Eosinophilia Severity Interpreter
Eosinophilia Severity Interpreter
Classify an absolute eosinophil count as mild, moderate or severe — and separate hypereosinophilia from hypereosinophilic syndrome, which the count alone can never do. A moderate count with a raised troponin outranks a severe one from a drug rash.
Eosinophilia Severity
Count + persistence + organ damage → classificationAEC 2.4 ×10⁹/L, single count, organs not yet assessed
The three questions, and the order they are answered in
0.5 to under 1.5 ×10⁹/L — mild
1.5 to under 5.0 — moderate
5.0 and above — severe
Hypereosinophilia (HE)
1.5 ×10⁹/L or above on two occasions at least a month apart, or marked tissue eosinophilia
Hypereosinophilic syndrome (HES)
HE plus organ or tissue damage attributable to it plus exclusion of other causes of that damage
- severity ≠ hypereosinophilia
- ‘moderate eosinophilia’ and ‘hypereosinophilia’ cover the same numeric range from 1.5 ×10⁹/L upwards, but the first is a description of one count and the second requires persistence. A single count of 3.0 is moderate eosinophilia and is not yet hypereosinophilia
- HE ≠ HES
- the difference is entirely the organ assessment. Persistent hypereosinophilia with no damage found is hypereosinophilia of undetermined significance; the same count with attributable damage is a syndrome that needs treating
- tissue eosinophilia is an alternative route
- marked tissue eosinophilia — conventionally more than 20% of nucleated cells in the marrow, or extensive tissue infiltration on biopsy — satisfies the hypereosinophilia criterion without the blood count ever reaching 1.5
- the 2024 update did not move the threshold
- the WHO and International Consensus Classification 2024 update on eosinophilic disorders retains hypereosinophilia as a peripheral blood eosinophil count above 1.5 ×10⁹/L. The bands on this page are current, not historical
Worked example
AEC 2.4 ×10⁹/L, single count, organs not yet assessed
2.4 ×10⁹/L is 1.5 or above and below 5.0 → moderate eosinophilia
It is also in the hypereosinophilia range — but hypereosinophilia needs a second count at least a month apart, and there is only one
So the label is moderate eosinophilia, not hypereosinophilia, and certainly not hypereosinophilic syndrome
What should not wait for the repeat is the organ assessment. Troponin and echocardiography now
Change the third input to 'damage attributable to eosinophils' and the result becomes eosinophil-mediated organ damage — a crit, on the same count
Drop the count to 0.9 with that same organ finding and the result is organ damage below the hypereosinophilia threshold, still a crit. The count does not govern
Severity, and the two definitions the numbers feed
| AEC (×10⁹/L) | AEC (/µL) | Severity | Relationship to hypereosinophilia |
|---|---|---|---|
| < 0.5 | < 500 | Not eosinophilic | — |
| 0.5 – 1.5 | 500 – 1,500 | Mild | Below the threshold |
| 1.5 – 5.0 | 1,500 – 5,000 | Moderate | In range — becomes hypereosinophilia if persistent |
| > 5.0 | > 5,000 | Severe | In range — becomes hypereosinophilia if persistent |
Where the sources disagree — the repeat interval
| Source | Criterion |
|---|---|
| StatPearls, Eosinophilia (NBK560929) | > 1,500/mm³ on two occasions at least one month apart, or marked tissue eosinophilia |
| StatPearls, Hypereosinophilic Syndrome (NBK599558) | > 1.5 ×10⁹/L on two examinations > 1 month apart; marrow eosinophils > 20% of nucleated cells |
| Merck Manual Professional Edition | > 1,500/µL persisting ≥ 6 months, or on 2 or more occasions at least 2 weeks apart |
| WHO / ICC 2024 update | Hypereosinophilia as a peripheral blood eosinophil count above 1.5 ×10⁹/L — threshold unchanged from earlier classifications |
Why a moderate count can outrank a severe one
| Picture | Count | Urgency |
|---|---|---|
| Eosinophilic myocarditis with a raised troponin | 1.8 ×10⁹/L | Immediate. Cardiac involvement is the complication that kills, and it is not predicted by the height of the count |
| DRESS with rash, fever and deranged liver enzymes | 6.0 ×10⁹/L | Urgent — but the treatment is stopping the drug, and the organ risk is what is driving it, not the 6.0 |
| Well patient, atopic, incidental finding | 6.2 ×10⁹/L | Needs a thorough assessment, but the count alone is not an emergency |
| Eosinophilic granulomatosis with polyangiitis, new mononeuritis | 0.9 ×10⁹/L | Urgent. Below the eosinophilia-to-hypereosinophilia threshold entirely, and a vasculitis |
Things that make the count lie
| Factor | Effect |
|---|---|
| Corticosteroids | Drop the count within hours. A normal eosinophil count on steroids is close to uninterpretable |
| Time of day | Eosinophils follow a rhythm inverse to cortisol, with a morning trough — which is when most outpatient blood is taken |
| Acute bacterial infection | Suppresses the count, potentially masking an underlying eosinophilia |
| Tissue sequestration | In organ-limited disease the cells are in the tissue, not the blood, and the blood count can be normal |
| Percentage instead of count | Every threshold here is an absolute count. 15% of a white cell count of 3.0 is 0.45 ×10⁹/L, which is not eosinophilia at all |
Three questions the count cannot answer on its own
An absolute eosinophil count above 0.5 ×10⁹/L is eosinophilia, and the conventional severity grades divide what follows into mild from 0.5 to 1.5, moderate from 1.5 to 5.0, and severe above 5.0. Those bands are stable across sources and the 2024 WHO and International Consensus Classification update on eosinophilic disorders leaves the key threshold — hypereosinophilia at above 1.5 ×10⁹/L — exactly where it was. So the numbers are not the difficult part.
The difficult part is that three different questions get asked of the same count and only one of them is arithmetic. The severity grade describes a single measurement. Hypereosinophilia describes a pattern: 1.5 ×10⁹/L or above on two occasions at least a month apart, or marked tissue eosinophilia, which can substitute for the blood criterion entirely. And hypereosinophilic syndrome adds a third requirement that has nothing to do with the count at all — organ or tissue damage attributable to the eosinophilia, with other causes of that damage excluded. A single count of 3.0 is moderate eosinophilia. It is not hypereosinophilia, and calling it hypereosinophilic syndrome skips two separate pieces of work.
The interval over which persistence must be shown is the one place the current sources genuinely differ. StatPearls, in both its eosinophilia and its hypereosinophilic syndrome chapters, says at least one month. The Merck Manual gives two alternatives: six months of persistence, or two occasions at least two weeks apart. This interpreter asks for the month, which is the stricter of the short options, and prints the alternatives rather than pretending the field has settled.
What matters more than any of it is that the blood count does not tell you about the tissue. The Merck Manual is unusually direct: ‘Cardiac involvement is not predicted by the degree or duration of eosinophilia’, some patients have very high counts with no tissue involvement, others have tissue damage at lesser elevations, and ‘there is no set level of eosinophilia at which organ damage occurs or at which treatment must be started’. Eosinophilic granulomatosis with polyangiitis, eosinophilic oesophagitis and eosinophilic pneumonia can all cause serious organ disease with a blood count below the hypereosinophilia threshold, because the pathology is where the cells have migrated rather than where they are being counted. A patient on corticosteroids will have a suppressed blood count and unaltered tissue disease.
None of which means the count is uninformative. A 2026 single-centre cohort found that the absolute eosinophil count correlated strongly with the number of organs involved, with an odds ratio of 50 for multi-organ involvement — while concluding that it should be read as ‘a risk marker that informs the urgency and breadth of investigation, not a stand-alone measure of tissue injury’. That is the right reading and it is how this interpreter is built. A high count means look harder and look sooner. It never means the search can be skipped, and a modest count never means the search was unnecessary.
Two practical points run underneath everything. Take the travel and residence history before the specialist tests, because helminth infection is the commonest cause worldwide — and exclude strongyloidiasis before giving corticosteroids, since immunosuppression can convert a chronic infection into fatal hyperinfection. And in anyone with persistent unexplained hypereosinophilia, screen for the tyrosine kinase gene fusions (PDGFRA, PDGFRB, FGFR1, JAK2), because that group is imatinib-responsive and is otherwise missed.
Frequently asked questions
What absolute eosinophil count counts as eosinophilia?
Above 0.5 ×10⁹/L, which is above 500 per microlitre. Mild eosinophilia runs from 0.5 to 1.5, moderate from 1.5 to 5.0, and severe above 5.0 ×10⁹/L. Every one of these thresholds is an absolute count, never a percentage.
What is the difference between hypereosinophilia and hypereosinophilic syndrome?
Hypereosinophilia is a count of 1.5 ×10⁹/L or above on two occasions at least a month apart, or marked tissue eosinophilia. Hypereosinophilic syndrome is that plus organ or tissue damage attributable to the eosinophilia, with other causes of the damage excluded. The difference is the organ assessment, not the number.
Did the 2024 WHO/ICC update change these thresholds?
No. The 2024 update on eosinophilic disorders retains hypereosinophilia as a peripheral blood eosinophil count above 1.5 ×10⁹/L, working from the 2022 WHO and International Consensus Classification. The bands used here are current.
Can there be organ damage with a count below 1.5?
Yes, and it is common. Eosinophilic granulomatosis with polyangiitis, eosinophilic oesophagitis, eosinophilic pneumonia and eosinophilic fasciitis can all cause substantial tissue disease at modest blood counts, because the cells are in the tissue. Prior corticosteroids will also suppress the blood count without touching the tissue.
Is a higher count more urgent?
Usually, but not reliably for any individual. The count correlates with the number of organs involved across a population, which is a reason to look harder and faster when it is high. But cardiac involvement specifically is not predicted by the degree or duration of eosinophilia, and there is no level at which organ damage begins or treatment must start.
Why does the repeat have to be a month apart?
Because the definition is about persistence, and a single count can reflect a transient reaction. The interval is not agreed: at least one month in the StatPearls and consensus-derived criteria, two weeks or six months of persistence in the Merck Manual. This interpreter uses the month.
What should be done before corticosteroids?
Exclude strongyloidiasis. Immunosuppression can convert a chronic Strongyloides infection into fatal hyperinfection syndrome, and the travel and residence history plus serology takes less time than the harm. Steroids also suppress the eosinophil count within hours, which destroys the information the count was carrying.
Related calculators
References
- Kovalszki A, Weller PF. Eosinophilia. StatPearls, NCBI Bookshelf NBK560929.
- Hypereosinophilic syndrome. StatPearls, NCBI Bookshelf NBK599558.
- Shomali W, Gotlib J. World Health Organization and International Consensus Classification of eosinophilic disorders: 2024 update on diagnosis, risk stratification, and management. Am J Hematol. 2024;99(5):946–968.
- Merck Manual Professional Edition. Hypereosinophilic syndrome.
- Optimizing diagnostic and management pathways for patients with eosinophilia of unknown origin: a multidisciplinary protocol. Front Med. 2025;12:1544047.
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.