In two recent assessments NICE concluded that the current evidence recommends FeNO should be routinely measured in all patients with suspected asthma and may have a role in asthma monitoring in selected patients. However, these conclusions have proved controversial. Acknowledging that there are differences in opinion, as a GP and clinical academic specialising in asthma, the following is a personal view based on clinical experience of using FeNO in routine general practice for more than 10 years.
What is FeNO and how is it measured?
FeNO stands for ‘fractional exhaled nitric oxide’. Nitric oxide is present in low levels in the ambient atmosphere and is produced in normal lungs at a constant low level, but production is greatly increased in inflamed airways, particularly in the eosinophilic, allergic pattern of inflammation classically characteristic of asthma. This is the type of inflammation that responds very well to standard doses of inhaled steroids, as opposed to the neutrophilic pattern of inflammation classically seen in COPD, which is far less steroid responsive.
FeNO can be measured as parts per billion (ppb) in expired air by blowing into analysers, known as ‘inflammometers’. We have long been aware of the crucial role of inflammation in asthma, and inhaled steroids are central to treatment but we have been unable to measure inflammation outside of research settings until recently.
FeNO can now be measured using simple and relatively inexpensive portable analysers by a steady blow into the mouthpiece, with a numerical readout (typically between 0 and 100ppb) in less than a minute. It’s simple to do, although basic training is needed for non-clinical staff. FeNO measurement is, however, simpler and quicker than spirometry, for example. My experience is that it’s possible in nearly all adults and most children aged over six years. There are several monitors commercially available, and costs for measurement are falling – currently the cost per test would typically be less than £10.
When should FeNO be used?
In diagnosis and deciding whether to prescribe inhaled steroids
NICE suggests that measuring FeNO should be standard in suspected asthma before starting inhaled steroids, as this will save the NHS money. We had access to FeNO in our practice in Gloucestershire, which we found to be useful and time saving. Patients generally present with non-specific symptoms such as breathlessness, chest tightness, cough and wheeze, and while some will have reversible obstruction apparent on spirometry (confirming diagnosis), this is usually not the case and its absence doesn’t ‘rule out’ asthma.
In those cases, peak flow monitoring may be helpful, although it is time- consuming for patient and staff, poorly adhered to and may again be inconclusive. Often we are driven to a ‘trial of treatment’ – by prescribing a steroid inhaler and seeing whether it helps symptoms. However, a raised FeNO before the steroid is started is an excellent predictor of a response to the therapy. We found that it was usually possible to make (or rule out) a diagnosis and start effective treatment after a single consultation, saving time and money.
A FeNO level above 40 makes asthma, and a response to inhaled steroid, very likely while a level under 20 makes it unlikely. An intermediate level warrants further investigation and we would usually start a trial of inhaled steroid as a form of diagnostic test and review, with a re-assessment of symptoms and FeNO at the next visit. A fall in FeNO levels and improved symptoms would be compelling, whereas a fall in symptoms alone has many possible explanations.
Smoking will tend to lower FeNO and blunt the signal to some extent, but we would still measure FeNO in smokers, although we’d encourage the patient to abstain for six hours and be aware that lower levels are expected.
An important point is that raised FeNO indicates inflamed airways and steroid responsiveness rather than asthma per se. So we often test FeNO in patients with a chronic undiagnosed cough even if they don’t have evidence of asthma. Chronic undiagnosed cough is referred to as eosinophilic bronchitis in specialist care and sometimes as ‘cough variant asthma’ – although this terminology is imprecise since asthma really means bronchoconstriction. We found that if the FeNO is raised then the cough will nearly always respond to inhaled steroids, and never responds if FeNO is low.
Similarly, we sometimes measure FeNO in patients with irreversible or partially reversible airways obstruction on spirometry, such as in those with COPD. In our experience, raised FeNO identified patients who responded well to inhaled steroids. We are all aware of the need to use inhaled steroids with care and caution in COPD, and also of the possibility of ‘asthma-COPD overlap syndrome’, or ACOS, in which inhaled steroids should be used early. In our experience, a raised FeNO in patients with COPD was a useful predictor of response and helpful in deciding whether or not to use inhaled steroids.
Monitoring – what treatment option in poorly controlled asthma?
The use of FeNO in monitoring asthma is less clear than in diagnosing it, and further research is urgently needed. We are currently awaiting a National Institute of Health Research (NIHR) funding decision on this very topic. However, we have found FeNO very helpful in selected cases. We would not measure FeNO in every patient every time because of cost, and our rule of thumb on whether to measure was ‘will my management decision be altered by the FeNO result’?
In an uncontrolled patient already on inhaled steroids, a raised FeNO (above 20ppb) would direct us to control inflammation. A low reading would not lead us to increase the inhaled steroid; instead we would think of other options such as increasing bronchodilators by switching to a combination inhaled steroid and long-acting ß-agonist (LABA), or adding a long-acting muscarinic receptor agonist (LAMA) if the patient is already on a LABA. Non-drug treatments may also be considered here; for example there is now strong evidence for teaching breathing control exercises to people who are symptomatic despite treatment.
A raised FeNO in a patient on inhaled steroids would indicate that inflammation was not fully controlled, but this can occur for several reasons. The commonest in my experience is non-compliance. The patient might not be taking the steroid, usually because they don’t believe it works – in that case, ask the patient and check repeat prescription records. The next most common reason is poor inhaler technique, meaning that medication isn’t getting in – observe their technique and consider a different inhaler device. If the patient is taking their inhaled steroid correctly, they need a bigger dose – so we would go for a high dose of steroid rather than a combined steroid and LABA, as the guidelines would suggest.
If FeNO remains high and control inadequate despite high-dose inhaled steroid, the patient has genuinely severe, steroid-resistant disease and should be referred to a difficult asthma clinic for consideration of a biological treatment such as anti-IgE or mepolizumab, as FeNO predicts response to these expensive but highly effective treatments.
We also found that demonstrating a raised FeNO to a poorly compliant patient could be an adherence aid. Showing objective evidence of inflammation seemed to make them more willing to give inhaled steroids a proper trial. A fall in FeNO (and improved subjective control) after taking their inhaled steroid for a month often seemed to convince patients that the medication was really needed.
Whether to reduce or stop inhaled steroids in a controlled patient
There is good evidence that some patients with good control can reduce or even come off inhaled steroids, although for many it’s the ongoing steroid that is providing control. If we were considering steroid reduction, we would measure FeNO, and only proceed if the level was below 20ppb. If so, we would reduce and remeasure FeNO after a month. If the level rises we would restart.
Is FeNO worthwhile in general practice?
Yes, and GPs managing asthma should have access to this test, but we shouldn’t have to pay for it out of our own pockets. We don’t pay for other important tests that improve care, and there would be an outcry if we did. I think the main reason the NICE proposals have proved so controversial is that there was no clarity about who would pick up the bill for the FeNO tests, and in the current climate, GPs are worried it would fall to them. The NICE model of how FeNO should be provided in routine care hasn’t been worked out properly.
I have no doubt that FeNO testing enabled me to provide better asthma care, and I don’t know any clinician who has used FeNO in their practice who doesn’t want to carry on, including the practices in the NICE pilot – providing they didn’t have to pay for it. I know that my hospital-based colleagues consider FeNO essential, and I can’t see any reason why those of us managing asthma in the community should not also have access to this test, particularly when NICE says it will save the NHS money.
It is up to the The Department of Health and Social Care to work out a model where the test is available to primary care. If a practice elects to take on measurement itself (which is more convenient for patients and more satisfying for the GPs), it should be financially rewarded for doing so. Similar models have been evolved for INR monitoring, for example. For those who don’t wish to do this, there should be alternative models, such as local ‘hubs’ – an interested practice could provide a service for local colleagues.
Although there are still important questions to be answered, I’d encourage anyone who can to use FeNO in their asthma clinic and make up their own mind about how useful it is.
Professor Mike Thomas is a former GP and a professor of primary care research at the University of Southampton. He has acted as an expert adviser on several UK NICE evaluations in the respiratory field, including the most recent asthma guidance, and was a member of the UK National Review of Asthma Deaths steering committee writing group
Competing interests: none declared
