SWISH Trial (Strategies for Weaning Infants on Supportive High Flow)

Overview

Bronchiolitis is a common type of chest infection that tends to affect babies and young children under a year old. In older children and adults, the same viruses that cause bronchiolitis lead to the 'common cold'. The symptoms of bronchiolitis are like a common cold and include a blocked or runny nose, a cough and a mildly raised temperature. Bronchiolitis affects the bronchioles which are the smaller breathing tubes in the lungs. They produce more mucus than usual and become swollen, leading to a cough and a runny nose. In more severe cases, the tubes become clogged up with mucus which causes breathing problems. In some babies, the breathing problems may present as breathing fast, with in-drawing of the muscles around the rib cage, and in rare cases, very young babies with bronchiolitis may stop breathing for brief periods ('apnoea'). The illness usually starts with a mild runny nose or cough, gets worse over three to five days or so, and then slowly gets better, usually lasting about 10 to 14 days in total. Around 2 in 100 infants with bronchiolitis will need to spend some time in hospital during the course of their illness. This is usually for one of two reasons: they need oxygen treatment to keep their oxygen saturations within acceptable levels or they cannot manage to feed from the breast or a bottle because of a blocked nose or difficulty breathing. Here at the Children's Hospital for Wales we are using 'High flow' to deliver oxygen. This is a relatively new concept on the general paediatric wards, and more established in a setting such as High Dependency Unit (HDU). However, we have been using it successfully on the wards for the last 3 years. High flow device delivering a mixture of oxygen and air at high flow to help open the child's airways so that their lungs can add oxygen to their blood. It is given through a set of prongs (short plastic tubes) inserted just inside the nostrils. Research has shown that the early use of high flow can reduce the chances of the child needing escalation of care to a high dependency unit or paediatric intensive care unit. The investigators are interested in studying the process of weaning high flow support once the child is over the worst of their illness. This will enable the investigators to use the most effective method of weaning babies from their high flow, and ready for discharge. This has the potential to reduce the number of hours spent in hospital for babies and their parents or guardians.

Bronchiolitis is the commonest respiratory infection in infancy leading to hospital admission. 46 per 1000 infants were admitted to hospital with bronchiolitis in England in 2011 and more recent studies suggest this number has remained static. This has a significant cost burden on the NHS. Respiratory support, mainly oxygenation, and keeping infants well hydrated are the mainstay of management in hospital. Respiratory support has traditionally been the domain of intensive care settings. This has been provided through an escalation of therapy from simple oxygen delivery by nasal cannula, to non-invasive ventilation with Continuous Positive Airway Pressure (CPAP) and finally to intubation with mechanical ventilation. These latter two strategies require highly skilled staff, so are costly, and are associated with a greater incidence of adverse events including ventilator-induced lung injury, barotrauma, and potential neurotoxicity associated with sedation. Over the last decade High Flow Nasal Cannula (HFNC) therapy has emerged as a new method to provide respiratory support for bronchiolitis. HFNC therapy works by delivering an increased volume of air and oxygen into the nasal passages than standard sub-nasal oxygen therapy, using a higher flow of humidified and heated gas. These increased flow rates exceed peak inspiratory flow and thereby result in more efficient delivery of oxygen to the terminal airways. Physiological studies have demonstrated reduced work of breathing and improved gas exchange. The PARIS study has demonstrated that HFNC can be used in a ward setting to reduce admission rates to the Paediatric Intensive Care Unit (PICU). 12% of the study population receiving HFNC needed escalation to PICU compared to 23% receiving standard care (oxygen therapy). The safety data from the PARIS study shows no difference in adverse effects between HFNC and standard oxygen therapy. Numerous studies over the last three decades have investigated the role of various medications in managing infants with bronchiolitis including adrenaline, steroids, salbutamol, and hypertonic saline; none of these studies have definitively changed the outcome of the disease nor the length of stay in hospital. Although the PARIS study showed a reduction in number of patients requiring escalation of care, it did not demonstrate any difference to the total length of stay in hospital. Aim Since HFNC is a relatively new method of providing respiratory support in bronchiolitis, there is lack of evidence on weaning strategy. The investigators primary aim is to examine the feasibility of different weaning strategies for infants, up to 12 months of age, supported on HFNC for bronchiolitis. The secondary outcomes include the time from decision to wean HFNC to the patient no longer requiring respiratory support and a safety assessment of the weaning strategies. This may lead to shorter total length of stay in hospital, without compromising their care. Infants with bronchiolitis who are on HFNC (the devices used will be Airvo 2, Fisher \& Paykel Healthcare) will be on a flow of 2L/kg/min (maximum 20L/min) and variable oxygen concentration to maintain target oxygen saturations \>90%. The participants will be randomised into one of two arms for weaning a) titrating oxygen to FiO2 21% and then stopping HFNC b) titrating oxygen to FiO2 30% and switching HFNC to low flow oxygen.