron deficiency might be the first sign of a bleeding lesion worth investigating. A ferritin of 40 is normal by one document and deficient by the other, which is exactly why the number alone settles nothing.

The American Society of Hematology's 2019 education review lands in a similar zone, noting that a ferritin at or below 30 micrograms per liter reaches about 92 percent sensitivity while holding specificity near 98 percent, which is why many clinicians treat the low-to-mid 30s and 40s as a gray band rather than a clean line.

Inflammation is the trap

Here is the part most likely to mislead. Ferritin is an acute-phase reactant. It rises during infection, chronic inflammation, malignancy, liver disease, and with advancing age, independent of how much iron is actually stored. The ASH review is explicit that this is what makes iron deficiency hard to detect in inflammatory conditions and in older patients: a person can have empty iron stores and a ferritin that looks reassuringly normal, or even elevated, purely because inflammation has propped the value up.

This is where interpreting tests together stops being a nicety and becomes the whole point. When inflammation is plausible, a normal ferritin cannot exclude iron deficiency, and clinicians shift their reading in two ways. First, they raise the ferritin threshold they will accept as reassuring. The WHO guideline explicitly recommends lifting the deficiency cutoff to 70 micrograms per liter in adults, and 30 micrograms per liter in children, when inflammation markers are elevated, or excluding those individuals from population estimates entirely. The AGA guideline similarly cautions that ferritin may not reflect iron stores in chronic inflammatory conditions or chronic kidney disease and that additional tests may be required. Second, they lean on transferrin saturation, which is less inflated by inflammation than ferritin. A low TSAT, often below 16 percent, or below 20 percent in the presence of inflammation, paired with a higher ferritin threshold near 100 micrograms per liter, is the ASH-described pattern for catching iron deficiency that a lone ferritin would hide. Checking an inflammatory marker such as C-reactive protein alongside iron studies is what tells you which set of thresholds you should even be using.

Reading the panel as a whole

A few patterns illustrate the logic. Low ferritin, low TSAT, low MCV is the internally consistent picture of iron deficiency, and it needs the least reinterpretation. Normal or high ferritin with low TSAT in someone with an inflammatory illness is the classic masked case, where storage looks adequate but supply is not reaching the marrow. A high MCV should redirect attention away from iron entirely and toward B12 or folate, because a big red cell is not an iron-poor red cell. And a normal MCV never rules out early iron deficiency, since cell size lags behind falling stores. The consistent thread is that each result adjusts how you weigh the others.

None of this is a formula for diagnosing yourself, and the thresholds cited here are appraisals of test performance, not treatment rules. The value in understanding the logic is more modest and more durable: it explains why a clinician orders several iron tests instead of one, why the same ferritin can mean different things in two different people, and why a single normal number is not the same as an answer.