Anaemia Type Classifier (MCV-Based)

Classify an anaemia from the MCV and then narrow it with the RDW, the reticulocyte count and the iron studies — starting from the reason the MCV misleads, which is that a mean cannot see two populations of cells.

Last updated 17 Sep 2026

MCV, RDW, reticulocytes and iron studies

Five inputs → anaemia type

MCV (fL)Mean cell volume in femtolitres. Microcytic is below 80 fL, normocytic 80 to 100 fL and macrocytic above 100 fL. On a modern analyser this is a measured quantity, not a calculated one — the instrument acquires a red-cell volume distribution by electrical impedance or light scatter and reports its mean, then derives the haematocrit from it. The formula from haematocrit and red cell count runs the other way and is the historical route.

RDW-CV (%)Red cell distribution width as a coefficient of variation, the form almost every analyser reports and the number printed as 'RDW' on a UK or US report. Normal is about 11.5% to 15%; above 15% is anisocytosis. This is an input rather than a note because it is the only thing on the report that can detect two populations of cells hiding inside one normal mean, which is the specific way this classification fails.

Absolute reticulocyte count (×10⁹/L)The absolute count, not the percentage. A percentage is a fraction of a red cell population that is itself reduced in anaemia, so it rises as the denominator falls even when the marrow is doing nothing; the absolute count needs no such correction. Below about 25 ×10⁹/L is hypoproliferative, 25 to 100 is the usual reference range and above 100 is a reticulocytosis. If your report gives only a percentage, multiply it by the red cell count.

Ferritin (µg/L)Micrograms per litre, numerically identical to ng/mL. Below 30 µg/L has high sensitivity for iron deficiency in a general population. The threshold rises with inflammation because ferritin is an acute-phase protein: with a raised CRP, a ferritin below 100 µg/L together with a transferrin saturation under 20% still suggests genuine iron deficiency. Always read it beside a CRP, which is why this page will not treat a ferritin of 80 as reassuring on its own.

Transferrin saturation (%)Serum iron divided by total iron-binding capacity, as a percentage. Below 20% indicates an iron supply insufficient for erythropoiesis, whether because the stores are empty or because iron is locked away by inflammation. It moves earlier than the ferritin in both directions, so it is the parameter that catches deficiency behind a normal-looking ferritin and the one that recovers first on treatment.

Normocytic mean, wide RDW, empty iron stores — a dimorphic pictureExample

MCV 88 fL, RDW-CV 18.5%, reticulocytes 45 ×10⁹/L, ferritin 18 µg/L, transferrin saturation 12%

The MCV × RDW grid, and where this page starts

The grid and its boundaries are published; the emphasis is this page's. The bold cell is where a dimorphic picture lands, and it is the cell a reader is most likely to misread, because the column heading says 'normocytic' and the patient may have two frank deficiencies. Reading down a column rather than along a row is the habit worth forming: the RDW changes the differential of a given MCV more than the MCV changes the differential of a given RDW.
	Microcytic, under 80 fL	Normocytic, 80 to 100 fL	Macrocytic, over 100 fL
Normal RDW, 11.5% to 15%	Thalassaemia; anaemia of chronic disease; lead poisoning	Anaemia of chronic disease; acute haemorrhage; hereditary spherocytosis; chronic liver disease; chronic lymphocytic leukaemia; chronic myeloid leukaemia	Aplastic anaemia; chronic liver disease; myelodysplastic syndrome; hypothyroidism; alcohol
Elevated RDW, over 15%	Iron deficiency; red cell fragmentation; sideroblastic anaemia	Early iron or folate deficiency; mixed deficiency; sickle cell anaemia; sideroblastic anaemia; myelofibrosis	Folate deficiency; vitamin B12 deficiency; myelodysplastic syndrome; immune haemolytic anaemia; cold agglutination

How a mean hides two populations — the page's own arithmetic

This table is arithmetic on the definition of a mean, not data from a study, and it is here because the arithmetic is the argument. The first and last rows are indistinguishable on the MCV and completely different patients. What separates them is the spread: the first row's population has a red cell distribution width far above 15%, the last row's does not. That is the entire case for reading the RDW as an input rather than as a footnote.
Small population	Large population	Mix	Resulting MCV	How it reads
65 fL	111 fL	50 : 50	88 fL	Normocytic — and both populations are frankly abnormal
70 fL	115 fL	50 : 50	92.5 fL	Normocytic
62 fL	120 fL	60 : 40	85.2 fL	Normocytic
70 fL	120 fL	80 : 20	80 fL	On the microcytic boundary
70 fL	120 fL	20 : 80	110 fL	Macrocytic — the larger population wins
88 fL	88 fL	one population	88 fL	Normocytic, and genuinely so

What each added test buys you

None of these replaces the MCV; each answers a question the MCV cannot be asked. The reticulocyte count is the one most often skipped and the one that most changes management in a normocytic anaemia, because it splits an under-producing marrow from a compensating one and those have entirely separate differentials. The film is the one most often skipped in favour of more numbers, and it is the only test here that can see a dimorphic population directly.
Test	What it answers that the MCV cannot	Threshold used here
RDW-CV	Whether there is one population of cells or more than one	Above 15% is anisocytosis; 11.5% to 15% is normal
Absolute reticulocyte count	Whether the marrow is under-producing or the cells are being lost or destroyed	Under 25 ×10⁹/L hypoproliferative; 25 to 100 normal; over 100 reticulocytosis
Ferritin	Whether the iron stores are empty — but it is an acute-phase protein	Under 30 µg/L; under 100 µg/L with inflammation and a saturation under 20%
Transferrin saturation	Whether iron is reaching the marrow, whatever the stores look like	Under 20% is a restricted supply
CRP	Whether the ferritin can be read at face value at all	Not an input here; send it with every ferritin
Blood film	What the individual cells look like, rather than a summary of them	No threshold — it is the only test on this list that is not a number

The MCV is a mean, and a mean cannot see two populations

Sorting an anaemia by mean cell volume is the oldest useful manoeuvre in haematology and it still works: it turns an undifferentiated low haemoglobin into a short list of tests worth sending. Microcytic below 80 fL points at iron deficiency, thalassaemia trait, anaemia of chronic disease and, less often, sideroblastic anaemia or lead. Macrocytic above 100 fL raises B12 and folate deficiency, alcohol, liver disease, hypothyroidism, myelodysplasia, drugs and a brisk reticulocytosis. Normocytic in between covers the anaemia of chronic disease, renal anaemia, acute blood loss, marrow failure and early deficiency states. The site's MCV calculator computes the index and the red cell indices interpreter reads the MCV, MCHC and RDW together; this page takes the MCV split as given and narrows it with the reticulocyte count and the iron studies, because those are what turn a category into a diagnosis.

It starts, though, from the failure. The MCV is a mean, which means it is a single number summarising a distribution, and a single number cannot represent a distribution with two peaks. When a patient is deficient in iron and in B12 or folate at the same time, they make a population of small iron-deficient cells and a population of large megaloblastic cells, and the analyser reports the average of the two. That average is frequently normal. The arithmetic is unforgiving: half a population at 65 fL and half at 111 fL gives a mean of 88 fL, a textbook normocytosis in a patient with two frank deficiencies. Published sources put it in exactly those terms — the counter averages the macrocytic B12-deficient cells and the microcytic iron-deficient cells, producing a normocytic value.

The instrument does not actually miss it, which is the useful part. The red cell distribution width is the coefficient of variation of the same volume distribution the MCV is the mean of, so it measures the spread the mean has discarded. A dimorphic population has a wide spread whatever its centre. A normocytic anaemia with an RDW above 15% is therefore not a reassuring result but a specific instruction: send B12 and folate, and look at a film, on which two populations of cells are usually obvious on inspection even when every index is unremarkable. That is why the RDW is a numeric input on this page rather than a caution at the bottom of it, and why the first rule it applies fires on a normal MCV.

Two clarifications about the RDW itself, because it is widely misread. It is not the width of a red cell; it is the width of the curve of red cell volumes. And the version almost every analyser prints, RDW-CV, is a percentage computed by dividing the standard deviation of that curve by the MCV, so it is arithmetically dependent on the MCV and rises in microcytosis for that reason alone. The other form, RDW-SD, is a width in femtolitres measured directly off the histogram and is independent of the MCV; the RDW-CV to RDW-SD calculator sets out the relationship between them, and the limits of estimating one from the other.

Once the MCV and the RDW have placed the patient, two things narrow it further and neither is an index. The absolute reticulocyte count separates a marrow that is not making cells from one that is making them and losing them, which is the single largest fork in a normocytic anaemia: under-production leads to renal function, inflammatory markers, thyroid function and a myeloma screen, while a reticulocytosis leads to haemolysis markers and a search for bleeding. And the iron studies decide whether the supply of iron is the problem — the ferritin for the stores, read beside a CRP because ferritin rises with inflammation, and the transferrin saturation for what is actually reaching the marrow. The site's iron studies panel interpreter and functional versus absolute iron deficiency interpreter take those four numbers further than this page does. And if the question is how severe the anaemia is rather than what type it is, the WHO anaemia severity classifier grades the haemoglobin itself.

Frequently asked questions

Can the MCV be normal when a patient has two deficiencies?

Yes, and it commonly is. Iron and vitamin B12 deficiencies occurring together produce a population of small cells and a population of large cells, and the analyser reports the mean of the two, which is frequently normal. Half a population at 65 fL and half at 111 fL averages to 88 fL. The red cell distribution width is what detects it, because it measures the spread of cell volumes rather than their average, and a blood film usually shows both populations directly.

What are the MCV cut-offs for microcytic, normocytic and macrocytic?

Microcytic is below 80 fL, normocytic 80 to 100 fL and macrocytic above 100 fL, with the reference interval for adults given as 80 to 100 fL. These boundaries are conventions rather than physiological thresholds, and a laboratory's own interval may differ slightly; the MCV also rises gradually with age.

Is the MCV measured or calculated?

On a modern analyser it is measured. The instrument acquires a red cell volume distribution by electrical impedance — the Coulter principle — or by light scatter, reports the mean of that distribution as the MCV, and then derives the haematocrit from MCV × red cell count. The formula MCV = haematocrit ÷ red cell count × 10 is the historical route, and it is what to use if you have a spun haematocrit and a cell count and no analyser.

What does a high RDW with a normal MCV mean?

That the cells vary in size even though their average size is normal, which means the population is probably not uniform. The published causes of that combination are early iron or folate deficiency, a mixed deficiency, sickle cell anaemia, sideroblastic anaemia and myelofibrosis; to those add a recent transfusion and an anaemia part-way through responding to treatment. It is an indication to send B12, folate and iron studies and to look at a film, not a reason for reassurance.

Why ask for the absolute reticulocyte count rather than the percentage?

Because the percentage is a fraction of a red cell population that is itself reduced in anaemia, so it rises as the denominator falls even when the marrow is doing nothing. The absolute count needs no correction. Below about 25 ×10⁹/L is hypoproliferative, 25 to 100 is the usual range, and above 100 is a reticulocytosis. If the anaemia is marked and the marrow is releasing reticulocytes early, a production index corrects for that too.

Does a normal ferritin exclude iron deficiency?

No. Ferritin is an acute-phase protein, so inflammation raises it independently of iron stores. Below 30 µg/L has high sensitivity for iron deficiency in a general population, but with a raised CRP a ferritin below 100 µg/L together with a transferrin saturation below 20% still suggests genuine deficiency. Always send a CRP with a ferritin, and read the transferrin saturation alongside it — it falls earlier and recovers sooner than the ferritin.

What distinguishes iron deficiency from thalassaemia trait on the indices?

Chiefly the RDW and the red cell count. Iron deficiency replaces the population progressively, so the cells vary in size and the RDW is wide; thalassaemia trait makes the whole population uniformly small, so the RDW is usually normal and the red cell count is high or normal despite the low haemoglobin. The Mentzer and Green & King indices formalise that comparison. Neither is diagnostic: haemoglobin electrophoresis or HPLC is, and the two conditions coexist often enough that a wide RDW does not exclude a trait.

Related calculators

References

Anemia. In: StatPearls. Treasure Island (FL): StatPearls Publishing (NBK499994) — the MCV boundaries of 80 and 100 fL, the reticulocyte production index thresholds of 2 and 3, the ferritin thresholds of 30 µg/L and of 100 µg/L with a transferrin saturation under 20% in inflammation, and the statement that combined iron and B12 deficiency can give a normal MCV because the counter averages the two populations.
Normal and abnormal complete blood count with differential. In: StatPearls (NBK604207), Table 2, Classification of Anemia by MCV and RDW — the MCV × RDW grid used on this page, with microcytic under 80 µm³, normocytic 80 to 100, macrocytic over 100, normal RDW 11.5% to 15% and elevated RDW over 15%.
Mean corpuscular volume. In: StatPearls (NBK545275) — the 80 to 100 fL reference interval, the calculated formula MCV = haematocrit ÷ (RBC ×10¹²/L) × 10, and the caution that the MCV may not give a complete picture in combined microcytic and macrocytic anaemia.
Snook J, Bhala N, Beales ILP, et al. British Society of Gastroenterology guidelines for the management of iron deficiency anaemia in adults. Gut. 2021;70(11):2030–2051 — the diagnostic use of ferritin and the obligations to investigate the gastrointestinal tract by age and sex.
World Health Organization. WHO Guideline on Use of Ferritin Concentrations to Assess Iron Status in Individuals and Populations. Geneva: WHO; 2020 (NBK569877) — the population thresholds of 12, 15 and 24 µg/L and the raised thresholds of 30 and 70 µg/L in the presence of infection or inflammation, given here for contrast with the clinical thresholds this page uses.
Camaschella C. Iron deficiency. Blood. 2019;133(1):30–39 — the sequence in which iron indices change and the interpretation of a restricted iron supply.

Medical Disclaimer: The tools and content provided here are for educational and reference purposes only. They are not intended to substitute for professional medical advice, diagnosis, or treatment. Clinical decisions should always be based on the comprehensive assessment of a qualified healthcare professional.