NIV in asthma

Audio clip: Adobe Flash Player (version 9 or above) is required to play this audio clip. Download the latest version here. You also need to have JavaScript enabled in your browser.

Asthma is common. It affects our patients, our family and our friends. Thankfully, severe and poorly controlled asthma is less common however, it is this asthma that we often see in the Emergency Department in the form of acute exacerbations.

It goes without saying that we need to be mindful that all patients with acute asthma can deteriorate quickly but how do we spot the individuals with greater risk of a near-fatal or fatal outcome? Risk factors outlined by the British Thoracic Society include non-adherence, previous hospital admission or ED attendance, and failure to measure pulmonary function.

When it comes to the medical management of acute asthma we are all familiar with the guidelines and the approach is fairly straightforward due to the limited number of agents available. That approach should involve an escalating strategy in line with the patient’s disposition based on respiratory effort, PEF, vital signs, and blood gases.

What’s the next course of action when our medical strategies don’t work? Should we run for the phone to bleep the ITU reg to get the patient tubed and admitted to ITU? Is there another option that might be useful such as a non-invasive strategy of intermittent positive pressure ventilation (NIPPV) as used in COPD and acute pulmonary oedema? On that matter the BTS/SIGN guidelines are less prescriptive and there is variable practice across the UK.

Screen Shot 2015-08-27 at 07.56.22


BTS Asthma Guidelines

If we are to consider using NIPPV in acute asthma we really should know how it works, in whom, and when it should be started. The nomenclature around NIPPV is, quite frankly, confusing. There are myriad terms many of which are used interchangeably and include NIV, NIPPV, NPPV, CPAP, BiPAP, ePAP, iPAP, and countless others.

In simple terms NIV is a form of ventilatory support that does not require the use of an artificial airway. Most commonly it is delivered by means of a tight-fitting face mask (or similar device, such as a hood) that allows oxygen to be delivered under pressure measured in cmH20.

In the context of type two (hypercapnic) respiratory failure that may occur in acute asthma, NIV is delivered using two different pressure support levels, often known as the inspiratory- and expiratory-positive airway pressures (iPAP and ePAP, respectively) and collectively referred to as BiPAP (bi-level positive airway pressure). Typical initial pressure levels would be 12/5 cmH20. Increases in iPAP serve to increase tidal volume and reduce CO2 levels; increases in ePAP serve to splint the airways and help to improve oxygenation. When only a single continuous positive airway pressure is used, this is known as CPAP with a typical starting pressure of 5 cmH20.

It’s pretty obvious that asthmatics have tight airways. This means when they exhale, they do so against a pressure known as intrinsic PEEP, also known as auto-PEEP. NIPPV is thought to offset this auto-PEEP and have a direct bronchodilatory effect independent of drug dispersion. As gas flow improves atelectatic lung re-expands and improves V/Q mismatch.

Put simply, NIPPV reduces the work of breathing and improves ventilation. Compared to intubation NIPPV has the advantage that it can be applied intermittently, sedation is not required, and there is reduced incidence of ventilator and nosocomial pneumonia.

With all of the above said, and clear guidelines advocating the use of NIPPV for COPD and acute pulmonary oedema, should we not simply be able to extrapolate its use to our critically unwell asthmatic?

In an overview of the topic up to 2008 Sutton and Ferguson published work in the EMJ asking ‘is NIV an effective intervention for patients with acute exacerbations of asthma?

They searched the literature including the Cochrane register, MEDLINE, EMBASE, and CINAHL to identify 282 papers or which nine were relevant and of sufficient quality and included three RCTs.

Each of the papers looked at slightly different outcomes including degree of shortness of breath, PEFR, oxygen saturations, heart and respiratory rate, blood pH, intubation rate, hospitalisation and change in FEV1 over time.

Generally, each of the papers concluded that use of NIV in asthma resulted in improvements in the primary outcomes compared to controls however, most studies including the RCTs had small sample sizes and methodological issues.

Sutton and Ferguson concluded that the published evidence up to 2008 did not support the use of NIV in acute asthma but did support further better-wrought RCTs.

The main paper we are going to discuss today comes from the Cochrane database and was published by lead author Lim in 2012.

The meta-analysis included RCTs that compared the treatment of asthma with usual medical care plus NIPPV versus usual medical care alone. Studies of adults presenting with acute severe asthma as the primary presentation to the ED were included using standard internationally defined definitions (BTS, American Thoracic Society). Studies looking at COPD or with a primary diagnosis of pneumonia were excluded.

Of 746 identified abstracts, five completed studies were included for meta-analysis whilst one study was awaiting publication. Of the five studies a detailed manuscript was unavailable for one; data extraction was based on the published abstract only. All RCTs were single centre and from different geographic regions including India, Brazil and Israel. A total of 206 patients were included across each of the five trials with a range from 21 to 63 patients in each.

In each of the RCTs patients were randomised to either intervention (NIPPV) or control group. Blinding was not undertaken due to the nature of the intervention and only one study used sham NIPPV in the control group. Four of the five studies utilised BiPAP for intervention whilst CPAP was used for intervention in one trial. Airway pressures were titrated in two studies, pre-determined in two studies, and not reported in one study. Duration of intervention ranged from nine minutes to 14 hours: two studies delivered intervention for less than one hour, two studies for more than one hour and one study intervention duration could not be determined.

Primary outcome measures were 1) need for endotracheal intubation (ETI) and 2) mortality. Secondary outcome measures included respiratory rate, pH, lung function measurements, length of hospital stay, length of ITU stay, treatment failure defined as the combination of mortality, ETI, and intolerance to allocated treatment, symptom score, and complications.

Only two of the five studies the 5 studies (86/203 participants) assessed mortality and tracheal intubation, and neither study had evidence of an effect for these primary out- comes.

In the two studies with mortality data, there were no instances of death in either; therefore meta-analysis could not be reported. With regards the need for tracheal intubation, two studies with 45 participants in the NIPPV arm and 41 in the control arm were assessed with no evidence of an effect between groups (RR 4.48; 95% CI 0.23 to 89.13). In the location subgroup analysis, there were no statistically significant differences in risk of tracheal intubation in both the ICU (Gupta 2010) and ward (Soroksky 2003) subgroups

In general, NIPPV provided favourable outcomes with respect of the majority of secondary outcomes. In summary:

  • Number of hospital admissions (one paper, n=33 (NIPPV n=3/17; control n=10/16): statistically significant in favour of NIPPV (RR 0.28, 95%CI 0.09-0.84)
  • Length of hospital stay (two studies, n=86): Meta-anlysis not performed due to methodological issues in each study. One found in favour of intervention and one in favour of contrl (NIPPV = 4 days, control 2.5 days)
  • Length of ICU stay in ours (two studies): One study in ITU (n=53) reported a statistically significant benefit but meta-analysis could not be performed due to methodological issues. The second study (n=30) did not find a statistically significant benefit (mean difference 0.3, 95%CI -0.63 to 1.23)
  • Treatment failure (two studies, NIPPV n=45; control, n=41): Overall, no statistically significant benefit was identified for this outcome on meta-analysis (RR 0.73; 95% CI 0.21 to 2.53)
  • None of the studies formally assessed the complications of NPPV. One study with 53 participants reported that NPPV was well tol- erated by all participants without serious adverse effects; however, frequent complaints of pain in the nasal bridge area were reported

Lung function tests (PEF, FVC, FEV1, MV, IC) were reported with mixed results:

  • PEF (four studies, n=153). Three out of four reported statistically improved PEF with NIPPV however, only two studies were meta-analysed due to data availability: NPPV was found to confer an absolute PEF improvement of 19.97% predicted, which was both clinically and statistically significant (MD 19.97%; 95% CI 15.01 to 24.93
  • FVC (three studies, n=90). Mixed results. Two available for meta-analysis (n=66) on the hospital ward and finding statistically and clinically significant results in favour of NIPPV (MD 12.27; 95% CI 4.38 to 20.16)
  • FEV1 (four studies, n=140). Two available for meta-analysis (n-66, ward patients) finding a statistically and clinically significant improvement of FEV1 as a % predicted (MD 14.02; 95% CI 7.73 to 20.32)
  • ABG (one study, n=53). Overall no statistically significant differ- ences were found between control and intervention groups.
  • Respiratory rate (five studies, n=203). Three studies (n=146) available for meta-analysis and indicated that the intervention provided a statistically but not clinically significant improvement in respiratory rate (MD -1.42; 95% CI -2.77 to – 0.07)

So what’s the take home message?

Whilst each of the individual studies concluded that the addition of NIPPV to standard medical therapy may be beneficial the results of meta-analysis do not give us any further clarity. The main problem with each of the studies is sample size. Further methodological problems cloud the picture further. What we can say is that there seems to be a trend towards clinically and statistically significantly improvement in secondary outcomes in favour of NIPPV.

Should we be trialling NIPPV in the ED for our critically unwell asthmatics? Probably not based on the current evidence. The cost of resource allocation and impact to the patient should not be underestimated and until further, larger, better-wrought trials are undertaken it is unlikely we will see NIPPV added to the guidelines.


Relevant resources

Setting up NIPPV via CORE EM

How to set up NIV from Shrewsbury Hospital


  1. Sutton L, Ferguson C. BET 3. Non-invasive positive pressure ventilation for patients with acute exacerbations of asthma. Emerg Med J 2009;26:59-60
  2. Lim WJ, Akram RM, Carson KV, et al. Non-invasive positive pressure ventilation for treatment of respiratory failure due to severe acute exacerbations of asthma. Cochrane Database Syst Rev. 2012 Dec 12;12:CD004360




Add this content to your CPD

Leave a Comment

Your email address will not be published.

This site uses Akismet to reduce spam. Learn how your comment data is processed.