FeNO can rule asthma in or out in primary care when spirometry is normal, if inhaled steroids are accounted for

Design

Systematic review and meta-analysis of diagnostic accuracy with a prospective primary care validation cohort

Population

30 studies (7,141 adults and 989 children) with suspected asthma and normal spirometry; DIVE cohort of 123 adults

Primary outcome

Area under the ROC curve for airway hyper-responsiveness on bronchial provocation testing

Effect

FeNO AUC 0.80 (95% CI 0.73 to 0.85) in adults; 65 ppb or more PPV 80%; below 20 ppb NPV 93%

A meta-analysis of 30 studies (22 in adults with 7,141 patients, nine in children with 989) looked at how well FeNO and blood eosinophil count identify airway hyper-responsiveness on bronchial provocation testing in patients aged 5 years and older with suspected asthma and normal spirometry. It was registered in PROSPERO. A single-centre primary care cohort (DIVE, 123 patients, 22 positive on methacholine testing) was used to test the findings prospectively.

FeNO had an AUC of 0.80 (95% CI 0.73 to 0.85) in adults and 0.75 (0.65 to 0.82) in children. In adults, a threshold of 48 ppb or more had a positive likelihood ratio of 10.1. At a 20% prevalence, 65 ppb or more gave a positive predictive value of 80%, and below 20 ppb a negative predictive value of 93%, with 18% of cases missed. Blood eosinophils were weak (AUC 0.65) and added nothing. In DIVE, FeNO above 65 ppb had a positive predictive value of 73% (wide interval 39 to 94) and below 20 ppb a negative predictive value of 92%. Accuracy fell within two weeks of inhaled corticosteroid exposure.

FeNO is not available everywhere, and cost and local reference ranges matter in India. The rule-out threshold still misses nearly one in five cases at 20% prevalence, so a negative result in a symptomatic patient should not end the work-up.

• Use FeNO as a rule-in test: a high value in a symptomatic patient with normal spirometry supports asthma
• Treat a very low FeNO as lowering, not excluding, the probability of asthma in a patient with convincing symptoms
• Test before starting inhaled corticosteroids, because accuracy fell within two weeks of exposure
• Do not rely on blood eosinophil count alone to diagnose asthma
• Use bronchial provocation or specialist review where the picture remains uncertain

Why it matters

Asthma is often diagnosed on symptoms alone because provocation testing is hard to access, and this gives usable thresholds.

Don't overread it

Thresholds depend on prevalence and recent steroid use, the primary care validation was in 123 patients at one centre, and a similar paediatric validation is still under way.

The statistics, in plain English

A positive predictive value of 80% means that of 100 people with a high FeNO at this prevalence, about 80 truly have airway hyper-responsiveness. A negative predictive value of 93% with 18% missed diagnoses means a low FeNO makes asthma less likely but leaves a real chance of missing it. Predictive values depend on prevalence, so they will differ in a different clinic population. The primary care cohort was small, which is why its confidence intervals are wide.

Read the rest in the app

You have read your two free briefings this month. The app carries all 27 specialties, every morning, free — and this finding is waiting in it.

Get Daily Doctor — free

Scan to keep reading on your phone. No account needed to start.