• Intensive Care
• Respiratory
• Audit

The National Institute for Health and Care Excellence (NICE) recommends the use of continuous positive airway pressure (CPAP) for children with impending respiratory failure due to bronchiolitis.1 Anecdotal reports, however, suggest that high flow nasal cannula therapy (HFNC) is becoming a popular alternative. Two recent systematic reviews, based mainly on observational studies, conclude that there is insufficient evidence at present to determine the effectiveness of HFNC in bronchiolitis.2 ,3

During winter 2014/2015, we conducted a regional audit to determine adherence to NICE guidance relating to the provision of respiratory support for bronchiolitic infants.

During a 3-month period (November–January), a convenience sample of district general hospitals (DGH) served by our regional retrieval service provided anonymised details for all infants aged <12 months with a diagnosis of bronchiolitis who received HFNC or CPAP. The audit was registered at Great Ormond Street Hospital. Data were obtained on 55 infants from five DGHs (range per DGH 2–19). Mean age was 18 weeks (range 1–52) and mean weight 6 kg (range 2.5–10.5). Most infants had no comorbidities (71%). One or more comorbidities were present in 16/55 (29%) of infants, most commonly prematurity (27%).

Fifty-one (93%) infants were started on HFNC as first-line treatment, with five of these requiring escalation to CPAP due to treatment failure as judged by the treating clinician (figure 1).

• Download figure
• Open in new tab
• Download powerpoint

Figure 1

Flow chart of infants admitted with bronchiolitis and started on respiratory support.

The indications for starting respiratory support, as per the treating clinician, were as follows: work of breathing 29/52 (56%); work of breathing and hypoxia 11/52 (21%); work of breathing and hypercarbia 4/52 (8%); work of breathing, hypoxia and hypercarbia 2/52 (4%); work of breathing and apnoea 2/52; hypoxia 2/52; hypercarbia and hypoxia 1/52 (2%) and hypercarbia 1/52.

The mean initial HFNC flow rate was 2 L/kg/min, although variation in practice was marked, with flows from 0.3 to 3.6 L/kg/min. Two hours after starting HFNC, there were statistically significant reductions in respiratory and heart rate; and for a subset of 16 patients, where paired data were available a statistically significant reduction in pCO₂ was demonstrated (table 1).

Our data provide a snapshot of current practice in five London DGHs. Although NICE guidance recommends CPAP, we found an overwhelming preponderance for HFNC as first-line treatment for respiratory failure. HFNC may be easier for staff to set up and more acceptable to children and parents. The main indication for starting HFNC appeared to be clinical signs of increased work of breathing rather than objective measurements of respiratory failure (hypoxia and hypercarbia) which may introduce significant clinical variability in when HFNC is started. However, data completeness may have been limited by the fact that collection was reliant on individual clinicians. The number of children started on CPAP as first-line treatment was too small to allow a meaningful comparison.

In conclusion, HFNC is being widely used for the treatment of bronchiolitis in the DGH setting and appears to have superseded CPAP. Evidence for its efficacy, however, is lacking. High-quality randomised controlled trials comparing HFNC with CPAP are urgently required, especially since clinical equipoise may be shifting rapidly.