Can a blood test reveal what kind of airway inflammation asthma has?
A routine blood eosinophil count identified eosinophilic airway inflammation about as well as exhaled nitric oxide, but blood neutrophils were a poor guide to neutrophilic asthma.
Source
Distribution of sputum cellular phenotype in a large asthma cohort: predicting factors for eosinophilic vs neutrophilic inflammation
Study at a glance
- Design
- Cross-sectional — Retrospective single-clinic series with same-day sputum induction, blood tests, exhaled NO and lung function; ROC and logistic regression to predict sputum phenotype, then tested in a separate later group of patients.
- N
- N=508 · 508 asthma patients with successful sputum induction (2005-2011); prediction equations were checked in a separate validation group of 178 asthmatics (138 for the eosinophil equation because IgE was missing in 40).
- Population
- Adults with asthma of all severities attending the University Asthma Clinic of Liege, Belgium.
- Outcome
- Sputum inflammatory phenotype (eosinophilic, neutrophilic, mixed, paucigranulocytic) and the accuracy of blood eosinophils, blood neutrophils and exhaled NO for predicting it.
Structured fields used in claim comparison tables when every cited study has a complete layer.
What they did
The researchers reviewed 508 asthma patients from a Belgian university clinic who produced a usable induced-sputum sample. On the same day each patient had exhaled nitric oxide (FeNO), spirometry and blood tests. Sputum with at least 3% eosinophils was classed as eosinophilic and at least 76% neutrophils as neutrophilic. They used ROC curves and multiple logistic regression to find blood and clinical markers of each phenotype, then tested the resulting equations in a separate group of 178 later patients.
What they found
42% of patients had eosinophilic, 40% paucigranulocytic, 16% neutrophilic and 3% mixed inflammation. Blood eosinophils correlated moderately with sputum eosinophils (r = 0.6); a cut-off of 220 per mm3 gave 77% sensitivity and 70% specificity, performing as well as FeNO. Blood eosinophil percentage, FeNO, IgE and a lower FEV1/FVC ratio independently predicted eosinophilic inflammation, and the combined equation showed moderate agreement in the validation group (kappa 0.59). Blood neutrophils correlated only weakly with sputum neutrophils, and older age and higher functional residual capacity were the only independent predictors of neutrophilic asthma; that equation agreed poorly in validation (kappa 0.24).
The limits
What it doesn't show
This is a retrospective series from one specialist clinic and includes only patients whose sputum induction succeeded, so it may not represent all people with asthma, particularly in primary care. Phenotype was measured once, although sputum inflammation can change over time and with treatment, and many patients were on inhaled steroids. The study shows which markers track sputum cells but did not test whether treating based on blood counts improves outcomes. The paper does not include a dedicated limitations section, and the neutrophilic threshold of 76% is higher than many other studies use, which affects the phenotype proportions.
Key terms
- Induced sputum
- Airway mucus coughed up after inhaling salty mist, used to count inflammatory cells in the airways without a bronchoscopy; considered the gold standard for inflammatory phenotype.
- Eosinophilic asthma
- Asthma with many eosinophils (here at least 3%) in the sputum; it tends to respond well to inhaled corticosteroids and anti-IL5 drugs.
- Paucigranulocytic asthma
- Asthma with neither raised eosinophils nor raised neutrophils in the sputum.
- FeNO
- Fraction of exhaled nitric oxide, a breath test used as a marker of eosinophilic airway inflammation.
- ROC curve / AUC
- A plot of sensitivity against false-positive rate across cut-offs; the area under it summarises how well a marker separates two groups (0.5 = chance, 1 = perfect).
- Cohen's kappa
- A measure of agreement between two classifications that corrects for agreement expected by chance.
Flashcards
0 of 11 answers reviewed
Research intelligence for this paper
See its role on concept claims, tensions it is part of, placement history, and related discoveries.
Quiz yourself
Which sputum phenotype was least common in this asthma population?
Common questions
Why does it matter which inflammatory phenotype someone's asthma has?
Eosinophilic inflammation predicts a good response to inhaled steroids and anti-IL5 drugs, whereas there is little evidence that steroids help short-term control when eosinophils are not raised. Knowing the phenotype can guide treatment better than the asthma label alone.
If blood eosinophils work as well as FeNO, which should clinicians use?
Both performed similarly. FeNO gives instant results and is more comfortable, while a blood count is widely available. They were independent predictors, so they may capture different mechanisms and could complement each other.
Why don't blood neutrophils predict airway neutrophils?
The authors suggest airway neutrophilia may depend on how activated neutrophils are and how they are drawn into the lung, rather than on how many circulate in the blood.
More on Respiratory medicine