Bronchiolitis is a common illness affecting the lower (smaller) respiratory airways in infants (younger than 24 months of age). Usually caused by a viral infection, it results in breathing problems, including cough, fast breathing, wheezing and can cause poor feeding. It is a major cause of hospitalisation in infants. Current treatment involves supporting infants to breath until the infection clears. An emerging method to support breathing is using blended, heated, humidified air and oxygen, through nasal cannulae (tubes) at flow rates higher than two litres per minute, which is the maximum for conventional dry oxygen delivery. This is known as high-flow nasal cannula therapy and it allows the comfortable delivery of high flow rates of an air/oxygen blend which may improve ventilation. This may lead to a reduced need for invasive respiratory support (e.g. intubation) and may have a clinical advantage over other treatments by preventing drying of the upper airway. This review assessed the effects of high-flow nasal cannula therapy, compared with other respiratory support, in the treatment of infants with bronchiolitis.
One study (19 participants) met our inclusion criteria. It showed that high-flow nasal cannula therapy is well tolerated as a treatment for bronchiolitis. Oxygen saturations (blood oxygen levels) were better at eight and 12 hours in participants receiving high-flow nasal cannula therapy than in those receiving oxygen therapy via a head box, but were similar between groups at 24 hours, although this may have been due to higher oxygen flow rates in the high-flow nasal cannula group. There was no clear evidence of a difference between the two groups in the duration of oxygen therapy, length of hospitalisation and time to discharge. No adverse events were reported in either group.
There is insufficient evidence to determine the effectiveness of high-flow nasal cannula therapy for treating bronchiolitis in infants. The included study provides some indication that HFNC therapy is feasible and well tolerated. However, our evidence is based on one low-quality, small study with uncertainty about the effects and some possibility of bias arising from the study methods. Further research is required to determine the role of high-flow nasal cannula therapy in the management of bronchiolitis in infants. The results of six ongoing studies identified will contribute to the evidence in future updates of this review.
The evidence is current to May 2013.
There is insufficient evidence to determine the effectiveness of HFNC therapy for treating infants with bronchiolitis. The current evidence in this review is of low quality, from one small study with uncertainty about the estimates of effect and an unclear risk of performance and detection bias. The included study provides some indication that HFNC therapy is feasible and well tolerated. Further research is required to determine the role of HFNC in the management of bronchiolitis in infants. The results of the ongoing studies identified will contribute to the evidence in future updates of this review.
Bronchiolitis is a common lower respiratory tract illness, usually of viral aetiology, affecting infants younger than 24 months of age and is a frequent cause of hospitalisation. It causes airway inflammation, mucus production and mucous plugging, resulting in airway obstruction. Effective pharmacotherapy is lacking and bronchiolitis is a major cause of morbidity and mortality.
Conventional treatment consists of supportive therapy in the form of fluids, supplemental oxygen and respiratory support. Traditionally oxygen delivery is as a dry gas at 100% concentration via low-flow nasal prongs. However, the use of heated, humidified, high-flow nasal cannula (HFNC) therapy enables delivery of higher inspired gas flows of an air/oxygen blend, up to 12 L/min in infants and 30 L/min in children. Its use provides some level of continuous positive airway pressure to improve ventilation in a minimally invasive manner. This may reduce the need for invasive respiratory support thus potentially lowering costs, with clinical advantages and fewer adverse effects.
To assess the effects of HFNC therapy compared with conventional respiratory support in the treatment of infants with bronchiolitis.
We searched CENTRAL (2013, Issue 4), MEDLINE (1946 to May week 1, 2013), EMBASE (January 2010 to May 2013), CINAHL (1981 to May 2013), LILACS (1982 to May 2013) and Web of Science (1985 to May 2013). In addition we consulted ongoing trial registers and experts in the field to identify ongoing studies, checked reference lists of relevant articles and searched conference abstracts.
We included randomised controlled trials (RCTs) or quasi-RCTs which assessed the effects of HFNC (delivering oxygen or oxygen/room air blend at flow rates greater than 4 L/min) compared to conventional treatment in infants (< 24 months) with a clinical diagnosis of bronchiolitis.
Two review authors independently used a standard template to assess trials for inclusion and extract data on study characteristics, 'Risk of bias' elements and outcomes. We contacted trial authors to request missing data. Outcome measures included the need for invasive respiratory support and time until discharge, clinical severity measures, oxygen saturation, duration of oxygen therapy and adverse events.
We included one RCT which was a pilot study with 19 participants that compared HFNC therapy with oxygen delivery via a head box. In this study, we judged the risk of selection, attrition and reporting bias to be low, and we judged the risk of performance and detection bias to be unclear due to lack of blinding. The median oxygen saturation (SpO2) was higher in the HFNC group at eight hours (100% versus 96%, P = 0.04) and at 12 hours (99% versus 96%, P = 0.04) but similar at 24 hours. There was no clear evidence of a difference in total duration of oxygen therapy, time to discharge or total length of stay between groups. No adverse events were reported in either group and no participants in either group required further respiratory support. Five ongoing trials were identified but no data were available in May 2013. We were not able to perform a meta-analysis.