Full blood count and blood-film interpretation in children — branching viva

Opening — framing the printout

The examiner begins: here is a full blood count with a haemoglobin of 100. Before I ask you anything, what is the single most important piece of information missing, and why does it change your interpretation? [1]

I would say the age of the child. A haemoglobin of 100 is read against an age-specific paediatric range, and the same number is normal for a two-month-old infant at the physiological nadir of infancy and abnormal for a ten-year-old. I cannot interpret any paediatric value without the age band, and I would not apply adult ranges. [1] [4]

Branch A — the physiological nadir of infancy

Now I tell you the child is a thriving, asymptomatic two-month-old breastfed infant. Is this anaemia? [1]

No. The haemoglobin of 100 sits at the physiological anaemia of infancy, the trough that occurs around two to three months in a term baby as blood volume expands faster than red cell mass, fetal haemoglobin is switched off, and the short neonatal red cell lifespan limits supply. The mean cell volume would be low-normal for age. If the child is thriving and asymptomatic with a normal film, no work-up is needed; prematurity, symptomatic pallor or an abnormal film would redirect me to a haemolysis or iron work-up. [1] [4]

Branch B — the mean-cell-volume sort

Now the child is an eighteen-month-old with a haemoglobin of 72 and a mean cell volume of 58. Talk me through the interpretation. [6]

I would sort this anaemia by size first. A small mean cell volume makes it microcytic, which points me to iron deficiency, thalassaemia trait, lead poisoning and chronic inflammation. The high red cell distribution width and the dietary history would lead me to iron deficiency as the most likely cause, and I would confirm with iron studies — a low ferritin and transferrin saturation with a high total iron-binding capacity. I would treat with oral elemental iron at 3 to 6 milligrams per kilogram per day, expect a reticulocyte rise within seven to ten days, and continue for three months after correction. [6]

Branch C — the thalassaemia distinction

The ferritin comes back replete. How do you now distinguish thalassaemia trait from iron deficiency? [6]

Thalassaemia trait shows microcytosis out of proportion to a mild anaemia, a normal or near-normal red cell distribution width because the cells are uniformly small, and a high red cell count. The Mentzer index — the mean cell volume divided by the red cell count — is above 13 in thalassaemia trait and below 13 in iron deficiency. I would confirm with haemoglobin electrophoresis, which shows a raised haemoglobin A2 in beta-thalassaemia trait. [6]

Branch D — the child with thrombocytopenia

Now here is a four-year-old, well, two weeks after a viral illness, with a platelet count of 12 and petechiae but a normal haemoglobin and white cell count. What is this? [9]

This is immune thrombocytopenia, provided the blood film shows no blasts and no schistocytes and the other cell lines are normal. The post-viral timing and the isolated thrombocytopenia in a well child are characteristic. Before I commit to the diagnosis, I would confirm the count is genuine and not EDTA-dependent platelet clumping, by reviewing the film for clumps and repeating the sample in citrate. I would manage to the bleeding phenotype: observe if the child is well with no bleeding, reserve steroids or intravenous immunoglobulin for bleeding or very low counts, guided by the American Society of Hematology 2019 guidance. [8] [9]

Closing — the printout that does not fit

Finally, a well child with no petechiae is reported to have a platelet count of 8. What is your move? [8]

When the number does not fit the child, I suspect an artefact. EDTA-dependent platelet clumping is the commonest cause of a spuriously low platelet count, and treating it as real would expose a well child to transfusion or marrow biopsy. I would repeat the sample in an alternative anticoagulant, review the film for clumps, and only act once the count is confirmed. [8] [10]